Gene Therapy of Cancer (2nd Edition)

S E C O N D E D I T I O N GENE THERAPY OF CANCER This Page Intentionally Left Blank S E C O N D E D I T I O N G...

Author: Edmund C. Lattime (Editor) | Stanton L. Gerson (Editor)

92 downloads 1511 Views 16MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

S E C O N D

E D I T I O N

GENE THERAPY OF CANCER

This Page Intentionally Left Blank

S E C O N D

E D I T I O N

GENE THERAPY OF CANCER Translational Approaches from Preclinical Studies to Clinical Implementation Edited by

EDMUND C. LATTIME, PhD Departments of Surgery and Molecular Genetics & Microbiology The Cancer Institute of New Jersey UMDNJ–Robert Wood Johnson Medical School Piscataway, New Jersey

STANTON L. GERSON, MD Division of Hematology/Oncology Department of Medicine and Ireland Cancer Center at Case Western Reserve University and University Hospitals of Cleveland Cleveland, Ohio

San Diego

San Francisco

New York

Boston

London

Sydney

Tokyo

C 2001 PhotoDisc, Inc. Cover photo credit: Images ∞ This book is printed on acid-free paper.

C 2002, 1999 by ACADEMIC PRESS Copyright

All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Requests for permission to make copies of any part of the work should be mailed to: Permissions Department, Academic Press, 6277 Sea Harbor Drive, Orlando, Florida 32887-6777 Academic Press An Elsevier Science Imprint 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.academicpress.com Academic Press 32 Jamestown Road, London NW1 7BY, UK http://www.academicpress.com Library of Congress Catalog Card Number: 2001099437 International Standard Book Number: 0-12-437551-0 PRINTED IN THE UNITED STATES OF AMERICA 02 03 04 05 06 07 EB 9 8 7 6 5 4

3

2

1

To Holly and Deb, who have given us the enthusiastic support needed to pursue this second edition, and to our children, Sarah, Ruth, James, and David who have known of gene therapy from their earliest days. We hope they experience with us the fulfillment of the latent promises of this field in all its forms.


Contents

Contributors Preface xix

xv

Preclinical Models and Clinical Cancer Gene Therapy Trials 33 V. Plasmid Expression Vectors 37 VI. Future Directions 43 References 45 P A R T

I

3. Parvovirus Vectors for the Gene Therapy of Cancer

VECTORS FOR GENE THERAPY OF CANCER

K. K. WONG, JR. AND SASWATI CHATTERJEE

I. Introduction 53 II. Biology of Parvoviridae and Vector Development 54 III. Applications of Recombinant Parvovirus Vectors to Cancer Gene Therapy 61 IV. Perspectives, Problems, and Future Considerations 71 References 71

1. Retroviral Vector Design for Cancer Gene Therapy CHRISTOPHER BAUM, WOLFRAM OSTERTAG, CAROL STOCKING, AND DOROTHEE VON LAER

I. II. III. IV. V. VI.

Introduction 3 Applications for Retroviral Vectors in Oncology Biology of Retroviruses 6 Principles of Retroviral Vector Systems 9 Advances in Retroviral Vector Tailoring 11 Outlook 22 References 23

4

4. Antibody-Targeted Gene Therapy C. LAMPERT, A. M. McCALL, AND L. M. WEINER

I. Introduction 81 II. Background: Monoclonal Antibodies and Cancer Therapy 81 III. Recent Advances: Monoclonal-Antibody-Mediated Targeting and Cancer Gene Therapy 84 IV. Future Directions 91 References 92

2. Noninfectious Gene Transfer and Expression Systems for Cancer Gene Therapy MARK J. COOPER

5. Ribozymes in Cancer Gene Therapy

I. Introduction 31 II. Advantages and Disadvantages of Infectious, Viral-Based Vectors for Human Gene Therapy 31 III. Rationale for Considering Noninfectious, Plasmid-Based Expression Systems 33 IV. Gene Transfer Technologies for Plasmid-Based Vectors:

¨ CARMELA BEGER, MARTIN KRUGER, AND FLOSSIE WONG-STAAL

I. Introduction 95 II. Ribozyme Structures and Functions

vii

96

viii

Contents

III. Cancer Disease Models for Ribozyme Application IV. Challenges and Future Directions 102 References 103

98

P A R T

IIa VACCINE STRATEGIES

6. The Advent of Lentiviral Vectors: Prospects for Cancer Therapy ` MICHEL SADELAIN AND ISABELLE RIVIERE

I. Introduction 109 II. Structure and Function of Lentiviruses 110 III. Features that Distinguish Lentiviral from Oncoretroviral Vectors 111 IV. Manufacture of Lentiviral Vectors 113 V. Possible Applications of Lentiviral Vectors in Cancer Therapy 117 VI. Conclusions 119 References 120

P A R T

II IMMUNE TARGETED GENE THERAPY

8. Development of Epitope-Specific Immunotherapies for Human Malignancies and Premalignant Lesions Expressing Mutated ras Genes SCOTT I. ABRAMS

I. Introduction 145 II. Cellular Immune Response and Antigen Recognition 146 III. Pathways of Antigen Processing, Presentation, and Epitope Expression 146 IV. T-Lymphocyte Subsets 147 V. ras Oncogenes in Neoplastic Development 147 VI. Cellular Immune Responses Induced by ras Oncogene Peptides 149 VII. Identification of Mutant ras CD4+ and CD8+ T-Cell Epitopes Reflecting Codon 12 Mutations 149 VIII. Anti-ras Immune System Interactions: Implications for Tumor Immunity and Tumor Escape 156 IX. Paradigm for Anti-ras Immune System Interactions in Cancer Immunotherapy 158 X. Future Directions 159 References 160

7. Immunologic Targets for the Gene Therapy of Cancer

P A R T

SUZANNE OSTRAND-ROSENBERG, VIRGINIA K. CLEMENTS, SAMUDRA DISSANAYAKE, MILEKA GILBERT, BETH A. PULASKI, AND LING QI

I. Introduction 128 II. Cellular (T-Lymphocyte-Mediated) Versus Humoral (Antibody-Mediated) Immune Responses to Tumor Cells 128 III. Response of CD4+ and CD8+ T Lymphocytes to Tumor Antigens Presented in the Context of Molecules Encoded by the Major Histocompatibility Complex 129 IV. Response of Tumor-Bearing Individuals to Tumor Antigens 132 V. Tumor-Associated Peptides as Candidate Targets for Tumor-Specific Lymphocytes 133 VI. Immunotherapeutic Strategies for the Treatment of Cancer 135 VII.Conclusions 138 References 138

IIb DENDRITIC CELL-BASED GENE THERAPY 9. Introduction to Dendritic Cells PATRICK BLANCO, A. KAROLINA PALUCKA, AND JACQUES BANCHEREAU

I. II. III. IV.

Introduction 167 Features of Dendritic Cells 167 Dendritic Cell Subsets 169 Functional Heterogeneity of Dendritic Cell Subsets 171 V. Dendritic Cells in Tumor Immunology 172 VI. Dendritic Cells and Gene Therapy 173 VII. Conclusions 174 References 174

ix

Contents

10. DNA and Dendritic Cell-Based Genetic Immunization Against Cancer LISA H. BUTTERFIELD, ANTONI RIBAS, AND JAMES S. ECONOMOU

I. Introduction 179 II. Background 179 III. Recent Advances: Methods of Genetic Immunization 183 IV. Preclinical Development and Translation to the Clinic 190 V. Proposed and Current Clinical Trials 190 VI. Future Directions 191 References 191

11. RNA-Transfected Dendritic Cells as Immunogens

V. Future Directions 215 VI. Conclusions 218 References 219

13. The Use of Particle-Mediated Gene Transfer for Immunotherapy of Cancer MARK R. ALBERTINI, DAVID M. KING, AND ALEXANDER L. RAKHMILEVICH

I. II. III. IV. V. VI.

Introduction 225 Background 225 Recent Advances 228 Issues Regarding Evaluation in Clinical Trials 234 Recent Clinical Trials 234 Potential Novel Uses and Future Directions 235 References 235

MICHAEL A. MORSE, SMITA K. NAIR, AND H. KIM LYERLY P A R T

I. Introduction 199 II. Advantages of Loading Dendritic Cells with Genetic Material 199 III. Viral Versus Nonviral Methods of Gene Transfer 200 IV. RNA Versus DNA Loading of Dendritic Cells 200 V. RNA Loading of Dendritic Cells 201 VI Amplification of RNA Used to Load Dendritic Cells 201 VII. Uses of RNA-Loaded Dendritic Cells 201 VIII. Future Directions 202 References 202

P A R T

IIc CYTOKINES AND CO-FACTORS 12. In Situ Immune Modulation Using Recombinant Vaccinia Virus Vectors: Preclinical Studies to Clinical Implementation EDMUND C. LATTIME, LAURENCE C. EISENLOHR, LEONARD G. GOMELLA, AND MICHAEL J. MASTRANGELO

IId GENETICALLY MODIFIED EFFECTOR CELLS FOR IMMUNE-BASED IMMUNOTHERAPY 14. Applications of Gene Transfer in the Adoptive Immunotherapy of Cancer KEVIN T. McDONAGH AND ALFRED E. CHANG

I. Introduction 241 II. Use of Gene-Modified Tumors to Generate AntitumorReactive T Cells 242 III. Genetic Manipulation of T Cells to Enhance Antitumor Reactivity 246 IV. Genetic Modulation of Dendritic Cells 250 V. Summary 251 References 251

15. Update on the Use of Genetically Modified Hematopoietic Stem Cells for Cancer Therapy ´ EDSEL U. KIM, LEE G. WILKE, AND JAMES J. MULE

I. Introduction 207 II. Generation of Cell-Mediated Immune Responses 208 III. Cytokine Gene Transfer Studies in Antitumor Immunity 210 IV. In Situ Cytokine Gene Transfer to Enhance Antitumor Immunity 210

I. Introduction 257 II. Human Hematopoietic Stem Cells as Vehicles of Gene Transfer 258 III. Preclinical Studies of Gene Transfer into Hematopoietic Stem Cells 259

x

Contents

IV. Applications of Genetically Manipulated Hematopoietic Stem Cells to the Therapy of Human Cancer 262 V. Conclusions 268 References 268

III. p53 302 IV. Conclusions 308 References 308

19. Antisense Downregulation of the Apoptosis-Related Bcl-2 and Bcl-xl Proteins: A New Approach to Cancer Therapy

P A R T

III ONCOGENE-TARGETED GENE THERAPY 16. Clinical Applications of Tumor-Suppressor Gene Therapy

IRINA V. LEBEDEVA AND C.A. STEIN

I. The Bcl Family of Proteins and their Role in Apoptosis 315 II. Downregulation of Bcl-2 Expression: Antisense Strategies 316 References 324

RAYMOND D. MENG AND WAFIK S. EL-DEIRY

I. II. III. IV. V.

Introduction 273 p53 273 BRCA1 275 Onyx-015 Adenoviruses 275 Summary and Future Work 276 References 277

17. Cancer Gene Therapy with Tumor Suppressor Genes Involved in Cell-Cycle Control RAYMOND D. MENG AND WAFIK S. EL-DEIRY

I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII. XIII.

Introduction 279 p21WAF1/CIP1 280 p16INK4 284 Rb 285 p14ARF 286 p27Kip1 286 E2F-1 287 PTEN 288 BRCA1 288 VHL 289 FHIT 289 Apoptosis-Inducing Genes Conclusions 291 References 291

20. Gene Therapy for Chronic Myelogenous Leukemia CATHERINE M. VERFAILLIE AND ROBERT CH ZHAO

I. Molecular Mechanisms Underlying Ph+ Leukemias 331 II. Therapy 331 III. Gene-Disruption Methods 332 IV. Anti-bcr-abl Targeted Therapies 332 V. Anti-bcr-abl Drug-Resistance Gene Therapy for CML 332 VI. Conclusion 334 References 334

P A R T

IV

289

MANIPULATION OF DRUG RESISTANCE MECHANISMS BY GENE THERAPY 21. Transfer of Drug-Resistance Genes into Hematopoietic Progenitors

18. Cancer Gene Therapy with the p53 Tumor Suppressor Gene

OMER N. KOC ¸ , STEVEN P. ZIELSKE, JUSTIN C. ROTH, JANE S. REESE, AND STANTON L. GERSON

RAYMOND D. MENG AND WAFIK S. EL-DEIRY

I. Introduction 299 II. Vectors for Gene Therapy

300

I. Introduction 341 II. Rationale for Drug-Resistance Gene Therapy 342

xi

Contents

III. Methyltransferase-Mediated Drug Resistance 344 IV. Cytidine Deaminase 348 V. Glutathione-S-Transferase 348 VI. Dual-Drug-Resistance Approach VII. Clinical Trials 350 VIII. Conclusion 351 References 351

24. Protection from Antifolate Toxicity by Expression of Drug-Resistant Dihydrofolate Reductase R. SCOTT McIVOR

349 I. II. III. IV.

22. Multidrug-Resistance Gene Therapy in Hematopoietic Cell Transplantation

VI.

RAFAT ABONOUR, JAMES M. CROOP, AND KENNETH CORNETTA

VII.

I. Introduction 355 II. P-Glycoprotein 356 III. Targeting Hematopoietic Progenitor Cells for Genetic Modification 356 IV. Expression of P-Glycoprotein in Murine Hematopoietic Progenitors 357 V. Expression of P-Glycoprotein in Human Hematopoietic Progenitors 358 VI. Results of Early Phase I Studies Using MDR1-Transduced Hematopoietic Cells 359 VII. Overcoming Transduction Inefficiency 360 VIII. MDR1 Gene Transfer into Humans: Recent Progress 361 IX. Implication and Future of MDR1 Gene Therapy in Humans 361 References 362

23. Development and Application of an Engineered Dihydrofolate Reductase and Cytidine-Deaminase-Based Fusion Genes in Myeloprotection-Based Gene Therapy Strategies OWEN A. O’CONNOR, TULIN BUDAK-ALPDOGAN, AND JOSEPH R. BERTINO

I. Introduction 365 II. Fusion Genes 368 III. Development of Clinically Applicable Gene Transfer Approaches 370 IV. Preclinical Evidence for Myeloprotection Strategies 371 V. Clinical Applications of Myeloprotection Strategies 373 VI. Challenges 377 References 378

V.

VIII.

Introduction 383 Drug-Resistant Dihydrofolate Reductases 384 Protection from Antifolate Toxicity In Vitro 385 Protection from Antifolate Toxicity In Vivo: Retroviral Transduction Studies 386 Dihydrofolate Reductase Transgenic Mouse System for In Vivo Drug-Resistance Studies 386 Antitumor Studies in Animals Expressing Drug-Resistant Dihydrofolate Reductase 387 Antifolate-Mediated In Vivo Selection of Hematopoietic Cells Expressing Drug-Resistant Dihydrofolate Reductase 388 Summary and Future Considerations 388 References 389

25. A Genomic Approach to the Treatment of Breast Cancer K.V. CHIN, DEBORAH TOPPMEYER, THOMAS KEARNEY, MICHAEL REISS, EDMUND C. LATTIME, AND WILLIAM N. HAIT

I. Introduction 393 II. Toward a Genomic Approach to Therapy 393 III. The Use of DNA Microarrays to Understand Drug Resistance 396 IV. Effects of Genomic-Based Approaches on the Management of Breast Cancer Patients 398 References 399

P A R T

V ANTI-ANIOGENESIS AND PRO-APOPTOTIC GENE THERAPY 26. Antiangiogenic Gene Therapy STEVEN K. LIBUTTI AND ANDREW L. FELDMAN

I. Introduction 405 II. Angiogenesis and its Role in Tumor Biology 405 III. Antiangiogenic Therapy of Cancer and the Role of Gene Therapy 406 IV. Preclinical Models of Antiangiogenic Gene Therapy 407

xii

Contents

V. Inhibiting Proangiogenic Cytokines 412 VI. Endothelial Cell-Specific Gene Delivery 414 VII. Future Directions in Antiangiogenic Gene Therapy 415 References 415

27. VEGF-Targeted Antiangiogenic Gene Therapy CALVIN J. KUO, FILIP A. FARNEBO, CHRISTIAN M. BECKER, AND JUDAH FOLKMAN

I. Introduction 421 II. Angiogenesis and Tumor Growth 422 III. Gene Therapy for Delivery of Antiangiogenic Factors 422 IV. Antiangiogenic Gene Therapy in the Experimental and Clinical Settings 423 V. Vascular Endothelial Growth Factor and Receptors 423 VI. Vascular Endothelial Growth Factor and Angiogenesis 424 VII. Vascular Endothelial Growth Factor Inhibition by Gene Transfer 425 VIII. Issues Regarding Clinical Translation of Antiangiogenic Gene Therapy 428 IX. Conclusion 432 References 432

28. Strategies for Combining Gene Therapy with Ionizing Radiation to Improve Antitumor Efficacy

III. Biology of Human Adenovirus 451 IV. Mechanisms of Adenovirus-Mediated Cell Killing 451 V. Approaches to Optimizing Tumor-Selective Adenovirus Replication 452 VI. Background: dl1520 (ONYX-015) 452 VII. Clinical Trial Results with Wild-Type Adenovirus: Flawed Study Design 453 VIII. A Novel Staged Approach to Clinical Research with Replication-Selective Viruses: dl1520 (ONYX-015) 454 IX. Results from Clinical Trials with dl1520 (ONYX-015) 455 X. Results from Clinical Trials with dl1520 (ONYX-015): Summary 459 XI. Future Directions 460 XII. Summary 462 References 462

30. E1A Cancer Gene Therapy DUEN-HWA YAN, RUPING SHAO, AND MIEN-CHIE HUNG

I. Introduction 465 II. HER2 Overexpression and E1A-Mediated Antitumor Activity 465 III. Mechanisms of E1A-Mediated Anti-Tumor Activity 467 IV. E1A Gene Therapy: Preclinical Models 470 V. E1A Gene Therapy: Clinical Trials 472 VI. Conclusion 473 References 474 P A R T

DAVID H. GORSKI, HELENA J. MAUCERI, AND RALPH R. WEICHSELBAUM

I. Introduction 435 II. Strategies Using Gene Therapy to Increase the Efficacy of Radiation Therapy 436 III. Enhancing the Replicative Potential of Antitumor Viruses with Ionizing Radiation 440 IV. Transcriptional Targeting of Gene Therapy with Ionizing Radiation (Genetic Radiotherapy) 441 V. Summary and Future Directions 443 References 444

VI PRODRUG ACTIVATION STRATEGIES FOR GENE THERAPY OF CANCER 31. Preemptive and Therapeutic Uses of Suicide Genes for Cancer and Leukemia FREDERICK L. MOOLTEN AND PAULA J. MROZ

29. Virotherapy with Replication-Selective Oncolytic Adenoviruses: A Novel Therapeutic Platform for Cancer DAVID KIRN

I. Introduction 449 II. Attributes of Replication-Selective Adenoviruses for Cancer Treatment 451

I. II. III. IV.

Introduction 481 Therapeutic Uses of Suicide Genes 482 Preemptive Uses of Suicide Genes in Cancer 483 Creation of Stable Suicide Functions by Combining Suicide Gene Transduction with Endogenous Gene Loss 485 V. Preemptive Uses of Suicide Genes to Control Graft-Versus-Host Disease in Leukemia 487

xiii

Contents

VI. Future Prospects for Preemptive Use of Suicide Genes 488 References 489

32. Treatment of Mesothelioma Using Adenoviral-Mediated Delivery of Herpes Simplex Virus Thymidine Kinase Gene in Combination with Ganciclovir DANIEL H. STERMAN AND STEVEN M. ALBELDA

I. Introduction 493 II. Clinical Use of HSV-TK in the Treatment of Localized Malignancies 494 III. Challenges and Future Directions 499 References 501

33. The Use of Suicide Gene Therapy for the Treatment of Malignancies of the Brain KEVIN D. JUDY AND STEPHEN L. ECK

I. II. III. IV.

Introduction 505 Retrovirus Vector for HSV-TK 506 Adenovirus Vector for HSV-TK 509 Herpes Simplex Virus Vectors Expressing Endogenous HSV-TK 510

V. Promising Preclinical Studies References 511

510

34. Case Study of Combined Gene and Radiation Therapy as an Approach in the Treatment of Cancer BIN S. TEH, MARIA T. VLACHAKI, LAURA K. AGUILAR, BRIAN MILES, GUSTAVO AYALA, DOV KADMON, THOMAS WHEELER, TIMOTHY C. THOMPSON, E. BRIAN BUTLER, AND ESTUARDO AGUILAR-CORDOVA

I. Introduction 513 II. Background of the Field 514 III. Recent Advances in Herpes Simplex Virus-Thymidine Kinase Suicide Gene Therapy 515 IV. Combined Herpes Simplex Virus-Thymidine Kinase Suicide Gene Therapy and Radiotherapy 516 V. Issues Regarding Clinical Trials, Translation into Clinical Use, Preclinical Development, Efficacy, Endpoints, and Gene Expression 521 VI. Potential Novel Uses and Future Directions 522 References 523 Index

525


Contributors

Joseph R. Bertino (365) Department of Medicine, Division of Hematologic Oncology and Lymphoma, and Programs of Molecular Pharmacology and Therapeutics, MemorialSloan Kettering Cancer Center, New York, New York 10021

Numbers in parentheses indicate the pages on which the authors’ contribution begin.

Rafat Abonour (355) Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana 46202

Patrick Blanco (167) Baylor Institute for Immunology Research, Dallas, Texas 75204

Scott I. Abrams (145) Laboratory of Tumor Immunology and Biology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892

Tulin Budak-Alpdogan (365) Department of Medicine, Programs of Molecular Pharmacology and Therapeutics, Memorial-Sloan Kettering Cancer Center, New York, New York 10021

Laura K. Aguilar (513) Harvard Gene Therapy Initiative, Harvard Medical School, Boston, Massachusetts 02115 Estuardo Aguilar-Cordova (513) Department of Radiology, Baylor College of Medicine, Houston, Texas 77030 and Harvard Gene Therapy Initiative, Harvard Medical School, Boston, Massachusetts 02115

E. Brian Butler (513) Department of Radiology, Baylor College of Medicine, Houston, Texas 77030 Lisa H. Butterfield (179) Division of Surgical Oncology, UCLA Medical Center, University of California, Los Angeles, California 90095

Steven M. Albelda (493) Thoracic Oncology Research Laboratory, Pulmonary, Allergy, and Critical Care Division, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19104

Alfred E. Chang (241) Department of Surgery, Division of Surgical Oncology, Comprehensive Cancer Center, University of Michigan, Ann Arbor, Michigan 48109

Mark R. Albertini (225) The University of Wisconsin, Comprehensive Cancer Center, Madison, Wisconsin 53792

Saswati Chatterjee (53) Division of Virology, City of Hope National Medical Center, Duarte, California 91010

Gustavo Ayala (513) Department of Pathology, Baylor College of Medicine, Houston, Texas 77030

K. V. Chin (393) Departments of Medicine and Pharmacology, The Cancer Institute of New Jersey, UMDNJ-Robert Wood Johnson Medical School, Piscataway, New Jersey 08901

Jacques Banchereau (167) Baylor Institute for Immunology Research, Dallas, Texas 75204 Christopher Baum (3) Medizinische Hochschule, Abt. Haematologie, 30625 Hannover, Germany Christian M. Becker (421) Department of Surgery, Children’s Hospital, Harvard Medical School, Boston, Massachusetts 02115

Virginia K. Clements (127) Department of Biological Sciences, University of Maryland, Baltimore, Maryland 21250

Carmela Beger (95) Department of Medicine, University of San Diego, La Jolla, California 92093

Mark J. Cooper (31) Copernicus Therapeutics, Inc., Cleveland, Ohio 44106

xv

xvi

Contributors

Kenneth Cornetta (355) Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana 46202 James M. Croop (355) Department of Pediatrics, Indiana University School of Medicine, Indianapolis, Indiana 46202 Samudra Dissanayake (127) Department of Biological Sciences, University of Maryland, Baltimore, Maryland 21250 Stephen L. Eck (505) HUP-Department of Neurosurgery, The University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19004 James S. Economou (179) Division of Surgical Oncology, and Department of Immunology, Microbiology, and Molecular Genetics, UCLA Medical Center, University of California, Los Angeles, California 90095 Laurence C. Eisenlohr (207) Department of Microbiology and Immunology, Thomas Jefferson University, Philadelphia, Pennsylvania 19107 Wafik S. El-Deiry (273, 279, 299) Laboratory of Molecular Oncology and Cell Cycle Regulation, Howard Hughes Medical Institute, Departments of Medicine and Genetics, Cancer Center and The Institute for Human Gene Therapy, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104 Filip A. Farnebo (421) Departments of Surgery and Genetics, Children’s Hospital, Harvard Medical School, Boston, Massachusetts 02115 Andrew L. Feldman (405) Surgery Branch, National Cancer Institute, Bethesda, Maryland 20892 Judah Folkman (421) Department of Surgery, Children’s Hospital, Harvard Medical School, Boston, Massachusetts 02115 Stanton L. Gerson (341) Division of Hematology/Oncology, Department of Medicine and Ireland Cancer Center at Case Western Reserve University and University Hospitals of Cleveland, Cleveland, Ohio 44106 Mileka Gilbert (127) Department of Biological Sciences, University of Maryland, Baltimore, Maryland 21250 Leonard G. Gomella (207) Department of Urology, Thomas Jefferson University, Philadelphia, Pennsylvania 19107 David H. Gorski (435) The Cancer Institute of New Jersey, UMDNJ-Robert Wood Johnson Medical School, New Brunswick, New Jersey 08901

William N. Hait (393) Departments of Medicine and Pharmacology, The Cancer Institute of New Jersey, UMDNJRobert Wood Johnson Medical School, Piscataway, New Jersey 08901 Mien-Chie Hung (465) Departments of Molecular and Cellular Oncology and Surgical Oncology, M. D. Anderson Cancer Center, The University of Texas, Houston, Texas 77030 Kevin D. Judy (505) HUP-Department of Neurosurgery, The University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19004 Dov Kadmon (513) Department of Urology, Baylor College of Medicine, Houston, Texas 77030 Thomas Kearney (393) Department of Surgery, The Cancer Institute of New Jersey, UMDNJ-Robert Wood Johnson Medical School, Piscataway, New Jersey 08901 Edsel U. Kim (257) Department of Otolaryngology, University of Michigan, Ann Arbor, Michigan 48109 David M. King (225) The University of Wisconsin, Comprehensive Cancer Center, Madison, Wisconsin 53792 David Kirn (449) Imperial Cancer Research Fund, Program for Viral and Genetic Therapy of Cancer, Imperial College School of Medicine, Hammersmith Hospital, London W11 OHS, United Kingdom Omer N. Ko¸c (341) Division of Hematology/Oncology, Department of Medicine and Ireland Cancer Center at Case Western Reserve University and University Hospitals of Cleveland, Cleveland, Ohio 44106 Martin Kruger ¨ (95) Department of Medicine, University of San Diego, La Jolla, California 92093 Calvin J. Kuo (421) Departments of Surgery and Genetics, Children’s Hospital, Harvard Medical School, Boston, Massachusetts 02115 C. Lampert (81) Department of Hematology and Medical Oncology, St. Peter’s University Hospital, New Brunswick, New Jersey 08901 Edmund C. Lattime (207, 393) Departments of Surgery and Molecular Genetics & Microbiology, The Cancer Institute of New Jersey, UMDNJ-Robert Wood Johnson Medical School, Piscataway, New Jersey 08901 Irina V. Lebedeva (315) Department of Medicine and Pharmacology, Columbia University, College of Physicians and Surgeons, New York, New York 10032 Steven K. Libutti (405) Surgery Branch, National Cancer Institute, Bethesda, Maryland 20892

Contributors

H. Kim Lyerly (199) Department of Surgery, Duke University Medical Center, Durham, North Carolina 27710 Michael J. Mastrangelo (207) Division of Medical Oncology, Department of Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania 19107

xvii

Wolfram Ostertag (3) Heinrich-Pette-Institut für Experimentelle Virologie und Immunologie an der Universität Hamburg, 20251 Hamburg, Germany Suzanne Ostrand-Rosenberg (127) Department of Biological Sciences, University of Maryland, Baltimore, Maryland 21250

Helena J. Mauceri (435) Department of Radiation and Cellular Oncology, University of Chicago Hospitals, Chicago, Illinois 60637

A. Karolina Palucka (167) Baylor Institute for Immunology Research, Dallas, Texas 75204

A. M. McCall (81) Fox Chase Cancer Center, Philadelphia, Pennsylvania 91010

Beth A. Pulaski (127) Department of Biological Sciences, University of Maryland, Baltimore, Maryland 21250

Kevin T. McDonagh (241) Department of Internal Medicine, Division of Hematology/Oncology, Comprehensive Cancer Center, University of Michigan, Ann Arbor, Michigan 48109

Ling Qi (127) Department of Biological Sciences, University of Maryland, Baltimore, Maryland 21250

R. Scott McIvor (383) Gene Therapy Program, Institute of Human Genetics, Department of Genetics, Cell Biology and Development, University of Minnesota, Minneapolis, Minnesota 55455 Raymond D. Meng (273, 279, 299) Laboratory of Molecular Oncology and Cell Cycle Regulation, Howard Hughes Medical Institute, Departments of Medicine and Genetics, Cancer Center and The Institute for Human Gene Therapy, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104

Alexander L. Rakhmilevich (225) The University of Wisconsin, Comprehensive Cancer Center, Madison, Wisconsin 53792 Jane S. Reese (341) Division of Hematology/Oncology, Department of Medicine and Ireland Cancer Center at Case Western Reserve University and University Hospitals of Cleveland, Cleveland, Ohio 44106 Michael Reiss (393) Department of Medicine, The Cancer Institute of New Jersey, UMDNJ-Robert Wood Johnson Medical School, Piscataway, New Jersey 08901

Brian Miles (513) Department of Urology, Baylor College of Medicine, Houston, Texas 77030

Antoni Ribas (179) Division of Surgical Oncology, and Division of Hematology/Oncology, UCLA Medical Center, University of California, Los Angeles, California 90095

Frederick L. Moolten (481) Edith Nourse Rogers Memorial Veterans Hospital, Bedford, Massachusetts 01730 and Boston University School of Medicine, Boston, Massachusetts 02118

Isabelle Rivière (109) Laboratory of Gene Transfer and Gene Expression, Department of Medicine and Immunology Program, Memorial Sloan-Kettering Cancer Center, New York, New York 10021

Michael A. Morse (199) Department of Medicine, Division of Medical Oncology and Transplantation, Duke University Medical Center, Durham, North Carolina 27710

Justin C. Roth (341) Division of Hematology/Oncology, Department of Medicine and Ireland Cancer Center at Case Western Reserve University and University Hospitals of Cleveland, Cleveland, Ohio 44106

Paula J. Mroz (481) Edith Nourse Rogers Memorial Veterans Hospital, Bedford, Massachusetts 01730 James J. Mulé (257) Department of Surgery, University of Michigan, Ann Arbor, Michigan 48109 Smita K. Nair (199) Department of Surgery, Duke University Medical Center, Durham, North Carolina 27710 Owen A. O’Connor (365) Department of Medicine, Division of Hematologic Oncology and Lymphoma, and Developmental Chemotherapy Services, Memorial-Sloan Kettering Cancer Center, New York, New York 10021

Michel Sadelain (109) Laboratory of Gene Transfer and Gene Expression, Department of Medicine and Immunology Program, Memorial Sloan-Kettering Cancer Center, New York, New York 10021 Ruping Shao (465) Department of Molecular and Cellular Oncology, M. D. Anderson Cancer Center, The University of Texas, Houston, Texas 77030 C. A. Stein (315) Department of Medicine and Pharmacology, Columbia University, College of Physicians and Surgeons, New York, New York 10032

xviii

Contributors

Daniel H. Sterman (493) Thoracic Oncology Research Laboratory, Pulmonary, Allergy, and Critical Care Division, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania 19104 Carol Stocking (3) Heinrich-Pette-Institut für Experimentelle Virologie und Immunologie an der Universität Hamburg, 20251 Hamburg, Germany Bin S. Teh (513) Department of Radiology, Baylor College of Medicine, Houston, Texas 77030 Timothy C. Thompson (513) Department of Urology, Baylor College of Medicine, Houston, Texas 77030

L. M. Weiner (81) Fox Chase Cancer Center, Philadelphia, Pennsylvania 91010 Thomas Wheeler (513) Department of Pathology, Baylor College of Medicine, Houston, Texas 77030 Lee G. Wilke (257) Department of Surgery, University of Michigan, Ann Arbor, Michigan 48109 K. K. Wong, Jr. (53) Division of Hematology and Bone Marrow Transplantation, and Division of Virology, City of Hope National Medical Center, Duarte, California 91010

Deborah Toppmeyer (393) Department of Medicine, The Cancer Institute of New Jersey, UMDNJ-Robert Wood Johnson Medical School, Piscataway, New Jersey 08901

Flossie Wong-Staal (95) Department of Medicine, University of San Diego, La Jolla, California 92093

Catherine M. Verfaillie (331) Stem Cell Institute, Cancer Center, and Division of Hematology, Oncology and Transplantation, Department of Medicine, University of Minnesota, Minneapolis, Minnesota 55455

Duen-Hwa Yan (465) Departments of Molecular and Cellular Oncology and Surgical Oncology, M. D. Anderson Cancer Center, The University of Texas, Houston, Texas 77030

Maria T. Vlachaki (513) Department of Radiology and Veterans Affairs Medical Center, Baylor College of Medicine, Houston, Texas 77030

Steven P. Zielske (341) Division of Hematology/Oncology, Department of Medicine and Ireland Cancer Center at Case Western Reserve University and University Hospitals of Cleveland, Cleveland, Ohio 44106

Dorothee von Laer (3) Chemotherapeutisches Forschungsinstitut, Georg-Speyer-Haus, 60596 Frankfurt, Germany Ralph R. Weichselbaum (435) Department of Radiation and Cellular Oncology, University of Chicago Hospitals, Chicago, Illinois 60637

Robert CH Zhao (331) Stem Cell Institute, Cancer Center, and Division of Hematology, Oncology and Transplantation, Department of Medicine, University of Minnesota, Minneapolis, Minnesota 55455

Preface

of clinical trials and appears to have accomplished its goal of achieving stem cell protection in both preclinical and clinical settings. Gene delivery remains an important aspect of gene therapy of cancer and a number of chapters focus on gene delivery systems using both viral and nonviral approaches. The reader will find this to be a comprehensive assessment of the current state of gene therapy of cancer offering state of the art research, a review of basic mechanisms and approaches, and a compilation of current clinical trial efforts. We have not encumbered this text with a number of the current controversies in gene therapy but would note the importance of these issues in the field. Conflict of interest remains an active area of discussion at every major cancer center in the country and has been the recent focus of the American Society of Gene Therapy. The careful monitoring of patients undergoing clinical trials in gene therapy will remain a priority for all clinical trialists in this field. Linking preclinical models to clinical endpoints is an important aspect of this focus and will enable intermediate assessments to define whether the gene therapy effect has been achieved prior to relying on clinical cancer response and will help drive both Phase I and Phase II clinical trial design. The next 5 years will be explosive in the next generation of preclinical and clinical developments of gene therapy of cancer. We hope this second edition will provide an important reference for investigators and observers alike in this exciting field.

The second edition of Gene Therapy of Cancer comes at a pivotal transition point in the development of this exciting technology. Much has occurred in the past 4 years to catapult preclinical and basic scientific concepts into therapeutic trials. In addition, while the outcome of the initial phase of clinical trials using gene therapy to target cancers has not yielded the amazing results initially hoped for, as with every new therapeutic venture in medicine, initial results provide the fodder for critical experiments, new targets, and new questions that propel the field forward. We have reorganized the presentations in the second edition to reflect the continued new emerging strategies that will ultimately lead to the success of this therapeutic approach and have added introductory chapters to a number of the sections with the goal of setting the contributions in their proper basic scientific context. Immune therapeutics takes on added emphasis given some of the recent breakthroughs in vaccine development and targeted delivery. Oncolytic virus therapeutics have also emerged in a very promising light with initial positive results observed in head and neck cancer leading to a number of preclinical advances. Therapies directed towards oncogenes, be it by expression of normal oncogenes, use of ribozymes and antisense therapeutics, and the use of E1A continue to be promising in preclinical and early clinical models. Hematopoietic stem cells are being used in gene therapy both in the antisense setting, for instance use of BCR/ABL antisense to block CML stem cell proliferation, and in genetically modified stem cells as immunotherapies. In addition, bone marrow protection by introduction of a drug-resistant gene into hematopoietic stem cells is entering its next phase

Edmund C. Lattime, PhD Stanton L. Gerson, MD

xix


P A R T I

VECTORS FOR GENE THERAPY OF CANCER


C

H

A

P

T

E

R

1 Retroviral Vector Design for Cancer Gene Therapy CHRISTOPHER BAUM

WOLFRAM OSTERTAG

CAROL STOCKING

Medizinische Hochschule Abt. Haematologie 30625 Hannover, Germany

Heinrich-Pette-Institut für Experimentelle Virologie und Immunologie an der Universität Hamburg, 20251 Hamburg, Germany

Heinrich-Pette-Institut für Experimentelle Virologie und Immunologie an der Universität Hamburg, 20251 Hamburg, Germany

DOROTHEE VON LAER Chemotherapeutisches Forschungsinstitut Georg-Speyer-Haus 60596 Frankfurt, Germany

ogy, including poor vector design [2]. Nevertheless, many former skeptics were turned to true believers, not only due to the enormous public and economical interest [3]. Thus, strong international competition in the field was generated, with an increasing number of researchers following valuable long-term concepts, including improvement of basic vector technology. An ideal vector should (1) allow efficient and selective transduction of the target cell of interest, (2) be maintained, (3) be expressed at levels necessary for achieving therapeutic effects, and, last but not least, (4) be safe in terms of avoiding unexpected side effects in the host. Viruses are a perfect tool for gene transfer as they have evolved to deliver their genome efficiently to target cells with subsequent high-level gene expression. Vector systems for therapeutic gene transfer have been developed from different virus groups, each system having specific advantages and drawbacks. Retroviruses have several unique features that render them highly suitable for vector development. Retroviral vectors are, therefore, the prevalent system for gene transfer in human cells. Retroviruses integrate and express their genome in a stable manner, thus allowing long-term manipulation with transferred genes. This is a prerequisite for many gene therapy applications, including some approaches in cancer gene therapy. Integration usually does not alter host cell functions and is well tolerated. In the retroviral genome, cis-acting elements, responsible for reverse transcription, integration, and packaging, can

I. Introduction 3 II. Applications for Retroviral Vectors in Oncology 4 III. Biology of Retroviruses 6 A. B. C. D.

Classification 6 Retroviral Genes and Their Products Retroviral cis Elements 6 Retroviral Life Cycle 7

6

IV. Principles of Retroviral Vector Systems A. Packaging Cells 9 B. Basic Vector Architecture

10

V. Advances in Retroviral Vector Tailoring A. Components Active in trans B. cis-Active Elements 16

VI. Outlook

9 11

11

22

References 23

I. INTRODUCTION In the past years, oncology was the center of gene therapy research [1]. However, despite generous support by, for example, the National Institutes of Health and related institutions in Europe, there is still a wide gap between the hopes raised and the results achieved. Most of the failures of gene therapy trials can be attributed to a discordant combination of overinterpreted clinical concepts and immature technol-

Gene Therapy of Cancer, Second Edition

3

C 2002 by Academic Press Copyright All rights of reproduction in any form reserved.

4

Christopher Baum et al.

be well separated from coding sequences. Such a genome structure facilitates the design of safe vectors and packaging cell lines. However, with the transition to applications in human gene therapy, severe limitations of conventional retroviral vector systems have become apparent. These include low and variable particle titers, lack of appropriate vector targeting to specific cell types and genomic loci, failure to transduce quiescent cells, and relatively inefficient, positiondependent transcription. Fortunately, substantial progress in vector development has been made, based on deeper understanding of the biology of retroviruses and target cells. Here, we review some of the work relevant to cancer gene therapy. We start with a short overview of potential applications of retroviral vectors in oncology. Then, we describe aspects of retrovirus biology relevant to gene therapy, to create a basis for discussing principles of and specific recent advances in retroviral vector design.

II. APPLICATIONS FOR RETROVIRAL VECTORS IN ONCOLOGY In oncology, several different strategies involving somatic gene transfer are currently considered (Table 1). We can

distinguish between diagnostic and therapeutic approaches; in either case, both healthy tissues or tumor cells may be targeted. Each strategy has special implications for vector design. Gene marking uses stable retroviral transduction of heterologous genetic sequences to analyze the biological (stem cell function, antiviral effects) or pathogenic (tumor cell contamination, graft-versus-host reaction) capacity of blood cell transplants [4–6]. Here, efficient transduction of long-lived hematopoietic cells is required. Moreover, long-term transgene expression is necessary for follow-up analyses involving phenotyping and preparative sorting of transduced cells, based on cell surface or cytoplasmic markers encoded by the vector [7–9]. Besides this entirely diagnostic approach, several therapeutic strategies target healthy tissues. These strategies are also relevant to gene therapy of some inborn genetic disorders or acquired viral infections, due to the use of marker genes that allow selection of transduced cells in vivo. Positive selection is established in the context of drug resistance gene transfer, negative selection in adoptive immunotherapy. Drug resistance gene transfer in nontumor tissues such as bone marrow is aimed at augmenting the therapeutic index of anticancer chemotherapy [10,11]. Protection at the level of hematopoietic progenitor cells reduces short-term toxicity, and protection at the stem cell level might even prevent

TABLE 1 Somatic Gene Transfer in Oncology and Implications for Vector Design Approach

Aim

Target cells

Vector requirements

Vector system

Gene marking

Diagnostic

Healthy hematopoietic or lymphocytic cells, tumor cells (both ex vivo)

Transduction of long-lived stem cells, stable gene expression

Retroviral vectors

Drug resistance gene transfer

Therapeutic (paradigm for positive selection of transduced cell in vivo)

Healthy hematopoietic cells (ex vivo)

Transduction of repopulating cells, stable and high gene expression

Retroviral vectors

Adoptive immunotherapy

Therapeutic (paradigm for negative selection of transduced cell in vivo)

Donor lymphocytes (ex vivo)

Transduction of lymphocytes, stable and high gene expression

Retroviral vectors

Mini-organs

Therapeutic

Healthy autologous or xenogenic cells (ex vivo)

Stable or inducible gene expression

Retroviral vectors

Suicide gene transfer

Therapeutic (but not systemic)

Tumor cells (usually in vivo)

Applicability in vivo, targeting to tumor cells; strong, but not necessarily stable gene expression

Retroviral vectors; alternatively herpes virus vectors or adenoviral vectors

Oncogene antagonism

Therapeutic (but not systemic)

Tumor cells (usually in vivo)

Applicability in vivo, targeting to tumor cells; strong, but not necessarily stable gene expression

Retroviral vectors; alternatively herpes virus vectors or adenoviral vectors

Tumor vaccination

Therapeutic

Tumor cells, antigenpresenting cells (ex vivo or in vivo)

Applicability in vivo; moderate, but not necessarily stable gene expression

Retroviral vectors; alternatively herpes virus vectors, adenoviral vectors, or physicochemically

Retroviral Vector Design for Cancer Gene Therapy

long-term toxicity and the mutagenicity of chemotherapy. The benefit for the patient will depend on the numbers of protected cells obtainable. These are expected to increase with each cycle of chemotherapy, because cells acquire a selective advantage upon expression of the drug resistance gene. Thus, this approach sets a paradigm for forced expansion of transduced cells in vivo. Similar to gene marking, this approach requires a number of technological improvements: First, helper functions of the vector systems and transduction conditions have to mediate efficient uptake (see Section V.A) and nuclear translocation (see Section V.B) in primitive hematopoietic cells (reviewed in Baum et al. [12]). Second, the vector needs to be equipped with cis-regulatory elements mediating dominant gene expression levels and thus strong penetrance of the phenotype (see Section V.B) [13,14]. Third, coexpression of a second gene (see Section V.B.7) is important in this approach, because coordinated transfer of two complementary drug resistance genes greatly widens its flexibility [15–18]. Finally, malignant cells must be excluded from productive transduction by cell purging or vector targeting (see Sections V.A.1.b and V.B.1). A paradigm for negative selection of transduced cells in vivo is established in adoptive immunotherapy. Here, ex vivo selected populations of allogenic donor lymphocytes are used to elicit an antiviral or antileukemic effect [19]. Gene transfer in lymphocytes serves for both positive and negative selection. After transduction, positive selection of gene-modified lymphocytes is performed ex vivo using cell surface markers. After reinfusion, concomitant expression of a negative selection marker (a suicide gene) is instrumental for treating eventually occurring graft-versus-host disease. Here, the key issue is to design vectors with reliable and persisting coexpression of two genes (see Sections V.B.2, V.B.4, and V.B.7). An extension of this approach is the transfer and expression of “designer” T-cell receptors in autologous or allogeneic T cells, in order to generate effector cells with a new, predefined target cell specificity [20]. Positive or negative selection and monitoring of transduced cells in vivo is crucial for the development of artificial mini-organs (derived from genetically manipulated cells [21]). In oncology, these are of interest for systemic delivery of tumor-antagonistic factors such as immunotoxins or inhibitors of angiogenesis. Further applications extend to genetic or acquired disorders that can be treated by delivery with enzymes, hormones, or ligands. Equipping mini-organs with regulatable promoters might allow the adjustment of supply according to individual clinical requirements (see Section V.B.5) [22]. Other therapeutic concepts rely on direct genetic manipulation of tumor cells. Some of these suffer from poor predictability, mostly due to the tremendous variability of tumor evolution among and within individual patients. Moreover, not all of these strategies acknowledge the systemic character of tumor diseases. Nevertheless, in selected pa-

5

tients, these strategies might offer interesting perspectives. Usually, vectors have to be applied in vivo to become effective, and sometimes even replication-competent vectors will be needed. Here, nonretroviral systems may offer important alternatives, given that the problem of instability of persistence or expression is of minor importance. Transfer of prodrug-converting enzyme genes (suicide gene transfer) is performed to render tumor cells susceptible to cytotoxic compounds requiring activation by a heterologous enzyme [23,24]. Alternatively, toxin genes may be used [25,26]. Another approach to control tumor cells by gene transfer is oncogene antagonism [27,28]. Here, one attempts to counteract tumor-promoting mutations of cellular genes. This is achieved by transducing tumor cells with wild-type copies of tumor suppressor genes or dominant negative proteins, antisense nucleotides, or ribozymes directed against oncogenes and their products. In compact tumor masses, some of these strategies may profit from the so-called bystander effect. This refers to cytostatic effects observed in nontransduced cells that result from delivery of proteins or activated cytotoxic drugs through direct intercellular exchange. However, this exchange might also dilute the effects in transduced cells [29]. With either approach, immune responses may be triggered that are expected to promote antitumor efficiency. Transfer of suicide or toxin genes,in contrast to oncogene antagonism, should exclude healthy tissues. Besides direct targeting of retroviral vectors to tumor cells, cellular vehicles may be used to deliver tumor-antagonistic gene products into tumor masses; these may be either cytotoxic T cells [26] or normal progenitor cells with homing capacity [30], tumor cells themselves [31], and possibly also endothelial cells or their precursors [32]. Application of retroviral vectors in vivo requires production of complement-resistant particles at high titers. Selectivity can be achieved already at the level of transduction, taking advantage of preferential infection of dividing cells by vectors based on murine retroviruses [33]. Specific targeting using engineered envelope proteins may, however, be superior (see Section V.A.1.b). At the level of transcriptional regulation, selectivity can be accomplished by insertion of promoters preferentially activated in tumor cells or tumor vasculature (see Section V.B.3). Depending on the tumor type, herpesviruses or adenoviruses (some of the latter specifically replicating in p53-negative cells) may represent alternative vectors [34]. Key aspects of targeting using specific receptors or promoters also apply to these nonretroviral systems. Finally, tumor vaccination is performed to evoke a systemic immune response to tumor-specific antigens [35,36]. This is accomplished by transfer and expression of genes that increase antigen presentation or improve effector cell functions. Target cells for transduction are tumor cells, antigenpresenting cells, or tumor-infiltrating T cells. Thus, tumor vaccination strategies are highly variable with respect to the

6


target cell population and to the type and numbers of activating genes to be transferred. Vectors may be applied either ex vivo or in vivo (after injection in tumor masses), and sustained gene expression in target cells is not necessarily required. For this approach, alternative vector systems (e.g., adenoviral or herpes vectors or physicochemical methods such as biolistics) may also be valuable. In all these different strategies, important variables to account for in vector design are the route of gene transfer, the target cell population, the efficiency and specificity of transduction, and the level, duration, and specificity of transgene expression (Table 1). Therefore, appropriate components of the vector system have to be identified for each application (“tailored vectors”). Fortunately, substantial progress has been made toward all aspects of vector design relevant to cancer gene therapy. These vector improvements are based on detailed insights into the biology of retrovirus–host interactions.

III. BIOLOGY OF RETROVIRUSES A. Classification Sequence data and genome structure are the basis for the classification of retroviruses (Table 2) [37]. Each group contains several virus strains that differ in biological properties, such as receptor utilization and pathogenicity. Until recently, most retroviral vector systems discussed for human gene therapy were based on murine leukemia viruses (MLVs). MLVs belong to the mammalian C-type retroviruses and are further classified according to the species distribution of their receptors. Ecotropic MLVs replicate only in rodent cells and xenotropic MLVs only in nonmurine cells, while polytropic and amphotropic MLVs can infect murine and nonmurine cells. The 10A1 strain has an overlapping but distinct host range, due to the use of the receptor for the gibbon ape leukemia virus (GALV) in addition to the amphotropic

TABLE 2 Retrovirus Genera Genus

Example viruses

Avian leukosis sarcoma

Rous sarcoma virus (RSV)

Mammalian C type

Murine leukemia virus (MLV), several strains: such as Moloney-, Harvey-, Abelson-, 407A-MLV. Feline leukemia virus (FeLV) Gibbon ape leukemia virus (GALV)

D-type viruses

Mason-Pfizer monkey virus (MPMV)

B-type viruses

Mouse mammary tumor virus (MMTV)

HTLV-BLV group

Human T cell leukemia virus (HTLV)-1 and 2

Lentivirus

Human immunodeficiency virus (HIV)-1 and -2

Spumavirus

Human foamy virus (HFV)

receptor [38]. Except for ecotropic viruses, gene transfer into human cells is possible with all groups of viruses mentioned. For historical reasons, most retroviral vectors applied thus far in human gene therapy have utilized the amphotropic receptor, although this is not the most efficient envelope for many targets.

B. Retroviral Genes and Their Products Retroviruses within a group share a very similar proviral structure. In the first three groups (see Table 2), including mammalian C-type retroviruses, the genome codes only for the virion structural proteins Gag, Pol, and Env (Fig. 1) [37]. The gag gene products constitute the viral matrix and package the two retroviral RNA genomes into a viral nucleocapsid. Encoded by the pol gene, the virion also includes several enzymes necessary for virus replication. These are the reverse transcriptase, the integrase, and the viral protease, which cleaves the Gag and Pol precursors into the individual proteins. Receptor utilization is determined mainly by the glycosylated env gene product SU, which is anchored in the viral envelope by the transmembrane protein TM. The viruses of the HTLV–BLV group, the lentiviruses, and the spumaviruses are more complex and also encode specific nonvirion proteins with different regulatory functions [39,40]. Examples are the viral transcriptional activators of gene expression, such as tax in HTLV, tat in HIV, and bel-1 in foamy viruses.

C. Retroviral cis Elements The cis-acting elements that regulate viral gene expression, reverse transcription, and integration of the provirus into the cellular DNA are organized very similarly in all retroviruses [37]. The provirus is flanked by the long terminal repeats (LTRs), carrying the terminal att sites, which are recognized by the integration machinery. The LTR is further divided into the three sections U3, R, and U5 (Fig. 1). The U3 region carries the viral enhancer and promoter elements. In the 3 LTR, initiation of transcription is suppressed, possibly due to interference with the 5 LTR. The polyadenylation signal resides in the R or U3 region. A recent report suggests that the retroviral splice donor may play an important role in suppressing the utilization of the polyadenylation signal of the 5 LTR [41]. Transcription of viral genomic RNA thus begins at the R region in the 5 LTR and ends with R in the 3 LTR. The RNA genome is thus flanked by identical redundant regions (R), which play an important role during reverse transcription (see Section II.D). The U5 region contains sequences necessary for reverse transcription and terminates with the att site. The untranslated leader comprises R and U5 regions of the 5 LTR and sequences upstream of gag, including 18 nucleotides that form the primer binding site (PBS). The PBS is perfectly complementary to the 3 terminus of the tRNA


7

FIGURE 1 Scheme of the proviral form of a replication-competent simple Ctype retrovirus. Sequences coding for trans-acting proteins are indicated above the drawing, cis-acting sequences below.

primer that initiates reverse transcription of the RNA genome into the minus strand of proviral DNA. Leader sequences downstream of the PBS contain the splice donor site for generating the subgenomic RNAs, as well as the packaging and dimerization signal, which directs incorporation of two viral RNA genomes into virions. Optimal packaging was originally reported to require additional sequences, such as the first 400 nucleotides of gag in MLV-based vectors (so-called gag+ vectors) [42], although more recent data suggest that viral coding sequences are dispensable for hightiter packaging of MLV vectors [43,44]. In human immunodeficiency virus (HIV), the minimal sequences sufficient for packaging have not yet been defined [45,46]; in addition to part of the leader, a region in the 5 end of the gag gene and sequences within env encompassing the Rev responsive element (RRE) appear to improve packaging [47]. The untranslated region between env and the 3 LTR contains the polypurine (PP) tract, a run of at least nine A and G residues. Synthesis of the plus strand of proviral DNA is initiated here.

transcriptase (RT) [37]. The nucleocapsid protein NC is also required for this process. Reverse transcription is initiated at the PBS. RT synthesizes the negative strand complementary to the U5 and R regions of the 5 LTR, while the RNAse H activity of RT degrades the genomic RNA. The nascent DNA strand is transferred to the 3 end of the RNA genome, where, starting with the U5 region, the negative strand is completed with concomitant degrading of the RNA genome. The PP tract escapes digestion and serves as a primer for

D. Retroviral Life Cycle The retroviral life cycle is illustrated in Fig. 2. Initially, retroviruses bind through the Env protein SU to a specific viral receptor on the cell surface. All known retroviral receptors are membrane proteins, and several have been cloned [48]. The receptors for amphotropic MLV (Pit-2) and for GALV (Pit-1) are phosphate transporters found on most human cells. Interaction of viral SU with the receptor exposes a fusion peptide in the TM and triggers fusion of the viral and cellular membranes with subsequent release of the nucleocapsid into the cytoplasm [49]. Here, the viral RNA genome is reverse transcribed into the proviral DNA by the viral reverse

FIGURE 2 Life cycle of a replication-competent retrovirus: (1) virion binding, (2) virion penetration and uncoating, (3) reverse transcription of RNA genome into proviral DNA, (4) nuclear transport of preintegration complex and integration of provirus, (5) transcription of genomic and subgenomic mRNA and translation of viral gene products, and (6) nucleocapsid assembly, budding, and maturation of virion.

8


FIGURE 3 Structure of provirus and viral genome of MLV. Provirus 1 is transcribed into the retroviral genome flanked by the R sequences. Two RNA genomes are packaged into a virion and released from the cell. After infection of the target cell, the genomic RNA is reverse transcribed into provirus 2. This again is flanked by two LTRs that contain the U5 region of the 5 LTR and the U3 region of the 3 LTR of provirus 1. Abbreviations are explained in Fig. 1.

plus-strand DNA synthesis, which proceeds through U3 and R. The plus strand is then transferred to the 5 end and transcription is completed. In brief, reverse transcription of the two RNA genomes generates a single provirus with two complete LTRs by duplicating the U3 and U5 regions of the RNA genome; importantly, 3 LTR and 5 LTR serve as templates for U3 and U5, respectively (Fig. 3). The infidelity of reverse transcription and recombinations occurring as a consequence of switching between the two RNA templates during reverse transcription lead to a high degree of variability of retroviruses. This represents a potential drawback for vector design, because unpredicatable errors may be introduced during vector infection [50]. However, other viral or nonviral methods of DNA transfer may be associated with even higher rates of recombination. In many instances, this results from the transfer of multiple copies of homologous DNA, which is excluded in retroviral systems. After reverse transcription, the nucleocapsid proteins remain tightly associated with the proviral DNA. This complex carries factors necessary for the integration of the viral DNA into the genome of the host cell. In MLV, this complex cannot pass through the nuclear pores, and nuclear transport requires mitosis with breakdown of the nuclear membrane [51,52]. Nuclear transport, however, is not the only factor limiting transduction of quiescent cells by vectors based on simple retroviruses [53]. In lentiviruses, such as HIV, mitosis is not required, and the preintegration complex is targeted to the intact nucleus with the help of the matrix, integrase, and possibly the Vpr protein [54,55]. Interestingly, a DNA flap generated during reverse transcription has been identified as a cis-acting component required for nulcear import of HIV-1

[56]. Despite the ability to transduce quiescent cells, integration of lenitviruses occurs more efficiently in metabolically activated cells and best in cycling cells [55]. Integration of the provirus is random with regard to position in the genome, with some preference for open chromatin [57]. Local structural features of host DNA rather than specific sequences influence the susceptibility to integration. Rarely, integration can produce alteration of the phenotype of an infected cell by activation or disruption of cellular genes [58]. Infection with actively replicating virus is accompanied by repetitive integration events in different cells and thus increases the possibility of proto-oncogene activation and the development of neoplasias [59]. In therapeutic retroviral gene transfer, the probability of such insertional mutagenesis has been minimized by the use of replication-incompetent retroviruses as vectors that integrate at low copy numbers (usually one or two per cell). For a single integration event, the risk of inducing tumor-promoting mutations is estimated to be in the range of 10−6 or lower [60]. Immune responses to altered cellular genes further reduce but do not exclude the likelihood of inducing tumors by retroviral vector integration. The integrated provirus is transcribed by the cellular RNA polymerase II. In simple retroviruses such as MLV, this process is solely dependent on the cellular transcription machinery. Between viral strains, binding sites for cellular transcription factors in the U3 region differ. Because the expression of many transcription factors is developmentally regulated, cell tropism and pathogenicity of retroviruses are influenced by the composition of cis elements in the LTR [61,62]. Complex retroviruses have a number of transcriptional and posttranscriptional transactivators which, in cooperation with cellular


factors, influence viral transcription levels, nuclear export of RNAs, and splicing patterns [40]. Viral transcripts are modified by cellular capping enzymes at the 5 end, and a poly(A) tail is added to the 3 end at the R–U5 border following specific polyadenylation signals. Retroviral transcripts enter one of three pathways: (1) The RNA is translated into the Gag or Gag–Pol precursor proteins. (2) The RNA is spliced into subgenomic RNA. (3) The fulllength viral RNA is packaged as a viral RNA genome into the virion and released from the cell. All subgenomic mRNAs are spliced from the same splice donor generally located in the leader. In simple retroviruses only the env transcript is spliced. Complex retroviruses have several splice acceptors in the 3 half of the genome, where several different smaller spliced transcripts are generated which code for regulatory proteins [37,40]. Initiation of translation of the gag–pol or the env message may also occur in a cap-independent manner, facilitated by an RNA structure that resembles an internal ribosomal entry signal [63]. The Gag precursor protein is always translated from the full-length viral RNA. Translation is continued past the stop codon to generate a Gag–Pro or Gag–Pro–Pol precursor protein at a low frequency (5–10%). The viruses of the avian sarcoma–leukosis virus (ASLV) complex are an exception, as gag has no stop codon and is always translated as a Gag–Pro polyprotein [64]. After virus assembly, the viral protease (PR) is activated by autocatalytic release from the precursor. PR then cleaves the Gag precursors into the matrix (MA), capsid (CA), and nucleocapsid (NC) proteins and several smaller peptides. The Pol precursor is cleaved by PR to yield the reverse transcriptase, which has an RNA- and DNA-directed polymerase and a ribonuclease H activity, and the smaller carboxyl terminal viral integrase. The Env protein is cleaved, most likely during transport to the cell surface, by a cellular protease into the viral surface protein (SU) and a transmembrane protein (TM). Viral genomic RNA is assembled into the nucleocapsid through specific interaction of the NC portion of the Gag

9

precursor and cis-active viral packaging sequences [45]. Band D-type retroviruses preassemble in the cytoplasm, while C-type viruses, such as MLV, assemble at the cytoplasmic membrane. The MA domain interacts with the inner surface of the cytoplasmic membrane and mediates budding of the virus [65]. The virus acquires the viral envelope by budding through membrane areas that contain Env proteins. However, viral glycoproteins are not necessary for virion formation, and in the absence of Env noninfectious, bald, enveloped particles are released. Proteolytic cleavage of viral polyproteins begins during the budding process and is completed in the released particles, thereby generating the mature infectious virions [66].

IV. PRINCIPLES OF RETROVIRAL VECTOR SYSTEMS First, we will discuss general rules for designing packaging cell line and vectors. In Section V, we approach specific aspects related to vector entry, integration, and expression.

A. Packaging Cells Retroviral vector systems are designed to mimic the infectious properties of retroviruses (stable transduction of target cells without inducing rearrangements and relatively stable gene expression) in replication-incompetent vector particles. The latter are produced from packaging cell lines. In these cells, viral coding regions are physically and functionally separated from the vector genome. The strict separation of cis-active and trans-active components serves purposes of safety, efficiency, and increased flexibility (Fig. 4). A modern, safety-modified packaging cell contains at least two expression constructs for viral genes, one encoding for gag–pol and one for env genes [67]. To prevent mobilization in retroviral particles, these mRNAs lack the packaging signal. In the case of MLV, this is easy to achieve, as retroviral

FIGURE 4 Separation of cis-acting sequences and trans-acting retroviral coding regions to generate safe packaging cell lines for release of replication-incompetent vectors. Abbreviations are explained in Fig. 1.

10


coding regions are not sufficient for packaging. In HIV, the extended and less well-defined packaging signal might represent a drawback to the development of helper cells (and vectors). High expression of viral proteins in packaging cells is best accomplished by allowing direct selection for the promoter driving the retroviral genes [68]. This can be managed by linkage with a dominant selectable marker, with coexpression obtained by reinitiation of transcription from the 3 untranslated region, internal ribosomal entry sites for reinitiation of translation, or alternative splice signals. Alternatively, inducible promoters can be used, representing the method of choice when particle components have cytopathic effects (as in the case of some proteins used for pseudotyping; see Section 5.A.1.a). Encoded from such “packaging constructs,” the packaging cell provides all trans-acting viral elements required for particle assembly, release, maturation, and transduction of the target cell. The vector RNA is also generated inside the packaging cell and contains those cis-active elements required for a single transduction: cap site and poly(A) signal for the genomic message; packaging signal for incorporation into particles; PBS, PP, and R–U5 sequences for reverse transcription; and att sites for integration. Moreover, it contains the transgene cassette(s) of interest including enhancer/promoter sequences for initiation of transcription (Fig. 4). Packaging cells not expressing RNAs with suitable packaging signals should not release infectious particles. However, depending on the cellular background, there is the potential risk of packaging endogenously expressed retroviral or retrovirus-like sequences. In the case of MLV-based systems, this risk is highest in a rodent background. Here, packaging and transfer of VL30-sequences are observed as frequent events, especially when vector titers are low [69,70], implying that this risk can be reduced when high-titer producer clones are selected for vector production. This is important because packaging of and recombination with endogenous retroviral elements are the most important events leading to the generation of replication-competent retroviruses from safety-modified packaging cells [71,72]. In immunocompromised primates and permissive mouse strains, replication-competent amphotropic MLV can induce lymphoma or leukemia [58,59]. Also, amphotropic retrovirus-induced spongiform encephalomyelopathy has been observed after inoculation in newborn mice [73] (for a thoughtful discussion on safety aspects of non-human, xenogenic viruses, see also Isacson and Brakefield [74]). More recently, many packaging cell lines have been developed or are under construction in a non-rodent background, such as human or canine, not known to express retroviral sequences packaged in MLV particles. In the human host, most xenogenic retroviruses are complement sensitive, precluding administration in vivo.

Complement sensitivity is defined at two levels: first, by specific Env sequences, and, second, by protein modifications characteristic to the species background of the producer cell [75,76]. Complete complement resistance is achieved by producing vector particles with alternative envelopes (e.g., derived from feline leukemia virus) in human packaging cells. However, repetitive administration is likely to be hindered by immunogenicity of retroviral proteins. Moreover, fundamental restrictions to successful transduction in vivo are present at the physicochemical level, such as particle concentration and motility (reviewed by Palsson and Andreadis [77]). Ex vivo, these can be overcome by suitable transduction protocols. To produce replication-incompetent vectors, vector genomes are introduced into packaging cells either by transfection or by retroviral transduction. Packaging systems have also been developed to release high-vector titers after transient transfection, or semi permanently from episomally replicating plasmids [78,79]. For clinical applications, stably transfected clonal packaging cell lines still represent the ultimate choice, as these allow vigorous preclinical testing of safety (especially the absence of replication-competent retroviruses [80]) and efficiency, as well as large-scale production of vector stocks from defined cell banks under conditions of good manufacturing practice (GMP). Titers released from retroviral packaging cell lines usually do not exceed 106 to 107 per milliliter of cell-free supernatant. Currently, this is the minimum required for ex vivo approaches, such as transduction of hematopoietic cells. When all components are improved, titers might be as high as 108 . Concentration of the fragile retroviral particles is alleviated when the membrane is stabilized with non-retroviral components (see Section V.A.1.a).

B. Basic Vector Architecture The flexibility of the retroviral genome offers a great degree of freedom for insertion of transgene cassettes. Still, retroviral vectors mimicking the basic architecture of their replication-competent ancestors (LTR–leader–gene(s)–LTR) have found the most widespread use (Fig. 5A and B) [81]. These vectors are usually very stable and also mediate reasonable transgene expression in the cellular system of interest, provided that appropriate enhancers are employed (see Section V.B). The gene is expressed either from within the gag region [42,82], exactly replacing gag [44], or from a subgenomic, spliced RNA which can lead to higher translation efficacy (MFG vector [43,83], GRS vector [44]). Efficiency as well as safety of retroviral vectors might be further improved when the leader is functionally inactivated or physically deleted in the target cell (Fig. 5C to F). Increased efficiency results from removing the long untranslated leader from the transcript, which might contain repressory elements for transcription or translation. This also excludes the


11

FIGURE 5 Flexibility in basic vector architecture. (A, B) LTR-controlled vectors including the packaging region () in the genomic transcript in transduced cells. gag and pol , residual fragments of viral genes gag and pol, respectively; destroyed start codon of gag. (C–F) Different forms vectors excluding from transcripts in transduced cells. Plasmid constructions are represented on the left (before reverse transcription), and the status of the proviral form after reverse transcription is shown on the right. In (E) and (F), the status in transduced cells after site-specific recombination is shown. The black filled triangles represent loxP sites recognized by the site-specific recombinase, Cre.

packaging region from vector transcripts in the target cell, thus precluding transmission of the vector in the hypothetical case of accidental superinfection with replication-competent retroviruses. There are several options for excluding the packaging signal. First, the transgene plus its enhancer-promoter can be placed in the U3 or R region of the LTR, resulting in a “double-copy” vector after completion of reverse transcription [84,85]. Duplication of the transgene cassette is expected to result in higher expression levels. Some sequences, however, are not compatible with this strategy, resulting in a high incidence of recombination. Second, self-inactivating or suicide vectors can be generated by deleting the enhancer– promoter or the promoter only in the U3 region of the LTR and placing the transgene of interest under control of an internal promoter, either in sense or in antisense orientation to the LTRs [86,87]. Third, in LTR-controlled vectors, sequences between PBS and the start codon of the transgene can be flanked by, for example, loxP sites, allowing conditional deletion upon expression of the bacteriophage recombinase cre [88]. Finally, reversion of double-copy vectors to a monocopy vector is possible with a self-contained loxP/cre vector [89]. Given that stability, titer, and expression characteristics of these more sophisticated constructions are better defined, they represent valuable alternatives to conventional, LTR-controlled vectors.

V. ADVANCES IN RETROVIRAL VECTOR TAILORING Besides the more general aspects of packaging cell line design and vector construction discussed above, specific advances can be noted with respect to distinct stages of the retroviral life cycle. Of special importance are those related to vector entry, integration (both defined by trans-active vector components), and expression (defined by cis-active elements) relevant to cancer gene therapy. This work should lead to vectors specifically tailored for clinical applications.

A. Components Active in trans 1. The Retroviral Envelope For many reasons, the amphotropic Env, hitherto used in most gene therapy trials involving retroviruses, is not a perfect choice for mediating vector entry. The cognate receptor, Pit-2, is too widely expressed to allow specific cell targeting, with the ironical exception of primitive hematopoietic cells, where expression is too poor to allow efficient transduction [90–93]. Moreover, the amphotropic Env is involved in an unexpected pathogenicity of replication-competent retroviruses: induction of spongiform encephalomyelopathy [73]. Alternative Env proteins like that of the 10A1 strain are also

12


associated with these potential drawbacks [94]. Moreover, the low stability of retroviral particles with conventional retroviral envelopes complicates vector concentration for in vivo delivery. Importantly, both vector stability and targeting can be improved by altering the retroviral envelope. The two major approaches are discussed below. a. Pseudotyped Retroviral Vectors Coinfection with two viruses generates hybrid virions, which contain the genome and core proteins of one virus and mixed envelope glycoproteins of both viruses. The host range of these “pseudotypes” is determined by both envelope proteins [48,95]. Pseudotyping can be used to alter the host range of retroviral vectors. Pseudotyped MLV-derived vectors thus can transduce cells that are normally resistant to MLV due to lack of functional amphotropic receptor (reviewed by Friedmann and Yee [96]). The mechanisms that determine whether a foreign viral envelope protein can be incorporated into the viral envelope are not well understood. The cytoplasmic anchor of the transmembrane (TM) Env protein was shown to guide MLV glycoproteins to the envelope of budding virus particles [97]. Thus, homologous Env proteins are efficiently incorporated into the MLV envelope, while heterologous proteins must be expressed at high densities in the cell membrane to allow pseudotype formation [98]. Pseudotyped retroviral vectors are generated by coexpression of vector RNA with retroviral Gag and Pol and the unrelated glycoprotein. Packaging systems for several pseudotypes of MLV have been developed. Pseudotypes that incorporate the glycoprotein of vesicular stomatitis virus (VSV-G) have an extremely broad host range [96,99]. The VSV-G protein enters the cell by interacting with an ubiquitous phospholipid component of cell membranes (Fig. 6) [100]. Mammalian, fish, and insect cells can be transduced

[96,99–103]. CD34+ hematopoietic progenitors were shown to be up to 10-fold more susceptible to a VSV-G than to an amphotropic pseudotype [103]. We found that VSV-G pseudotypes can infect hematopoietic stem cell lines and fibroblasts equally well, while transduction of stem cells with amphotropic vectors was at least 100-fold less efficient [93]. This observation reveals that the receptor deficiency of primitive hematopoietic cells to retroviral transduction [91,92] can be completely overcome by vector pseudotyping [93]. An additional advantage is that VSV-G confers great stability on the retroviral particle, and pseudotypes can be concentrated to high titers by ultracentrifugation. This is of interest also for in vivo applications. However, the host range of VSV-G is too broad to allow specific cell targeting, and the high immunogenicity of VSV-G is expected to preclude repetitive administrations in vivo. Another major drawback has been that the VSV-G protein is toxic for the cell. Pseudotypes are thus only produced for a limited period from already dying packaging cells [101]. Recently, stable packaging cell lines have been generated by placing the VSV-G gene behind an inducible promoter. In these lines the VSV-G gene is repressed but can be induced for vector production. However, vector production here is also accompanied by cell death [102]. Alternatively, pseudotypes of MLV vectors that incorporate Env proteins of the feline RD114 retrovirus, the gibbon ape leukemia virus (GALV) or the 10A1 mouse retrovirus transduce hematopoietic progenitors and lymphocytes more efficiently than amphotropic pseudotypes [104–107]. Other retroviral Env proteins have also been utilized. Examples are the glycoproteins of mink cell focus-forming (MCF) MLV-strains, HTLV-I, HIV-1, and human foamy virus (HFV) [38,108–111]. The tropism of these pseudotypes generally has not been properly evaluated to show advantage over amphotropic vectors. HFV pseudotypes are especially

FIGURE 6 VSV-G protein pseudotyped retrovirus. Pseudotyping affects particle stability, receptor targeting, and mode of entry. After uncoating, different envelope pseudotypes follow the same pathway.


promising, as it is assumed that all mammalian cell types are infectable. However, we have observed that the hematopoietic progenitor cell lines FDC-Pmix and FDC-P1 are not only partially resistant to amphotropic MLV but also to HFV infection [93,112]. Recently, we described a novel retroviral pseudotype with the glycoproteins GP-1/-2 of the lymphocytic choriomeningitis virus (LCMV). LCMV GP is not cell toxic and is efficiently incorporated into MLV as well as into lentiviral vectors. This pseudotype has a broad host range and can be concentrated to high titers [113]. Further perspectives in pseudotype development are opened by generating chimeric envelopes. For instance, efficient pseudotyping of MLV with HFV surface proteins is only possible for chimeric envelope proteins containing an unprocessed cytoplasmic tail of MLV TM fused to a truncated HFV envelope protein [108]. Using similar chimeric envelope proteins it may be possible to generate a large panel of different pseudotypes with unrelated viral or nonviral membrane proteins which normally are not incorporated into retroviral envelopes efficiently.

13

b. Ligand-Directed Targeting Ideally, vectors should be designed to selectively transduce specific target cells of interest present within mixed cell populations ex vivo or even intact organs in vivo. In oncogene antagonism and suicide gene transfer (see Section II), specific or at least preferential targeting to tumor cells has to be achieved in vivo. Here, binding of virus to non target cells will lead to considerable loss of the effective virus titer and also increases unwanted side effects. In contrast, drug resistance gene transfer to hematopoietic cells (see Section II) is performed ex vivo, with strict exclusion of malignant cells. Targeting retroviral transduction can be achieved at two levels: first, by colocalization of cells and viruses on a specific matrix (Fig. 7A), and, second, by equipping retroviral particles with cell-specific ligands (Fig. 7B to F). Colocalization of cells and viruses on a biochemical matrix can only be used ex vivo, alleviating vector–cell interactions at the physicochemical level. Colocalization can lead to higher transduction efficiency in cells with poor receptor representation, paradigmatically shown in fibronectin-assisted retroviral transduction of hematopoietic progenitor cells or

FIGURE 7 Targeting transduction via cellular receptors. The strategies discussed in Section V.A.1.b are schematically represented. Entry either occurs via the differentiation-specific cellular receptor (open symbol on cell surface) or still requires the natural retrovirus receptor (shaded symbol). (A) colocalization of virus and cell via matrix proteins; (B) molecular bridge between virus and cell (here, cross-linked antibodies); (C) a peptide in the binding domain of SU alters its tropism; (D) the binding domain of SU is replaced with a targeting ligand; (E) an N-terminal addition is linked to SU via a protease-cleavable linker or, as shown in (F), via a flexible linker.

14


lymphocytes [9,114,115]. However, it remains to be seen whether this approach can be elaborated for target-specific virus uptake—for example, by displaying ligands on the matrix which are selectively recognized by the target cell population of interest (Fig. 7A). The alternative approach is to retarget retroviral entry via specific cell surface molecules by manipulating the viral envelope. To this end, several strategies have been followed (reviewed by Cosset and Russell [116,117]). While specific binding is relatively easy to achieve, virus uptake with engineered envelopes often is much less efficient. One initial approach was to direct specific virus binding by creating a molecular bridge between the virion and the cell surface (Fig. 7B). Here, virus particles are coated with specific antibodies for the surface subunit of Env (SU), and the cells are incubated with an antibody specific for a membrane protein such as the epidermal growth factor receptor or the insulin receptor. Both antibodies are then linked by secondary antibodies or by biotin/streptavidin [118]. In other studies, ecotropic or avian retroviral envelope proteins were modified. Wild-type Env proteins of these viruses do not allow infection of human cells. Therefore, incorporation of specific binding epitopes can selectively retarget the virus to human cells with the complementary membrane protein of choice. Three general strategies have been followed: 1. Small peptides that specifically bind to cellular receptors are introduced into binding domains of the SU protein without affecting natural receptor recognition [119,120]. Human breast cancer cell lines, overexpressing human epidermal growth factor receptors (HER-2 and HER-4) could be specifically targeted by insertion of the heregulin peptide, a ligand for HER-2 and HER-4, into ecotropic SU of MLV [121]. However, titers and efficiency of transduction were too low for in vivo applications (Figure 7 C). 2. The complete binding domains of SU are substituted by alternative ligands for cellular receptors (Fig. 7D). Erythroid progenitor cells have been targeted by substituting SU binding domains with erythropoietin [119]. Chimeric SU proteins that contain single-chain antibodies (scAs) attached to the truncated retroviral Env proteins are a versatile system with the potential of targeting cells via specific epitopes of many different membrane proteins. An example is the scA B6.2, which binds to an antigen on breast and colon cancer cells [122]. This strategy works well with the Env of the avian spleen necrosis virus, whereas transduction with MLVderived chimeric Env proteins is inefficient and associated with low titers [123]. Differences in the flexibility of Env proteins and in the pathways involved in virus internalization might be responsible for this discrepancy. In wild-type Env, virion binding causes a conformational change in SU, thereby exposing fusion domains of the Env transmembrane unit (TM), finally leading to viral penetration. In the chimeric

envelope proteins described so far, conformation is generally altered and fusion processes are not triggered efficiently [124]; therefore, most chimeric Env proteins support efficient binding of virions, but postbinding events are impeded or even completely blocked. This indicates that additional alterations in TM might be required to improve uptake. An alternative would be to incorporate foreign viral envelope glycoproteins, such as the hemagglutinin glycoproteins from fowl plague virus which display a targeting ligand, into the retroviral envelope [125]. The postbinding functions of such nonretroviral envelope glycoproteins may be less sensitive to modification of the protein. Virion infectivity can also be increased by incorporating additional wild-type Env proteins that mediate fusion (“fusion helpers”) [126]. Such receptor cooperation is also of central importance for the third targeting strategy. 3. With the aim of improving virus penetration, specific binding domains have been added to the N-terminus of the complete amphotropic Env proteins [127–130] (Fig. 7E and F). Amphotropic Env mediates infection of many human cell types. N-terminal additions, however, can block binding to the amphotropic receptor. Instead, these viruses bind to another membrane receptor of choice, but penetration still might require the amphotropic receptor. Two types of linkers between the retroviral Env and the N-terminal targeting domain allow such a two-step entry mechanism. In one approach, the ligand is fused to the amphotropic Env via a protease-cleavable linker [127] (Fig. 7E). Particles displaying these proteins bind to but do not infect cells. After protease cleavage, the N-terminal extension is released and the amphotropic binding domain exposed. Bound virus can then enter the cell efficiently by the amphotropic receptor. An unsolved problem in vivo is the systemic application of protease. An alternative would be to characterize cellular membrane proteases with their specific target sequences. In another approach, cooperation between two receptors is mediated by a proline-rich linker between amphotropic Env and the additional binding domain (Fig. 7F). Here, it is assumed that binding of the added specific ligand to its receptor triggers a conformational change that exposes amphotropic binding domains, which then mediate efficient entry via the amphotropic receptor [131]. A problem with both approaches may be that the amphotropic receptor is not expressed at sufficient levels on all cell types, as evident from studies with early hematopoietic cells [93,107]. While most investigators concentrate on positive targeting, negative targeting of selected cells can also be desirable in cancer gene therapy. An example is drug resistance gene transfer (see Section II), where the transduction of malignant cell is potentially hazardous because clones resistant to chemotherapy might be generated. Here, the specific


blockade to retrovirus entry found with many engineered Env proteins can potentially be exploited to increase the safety of vectors. 2. Nuclear Transport and Integration a. Vectors Derived from Complex Retroviruses Nuclear transport of the preintegration complex is restricted in C-type retroviruses such as MLV which require mitosis and breakdown of the nuclear membrane for integration into the host cell genome. Unlike MLV vectors, lentiviral vectors can transduce nondividing, yet postmitotic, cells such as neurons and terminally differentiated macrophages [132– 135]. Malignant cells, especially in larger tumors where blood supply becomes limiting, are also often quiescent. Similarly, hematopoietic stem cells rarely cycle but can be transduced with lentiviral vectors [136]. Several lentiviral packaging systems have been developed that generally use the G protein of vesicular stomatitis virus (VSV-G) and not the retroviral Env as an envelope glycoprotein [136–138]. A major problem is that several gene products used in lentiviral packaging systems, such as the protease, VSV-G, and vpr, have proven to be toxic [139]. Therefore, vector titers in stable packaging cell lines have been low. Recently, however, an inducible packaging system that produces titers as high as 106 per milliliter has been described [140,141]. A problem for vector safety is that packaging sequences are dispersed throughout the HIV genome and are not clearly separated from coding regions [46,47]. Therefore, vectors and packaging constructs share common sequences with the potential to generate replication-competent viruses with pathogenic potential by homologous recombinations. In the latest generation of lentiviral vectors, this risk has been largely eliminated by reducing lentiviral genes to gag, pol, and rev, which are expressed from two separate plasmids [142]. However, there still remains a concern that individuals treated with HIV-derived vectors may exhibit serum conversion to HIV-1 [143]. Another alternative in the future may be vectors derived from animal lentiviruses such as simian immunodeficiency virus, feline immunodeficiency virus, and equine infectious anemia virus. Such vectors would have the ability to transduce quiescent cells but not the pathogenic potential of HIV [40,144–147]. Vectors derived from foamy viruses may have several advantages over lentiviral vectors. Foamy viruses are now generally considered to be apathogenic in humans, although this issue has been controversial in the past [148]. Foamy viruses have an increased packaging capacity (12 kb compared to 9–10 kb in MLV). They infect many mammalian cell types; therefore, the host range is generally considered to be broad [149,150]. However, infection of hematopoietic stem cell was found to be inefficient for cell-free vectors that carry the foamy virus envelope glycoprotein, and

15

foamy virus capsids require the cognate envelope protein for particle export [112,151]. This indicates that the viral host range may be more restricted than is generally believed. It has been postulated that HFV vectors transduce stationary cells more efficiently than MLV; however, this issue is still controversial [152,153]. Helper virus free vectors have been developed, but safety concerns remain, as the packaging sequences in the viral genome have not been defined clearly and a clear separation of viral trans and cis elements may not be possible [154,155]. Taken together, knowledge of the biology of foamy viruses and other complex retroviruses is still limited. Extensive studies will be necessary before the value of these vectors for human gene therapy can be assessed. b. Targeting Integrase Targeting integrase to selected genomic loci is desirable to completely avoid insertional mutagenesis, to select integration sites supporting long-term expression, and to reduce clonal variability of gene expression. In the context of cancer gene therapy, these considerations are of relevance for gene marking, drug resistance gene transfer, and adoptive immunotherapy (see Section II). Among the factors influencing site selection are overall DNA confirmation (open chromatin is a better target than heterochromatin), DNA sequence (in terms of local chemical or structural features rather than concrete motifs), DNA bending, and associated nuclear proteins (transcription factors, topoisomerases, replication proteins, matrix proteins) [156– 158]. Integrases from various retroviruses differ with respect to target site selection, depending on the central core domain [159]. Systems for targeting retroviral integration to specific sequences are based on fusion of the IN protein with DNAbinding domains of well-characterized transcription factors, resulting in preferred but not specific integration to cognate sites [160–162]. A more efficient alternative might be to exploit the specificity of some yeast retrotransposons (Ty1, Ty3) for genes transcribed by RNA-polymerase III which exist in multiple copies and for which integration of a transgene is not expected to be hazardous [163,164]. Also, some human LINE elements and related retrotransposable sequences from other species encode endonucleases that prefer DNA with certain structural features [165]. It seems attractive to exploit such endonucleases, which are functionally distinct from integrases, for vector packaging systems. A completely different approach is to block the viral integration process by elimination of att sites from the vector. Then, selection can be made for integration via homologous recombination; this process, however, is limited by the cloning capacity of retroviral vectors (9–10 kb) and by the extremely low frequency of gene targeting in somatic cells [166].

16


B. cis-Active Elements With the exception of gene marking, which theoretically can be performed without introducing active transcription units, all other applications for somatic gene transfer in oncology require a certain strength and duration of vector transcription (Table 1). Choosing the appropriate cis-acting elements guarantees full penetrance of the phenotype of interest and thus influences both safety and efficiency of the gene transfer. As opposed to physicochemical transfection methods, retroviruses are characterized by only moderate integration site dependence of gene expression. This implies that integration occurs at permissive loci or that retroviruses transfer genetic elements that can actively induce conformational or functional changes in their environment [167]. Such elements may reside in the enhancer region and involve yet poorly understood mechanisms, including secondary DNA structures [168]. Residual modulatory influences by the integration site usually lead to about 50-fold variation of gene expression levels among independent clones. However, depending on specific vector sequences and the genetic environment, complete extinction (silencing) of retroviral gene expression can also occur (reviewed by Lund et al. [169]). This is most evident in embryonal stem cells [170] but also has been observed in hematopoietic stem cells [171] and more mature tissues such as fibroblasts in vivo [172] and several somatic cell lines in vitro [169]. Therefore, modifying cis-acting elements in retroviral vectors can affect all aspects mentioned: differentiation-dependent gene expression levels, integration-site-dependent modifications, and incidence as well as kinetics of silencing. Thus, it is crucial to equip vectors with enhancer sequences fitting to the host’s transcriptional setting. In simple retroviruses such as MLV and derived vectors, two major targets for transcriptional control have been identified: the dominant enhancer–promoter is located in the U3 region of the LTR, but sequences of the nontranslated leader (especially PBS) also contribute [173,174] (Fig. 8). Retroviral enhancers display recognition sites for a variety of transcription factors intimately involved in differentiation processes of their natural target cell population. Precise consensus sequences, their numbers, and relative orientation are crucial for enhancer strength and specificity [175]. Most retroviral enhancers are poorly expressed in more primitive, uncommitted cells such as embryonic and hematopoietic stem cells [174]. This is mainly because these cells are not fully equipped with transcriptional activators or even express active repressors recognizing the retroviral enhancer. In permissive environments, such as in more mature hematopoietic cells, retroviral cis-elements generally act quite autonomously and in a dominant manner, resulting in efficient transcription levels. Here, up to 0.1% of cellular transcripts can be generated from single-copy integrations. But even in more mature cells,

FIGURE 8 Dominant cis-acting elements of a murine leukemia virus reside in the U3 region of the LTR (specified in more detail for the strain SFFVp) and in the primer binding site (PBS) of the untranslated leader. SD, splice donor; , packaging signal. For abbreviations of the enhancer boxes (gray) shown, refer to the text and Baum et al. [168].

differences in crucial enhancer elements greatly influence tissue tropism. 1. Early Hematopoietic Cells These represent a mixed cell population of primitive and uncommitted cells, with a latent, yet enormous potential for proliferation and step-wise differentiation following predefined genetic programs (reviewed by Morrison et al. [176] and Weissman [177]). This is the target cell population for drug resistance gene transfer (see Section II), where high levels of transgene expression are crucial for protection from chemotherapeutic side effects [13,14]. Many vectors utilize control elements of the Moloney MLV (MoMLV) or the related Harvey murine sarcoma virus. These elements are strongly recruited in activated T cells but are only moderately active in more mature myeloid and erythroid precursor cells and are repressed to low levels in stem cells. Further repression of MoMLV-based vectors results from inhibitory elements targeting the PBS, both in embryonic stem cells and in early hematopoietic cells [13]. Based on systematic studies of transcription control of murine retroviruses in embryonic and early hematopoietic cells, we developed a series of vectors better adapted to the needs of these cells. The complex genealogy of these vectors is illustrated in Fig. 9. cis-Active elements of myeloproliferative sarcoma virus (MPSV), differing from MoMLV by mutations in putative repressor sites and in one binding site for the transcription factor Sp1, perform better in hematopoietic progenitor and in


17

FIGURE 9 Genealogy of retroviral vectors developed for strong constitutive gene expression in early hematopoietic cells and embryonic stem cells.

embryonic stem cells. PCC4-cell passaged MPSV (PCMV) is an MPSV-variant that has lost one copy of the direct repeat of the enhancer. It arose by forced passage in embryonic carcinoma cells and contains the first retroviral enhancer known to be active in primitive embryonic stem cells [178]. When it was combined with the leader of an endogenous retrovirus displaying an alternative PBS sequence, a vector resulted that allowed LTR-driven gene expression in undifferentiated embryonic stem cells. This chimeric virus is known as murine embryonic stem cell virus (MESV) [173]. The MESV-backbone has been modified to include features of the MoMLV-based LX vectors [82] in the 5 untranslated

region (UTR) (packaging signal and untranslated gag sequences) and in the 3 UTR (complete deletion of env). These modifications were incorporated to increase packaging efficiency and vector safety but did not improve gene expression as compared to MESV (MSCV, murine stem cell vector [179]). Vectors based on MESV (including MSCV) have found widespread use in experimental hematology, being associated with moderate, yet reliable transgene expression in myelo-erythroid progenitor cells and lymphocytes [7]. In the MPSV–MESV hybrid vector (MPEV), the enhancer of MESV was replaced with the corresponding sequences of MPSV, roughly doubling gene expression levels due to the

18


presence of the second copy of the direct repeat. A similar vector has been developed by Kohn and colleagues and was named MD [180]. This group has shown that removal of an enhancer sequence located 5 to the direct repeat, containing a putative repressor site, may raise the probability for long-term expression in transplanted hematopoietic cells [171,182]. Enhancers of Friend–MCF viruses such as SFFVp (spleen focus-forming virus) were found to allow further increased gene expression levels in myeloerythroid cells [13]. An SFFVp-based vector can mediate sustained multilineage gene expression through serial transplantations in mice [183]. When the Friend–MCF-related U3 regions are combined with the nonrestrictive leader of MESV, novel vectors result which we named FMEV (Friend–MCF–MESV hybrid). These currently represent a reasonable choice for strong transgene expression in hematopoietic cells (Fig. 10A) [13,14]. The importance of improving enhancer strength became evident from comparative vector studies in the context of drug resistance gene transfer. Only MPEV, and, even better, FMEV mediated high-dose drug resistance. Background-free selection of primary hematopoietic cells was thus possible when the human multidrug resistance 1 (MDR1) gene was expressed [13,14] (Fig. 10B). Moreover, intact proliferation and differentiation of transduced hematopoietic progenitor cells were observed in the presence of myeloablative doses of chemotherapeutic agents, indicating complete detoxification [14] (Fig. 10C). FMEV also allows dominant selection with MDR1 when a second gene is coexpressed. This is remarkable because coexpression of a second gene leads to reduced MDR1 expression when compared with the monocistronic counterpart [16]. Strong gene expression from FMEV vectors can also be instrumental for studies employing cell surface markers [184] or cytoplasmic proteins such as green fluorescent protein [185]. In order to further increase the transcriptional strength and specificity of FMEV, we are performing a molecular analysis of Friend–MCF-type enhancers. At least three crucial motifs contributing to strong and relatively lineage-independent activity in hematopoietic cells were identified: recognition sites for the ubiquitous transactivator, Sp1; ETS family members; and AML1/PEBP [168] (Fig. 8). As expected, these are all important transcriptional regulators in hematopoietic cells [186]. Additional activation may result from E-Box binding basic helix–loop–helix factors [187] and Myb [188]. Similar recognition sites are represented in a number of endogenous promoters controlling differentiation-dependent cellular genes. Such cellular motifs can be successfully incorporated in retroviral vectors [189]. Variations in enhancer assembly (e.g., by developing hybrid enhancers composed of distinct modules of retroviral or endogenous enhancers) are expected to result in even higher gene expression levels. Other alterations may lead to more specific and lineage-restricted activity within the hematopoietic system. Thus, it seems possible to develop novel enhancers that are strongly recognized

FIGURE 10 Vector design determines phenotype, here shown for myeloprotection by drug resistance gene transfer. (A) Different types of retroviral vectors evaluated in the context of transfer of the multidrug resistance gene (MDR1). (B) Relative selective advantage conferred to primary human hematopoietic colony forming units (CFUs) kept under selection with the chemotherapeutic agent Taxol, recognized by the MDR1-encoded efflux pump, P-glycoprotein. Data are calculated from Eckart et al. [14] and expressed as cloning efficiency fold negative control (i.e., cells transduced with MP1N). PC-MDR (MESV-type) is only slightly better than V-MDR and therefore not shown. (C) Average colony morphology at selection with 15 ng Taxol/mL reveals importance of complete detoxification. This can only be achieved with vector backbones of strong transcriptional activity.

in hematopoietic progenitors but have low activity in tumor cells (e.g., are of epithelial origin). With such hematopoiesisspecific enhancers, transduction of non hematopoietic tumor cells would have no significant consequences in terms of inducing drug resistance.


2. T Lymphocytes Although they represent a mature blood cell population, T cells can be very long lived and have the capacity for limited clonal activation and expansion. In the switch between resting and the activated status, chromosomal organization and transcription factor equipment is reordered. Thus, stably integrating retroviral vectors are a perfect tool for genetic manipulation of T cells, but vector expression may vary depending on the cellular activation status [190,191]. All MLV-based vectors described in Fig. 10A and related constructs mediate sufficient expression in activated T lymphocytes for application in adoptive immunotherapy [190,192]. In T cells, however, MPEV is clearly stronger than FMEV. The enhancer of SL3-3, a highly lymphotropic MLV, is an interesting alternative [193]. As discussed for early hematopoietic cells, insights into the molecular mechanisms defining T lymphotropism of retroviral or endogenous enhancers is expected to create the basis for developing artificial transgene enhancers with increased T-cell specificity. Interestingly, as with some endogenous T-lymphocytic promoters, reversible downregulation of retroviral gene expression was observed in resting T cells. This might be prevented by inclusion of scaffold attachment regions in the vector (see Section V.B.4) [191]. 3. Tumor Cells Mechanisms of tumor-specific transcriptional controls are of interest for targeting of tumor cells in suicide gene transfer and oncogene antagonism, as outlined above (see Section II). Generally, the specificity of heterologous promoters in retroviral vectors is increased when more promiscuous retroviral enhancer sequences are deleted. Transcriptional targeting of tumors can be achieved using control elements of genes that are “tumor specific” or over-expressed in tumors. When targeting metastases, control elements of genes specific to the parental tissue of the tumor might also be sufficient. Also, hypoxia-responsive promoters have been proposed for tumor targeting [194]. A more indirect approach is the targeting of endothelial cells involved in tumor angiogenesis using “endotheliotropic” control regions. Thus, an ever-increasing number of candidate promoters is being proposed (reviewed by Sikora [23] and Miller and Whelan [195]). However, for most of them, evidence for tumor specificity in vivo is yet to be confirmed.

19

may be directly triggered by the integrated vector. Silencing involves functional reorganizations within the chromosome. As a result, vector sequences can be methylated in CpG islands, which may play a role for fixation of downregulation [196,197]. The speed and incidence of silencing depend on the cellular background, the genomic integration site, and (not well defined) on specific vector sequences, including transgene cDNAs (reviewed by Lund et al. [169]). This opens perspectives for active prevention of silencing by vector improvements. Studies with housekeeping promoters indicate that Sp1 binding sites can counteract silencing to some extent [198]. The retroviral enhancers of MoMLV, MPSV, PCMV, and Friend–MCF viruses differ with respect to number, affinity, and positioning of Sp1 binding sites [168,199]. The relevance to long-term expression remains to be shown. Furthermore, MESV-derived leader sequences or vectors containing other, even artificial, primer binding sites avoiding transcriptional repression in embryonic and hematopoietic cells (see Section V.B.1) might support long-term expression. However, silencing of MESV-leader-based vectors is also observed upon differentiation of embryonic stem cells permissive to vector expression in the undifferentiated state [170]. Insertion of scaffold attachment regions in retroviral vectors, as described by Bode and colleagues [200], may shield retroviral control regions from negative influences of the integration site and thus support transcriptional autonomy of a chromosomally integrated transgene [191], as demonstrated earlier for stably transfected plasmids [201]. Consequently, downregulation of retroviral enhancers in resting T cells [191] and irreversible silencing in transplanted hematopoietic cells can be prevented to some extent [202]. Similarly, insulator elements derived from the chicken HS4 element may reduce position dependence of retrovirally integrated transgenes [203], and even vectors lacking such elements can exhibit consistent long-term expression in hematopoietic cells [181,182]. However, most studies published so far were conducted under conditions that allowed more than one transgene integration in single repopulating cells. To clarify the probability of silencing from a single integrated transgene, further systematic analyses in appropriate primary cell systems and results from comparative clinical studies are still awaited. Importantly, results achieved with a given reporter cDNA may not necessarily be predictive for vectors containing different inserts, as coding sequences may also exhibit cis elements that influence the probability of gene silencing (see Section V.B.8).

4. Silencing Silencing not only reduces the efficiency but can also compromise the safety of gene transfer strategies. This is of special importance for negative selection of transduced cells (as required in suicide gene transfer, adoptive immunotherapy, or mini-organs; see Section II). Here, cells having silenced the vector will escape exogenous control. Silencing results from dominant negative influences of the integration site or

5. Regulatable Promoters Regulatable promoters are of interest for generating artificial mini-organs and also for drug resistance gene transfer (see Section II). Progress in regulatable promoter systems has been revewied by others [195]. Best documented in retroviral vectors is the tetracycline-regulated system, available both for conditional repression and induction of transgene

20


expression [204,205]. Moreover, a number of alternative artificial systems for conditional promoter induction or repression have already been described [195]. The applicability of synthetic inducer/promoter systems has been demonstrated in vivo using retroviral vectors expressing erythropoietin from a tetracycline-regulated cassette [21,22]. Further advances in regulated vectors are expected to address potential limitations of the systems: side effects of the drugs administered for regulation, immunogenicity of the synthetic transactivators or repressors employed, toxic squelching effects eventually occurring from overexpressed synthetic transcription factors, clonal variabilities in inducibility related to the retroviral integration site, differentiation dependence of regulation, and maintenance of regulation over time. 6. RNA Elements Nuclear and cytoplasmic processing of newly transcribed RNA is dependent on cis-regulatory RNA elements, which determine the rate of splicing, polyadenylation, nuclear export, RNA stability, and initiation of translation. Most of these processes are functionally coupled. Considering that cytoplasmic accumulation and translation of many cellular RNAs are rate limiting and can be dependent on the presence of appropriate introns, export signals, and/or polyadenylation tails [206,207], there is a growing interest in sequences that improve posttranscriptional processing of a given RNA. At least three categories of RNA modules may enhance expression from retroviral gene transfer vectors on a posttranscriptional level: splice signals that create an intron in the 5 untranslated region [43,44,83]; constitutive RNA transport elements, originally discovered in D-type retroviruses [208]; and last, but not least, the posttranscriptional regulatory element of woodchuck hepatitis virus [209]. Importantly, enhancement of expression depends not only on the specific element, but also on the gene and promoter of interest, implying context-dependent activity of RNA elements [210].

Proper combinations of RNA elements can enhance expression of a given cDNA by more than one order of magnitude, and expression of some coding sequences may even be absolutely dependent on the presence of either a constitutive transport element (CTE) or an intron [210]. Thus, comparative analyses are recommended to improve the performance of a given vector by inclusion of RNA elements. Moreover, the efficiency and safety of retroviral gene vectors may be increased by redesigning 5 untranslated regions to avoid aberrant start codons located 5 of the cDNA of interest [44]. 7. Coexpression Strategies Vectors expressing more than one transgene greatly widen the perspective of most cancer gene therapy approaches. Depending on the specific application, coexpression is used to combine two selectable marker genes, a selectable marker gene with a nonselectable gene, or two nonselectable genes (Table 3). There are several options for simultaneously expressing different biological functions from a single vector (Fig. 11). In general, type and positioning of transgenes, as well as cellular background and specific experimental conditions (especially the stringency of selection applied) greatly influence the efficacy of the coexpression strategy. Therefore, as with the inclusion of other cis elements, systematic comparative studies appear desirable for each coexpression vector developed for a specific clinical use [16,211]. a. Internal Promoters To express genes from retroviral vectors, promoters can be placed not only in the LTR but also, in either orientation, in the sequences between the leader and the 3 LTR (Fig. 11A). These internal promoters can be used in vectors where the U3 promoter has been deleted or in addition to an LTR-controlled transcription unit. However, when two promoters are located close to each other, there is the potential of promoter interference, leading to shutdown of one promoter to the advantage

TABLE 3 Reasons for Expressing Two or More Genes from a Single Vector Combination Selectable marker gene only.

Approach (see Section II) Drug resistance gene transfer

Example Complementary drug resistance genes (to widen spectrum of resistance). Drug resistance gene(s) plus suicide gene (to remove transduced cells in case of pathogenicity).

Selectable marker gene plus nonselectable gene. Nonselectable genes only.

Adoptive immunotherapy

Surface marker plus suicide gene (to select transduced cells before reinfusion).

Suicide gene transfer

Two suicide genes (improves efficacy).

Mini-organs

Suicide gene plus therapeutic gene of interest (to remove transduced cells in case of pathogenicity).

Oncogene antagonism

Suicide gene plus anti-oncogene (improves efficacy).

Oncogene antagonism

Complementary anti-oncogenes (improves efficacy).

Tumor vaccination

Cooperating immunostimulatory genes (improves efficacy).


21

FIGURE 11 Strategies for coexpression of two genes from a retroviral vector (see Section V.B.7). Open arrows indicate mRNAs, and the bold arrow represents the genomic message of the vector. , packaging signal; IP, internal promoter (orientation indicated by the filled arrow); S.D. and S.A., splice donor and splice acceptor, respectively; IRES, internal ribosomal entry site.

of its neighbor [212–214]. The stronger promoter (or the promoter selected for) either tends to exploit enhancer sequences of the neighboring promoter or inhibits formation of the Pol-II initiation complex at the internal promoter. Here, separation by transcriptional termination signals would be a possible solution [215]; however, this is inappropriate in retroviral vectors, as it would lead to premature termination of genomic messages in packaging cells. Placing the internal promoter in antisense orientation to the LTR might reduce interference at the transcriptional level, but doing so necessarily generates antisense RNA, which is expected to disturb translation of the cotransferred gene. Therefore, vectors containing internal promoters might generate unwanted effects, especially under conditions of dominant selection for only one promoter, as used in adoptive immunotherapy. b. Alternative Splicing For reasons not entirely understood, retroviral splice donor and splice acceptor sequences are only partly recognized in

host cells. This leads to a defined ratio of genomic and subgenomic messages and can be exploited for constructing splicing vectors that sometimes, but not always, yield good results [16,211,216]. Generating a spliced, subgenomic message can be associated with improved nuclear export, increased halflife of cytoplasmic RNA, or improved translation efficacy. Importantly, type and positioning of the transgenes will affect the efficacy of alternative splicing. Finally, cDNAs inserted in splice vectors must be free of cryptic splice signals (Fig. 11B). c. Internal Ribosome Entry The internal ribosomal entry site (IRES) was originally described in picorna viruses. The IRES is a complex domain of the RNA (the size of a few hundred base pairs), generating a specific structure allowing cap-independent initiation of translation. When introduced in front of the start codon of the transgene, bi- or even oligocistronic vectors can be generated [213,217–219]. Compared to internal promoters and alternative splicing, IRES control has the advantage of

22


exploiting a single mRNA for translation of two (or more) proteins (Fig. 11C). However, not every cDNA is fully compatible with translation via an IRES, and sometimes alternative and mutually exclusive rather than simultaneous initiation of translation might predominate. Early reports state that IRES-dependent initiation of translation occurs as efficiently as that from capped RNAs, with capped RNAs referring to transgenes expressed from within the gag region of vector mRNAs [213,217–219]. However, it was demonstrated that MLVs also use an IRES-related mechanism for translating gag–pol and env messages [63], and there is accumulating evidence that initiation of translation from within the gag region is suboptimal [44,83]. Accordingly, we and others found that IRES-dependent gene expression may be significantly reduced [211,220]. Also, virus titers may be suboptimal in the presence of a nonretroviral IRES, and even expression of the gene located 5 to the IRES may be compromised [16,211]. Moreover, it remains to be elucidated as to what extent IRESdependent translation may show reduced fidelity with respect to the choice of initiation codon and whether it is subject to differentiation-dependent control. d. Positioning in Untranslated Vector Regions Sometimes it is sufficient to express RNA without translation, as in approaches utilizing antisense RNA or ribozymes for oncogene antagonism (see Section II). These therapeutic RNAs can be located in untranslated vector regions, preferably in the 3 untranslated region of another gene coexpressed from the vector (Fig. 11D). The same strategy cannot be recommended for open reading frames: Spontaneous reinitiation of translation from the 3 untranslated region of a gene occurs at greatly reduced efficiency [68]. e. Fusion Proteins and Protein Cleavage Multifunctional fusion proteins are a good choice for coexpression provided that the domains of interest are active in similar subcellular localizations (Fig. 11E). Some cytosolic proteins might also function when expressed as the cytoplasmic tail of a membrane-anchored fusion protein [221]. It needs to be determined whether the efficacy of the fusion protein is comparable to those of the individual components. A potential risk of this approach is that the fusion site might give rise to an immunogenic peptide. An interesting extension of this approach is the inclusion of a cleavable linker between the protein domains of interest. The 2A proteinase of the foot and mouth disease virus (FMDV), a short peptide that has the interesting property of inducing cotranslational protein separation when inserted in the frame between two protein domains, can be successfully introduced in retroviral vectors [222]. Further interesting features of the FMDV 2A proteinase are that it does not disturb virus titers and it allows coexpression of two proteins that have different subcellular localizations [223]. Thus, vectors may be generated that express two or more proteins at coordinated levels.

8. cis Elements in cDNAs Even cDNAs can contain cis-acting elements, active either at the transcriptional [224] or posttranscriptional level [225–227]. This aspect of vector design is often neglected but can have profound influence on overall vector performance. A cDNA may harbor silencer elements or contain enhancers influencing levels as well as tissue-tropism of vector expression [224]. Some cDNAs may be unstable when expressed from retroviral vectors [225]. The retroviral life cycle implies that aberrant signals for splicing, termination, and polyadenylation; primer binding; or cryptic PP tracts will reduce vector titers or give rise to rearranged vector copies, with unpredictable immunological or toxicological consequences. Examples relevant to cancer gene therapy are the drug resistance genes MDR1 [16,226] and thymidine kinase of herpes simplex virus [227]. A stable, selectable marker gene coexpressed with the unstable sequence can serve as a tool to tag hot spots of recombination, providing the basis for cDNA improvement [16]. Thus, a lot of fine-tuning work may be required to develop stable and, hence, safe vectors suitable for actual clinical use. Evolution has done that work for retroviral genes. Vector designers usually follow empirical approaches, not always the most elegant and effective way to success.

VI. OUTLOOK Retroviral vector systems have dominated cancer gene therapy research in the past years, and they will certainly continue to play an important role. However, in future clinical trials it will be of outstanding importance to use specifically tailored and highly effective vectors. Only then can the perspectives of gene therapy concepts be evaluated. Based on a deeper understanding of the biology of retroviruses and their target cells, improved vector systems have already been created and now await clinical testing to assess efficacy and safety. Key developments include the advent of complex retrovirus-based systems for transduction of nondividing cells, pseudotyping and envelope engineering to widen or specify the host range at the level of transduction, and higher diversity in enhancer choice based on deeper insights into the transcriptional control of retroviral transgenes. Especially, further progress in the field of transductional and transcriptional targeting will have substantial impact on the therapeutic quality of cancer gene therapy approaches. So far, vector design has been dominated and also limited by deductive analyses of virus–host interactions. Future vector design should also follow a more evolutionary approach, taking advantage of the inherent genetic variability of viruses; therefore, we need to establish intelligent systems for selecting and screening improved mutants. For widely applicable oncologic strategies, tailoring can be performed as an international, multicenter effort. Unfortunately, for more


specialized applications with small patient numbers this will be unaffordable. Here, concentration in specific centers of expertise might represent a solution. Importantly, many aspects of vector tailoring worked out using simple retroviral vectors will also be applicable to lentiviral vectors and nonretroviral systems (based on adenovirus, adeno-associated viruses, herpes viruses, or physicochemical methods), which are emerging as important alternatives for some approaches in cancer gene therapy and will substantially widen the perspectives of the field.

References 1. Mulligan, R. C. (1993). The basic science of gene therapy. Science 260, 926–932. 2. Friedmann, T. (1996). Human gene therapy—an immature genie, but certainly out of the bottle. Nat. Med. 2, 144–145. 3. Dickman, S. (1997). Richard Mulligan: from skeptic to true believer. Curr. Biol. 7, R601–R602. 4. Brenner, M. K. (1994). Genetic marking and manipulation of hematopoietic progenitor cells using retroviral vectors. Immunomethods 5, 204–210. 5. Dunbar, C. E. (1996). Gene transfer to hematopoietic stem cells: implications for gene therapy of human disease. Annu. Rev. Med. 47, 11–20. 6. Rooney, C. M., Smith, C. A., Ng, C. Y., Loftin, S., Li, C., Krance, R. A., Brenner, M. K., and Heslop, H. E. (1995). Use of gene-modified virusspecific T lymphocytes to control Epstein–Barr-virus-related lymphoproliferation. Lancet 345, 9–13. 7. Pawliuk, R., Eaves, C. J., and Humphries, K. R. (1997). Sustained high-level reconstitution of the hematopoietic system by preselected hematopoietic cells expressing a transduced cell-surface antigen. Hum. Gene Ther. 8, 1595–1604. 8. Phillips, K., Gentry, T., McCowage, G., Gilboa, E., and Smith, C. (1996). Cell-surface markers for assessing gene transfer into human hematopoietic cells. Nat. Med. 2, 1154–1156. 9. Fehse, B., Uhde, A., Fehse, N. Eckert, H. G., Clausen, J., Rüger, R., Koch, S., Ostertag, W., Zander, A. R., and Stockschläder, M. (1997). Selective immunoaffinity-based enrichment of CD34+ cells transduced with retroviral vectors containing an intracytoplasmatically truncated version of the human low-affinity nerve growth factor receptor (LNGFR) gene. Hum. Gene Ther. 8, 1815–1827. 10. Gottesman, M. M., Germann, U. A., Aksentijevich, I., Sugimoto, Y., Cardarelli, C. O., and Pastan, I. (1994). Gene transfer of drug resistance genes. Implications for cancer therapy. Ann. N.Y. Acad. Sci. 716, 126–138. 11. Baum, C., Fairbairn, L., Hildinger, M., Lashford, L. S., HegewischBecker, S., and Rafferty, J. R. (1999). New perspectives for cancer chemotherapy by genetic protection of hematopoietic cells. Expert reviews in molecular medicine. http://www-ermm.cbcu.cam.ac.uk. 12. Baum, C. (1997). Gene transfer and transgene expression in hematopoietic cells. In: Concepts in Gene Therapy (M. Strauss, and J. A. Barranger, eds.), pp. 233–265. DeGruyter, Berlin. 13. Baum, C., Hegewisch-Becker, S., Eckert, H.-G., Stocking, C., and Ostertag, W. (1995). Novel retroviral vectors for efficient expression of the multidrug resistance (mdr-1) gene in early hematopoietic cells. J. Virol. 69, 7541–7547. 14. Eckert, H.-G., Stockschläder, M., Just, U., Hegewisch-Becker, S., Grez, M., Uhde, A., Zander, A., Ostertag, W., and Baum, C. (1996). Highdose multidrug resistance in primary human hematopoietic progenitor cells transduced with optimized retroviral vectors. Blood 88, 3407– 3415.

23

15. Galipeau, J., Benaim, E., Spencer, H. T., Blakley, R., and Sorrentino, B. P. (1997). A bicistronic retroviral vector for protecting hematopoietic cells against antifolates and P-glycoprotein effluxed drugs. Hum. Gene Ther. 8, 1773–1783. 16. Hildinger, M., Fehse, B., Hegewisch-Becker, S., John, J., Rafferty, J. R., Ostertag, W., and Baum, C. (1998). Dominant selection of hematopoietic progenitor cells with retroviral MDR1 co-expression vectors. Hum. Gene Ther. 9, 33–42. 17. Sauerbrey, A., McPherson, J. P., Zhao, S. C., Banerjee, D., and Bertino, J. R. (1999). Expression of a novel double-mutant dihydrofolate reductase-cytidine deaminase fusion gene confers resistance to both methotrexate and cytosine arabinoside. Hum. Gene Ther. 10, 2495– 2504. 18. Jelinek, J., Rafferty, J. A., Cmejla, R., Hildinger, M., Chinnasamy, D., Lashford, L. S., Ostertag, W., Margison, G. P., Dexter, T. M., Fairbairn, L. J., and Baum, C. (1999). A novel dual function retrovirus expressing multidrug resistance 1 and O 6 -alkylguanine-DNA-alkyltransferase for engineering resistance of haemopoietic progenitor cells to multiple chemotherapeutic agents. Gene Ther. 6, 1489–1493. 19. Bonini, C. et al. (1997). HSV-TK gene transfer into donor lymphocytes for control of allogeneic graft-versus-leukemia. Science 176, 1719– 1724. 20. Clay, T. M., Custer, M. C., Spiess, P. J., and Nishimura, M. I. (1999). Potential use of T cell receptor genes to modify hematopoietic stem cells for the gene therapy of cancer. Pathol. Oncol. Res. 5, 3–15. 21. Bohl, D., and Heard, J.-M. (1997). In vivo secretion of therapeutic proteins from neo-organs. In: Concepts in Gene Therapy (M. Strauss and J. A. Barranger, eds.), pp. 297–314. De Gruyter, Berlin. 22. Bohl, D., and Heard, J. M. (1997). Modulation of erythropoietin delivery from engineered muscles in mice. Hum. Gene Ther. 8, 195–204. 23. Greco, O., and Dachs, G.U. (2001). Gene directed enzyme/prodrug therapy of cancer: historical appraisal and future prospectives. J. Cell. Physiol. 187, 22–36. 24. Harris, J. D., Gutierrez, A. A., Hurst, H. C., Sikora, K., and Lemoine, N. R. (1994). Gene therapy for cancer using tumour-specific prodrug activation. Gene Ther. 1, 170–175. 25. Martin, V., Cortes, M. L., de Felipe, P., Farsetti, A., Calcaterra, N. B., and Izquierdo, M. (2000). Cancer gene therapy by thyroid hormonemediated expression of toxin genes. Cancer Res. 60, 3218–3224. 26. Vallera, D. A., Jin, N., Baldrica, J. M., Panoskaltsis-Mortari, A., Chen, S. Y., and Blazar, B. R. (2000). Retroviral immunotoxin gene therapy of acute myelogenous leukemia in mice using cytotoxic T cells transduced with an interleukin 4/diphtheria toxin gene. Cancer Res. 60, 976–984. 27. Roth, J. A., Nquyen, D., Lawrence, D. D., Kemp, B. L., Carrasco, C. H. et al. (1996). Retrovirus-mediated wild-type p53 gene transfer to tumors of patients with lung cancer. Nat. Med. 2, 985–991. 28. Yang, Z. Y., Perkins, N. D., Ohno, T., Nabel, E. G., and Nabel, G. J. (1995). The p21 cyclin-dependent kinase inhibitor suppresses tumorigenicity in vivo. Nat. Med. 1, 1052–1056. 29. Wygoda, M. R., Wilson, M. R., Davis, M. A., Trosko, J. E., Rehemtulla, A., and Lawrence, T. S. (1997). Protection of herpes simplex virus thymidine kinase-transduced cells from ganciclovir-mediated cytotoxicity by bystander cells: the Good Samaritan effect. Cancer Res. 57, 1699–1703. 30. Benedetti, S., Pirola, B., Pollo, B., Magrassi, L., Bruzzone, M. G., Rigamonti, D., Galli, R., Selleri, S., Di Meco, F., De Fraja, C., Vescovi, A., Cattaneo, E., and Finocchiaro, G. (2000). Gene therapy of experimental brain tumors using neural progenitor cells. Nat. Med. 6, 447– 450. 31. Tamura, M., Ikenaka, K., Tamura, K., Yoshimatsu, T., Miyao, Y., Kishima, H., Mabuchi, E., and Shimizu, K. (1998). Transduction of glioma cells using a high-titer retroviral vector system and their subsequent migration in brain tumors. Gene Ther. 5, 1698–1704. 32. Gomez-Navarro, J., Contreras, J. L., Arafat, W., Jiang, X. L., Krisky, D., Oligino, T., Marconi, P., Hubbard, B., Glorioso, J. C., Curiel, D. T., and

24

33.

34.

35. 36.

37.

38.

39.

40.

41.

42.

43.

44.

45. 46.

47.

48. 49. 50.

51.

52.

53.

Christopher Baum et al. Thomas, J. M. (2000). Genetically modified CD34+ cells as cellular vehicles for gene delivery into areas of angiogenesis in a rhesus model. Gene Ther. 7, 43–52. Hurford, R. J., Dranoff, G., Mulligan, R. C., and Tepper, R. I. (1995). Gene therapy of metastatic cancer by in vivo retroviral gene targeting. Nat. Genet. 10, 430–435. Bischoff, J. R., Kirn, D. H., Willimas, A., Heise, C., Horn, S., Muna, M., Ng, L., Sampson-Johannes, A., Fattaey, A., and McCormick, F. (1996). An adenovirus mutant that replicates selectively in p53-deficient human tumor cells. Science 274, 373–376. Blankenstein, T., Cayeux, S., and Qin, Z. (1996). Genetic approaches to cancer immunotherapy. Rev. Physiol. Biochem. Pharmacol. 129, 1–49. Simons, J. W., and Mikhak, B. (1998). Ex-vivo gene therapy using cytokine-transduced tumor vaccines: molecular and clinical pharmacology. Semin. Oncol. 25, 661–676. Coffin, J. M. (1996). Retroviridae: the viruses and their replication. In: Field’s Virology (B. N. Fields, D. M. Knipe, and P. M. Howley, eds.), pp. 763–844. Lippincott-Raven, Philadelphia. Miller, A. D., and Chen, F. (1996). Retrovirus packaging cells based on 10A1 murine leukemia virus for production of vectors that use multiple receptors for cell entry. J. Virol. 70, 5564–5571. Mergia, A., Shaw, K. E., Lowe, E., Barry, P. A., and Luciw, P. A. (1990). Simian foamy virus type 1 is a retrovirus which encodes a transcriptional transactivator. J. Virol. 64, 3598–3604. Luciw, P. A. (1996). Human immunodeficiency viruses and their replication. In: Field’s Virology (B. N. Fields, D. M. Knipe, and P. M. Howley, eds.), pp. 1881–1975. Lippincott-Raven, Philadelphia. Ashe, M. P., Furger, A., and Proudfoot, N. J. (2000). Stem-loop 1 of the U1 snRNP plays a critical role in the suppression of HIV-1 polyadenylation. RNA 6, 170–177. Bender, M. A., Palmer, T. D., Gelinas, R. E., and Miller, A. D. (1987). Evidence that the packaging signal of Moloney murine leukemia virus extends into the gag region. J. Virol. 61, 1639–1646. Kim, S. H., Yu, S. S., Park, J. S., Robbins, P. D., An, C. S., and Kim, S. (1998). Construction of retroviral vectors with improved safety, gene expression, and versatility. J. Virol. 72, 994–1004. Hildinger, M., Abel, K. L., Ostertag, W., and Baum, C. (1999) Design of 5 untranslated sequences in retroviral vectors developed for medical use. J. Virol. 73, 4083–4089. Berkowitz, R., Fisher, J., and Goff, S. P. (1996). RNA packaging. Curr. Top. Microbiol. Immunol. 214, 177–218. Berkowitz, R., Hammarskjold, M.-L., Helga-Maria, C., Rekosh, D., and Goff, S. (1995). 5 regions of HIV-1 RNAs are not sufficient for encapsidation: implications for the HIV-1 packaging signal. Virology 212, 718–723. Richardson, J. H., Child, L. A., and Lever, A. M. (1993). Packaging of human immunodeficiency virus type 1 RNA requires cis-acting sequences outside the 5 leader region. J. Virol. 67, 3997–4005. Weiss, R. A. (1993). Pseudotyped viruses and envelope composition. In: The Retroviridae (J. A. Levy, ed.), pp. 5–8. Plenum Press, New York. Hunter, E., and Swanstrom, R. (1990). Retrovirus envelope glycoproteins. Curr. Top. Microbiol. Immunol. 157, 187–253. Temin, H. M. (1993). Retrovirus variation and reverse transcription: abnormal strand transfers result in retrovirus genetic variation. Proc. Natl. Acad. Sci. USA 90, 6900–6903. Roe, T., Reynolds, T., Yu, G., and Brown, P. O. (1993). Integration of murine leukemia virus DNA depends on mitosis. EMBO J. 12, 2099– 2108. Miller, D. G., Adam, M. A., and Miller, A. D. (1990). Gene transfer by retrovirus vectors occurs only in cells that are actively replicating at the time of infection. Mol. Cell. Biol. 10, 4239–4242. Lieber, A., Kay, M. A., and Li, Z. Y. (2000). Nuclear import of Moloney murine leukemia virus DNA mediated by adenovirus preterminal protein is not sufficient for efficient retroviral transduction in nondividing cells. J. Virol. 74, 721–734.

54. Bukrinsky, M. I., Haggerty, S., Dempsey, P., Sharova, N., Adzhubei, A., Spitz, L., Lewis, P., Goldfarb, D., Emerman, M., and Stevenson, M. (1993). A nuclear localization signal within HIV-1 matrix protein that governs infection of non-dividing cells. Nature 365, 666–669. 55. Fouchier, R. A., and Malim, M. H. (1999). Nuclear import of human immunodeficiency virus type-1 preintegration complexes. Adv. Virus Res. 52, 275–299. 56. Zennou, V., Petit, C., Guetard, D., Nerhbass, U., Montagnier, L., and Charneau, P. (2000) HIV-1 genome nuclear import is mediated by a central DNA flap. Cell 101, 173–185. 57. Rohdewohld, H., Weiher, H., Reik, W., Jaenisch, R., and Breindl, M. (1987). Retrovirus integration and chromatin structure: Moloney murine leukemia proviral integration sites map near DNase Ihypersensitive sites. J. Virol. 61, 336–343. 58. Jonkers, J., and Berns, A. (1996). Retroviral insertional mutagenesis as a strategy to identify cancer genes. Biochim. Biophys. Acta 1287, 29–57. 59. Donahue, R. E., Kessler, S. W., Bodine, D., McDonagh, K., Dunbar, C., Goodman, S., Agricola, B., Byrne, E., Raffeld, M., Moen, R. et al. (1992). Helper virus induced T cell lymphoma in nonhuman primates after retroviral mediated gene transfer. J. Exp. Med. 176, 1125– 1135. 60. Stocking, C., Bergholz, U., Friel, J., Klingler, K., Wagener, T., Starke, C., Kitamura, T., Miyajima, A., and Ostertag, W. (1993). Distinct classes of factor-independent mutants can be isolated after retroviral mutagenesis of a human myeloid stem cell line. Growth Factors 8, 197–209. 61. Tsichlis, P. N., and Lazo, P. A. (1991). Virus-host interactions and the pathogenesis of murine and human oncogenic retroviruses. In: Retroviral Insertion and Oncogene Activation (H. G. Kung and P. K. Vogt, eds.), pp. 95–173. Springer-Verlag, Berlin. 62. Ostertag, W., Stocking, C., Johnson, G. R., Kluge, N., Kollek, R., Franz, T., and Hess, N. (1987). Transforming genes and target cells of murine spleen focus-forming viruses. Adv. Cancer Res. 48, 193–355. 63. Deffaud, C., and Darlix, J. L. (2000). Characterization of an internal ribosomal entry segment in the 5 leader of murine leukemia virus env RNA. J. Virol. 74, 846–850. 64. Luciw, P. A., and Leung, N. J. (1992). Mechanisms of retrovirus replication. In: The Retroviridae 1 (J. A. Levy, ed.), pp. 159–298. Plenum Press, New York. 65. Kräusslich, H.-G., and Welker, R. (1996). Intracellular transport of capsid components. In: Morphogenesis and Maturation of Retroviruses (H. G. Kräusslich, ed.), pp. 25–64. Springer-Verlag, Berlin. 66. Einfeld, D. (1996). Maturation and assembly of retroviral glycoproteins. In: Morphogenesis and Maturation of Retroviruses (H. G. Kräusslich, ed.), pp. 133–176. Springer-Verlag, Berlin. 67. Miller, A. D. (1990). Retrovirus packaging cells. Hum. Gene Ther. 1, 5–14. 68. Cosset, F.-L., Takeuchi, Y., Battini, J.-L., Weiss, R. A., and Collins, M. K. L. (1995). High-titer packaging cells producing recombinant retrovirus resistant to human serum. J. Virol. 69, 7430–7436. 69. Wagener, T., Stocking, C., and Ostertag, W. (1995). unpublished data. 70. Hatzoglou, M., Hodgson, C. P., Mularo, F., and Hanson, R. W. (1990). Efficient packaging of a specific VL30 retroelement by psi 2 cells which produce MoMLV recombinant retroviruses. Hum. Gene Ther. 1, 385–397. 71. Chong, H., and Vile, R. G. (1996). Replication-competent retrovirus produced by a ‘split-function’ third generation amphotropic packaging cell line. Gene Ther. 3, 624–629. 72. Vanin, E. F., Kaloss, M., Broscius, C., and Nienhuis, A. W. (1994). Characterization of replication-competent retroviruses from nonhuman primates with virus-induced T-cell lymphomas and observations regarding the mechanism of oncogenesis. J. Virol. 68, 4241–4250. 73. Münk, C., Lohler, J., Prassolov, V., Just, U., Stockschlader, M., and Stocking, C. (1997). Amphotropic murine leukemia viruses induce


74. 75.

76.

77.

78.

79.

80.

81.

82. 83.

84.

85.

86.

87.

88.

89.

90.

91.

spongiform encephalomyelopathy. Proc. Natl. Acad. Sci. USA 94, 5837–5842. Isacson, O., and Brakefield, X. O. (1997). Benefits and risks of hosting animal cells in the human brain. Nat. Med. 3, 964–969. Takeuchi, Y., Cosset, F. L., Lachmann, P. J., Okada, H., Weiss, R. A., and Collins, M. K. (1994). Type C retrovirus inactivation by human complement is determined by both the viral genome and the producer cell. J. Virol. 68, 8001–8007. Takeuchi, Y., Porter, C. D., Strahan, K. M., Preece, A. F., Gustafsson, K., Cosset, F. L., Weiss, R. A., and Collins, M. K. (1996). Sensitization of cells and retroviruses to human serum by (alpha 1–3) galactosyltransferase. Nature 379, 85–88. Palsson, B., and Andreadis, S. (1997). The physico-chemical factors that govern retrovirus-mediated gene transfer. Exp. Hematol. 25, 94– 102. Kinsella, T. M., and Nolan, G. P. (1996). Epsiomal vectors rapidly and stably produce high-titer recombinant retrovirus. Hum. Gene Ther. 7, 1405–1413. Grignani, F., Kinsella, T., Mencarelli, A., Valtieri, M., Riganelli, D., Grignani, F., Lanfrancone, L., Peschle, C., Nolan, G. P., and Pelicci, P. G. (1998). High-efficiency gene transfer and selection of human hematopoietic progenitor cells with a hybrid EBV/retroviral vector expressing the green fluorescence protein. Cancer Res. 58, 14–19. Wilson, C. A., Ng, T. H., and Miller, A. E. (1997). Evaluation of recommendations for replication-competent retrovirus testing associated with use of retroviral vectors. Hum. Gene Ther. 8, 869–874. Correll, P. H., Colilla, S., and Karlsson, S. (1994). Retroviral vector design for long-term expression in murine hematopoietic cells in vivo. Blood 84, 1812–1822. Miller, A. D., and Rosman, G. J. (1989). Improved retroviral vectors for gene transfer and expression. Biotechniques 7, 980–982. Krall, W. J., Skelton, D. C., Yu, X.-J., Riviere, I., Lehn, P., Mulligan, R. C., and Kohn, D. B. (1996). Increased levels of spliced RNA account for augmented expression from the MFG retroviral vector in hematopoietic cells. Gene Ther. 3, 37–48. Hantzopulos, P. A., Sullenger, B. A., Ungers, G., and Gilboa, E. (1989). Improved gene expression upon transfer of the adenosine deaminase minigene outside the transcriptional unit of a retroviral vector. Proc. Natl. Acad. Sci. U.S.A. 86, 3519–3523. Adam, M. A., Osborne, W. R., and Miller, A. D. (1995). R-region cDNA inserts in retroviral vectors are compatible with virus replication and high-level protein synthesis from the insert. Hum. Gene Ther. 6, 1169–1176. Yu, S. F., von Ruden, T., Kantoff, P. W., Garber, C., Seiberg, M., Ruther, U., Anderson, W. F., Wagner, E. F., and Gilboa, E. (1986). Selfinactivating retroviral vectors designed for transfer of whole genes into mammalian cells. Proc. Natl. Acad. Sci. USA 83, 3194–3198. Olson, P., Nelson, S., and Dornburg, R. (1994). Improved selfinactivating retroviral vectors derived from spleen necrosis virus. J. Virol. 68, 7060–7066. Bergemann, J., Kuhlcke, K., Fehse, B., Ratz, I., Ostertag, W., and Lother, H. (1995). Excision of specific DNA-sequences from integrated retroviral vectors via site-specific recombination. Nucleic Acids Res. 23, 4451–4456. Russ, A. P., Friedel, C., Grez, M., and von Melchner, H. (1996). Self-deleting retrovirus vectors for gene therapy. J. Virol. 70, 4927– 4932. Beck-Engeser, G., Stocking, C., Just, U., Albritton, L., Dexter, M., Spooncer, E., and Ostertag, W. (1991). Retroviral vectors related to the myeloproliferative sarcoma virus allow efficient expression in hematopoietic stem and precursor cell lines, but retroviral infection is reduced in more primitive cells. Hum. Gene Ther. 2, 61–70. Crooks, G. M., and Kohn, D. B. (1993). Growth factors increase amphotropic retrovirus binding to human CD34+ bone marrow progenitor cells. Blood 82, 3290–3297.

25

92. Orlic, D., Girard, L. J., Jordan, C. T., Anderson, S. M., Cline, A. P., and Bodine, D. M. (1996). The level of mRNA encoding the amphotropic retrovirus receptor in mouse and human hematopoietic stem cells is low and correlates with the efficiency of retrovirus transduction. Proc. Natl. Acad. Sci. USA 93, 11097–11102. 93. von Laer, D., Thomsen, S., Vogt, B., Donath, M., Kruppa, J., Rein, A., Ostertag, W., and Stocking, C. (1998). Entry of amphotropic and 10A1 pseudotyped murine retroviruses is restricted in hematopoietic stem cell lines. J. Virol. 72, 1424–1430. 94. Münk, C., Thomsen, S., Stocking, C., and Lohler, J. (1998). Murine leukemia virus recombinants that use phosphate transporters for cell entry induce similar spongiform encephalomyelopathies in newborn mice. Virology 252, 318–323. 95. Rubin, H. (1965). Genetic control and cellular susceptibility to pseudotypes of Rous sarcoma virus. Virology 26, 270–282. 96. Friedmann, T., and Yee, J.-K. (1995). Pseudotyped retroviral vectors for studies of human gene therapy. Nat. Med. 1, 275–277. 97. Januszeski, M. M., Cannon, P. M., Chen, D., Rozenberg, Y., and Anderson, W. F. (1997). Functional analysis of the cytoplasmic tail of Moloney murine leukemia virus envelope protein. J. Virol. 71, 3613– 3619. 98. Suomalainen, M., and Garoff, H. (1994). Incorporation of homologous and heterologous proteins into the envelope of Moloney murine leukemia virus. J. Virol. 68, 4879–4889. 99. Emi, N., Friedmann, T., and Yee, J.-K. (1991). Pseudotype formation of murine leukemia virus with the G protein of vesicular stomatitis virus. J. Virol. 65, 1202–1207. 100. Conti, C., Mastromarino, P., and Orsi, P. (1991). Role of membrane phospholipids and glycolipids in cell-to-cell fusion of VSV. Comp. Immun. Microbiol. Infect. Dis. 14, 303–313. 101. Burns, J. C., Friedmann, T., Driever, W., Burrascano, M., and Yee, J.-K. (1993). Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: concentration to very high titer and efficient gene transfer into mammalian and nonmammalian cells. Proc. Natl. Acad. Sci. USA 90, 8033–8037. 102. Yang, Y. P., Vanin, E. F., Whitt, M. A., Fornerod, M., Zwart, R., Schneiderman, R. D., Grosveld, G., and Nienhuis, A. W. (1995). Inducible, high-level production of infectious murine leukemia retroviral vector particles pseudotyped with vesicular stomatitis virus G envelope protein. Hum. Gene Ther. 6, 1203–1213. 103. Akkina, R. K., Walton, R. M., Chen, M. L., Li, Q.-X., Planelles, V., and Chen, I. S. Y. (1996). High-efficiency gene transfer into CD34+ cells with a human immunodeficiency virus type 1-based retroviral vector pseudotyped with vesicular stomatitis virus envelope glycoprotein G. J. Virol. 70, 2581–2585. 104. von Kalle, C., Kiem, H.-P., Goehle, S., Darovsky, B., Heimfeld, S., Torok-Storb, B., Storb, R., and Schuening, F. G. (1994). Increased gene transfer into human hematopoietic progenitor cells by extended in vitro exposure to a pseudotyped retroviral vector. Blood 84, 2890–2897. 105. Bunnell, B. A., Mesler Muul, L., Donahue, R. E., Blaese, R. M., and Morgan, R. A. (1995). High-efficiency retroviral-mediated gene transfer into human and nonhuman primate peripheral blood lymphocytes. Proc. Natl. Acad. Sci. USA 92, 7739–7743. 106. Uckert, W., Becker, C., Gladow, M., Klein, D., Kammertoens, T., Pedersen, L., and Blankenstein, T. (2000). Efficient gene transfer into primary human CD8+ T lymphocytes by MuLV-10A1 retrovirus pseudotype. Hum. Gene Ther. 11, 1005–1014. 107. Barrette, S., Douglas, J., Orlic, D., Anderson, S. M., Seidel, N. E., Miller, A. D., and Bodine, D. M. (2000). Superior transduction of mouse hematopoietic stem cells with 10A1 and VSV-G pseudotyped retrovirus vectors. Molec. Ther. 1, 330–338. 108. Lindemann, D., Bock, M., Schweizer, M., and Rethwilm, A. (1997). Efficient pseudotyping of murine leukemia virus particles with chimeric human foamy virus envelope proteins. J. Virol. 71, 4815– 4820.

26


109. Wilson, C., Reitz, M. S., Okayama, H., and Eiden, M. V. (1989). Formation of infectious hybrid virions with gibbon ape leukemia virus and human T-cell leukemia virus retroviral envelope glycoproteins and the Gag and Pol proteins of Moloney murine leukemia virus. J. Virol. 63, 2374–2378. 110. Mammano, F., Salvatori, F., Indraccolo, S., De Rossi, A., ChiecoBianchi, L., and Göttlinger, H. G. (1997). Truncation of the human immunodeficiency virus type 1 envelope glycoprotein allows efficient pseudotyping of Moloney murine leukemia virus particles and gene transfer into CD4+ cells. J. Virol. 71, 3341–3345. 111. Loiler, S. A., DiFronzo, N. L., and Holland, C. A. (1997). Gene transfer to human cells using retrovirus vectors produced by a new polytropic packaging cell line. J. Virol. 71, 4825–4828. 112. von Laer, D., Lindemann, D., Roscher, S., Herwig, U., Friel, J., and Herchenröder, O. (2001). Low level expression of functional foamy virus receptor on hematopoietic progenitor cells, Virology 288, 139– 144. 113. Miletic, H., Bruns, M., Tsiakas, K., Vogt, B., Rezai, R., Baum, C., Kuhlke, K., Cosset, F. L., Ostertag, W., Lother, H., and von Laer, D. (1999). Retroviral vectors pseudotyped with lymphocytic choriomeningitis virus. J. Virol. 73, 6114–6116. 114. Hanenberg, H., Xiao, L. X., Dilloo, D., Hashino, K., Kato, I., and Williams, D. A. (1996). Colocalization of retrovirus and target cells on specific fibronactin fragments increases genetic transduction of mammalian cells. Nat. Med. 2, 876–882. 115. Moritz, T., Patel, V. P., and Williams, D. A. (1994). Bone marrow extracellular matrix molecules improve gene transfer into human hematopoietic cells via retroviral vectors. J. Clin. Invest. 93, 1451– 1457. 116. Cosset, F. L., and Russell, S. J. (1996). Targeting retrovirus entry. Gene Ther. 3, 946–956. 117. Russell, S. J., and Cosset, F. L. (1999). Modifying the host range properties of retroviral vectors. J. Gene Med. 1, 300–311. 118. Etienne-Julan, M., Roux, P., Carillo, S., Jeanteur, P., and Piechaczyk, M. (1992). The efficiency of cell targeting by recombinant retroviruses depends on the nature of the receptor and the composition of the artificial cell-virus linker. J. Gen. Virol. 73, 3251–3255. 119. Kasahara, N., Dozy, A. M., and Kan, Y. W. (1994). Tissue-specific targeting of retroviral vectors through ligand-receptor interactions. Science 266, 1373–1376. 120. Valsesia, W. S., Drynda, A., Deleage, G., Aumailley, M., Heard, J. M., Danos, O., Verdier, G., and Cosset, F. L. (1994). Modifications in the binding domain of avian retrovirus envelope protein to redirect the host range of retroviral vectors. J. Virol. 68, 4609–4619. 121. Xiaoliang, H., Kasahara, N., and Wai Kan, Y. (1995). Ligand-directed retroviral targeting of human breast cancer cells. Proc. Natl. Acad. Sci. USA 92, 9747–9751. 122. Chu, T. H., and Dornburg, R. (1995). Retroviral vector particles displaying the antigen-binding site of an antibody enable cell-typespecific gene transfer. J. Virol. 69, 2659–2663. 123. Jiang, A., and Dornburg, R. (1999). In vivo cell type-specific gene delivery with retroviral vectors that display single chain antibodies. Gene Ther. 6, 1982–1987. 124. Zhao, Y., Zhu, L., Lee, S., Li, L., Chang, E., Soong, N. W., Douer, D., and Anderson, W. F. (1999). Identification of the block in targeted retroviral-mediated gene transfer. Proc. Natl. Acad. Sci. USA 96, 4005– 4010. 125. Hatziioannou, T., Valsesia-Wittmann, S., Russell, S. J., and Cosset, F. L. (1998). Incorporation of fowl plague virus hemagglutinin into murine leukemia virus particles and analysis of the infectivity of the pseudotyped retroviruses. J. Virol. 72, 5313–5317. 126. Chu, T. H., and Dornburg, R. (1997). Toward highly efficient cell-typespecific gene transfer with retroviral vectors displaying single-chain antibodies. J. Virol. 71, 720–725.

127. Nilson, B. H., Morling, F. J., Cosset, F. L., and Russell, S. J. (1996). Targeting of retroviral vectors through protease-substrate interactions. Gene Ther. 3, 280–286. 128. Somia, N. V., Zoppe, M., and Verma, I. M. (1995). Generation of targeted retroviral vectors by using single-chain variable fragment: an approach to in vivo gene delivery. Proc. Natl. Acad. Sci. USA 92, 7570– 7574. 129. Marin, M., Noel, D., Valsesia, W. S., Brockly, F., Etienne, J. M., Russell, S., Cosset, F. L., and Piechaczyk, M. (1996). Targeted infection of human cells via major histocompatibility complex class I molecules by Moloney murine leukemia virus-derived viruses displaying singlechain antibody fragment-envelope fusion proteins. J. Virol. 70, 2957– 2962. 130. Russell, S. J., Hawkins, R. E., and Winter, G. (1993). Retroviral vectors displaying functional antibody fragments. Nucleic Acids Res. 21, 1081– 1085. 131. Valsesia-Wittmann, S., Morling, F. J., Hatziioannou, T., Russell, S. J., and Cosset, F. L. (1997). Receptor co-operation in retrovirus entry: recruitment of an auxiliary entry mechanism after retargeted binding. EMBO J. 16, 1214–1223. 132. Naldini, L., Blömer, U., Gallay, P., Ory, D., Mulligan, R., Gage, F. H., Verma, I. M., and Trono, D. (1996). In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 272, 263–267. 133. Naldini, L., Blömer, U., Gage, F. H., Trono, D., and Verma, I. M. (1996). Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc. Natl. Acad. Sci. USA 93, 1382–1388. 134. Reiser, J., Harmison, G., Kluepfel-Stahl, S., Brady, R. O., Karlsson, S., and Schubert, M. (1996). Transduction of nondividing cells using pseudotyped defective high-titer HIV type particles. Proc. Natl. Acad. Sci. USA 93, 15266–15271. 135. Blömer, U., Naldini, L., Kafri, T., Trono, D., Verma, I. M., and Gage, F. H. (1997). Highly efficient and sustained gene transfer in adult neurons with a lentivirus vector. J. Virol. 71, 6641–6649. 136. Miyoshi, H., Smith, K. A., Mosier, D. E., Verma, I. M., and Torbett, B. E. (1999). Transduction of human CD34+ cells that mediate longterm engraftment of NOD/SCID mice by HIV vectors. Science 283, 682–686. 137. Carroll, R., Lin, J.-T., Dacquel, E. J., Mosca, J. D., Burke, D. S., and St. Louis, D. C. (1994). A human immunodeficiency virus type 1 (HIV1)-based retroviral vector system utilizing stable HIV-1 packaging cell lines. J. Virol. 68, 6047–6051. 138. Richardson, J. H., Kaye, J. F., Child, L. A., and Lever, A. M. L. (1995). Helper virus-free transfer of human immunodeficiency virus type 1 vectors. J. Gen. Virol. 76, 691–696. 139. Konvalinka, J., Litterst, M. A., Welker, R., Kottler, H., Rippmann, F., Heuser, A. M., and Krausslich, H. G. (1995). An active-site mutation in the human immunodeficiency virus type 1 proteinase (PR) causes reduced PR activity and loss of PR-mediated cytotoxicity without apparent effect on virus maturation and infectivity. J. Virol. 69, 7180–7186. 140. Corbeau, P., Kraus, G., and Wong-Staal, F. (1996). Efficient gene transfer by a human immunodeficiency virus type 1 (HIV-1)-derived vector using a stable HIV packaging cell line. Proc. Natl. Acad. Sci. USA 93, 14070–14075. 141. Kafri, T., van Praaag, H., Ouyang, L., Gage, F. H., and Verma, I. M. (1999). A packaging cell line for lentivirus vectors. J. Virol. 73, 576– 584. 142. Dull, T., Zufferey, R., Kelly, M., Mandel, R. J., Nguyen, M., Trono, D., and Naldini, L. (1998). A third-generation lentivirus vector with a conditional packaging system. J. Virol. 72, 8463–8471. 143. Romano, G., Michell, P., Pacilio, C., and Giordano, A. (2000). Latest developments in gene transfer technology: achievements, perspectives,


144.

145. 146.

147.

148.

149. 150.

151.

152. 153. 154. 155.

156.

157.

158.

159.

160.

161.

162.

and controversies over therapeutic applications. Stem Cells 18, 19–39. Poeschla, E. M., Wong-Staal, F., and Looney, D. J. (1998). Efficient transduction of nondividing human cells by feline immunodeficiency virus lentiviral vectors. Nat. Med. 4, 354–357. Olsen, J. C. (1998). Gene transfer vectors derived from equine infectious anemia virus. Gene Ther. 5, 1481–1487. Schnell, T., Foley, P., Wirth, M., Munch, J., and Uberla, K. (2000). Development of a self-inactivating, minimal lentivirus vector based on simian immunodeficiency virus. Hum. Gene Ther. 11, 439– 447. Metharom, P., Takyar, S., Xia, H. H., Ellem, K. A., Macmillan, J., Shepherd, R. W., Wilcox, G. E., and Wei, M. Q. (2000). Novel bovine lentiviral vectors based on Jembrana disease virus. J. Gene Med. 2, 176–185. Schweizer, M., Turek, R., Hahn, H., Schliephake, A., Netzer, K.-O., Eder, G., Reinhardt, M., Rethwilm, A., and Neumann-Haefelin, D. (1995). Markers of foamy virus infections in monkeys, apes and accidentally infected humans: appropriate testing fails to confirm suspected foamy prevalence in humans. AIDS Res. Hum. Retroviruses 11, 161– 170. Hooks, J. J., and Gibbs, C. J. J. (1995). The foamy viruses. Bacteriolo. Revi. 39, 169–185. Mikovits, J. A., Hoffman, P. M., Rethwilm, A., and Ruscetti, F. W. (1996). In vitro infection of primary and retrovirus-infected human leukocytes by human foamy virus. J. Virol. 70, 2774–2780. Pietschmann, T., Heinkelein, M., Heldmann, M., Zentgraf, H., Rethwilm, A., and Lindemann, D. (1999). Foamy virus capsids require the cognate envelope protein for particle export. J. Virol. 73, 2613–2621. Bieniasz, P. D., Weiss, R. A., and McClure, M. O. (1995). Cell cycle dependence of foamy retrovirus infection. J. Virol. 69, 7295–7299. Russell, D. W., and Miller, A. D. (1996). Foamy virus vectors. J. Virol. 70, 217–222. Trobridge, G. D., and Russell, D. W. (1998). Helper-free foamy virus vectors. Hum. Gene Ther. 9, 2517–2525. Heinkelein, M., Thurow, J., Dressler, M., Imrich, H., NeumannHaefelin, D., McClure, M. O., and Rethwilm, A. (2000). Complex effects of deletions in the 5 untranslated region of primate foamy virus on viral gene expression and RNA packaging. J. Virol. 74, 3141– 3148. Sandmeyer, S. B., Hansen, L. J., and Chalker, D. L. (1990). Integration specificity of retrotransposons and retroviruses. Annu. Rev. Genet. 24, 491–518. Withers-Ward, E. S., Kitamura, Y., Barnes, J. P., and Coffin, J. M. (1994). Distribution of targets for avian retrovirus DNA integration in vivo. Genes Dev. 8, 1473–1487. Muller, H. P., and Varmus, H. E. (1994). DNA bending creates favored sites for retroviral integration: an explanation for preferred insertion sites in nucleosomes. EMBO J. 13, 4704–4714. Shibagaki, Y., and Chow, S. A. (1997). Central core domain of retroviral integrase is responsible for target site selection. J. Biol. Chem. 272, 8361–8369. Katz, R. A., Merkel, G., and Skalk, A. M. (1996). Targeting of retroviral integrase by fusion to a heterologous DNA binding domain: in vitro activities and incorporation of a fusion protein into viral particles. Virology 217, 178–190. Goulavic, H., and Chow, S. A. (1996). Directed integration of viral DNA mediated by fusion proteins consisting of human immunodeficiency virus type 1 integrase and Escherichia coli LexA protein. J. Virol. 70, 37–46. Bushman, F. D., and Miller, M. D. (1997). Tethering human immunodeficiency virus type 1 preintegration complexes to target DNA promotes integration at nearby sites. J. Virol. 71, 458–464.

27

163. Dildine, S. L., and Sandmeyer, S. B. (1997). Integration of the yeast retrovirus-like element Ty3 upstream of a human tRNA gene expressed in yeast. Gene 194, 227–233. 164. Devine, S. E., and Boeke, J. D. (1996). Integration of the yeast retrotransposon Ty1 is targeted to regions upstream of genes transcribed by RNA polymerase III. Genes Dev. 10, 620–633. 165. Feng, Q., Moran, J., Kazazian, H., and Boeke, J. D. (1996). Human L1 retrotransposon encodes a conserved endonuclease required for retrotransposition. Cell 87, 905–916. 166. Ellis, J., and Bernstein, A. (1989). Gene targeting with retroviral vectors: recombination by gene conversion into regions of nonhomology. Mol. Cell. Biol. 9, 1621–1627. 167. Pazin, M. J., Sheridan, P. L., Cannon, K., Cao, Z., Keck, J. G., Kadonga, J. T., and Jones, K. A. (1996). NF-kappa B-mediated chromatin reconfiguration and transcriptional activation of the HIV-1 enhancer in vitro. Genes Dev. 10, 37–49. 168. Baum, C., Itoh, K., Meyer, J., Laker, C., Ito, Y., and Ostertag, W. (1997). The potent enhancer activity of the polycythemic strain of spleen focusforming virus in hematopoietic cells is governed by a binding site for Sp1 in the upstream control region and by a unique enhancer core motif, creating an exclusive target for PEBP/CBF. J. Virol. 71, 6323–6331. 169. Lund, A. H., Duch, M., and Pedersen, F. S. (1996). Transcriptional silencing of retroviral vectors. J. Biomed. Sci. 3, 365–378. 170. Laker, C., Meyer, J., Schopen, A., Friel, J., Heberlein, C., Ostertag, W., and Stocking, C. (1998). Host cis-mediated extinction of a retrovirus permissive for expression in embryonal stem cells during differentiation. J. Virol. 72, 339–348. 171. Challita, P.-M., and Kohn, D. B. (1994). Lack of expression from a retroviral vector after transduction of murine hematopoietic stem cells is associated with methylation in vivo. Proc. Natl. Acad. Sci. USA 91, 2567–2571. 172. Scharfmann, R., Axelrod, J. H., and Verma, I. (1991). Long-term in vivo expression of retrovirus-mediated gene transfer in mouse fibroblast implants. Proc. Natl. Acad. Sci. USA 88, 4626–2630. 173. Grez, M., Akgün, E., Hilberg, F., and Ostertag, W. (1990). Embryonic stem cell virus, a recombinant murine retrovirus with expression in embryonic stem cells. Proc. Natl. Acad. Sci. USA 87, 9202–9206. 174. Stocking, C., Grez, M., and Ostertag, W. (1993). Regulation of retrovirus infection and expression in embryonic and hematopoietic stem cells. In: Virus Strategies: Molecular Biology and Pathogenesis (W. Doerfler and P. Böhm, eds), pp. 433–455. VCH Verlagsgesellschaft, Weinheim. 175. Speck, N. A., Renjifo, B. V., Golemis, E., Fredrickson, T. N., Hartley, J. W., and Hopkins, N. (1990). Mutations of the core or adjacent LVb elements of the Moloney leukemia virus enhancer alters disease specificity. Genes Dev. 4, 223–242. 176. Morrison, S. J., Uchida, N., and Weissman, I. L. (1995). The biology of hematopoietic stem cells. Annu. Rev. Cell Dev. Biol. 11, 35–71. 177. Weissman, I. L. (2000). Stem cells: units of development, units of regeneration, and units in evolution. Cell 100, 157–168. 178. Hilberg, F., Stocking, C., Ostertag, W., and Grez, M. (1987). Functional analysis of a retroviral host range mutant: altered long terminal repeat sequences allow expression in embryonal carcinoma cells. Proc. Natl. Acad. Sci. USA 84, 5232–5236. 179. Hawley, R. G., Lieu, F. H., Fong, A. Z., and Hawley, T. S. (1994). Versatile retroviral vectors for potential use in gene therapy. Gene Ther. 1, 136–138. 180. Challita, P. M., Skelton, D., el-Khoueiry, A., Yu, X. J., Weinberg, K., and Kohn, D. B. (1995). Multiple modifications in cis elements of the long terminal repeat of retroviral vectors lead to increased expression and decreased DNA methylation in embryonic carcinoma cells. J. Virol. 69, 748–755. 181. Robbins, P. B., Skelton, D. C., Yu, X. J., Halene, S., Leonard, E. H., and Kohn, D. B. (1998). Consistent, persistent expression from modified

28

182.

183.

184.

185.

186. 187.

188.

189.

190.

191.

192.

193.

194.

195.

196.

197. 198.

199.

Christopher Baum et al. retroviral vectors in murine hematopoietic stem cells. Proc. Natl. Acad. Sci. USA 95, 10182–10187. Halene, S., Wang, L., Cooper, R. M., Bockstoce, D. C., Robbins, P. B., and Kohn, D. B. (1999). Improved expression in hematopoietic and lymphoid cells in mice after transplantation of bone marrow transduced with a modified retroviral vector. Blood 94, 3349–3357. Tumas, D. B., Spangrude, G. J., Brooks, D. M., Williams, C. D., and Chesebro, B. (1996). High-frequency cell-surface expression of a foreign protein in murine hematopoietic stem cells using a new retroviral vector. Blood 87, 509–517. Hildinger, M., Eckert, H. G., Schilz, A. J., John, J., Ostertag, W., and Baum, C. (1998). FMEV vectors: both retroviral long terminal repeat and leader are important for high expression in transduced hematopoietic cells. Gene Ther. 5, 1575–1579. van Hennik, P. B., Verstegen, M. M., Bierhuizen, M. F., Limon, A., Wognum, A. W., Cancelas, J. A., Barquinero, J., Ploemacher, R. E., and Wagemaker, G. (1998). Highly efficient transduction of the green fluorescent protein gene in human umbilical cord blood stem cells capable of cobblestone formation in long-term cultures and multilineage engraftment of immunodeficient mice. Blood 92, 4013–4022. Shivdasani, R. A., and Orkin, S. (1996). The transcriptional control of hematopoiesis. Blood 87, 4025–4039. Nielsen, A. L., Pallisgaard, N., Pedersen, F. S., and Jorgensen, P. (1994). Basic helix–loop–helix proteins in murine type C retrovirus transcriptional regulation. J. Virol. 68, 5638–5647. Zaiman, A. L., and Lenz, J. (1996). Transcriptional activation of a retrovirus enhancer by CBF (AML1) requires a second factor: evidence for cooperativity with c-Myb. J. Virol. 70, 5618–5629. Malik, P., Krall, W. J., Yu, X. J., Zhou, C., and Kohn, D. B. (1995). Retroviral-mediated gene expression in human myelomonocytic cells: a comparison of hematopoietic cell promoters to viral promoters. Blood 86, 2993–3005. Plavec, I., Voyovich, A., Moss, K., Webster, D., Hanley, M. B., Escaich, S., Ho, K. E., Boehnlein, E., and DiGiusto, D. L. (1996). Sustained retroviral gene marking and expression in lymphoid and myeloid cells derived from transduced hematopoietic progenitor cells. Gene Ther. 3, 717–724. Agarwal, M., Austin, T. W., Morel, F., Chen, J., Bohnlein, E., and Plavec, I. (1998). Scaffold attachment region-mediated enhancement of retroviral vector expression in primary T cells. J. Virol. 72, 3720– 3728. Onodera, M., Nelson, D. M., Yachie, A., Jagadeesh, G. J., Bunnell, B. A., Morgan, R. A., and Blaese, R. M. (1998). Development of improved adenosine deaminase retroviral vectors. J. Virol. 72, 1769–1774. Couture, L. A., Mullen, C. A., and Morgan, R. A. (1994). Retroviral vectors containing chimeric promoter/enhancer elements exhibit celltype-specific gene expression. Hum. Gene Ther. 5, 667–677. Dachs, G. U., Patterson, A. V., Firth, J. D., Ratcliffe, P. J., Townsend, K. M., Stratford, I. J., and Harris, A. L. (1997). Targeting gene expression to hypoxic tumor cells. Nat. Med. 3, 515–520. Miller, N., and Whelan, J. (1997). Progress in transcriptionally targeted and regulatable vectors for genetic therapy. Hum. Gene Ther. 8, 803– 815. Gautsch, J. W., and Wilson, M. C. (1983). Delayed de novo methylation in teratocarcinoma suggests additional tissue-specific mechanisms for controlling gene expression. Nature 301, 32–37. Bird, A. P. (1986). CpG-rich islands and the function of DNA methylation. Nature 321, 209–213. Macleod, D., Charlton, J., Mullins, J., and Bird, A. P. (1994). Sp1 sites in the mouse aprt gene promoter are required to prevent methylation of the CpG island. Genes Dev. 8, 2282–2292. Grez, M., Zörnig, M., Nowock, J., and Ziegler, M. (1991). A single point mutation activates the Moloney murine leukemia virus long terminal repeat in embryonal stem cells. J. Virol. 65, 4691–4698.

200. Schubeler, D., Mielke, C., Maass, K., and Bode, J. (1996). Scaffold/matrix-attached regions act upon transcription in a contextdependent manner. Biochemistry 35, 11160–11169. 201. Phi-Van, L., von Kries, J. P., Ostertag, W., and Strätling, W. H. (1990). The chicken lysozyme matrix attachment region increases transcription from a heterologous promoter in heterologous cells and dampens position effects on the expression of transfected cells. Mol. Cell. Biol. 10, 2302–2307. 202. Dang, Q., Auten, J., and Plavec, I. (2000). Human beta interferon scaffold attachment region inhibits de novo methylation and confers longterm, copy-number-dependent expression to a retroviral vector. J. Virol. 74, 2671–2678. 203. Rivella, S., Callegari, J. A., May, C., Tan, C. W., and Sadelain, M. (2000). The cHS4 insulator increases the probability of retroviral expression at random chromosomal integration sites. J. Virol. 74, 4679– 4687. 204. Gossen, M., Bonin, A. L., and Bujard, H. (1993). Control of gene activity in higher eukaryotic cells by prokaryotic regulatory elements. Trends Biochem. Sci. 18, 471–475. 205. Gossen, M., Freundllieb, S., Bender, G., Muller, G., Hillen, W., and Bujard, H. (1995). Transcriptional activation by tetracyclines in mammalian cells. Science 268, 1766–1769. 206. Huang, Y., Wimler, K. M., and Carmichael, G. G. (1999). Intronless mRNA transport elements may affect multiple steps of pre-mRNA processing. Embo J. 18, 1642–1652. 207. Luo, M. J., and Reed, R. (1999). Splicing is required for rapid and efficient mRNA export in metazoans. Proc. Natl. Acad. Sci. USA 96, 14937–14942. 208. Pollard, V. W., and Malim, M. H. (1998). The HIV-1 Rev protein. Annu. Rev. Microbiol. 52, 491–532. 209. Zufferey, R., Donello, J. E., Trono, D., and Hope, T. J. (1999). Woodchuck hepatitis virus post-transcriptional regulatory element enhances expression of transgenes delivered by retroviral vectors. J. Virol. 73, 2886–2892. 210. Schambach, A., Wodrich, H., Hildinger, M., Bohne, J., Kräusslich, H.-G., and Baum, C. (2000). Context-dependence of different modules for post-transcriptional enhancement of gene expression from retroviral vectors. Mol. Ther. 2, 435–445. 211. Hildinger, M., Schilz, A., Eckert, H. G., Bohn, W., Fehse, B., Zander, A., Ostertag, W., and Baum, C. (1999). Bicistronic retroviral vectors for combining myeloprotection with cell-surface marking. Gene Ther. 6, 1222–1230. 212. Emerman, M., and Temin, H. M. (1984). Genes with promoters in retrovirus vectors can be independently suppressed by an epigenetic mechanism. Cell 39, 449–467. 213. Ghattas, I. R., Sanes, J. R., and Majors, J. E. (1991). The encephalomyocarditis virus internal ribosome entry site allows efficient coexpression of two genes from a recombinant provirus in cultures cells and in embryos. Mol. Cell. Biol. 11, 5848–5849. 214. Eggermont, J., and Proudfoot, N. J. (1993). Poly(A) signals and transcriptional pause sites combine to prevent interference between RNA polymerase II promoters. EMBO J. 12, 2539–2548. 215. Proudfoot, N. J. (1986). Transcriptional interference and termination between duplicated alpha-globin gene constructs suggest a novel mechanism for gene regulation. Nature 322, 562–565. 216. Ahlers, N., Hunt, N., Just, U., Laker, C., Ostertag, W., and Nowock, J. (1994). Selectable retrovirus vectors encoding Friend virus gp55 or erythropoietin induce polycythemia with different phenotypic expression and disease progression. J. Virol. 68, 7235– 7243. 217. Adam, M. A., Ramesh, N., Miller, A. D., and Osborne, W. R. (1991). Internal initiation of translation in retroviral vectors carrying picornavirus 5 nontranslated regions. J. Virol. 65, 4985– 4990.

Retroviral Vector Design for Cancer Gene Therapy 218. Boris-Lawrie, K. A., and Temin, H. M. (1993). Recent advances in retrovirus vector technology. Curr. Opin. Genet. Dev. 3, 102–109. 219. Morgan, R. A., Couture, L., Elroy-Stein, O., Ragheb, J., Moss, B., and Anderson, W. F. (1992). Retroviral vectors containing putative internal ribosome entry sites: development of a polycistronic gene transfer system and applications to human gene therapy. Nucleic Acids Res. 20, 1293–1299. 220. Mizuguchi, H., Xu, Z.,Ishii-Watabe, A., Uchida, E., and Hayakawa, T. (2000). IRES-dependent second gene expression is significantly lower than cap-dependent first gene expression in a bicistronic vector. Molecu. Ther. 1, 376–382. 221. Germann, U. A., Chin, K.-V., Pastan, I., and Gottesman, M. M. (1990). Retroviral transfer of a chimeric multidrug resistance-adenosine deaminase gene. FASEB J. 4, 1501–1506. 222. de Felipe, P., Martin, V., Cortes, M. L., Ryan, M., and Izquierdo, M. (1999). Use of the 2A sequence from foot-and-mouth disease virus in the generation of retroviral vectors for gene therapy. Gene Ther. 6, 198–208. 223. Klump, H., Schiedlmeier, B., Vogt, B., Ryan, M., Ostertag, W., and Baum, C. (2000). Retroviral vector-mediated expression of HOXB4 in

224.

225.

226.

227.

29

hematopoietic cells using a novel coexpression strategy. Gene Ther. 8, 811–817. Artelt, P., Grannemann, R., Stocking, C., Friel, J., Bartsch, J., and Hauser, H. (1991). The prokaryotic neomycin-resistance-encoding gene acts as a transcriptional silencer in eukaryotic cells. Gene 99, 249–254. Bunting, K. D., Webb, M., Giorgianni, G., Galipeau, J., Blakley, R. L., Townsend, A., and Sorrentino, B. P. (1997). Coding region-specific destabilization of mRNA transcripts attenuates expression from retroviral vectors containing class 1 aldehyde dehydrogenase cDNAs. Hum. Gene Ther. 8, 1531–1543. Sorrentino, B. P., McDonagh, K. T., Woods, D., and Orlic, D. (1995). Expression of retroviral vectors containing the human multidrug resistance 1 cDNA in hematopoietic cells of transplanted mice. Blood 86, 491–501. Garin, M. I., Garrett, E., Tiberghien, P., Apperley, J. F., Chalmers, D., Melo, J. V., and Ferrand, C. (2001). Molecular mechanism for ganciclovir resistance in human T lymphocytes transduced with retroviral vectors carrying the herpes simplex virus thymidine kinase gene. Blood. 97, 122–129.


C

H

A

P

T

E

R

2 Noninfectious Gene Transfer and Expression Systems for Cancer Gene Therapy MARK J. COOPER Copernicus Therapeutics, Inc. Cleveland, Ohio 44106

I. Introduction 31 II. Advantages and Disadvantages of Infectious, Viral-Based Vectors for Human Gene Therapy 31 III. Rationale for Considering Noninfectious, Plasmid-Based Expression Systems 33 IV. Gene Transfer Technologies for Plasmid-Based Vectors: Preclinical Models and Clinical Cancer Gene Therapy Trials 33

[37,38]; anti-sense constructs to insulin-like growth factor I [39,40], and polynucleotide vaccines [41–45]; replacement of wild-type tumor suppressor genes, such as p53 [44–49]; and anti-sense blockade of oncogenes, such as K-ras [50– 52]. In order to move gene therapy into the mainstream of cancer therapeutics, however, it will ultimately be necessary to devise strategies to administer a gene therapy reagent to a patient in the familiar context of a pharmaceutical and to perform gene transfer in vivo. Currently utilized viral-based gene therapy vectors, including retroviral, adenoviral, and adenoassociated viral vectors, fail to realize this potential due to limitations in their expression characteristics, lack of specificity in targeting tumor cells for gene transfer, immunogenicity and other acute and chronic toxicities, and safety concerns regarding induction of secondary malignancies and recombination to form replication-competent virus. These limitations have refocused efforts to develop noninfectious, gene transfer technologies for in vivo gene delivery of plasmid-based expression vectors. These vectors exist as extrachromosomal elements in populations of transiently transfected tumor cells. As discussed later, incorporation of transcription control sequences, including tissue-specific enhancers and inducible promoters, and elements permitting controlled vector replication in tumor cells has the potential to yield cancer gene therapy vectors that are both safe and effective for direct in vivo gene transfer.

A. Direct Injection of DNA 33 B. Particle-Mediated Gene Delivery 34 C. Gene Transfer of DNA Precipitated with Calcium Phosphate 35 D. Liposome-Mediated Gene Delivery 35 E. Ligand/DNA Conjugates 36

V. Plasmid Expression Vectors

37

A. Tissue-Specific Promoters 38 B. Inducible Promoters 38 C. Replicating Plasmid Vectors: Episomes

VI. Future Directions

40

43

Acknowledgments 45 References 45

I. INTRODUCTION Gene therapy provides a significant opportunity to devise novel strategies for the control or cure of cancer. Current approaches to cancer gene therapy typically employ viralbased vectors to express suitable target genes in human cancer cells either ex vivo or in vivo [1–4]. Therapeutic gene targets currently being evaluated include susceptibility genes, such as herpes simplex thymidine kinase followed by ganciclovir treatment [5–15]; genes that target the immune system to eliminate cancer cells, such as cytokines [16–35], costimulatory molecules [36], foreign histocompatibility genes


II. ADVANTAGES AND DISADVANTAGES OF INFECTIOUS, VIRAL-BASED VECTORS FOR HUMAN GENE THERAPY A number of viruses that infect humans, including retrovirus, herpes virus, adenovirus, and adeno-associated virus,

31


32

Mark J. Cooper

have been modified to generate efficient expression vectors. These vectors either integrate into genomic DNA or persist as extrachromosomal elements and have distinct expression characteristics, as summarized in Table 1. The primary advantage of these vectors is the ability to infect a high percentage of target cells in vitro and, in some cases, in vivo [7,53,54]. Whereas retroviral vectors yield one or several integrated proviral copies per cell, other vectors can introduce higher copy numbers of transcriptional cassettes, thereby enhancing transient levels of gene expression. Some viral-based vectors, such as those derived from recombinant adenovirus, may replicate in transduced cells at a low level, and this feature has usually been interpreted as an undesired feature raising safety concerns regarding unregulated, systemic gene transfer [4,55–57]. More recently, E1B-attenuated adenoviral vectors have been developed that replicate in tumor cells, resulting in tumor cell lysis and virus propagation within the tumor [58,59]. Although replication of this adenovirus construct was initially thought to be restricted to p53-negative tumor cells [58], other studies demonstrate that virus replication is independent of p53 status [60–62]. Intratumoral injections of these vectors have produced localized tumor regression in patients with recurrent head and neck cancers [63,64]. Although viral-based vectors may be particularly useful for gene transfer ex vivo, this approach requires costly manipulations of tumor biopsies to yield either transient [65] or stably selected and characterized transfectants [19]. Additionally, the latter approach may prove to be a particularly poor choice for gene targets that stimulate the immune system to eliminate tumor cells, as representation of tumor heterogeneity is likely lost prior to gene transfer. While high-level infectivity of viral-based vectors remains an attractive feature, multiple safety concerns and technical features limit their applications, including: (1) safety con-

cerns regarding integration of vector DNA into host cell genomic DNA, which may induce secondary malignancies by activation of proto-oncogenes or inactivation of tumor suppressor genes [66]; (2) potential for recombination events to produce an infectious virus able to replicate in vivo (recombination could occur either in vitro during vector preparation, or possibly in vivo, particularly when using vectors derived from pathogenic human viruses, such as adenovirus) [2,3,55– 57,67,68]; (3) presentation of viral antigens on the surface of infected human cells, resulting in T-cell recognition and destruction of transduced cells [69]; (4) lack of specificity of cell types recognized by endogenous viral coat proteins, resulting in unintended transduction of nontargeted cell types in vivo; (5) heterogeneity of expression of viral coat protein receptors by tumor cell targets, thereby limiting the tumor cell population that can be transduced (viral receptor-negative cells may be selected for during treatment); (6) the fact that retroviral vectors will not express target genes in nonreplicating tumor cells [70]; (7) technical limitations regarding strategies to produce higher levels of gene expression in an infected cell; (8) difficulties in reproducibly producing, concentrating, delivering, and storing high titer viral vectors for clinical use; (9) complement-mediated mechanisms of inactivation may limit use of some viral-based vectors in vivo [71]; (10) the potential for some virally encoded proteins to yield undesired toxic effects in addition to immune recognition, leading to altered cell functions or transformation [2,4]; and (11) immunogenicity of viral-based vectors, resulting in incrementally decreased effectiveness during repeated treatments in vivo [2,4,72–76]. These safety concerns and limitations in the ability of infectious, viral-based vectors to yield maintained, high-level gene expression in transiently transfected tumor cells have led to the development of alternative, noninfectious gene expression and gene transfer technologies, as reviewed later.

TABLE 1 Infectious, Viral-Based Vectors for Cancer Gene Therapy Integration or extrachromosomal distribution

Expression limited to cells undergoing replication at time of infection

Retrovirus

I

Yes

70

Adenovirus

E

No

55–57, 293

Adeno-associated virus

Ia

Yesb

294–296

Herpes simplex virus

E

No

297

Vaccinia virus

E

No

298

Autonomous parvovirus (LuIII)

E

Yes

299

Vector

Note: Abbreviations: I, integration; E, extrachromosomal. in replicating cells, transient extrachromosomal persistence in stationary phase cells. b 90% of expression limited to cells traversing S phase.

a Integration

Ref.

Noninfectious Gene Transfer and Expression Systems for Cancer Gene Therapy

III. RATIONALE FOR CONSIDERING NONINFECTIOUS, PLASMID-BASED EXPRESSION SYSTEMS Initial assumptions regarding requirements for effective cancer gene therapy have changed since demonstration of a significant “innocent bystander” effect using gene targets that confer antibiotic susceptibility, such as herpes simplex virus thymidine kinase followed by ganciclovir treatment [8], or genes that activate the immune system to recognize and kill tumor cells [16–43]. It may therefore not be necessary to transfect 50–100% of tumor cells in order to produce a cure. These findings provide an important rationale to consider nonviral-based vectors for gene therapy applications, particularly constructs that yield high levels of gene expression per transfected cell. Moreover, new technical advances in receptor-mediated gene delivery of plasmid-based vectors now yield transient transfection efficiencies in vivo that approximate those observed using viral-based vectors [77–80].

IV. GENE TRANSFER TECHNOLOGIES FOR PLASMID-BASED VECTORS: PRECLINICAL MODELS AND CLINICAL CANCER GENE THERAPY TRIALS Several gene transfer methods yield efficient transient transfection efficiencies following either in vitro or in vivo applications, as listed in Table 2. Although some of these methods are limited by the target cell type transfected or by the specificity of gene transfer, receptor-mediated gene transfer technologies have the potential to yield efficient and specific gene delivery to targeted tumor cells in vivo and therefore may have widespread utility.

A. Direct Injection of DNA Perhaps the simplest formulation for in vivo gene transfer of plasmid vectors into cells is by direct administration of suTABLE 2 Gene Transfer Technologies for Plasmid-Based Vectors Gene transfer limited to specific tissues

Ability to target tumor cells

Direct injection of naked DNA

Yes

No

Particle bombardment

Yes

No

Calcium phosphate

No

No

Liposome/DNA complexes

No

Yes

Ligand/DNA conjugates

No

Yes

Gene transfer method

33

percoiled DNA into tissues. Early studies demonstrated that DNA can be directly introduced into cells in vivo by simply injecting target organs with viral DNA. For example, when polyoma virus [81,82] or ground squirrel hepatitis virus [83] DNA were directly injected into mice or ground squirrels, respectively, the animals developed systemic infection, and active virus particles were recovered. In these studies, however, very inefficient initial levels of in vivo gene transfer of purified virion DNA could be detected due to amplification of the gene transfer mechanism via systemic virus infection. In related studies, gene expression was observed in the liver and spleen of newborn rats 2 days following intraperitoneal injection of calcium-phosphate-precipitated plasmid DNA encoding the chloramphenicol acetyltransferase reporter gene [84]. More recently, direct injection of naked plasmid DNA was shown to yield significant levels of gene expression in rat skeletal and cardiac muscle, but not in kidney, lung, liver, or brain [85,86]. For example, direct injection of 25 μg of p-CMVint-lux plasmid DNA encoding the luciferase marker gene driven by the CMV immediate-early promoter into the rectus femoris muscle of mice yielded peak gene expression at day 14, and expression was detectable for up to 120 days [87]. The mechanism by which plasmid DNA is taken up by muscle cells is unclear but does not seem to be related to direct cell injury to the sarcolemmal membrane [88]. In more recent studies, significant gene expression has also been observed following direct injection of naked plasmid DNA into rat or cat liver [89] and rabbit thyroid follicular cells [90], expanding the tissue types that can be transfected using this method. Gene expression in transfected muscle cells is sufficient to produce antiviral immunity. For example, mice having their quadriceps muscles injected with a plasmid encoding influenza A nucleoprotein developed humoral and cytotoxic T-cell responses to this antigen and were protected from subsequent challenge with influenza A virus [91]. In a similar fashion, direct intramuscular gene transfer of plasmid DNA encoding HIV envelop protein (gp160) in mice confers humoral and cell-mediated immunity against recombinant envelop protein, and sera from these animals neutralizes HIV infectivity in vitro [92]. Direct injection of plasmid DNA also results in efficient gene delivery to subcutaneous tissues, including keratinocytes, fibroblasts, and dendritic cells [93]. This later approach may be superior to direct muscle injection for the development of cytotoxic T-cell immunity, perhaps because of antigen presentation by macrophages and dendritic cells in the subcutaneous tissues [93]. Intradermal gene transfer efficiencies can also be enhanced by delivering electric pulses to subcutaneous tissues after a local injection [94,95] or by use of liquid or powder sprays or particle bombardment methods of DNA transfer [96,97]. Intramuscular or intradermal gene transfer of plasmid vectors encoding tumor-associated antigens may yield effective cancer vaccines. This approach requires prior knowledge

34

Mark J. Cooper

of potential tumor-associated antigens in a given patient’s tumor that presumably have not yet been adequately presented to the host immune system. A variety of tumorassociated antigens have been identified that have the potential to stimulate a cytotoxic T-cell response and are therefore candidate antigens for tumor vaccines. In human melanoma, such tumor-associated antigens include p97, MAGE-1, MAGE-2, MAGE-3, Melan-A, MART-1, gp100, and tyrosinase [98–107]. Cytotoxic T-cell responses also have been demonstrated against mucin products of the MUC-1 gene in patients with pancreatic and breast carcinomas, and antigenicity appears to be related to underglycosylated forms of the protein found in tumor cells [108]. In ideal circumstances, tumor-associated antigens would only be expressed by the tumor and not normal tissues, and a cancer vaccine would generate a tumor-specific immune response. Because peptide fragments of cellular proteins are displayed on the cell surface in conjunction with major histocompatibility antigens by TAP transporter proteins [109], the immune system is able to survey for the presence of gene mutations that generate novel peptide antigens. Tumor-specific cytotoxic T-cell immunity has been demonstrated against peptide fragments from oncogenes or tumor suppressor genes that are mutated during the generation of the malignancy. These vaccines include peptides encoding point mutations in ras genes [110– 113] and p53 [114] and the unique breakpoint in the bcr–abl fusion gene [115]. One example of a successful cancer vaccine model is development of antitumor immunity to tumor cells expressing human carcinoembryonic antigen (CEA). CEA is expressed at high levels in several types of human adenocarcinomas, including colon, breast, gastric, pancreatic, and non-small-cell lung carcinomas [41,42,116]. CEA is also expressed at high levels in human fetal gut and at low levels in normal colonic mucosal cells, but it is not expressed in murine tissues [41]. Therefore, mice immunized with CEA protein would be expected to develop antitumor immunity to syngeneic tumor cells expressing human CEA. This result has been demonstrated by using a recombinant vaccinia virus vector encoding human CEA cDNA to immunize mice [41]. These studies used a murine colon carcinoma cell line, MC38, that had been transduced with a retroviral vector encoding CEA cDNA, generating the modified MC38–CEA-2 cell line. Vacciniavector-immunized syngeneic C57BL/6 mice developed humoral and cell-mediated immunity to CEA, and MC38–CEA2 cells injected in immunized animals were rejected [41]. To extend these studies, Curiel and colleagues have demonstrated that C57BL/6 mice can develop antitumor immunity to MC38–CEA-2 cells by directly injecting plasmid DNA encoding CEA cDNA into striated muscle [43]. In these studies, the tongues of C57BL/6 mice were injected weekly with 100 μg of plasmid DNA encoding CEA. After four doses, these animals produced anti-CEA antibodies and developed cell-mediated immunity to MC38–CEA-2 cells.

Importantly, these immunized mice rejected MC38–CEA-2 cells that were subcutaneously inoculated in these animals 1 week following the last immunization. These results demonstrate the ability to generate an effective cancer vaccine by expressing a tumor-associated antigen following direct in vivo gene transfer of plasmid DNA. Further issues that need to be addressed by the use of polynucleotide vaccines include choice of specific tumorassociated antigens likely to produce antitumor immunity in cancer patients of a given tumor type and the clinical setting in which this approach is likely to be effective. For example, administration of a tumor vaccine in an adjuvant setting following initial surgical removal of the primary mass may improve conditions for success by selecting a patient population that has not yet received immune-suppressive cytotoxic chemotherapy and whose tumor burden is small. In addition, analysis of tumor tissue for expression of relevant tumor antigens may be quite important, as multiple tumor-associated antigens may need to be targeted to address clonal evolution of heterogeneous populations of tumor cells. Alternatively, antitumor immunity can be achieved by addressing mechanisms of tumor cell immune tolerance [117]. Such approaches are independent of identification of tumorspecific antigens. For examples, many tumors lack expression of the transporters associated with antigen processing (TAP) [118,119], resulting in insufficient presentation of tumorassociated antigens with nascently produced class I MHC molecules. Direct TAP1 gene transfer to such tumors in an animal model results in prolonged survival of tumor-bearing mice [120]. Intratumor gene transfer of other components required for generation of cytotoxic T-cells, including class I MHC molecules [121,122] and β2 -microglobulin [123], may be effective in some tumors. Other approaches include gene transfer of the B7.1 costimulatory molecule and anti-sense or ribozyme strategies to decrease local production of immunosuppressive cytokines and receptors [36,124–132].

B. Particle-Mediated Gene Delivery An alternative approach for delivery of plasmid constructs into human cells in vivo is to coat metallic particles with a DNA vector and then introduce the particles directly into tissues using a “gene gun” to accelerate the particles to a high velocity [133]. Subcutaneous tissues can be directly transfected in vivo because the particles can penetrate to this depth. Visceral tissues have also been transfected in vivo in animals, although this approach requires an operative procedure to bring the tissue of interest in close approximation to the gene gun instrument. Nevertheless, particle-mediated gene transfer of a plasmid vector encoding influenza virus hemagglutinin subtype 1 has been demonstrated to immunize mice against challenge with a lethal inoculum of influenza virus [96]. This approach has significant potential for development of cancer vaccines, because efficient gene transfer of polynucleotide


vaccines into subcutaneous tissues may be particularly effective in presenting antigens to the immune system [93]. Cancer preclinical models using particle-mediated gene transfer into subcutaneous tumor explants have also demonstrated improved survival of tumor-bearing mice using a variety of cytokine targets, including IL-2, IL-6, and IFN-γ [134]. In addition to ballistic gene transfer using metallic particles, devices have been developed for needleless intradermal DNA delivery using powders and liquid sprays [96,97].

C. Gene Transfer of DNA Precipitated with Calcium Phosphate Plasmid DNA precipitated with calcium phosphate can efficiently transfect cells in tissue culture, as reported by Graham and Van der Eb in 1973 [135]. More then a decade ago, this technique was also used for in vivo gene transfer of viral and plasmid DNA into liver and spleen by either direct inoculation into the tissue bed or intraperitoneal instillation [81–84]. Despite these initial promising results and the ease of preparing these DNA precipitates, this method has largely been supplanted by alternative approaches that are thought to yield superior in vivo transfection efficiencies. Nevertheless, this method has recently been employed in preclinical cancer gene therapy studies evaluating introduction of herpes simplex virus thymidine kinase (HSV-TK) into melanoma explants [136]. In these studies, plasmid DNA encoding HSV-TK was precipitated with calcium phosphate and directly injected into established B16 melanoma tumor explants in syngeneic C57/BL mice. After administration of intraperitoneal ganciclovir, treated animals achieved a partial tumor regression.

D. Liposome-Mediated Gene Delivery Polycationic lipids can be mixed with plasmid DNA to form liposome structures that are thought to fuse with the target cell membrane and thereby mediate gene delivery [137]. Several lipid preparations have been formulated for this application, including mixtures of dioleoyl phosphatidylethanolamine (DOPE) with DOTMA (lipofectin), DOSPA (lipofectamine), DDAB (lipofectace), DOGS (transfectam), DOTAP, DMRIE, and DC cholesterol (reviewed in Felgner et al. [138]). This approach can yield very high transfection efficiencies in vitro and can also be used for direct in vivo gene transfer. Plasmid DNA has been delivered to tumor explants in syngeneic mice by injecting the tumor nodule with liposome/DNA complexes, achieving a transient transfection efficiency of approximately 1–10% [37]. A particular advantage of this approach is the ease of preparing DNA/liposome complexes, the stability of the individual components, and the versatility to transfect a variety of tumor types. The liposome/DNA complex can be directly injected into a palpable tumor nodule [38]. Alternatively, visceral

35

tumor masses can be directly instilled with liposome/DNA complexes by employing radiologic procedures, such as CAT scans, to identify the location of the tumor and assist in percutaneous tumor injection [139,140]. Alternative approaches include use of bronchoscopy, cystoscopy, endoscopy, or laparoscopy to directly inject liposome/DNA complexes into visualized tumor masses. Liposome/DNA complexes have minimal systemic toxicities [141,142] and can be administered repeatedly to the same patient with expectations of equivalent efficiencies of gene transfer. Liposome/DNA complexes administered intravenously also can deliver plasmid vectors into multiple tissue types. In 1983, Nicolau et al. injected rats intravenously with a plasmid vector encoding rat preproinsulin I complexed with liposomes composed of phosphatidylcholine, phosphatidylserine, and cholesterol [143]. In these studies, radioactive labeled liposomes were shown to be taken up specifically by liver and spleen, and 6 hours after injection treated animals experienced a fall in serum glucose and an increase in serum, liver, and splenic insulin levels relative to control animals. More recently, Zhu and Debs demonstrated gene expression in diverse tissue types, including liver, spleen, kidney, lung, heart, lymph nodes, and bone marrow, following intravenous administration of chloramphenicol acetyltransferase reporter plasmids complexed with liposomes composed of DOTMA and DOPE lipids [144]. Gene expression was detected for up to 9 weeks following gene transfer. This widespread gene delivery raises the possibility of using intravenous administration of liposome/DNA complexes to introduce therapeutic genes in multiple foci of metastatic disease. For example, a study employing a p53 mutant human breast cancer xenogeneic model suggests that intravenous administration of liposome/DNA complexes encoding wild-type p53 may reduce the size of primary tumor explants and decrease the development of metastatic disease to lungs [145].Although the liposome formulations described above do not specifically target tumor cells, ongoing studies suggest that it may be possible to increase the specificity of liposome-mediated gene transfer by conjugating ligands for cell surface receptors to lipid moieties. In recent studies, receptor-mediated gene transfer in vitro has been demonstrated for liposome preparations targeting the folate, erbB-2, transferrin, and mannose receptors [146–153]. Additionally, so-called “stealth” liposomes have been developed to avoid rapid clearance by reticuloendothelial cells following an intravenous injection [148,154–158]. In preclinical models, tumor-bearing animals treated with chemotherapeutic agents encapsulated in stealth liposomes had improved survival compared to control groups treated with free drug alone [157]. Stealth liposomes typically include polyethylene glycol to increase serum half-life, and these preparations pool in tissues, such as tumors, that have increased vascular permeability, resulting in passive targeting of the complexes [159–163]. As much as 3–6% of the dose of DNA was reported to localize in the tumor nodule, although

36

Mark J. Cooper

DNA transfer into tumor cells was inefficient [164,165]. Because liposomes incorporating polyethylene glycol tend to be non-fusogenic, pH-sensitive and kinetically unstable linkages are being developed to reversibly release polyethylene glycol from the lipsome, thereby enhancing their gene transfer properties [166,167]. An active area of current research is to develop targeted and stealthy liposome preparations suitable for intravenous delivery of plasmid constructs for cancer gene therapy. Several preclinical models have demonstrated antitumor responses when mixtures of liposomes and plasmid vectors have been directly transferred into established tumor explants in syngeneic mice. For example, plasmids encoding the murine class I H-2Ks gene have been complexed with cationic liposomes and injected into established CT26 colon carcinoma (H-2Kd ) and MCA 106 fibrosarcoma (H-2Kb ) cells. As reported by Plautz and Nabel, a cytotoxic T-cell response to H-2Ks antigen was induced, and animals preimmunized to H-2Ks antigen demonstrated significant antitumor activity, with some animals achieving long-term survival [37]. In addition, this antitumor activity was cell line specific, because animals bearing MCA 106 tumors previously cured following injection with H-2Ks plasmid rejected secondary tumor challenges with parental MCA 106 cells but not syngeneic B16BL/6 melanoma cells. These findings suggest that expression of foreign class I histocompatibility antigens by these tumor cells resulted in recognition of heretofore unrecognized tumor-associated antigens by cytotoxic T cells. This hypothesis would account for the observed efficient tumor elimination and prolonged survival despite the fact that only a modest percentage of the tumor cells were transiently transfected following direct tumor inoculation by DNA/liposome complexes. In a pilot study at the University of Michigan, Nabel and colleagues have extended their preclinical model to a clinical cancer gene therapy protocol by evaluating liposomemediated gene transfer of plasmids encoding HLA-B7 in patients with metastatic melanoma. In these studies, liposome/DNA complexes were directly injected into subcutaneous, nodal, and visceral masses, and one out of the first five patients evaluated demonstrated a significant response [38]. These encouraging findings have led to several active trials evaluating the expression of a bicistronic plasmid encoding HLA-B7 and β2 -microglobulin in patients with metastatic colon cancer, renal cancer, and melanoma [139,140,168]. In these trials, plasmid DNA complexed with liposomes composed of DIMRIE and DOPE lipids are directly injected into tumor masses. In initial reports, HLA-B7 gene expression has been shown in tumor biopsies after gene transfer [139,140], and antitumor immunity has been observed in local tumorinfiltrating lymphocytes [169]. These clinical trials are currently in progress, and additional data regarding generation of T-cell immunity to HLA-B7 target cells, tumor responses, survival, and toxicities of the treatment are pending.

Due to a lack of cytosine methylase activity in bacteria, typical preparations of plasmid DNA have unmethylated cytosine nucleotides (CpG islands). Such unmethylated CpG islands possess potent adjuvant and immunomodulatory effects and can produce locally elevated levels of cytokines [170–177]. Such unmethlyated CpG islands play a significant role in the effectiveness of DNA vaccines administered as an intramuscular or intradermal injection by stimulating a potent TH1 response. When administered systemically, plasmid DNA alone does not produce systemic levels of cytokines, but complexes of DNA and cationic liposomes can produce elevated levels of TNF-α, IL-12, and IFN-γ [178]. Moreover, an antitumor effect was observed when mice bearing pulmonary metastases received intravenous injections of liposome/plasmid DNA complexes lacking a therapeutic gene; these effects were comparable to liposome/DNA complexes encoding IL-12 or p53 [178]. Similar results were observed in subcutaneous and intraperitoneal syngeneic tumor models [178,179]. This potent antitumor effect was found in immunocompetent mice but not SCID, athymic, or SCID/Beige mice [178,179]. Moreover, these antitumor effects of liposome/DNA complexes lacking a therapeutic gene were inhibited by prior methylation of CpG motifs in the plasmid using SSI methylase [178]. These interesting results may explain, in part, the ability of intravenously administered cationic liposome/DNA complexes to generate greater antitumor effects than predicted based on their more modest gene transfer efficiency.

E. Ligand/DNA Conjugates Negatively charged plasmid DNA molecules and polycations, such as poly(L-lysine), can form complex structures consisting of either unimolecular or multimolecular complexes (with respect to the DNA) [77,78]. To enable efficient and cell-specific gene transfer, the poly(L-lysine) polymer can be modified by covalently attaching ligands that can subsequently bind to specific cellular receptors [180]. If the DNA/poly(L-lysine) complex contains a suitable ligand, then the DNA/poly(L-lysine) complex can be internalized in the cell when the receptor undergoes endocytosis. Most of these early DNA/poly(L-lysine) formulations were multimolecular complexes, approximately 100–200 nm in diameter [77], which may have limited their ability to enter cells via receptor-mediated endocytosis. Additionally, efficient expression of the internalized plasmid requires several additional steps, including exit from the endosome prior to destruction of the DNA by fusion of the endosome with lysosomes and transfer of the plasmid DNA to the nucleus [77,181]. Initial formulations of poly(L-lysine)/DNA complexes for in vivo gene transfer targeted the liver asialoglycoprotein receptor for gene delivery using asialoorosomucoid covalently linked to poly(L-lysine) [180]. Gene expression was transient,


although preferential gene transfer to the liver was observed. In later studies, gene expression was improved by performing a partial hepatectomy in association with receptor-mediated gene transfer [182]. Further improvements in gene expression were achieved by using endosomolytic agents, such as defective adenovirus particles or peptides derived from the N-terminal region of influenza virus hemagglutinin HA-2 protein, to enable transferrin-conjugated poly(L-lysine)/DNA complexes to exit the endosome and enter the cytoplasm for eventual transfer to the nucleus [183–185]. This modification has been shown to achieve transient gene expression in lung following direct instillation of ligand/DNA complexes into the airway of rats [186]. Gene transfer in vitro has been demonstrated in primary intestinal mucosal cells and the transformed Caco2 colon adenocarcinoma cell line [187], suggesting an approach for gene delivery into tumor cells. However, most of these studies employed rapidly dividing cells in which the nuclear membrane barrier is broken down during mitosis, thereby permitting plasmid to enter the nucleus. Because the nuclear membrane severely restricts transfer of large DNA complexes into the nucleus [188–196], the relevance of these findings for in vivo gene transfer in humans is questioned. Recent studies have focused on formulations of condensed, unimolecular DNA/poly(L-lysine) complexes that efficiently enter the cell via receptor-mediated endocytosis [77,78]. Such complexes consist of a single molecule of plasmid DNA and are spheroids approximately 15–20 nm in diameter; these preparations achieve efficient and specific gene transfer following intravenous gene delivery. For example, condensed, unimolecular galactosylated DNA/poly(L-lysine) complexes encoding human factor IX cDNA efficiently target the hepatic asialoglycoprotein receptor, and transfected rats have detectable human factor IX in their serum for up to 140 days [78]. This result was achieved without the need for partial hepatectomy. Condensed DNA/poly(L-lysine) complexes have also been prepared by coupling the FAB fragment of an antibody recognizing the polymeric immunoglobulin receptor [79]. These complexes have a diameter of approximately 25 nm and yield efficient gene transfer into target rat lung epithelial cells following intravenous administration. Approximately 18% of tracheal epithelial cells were transfected as monitored by expression of the beta-galactosidase marker gene following a single intravenous injection of 300 μg of plasmid DNA formulated in these condensed complexes [79]. Expression was specific for tissues expressing the polymeric immunoglobulin receptor. In other studies, the mannose receptor on macrophages has been targeted for in vivo gene delivery by formulating condensed mannosylated DNA/poly (L-lysine) complexes [80]. In these studies, efficient and specific gene transfer was shown to correlate with the formulation of unimolecular, condensed DNA/poly(L-lysine) complexes. Recently, the serpin enzyme complex receptor (SECR) also has been targeted for gene transfer using con-

37

FIGURE 1 Electron micrograph of condensed DNA complexes in normal saline. A 6.7-kbp plasmid was compacted using a polyethylene glycolsubstituted polymer consisting of cysteine followed by 30 lysines. Spheroidal particles were observed having an average size of 20 ± 2.5 nm (as assayed by dynamic light scattering analysis). The bar represents 100 nm.

densed DNA/poly(L-lysine) particles [197–199]. Gene transfer in vitro correlated closely with the level of cell surface SECR expression, and gene expression in vivo following an intravenous injection correlated with SECR-expressing tissues. Together, these studies suggest that coupling poly (L-lysine) to ligands that recognize cellular receptors preferentially expressed by tumor cells may provide an efficient and specific approach for in vivo gene transfer of plasmid vectors into cancer cells. Recently, unimolecularly compacted plasmid DNA complexes have been optimized for stability in physiologic saline and serum at 37◦ C [200]. These complexes consist of a single molecule of DNA and sufficient polylysine carrier molecules to prepare essentially charge-neutral particles. Based on electron microscopy (Fig. 1) and dynamic light scattering, these complexes have the minimum possible size as predicted by the partial specific volume of DNA and polycation [201]. These preparations of compacted DNA readily transfect nondividing, postmitotic cells [181] and yield very high levels of transgene expression when directly instilled into the lung [202]. Modifications of these complexes to include ligands for receptors that are highly and preferentially expressed by tumor cells may result in an effective and nontoxic gene transfer platform for systemic cancer therapy.

V. PLASMID EXPRESSION VECTORS Unlike viral-based infectious vectors, plasmid vectors must be introduced into cells by specific gene transfer technologies, as reviewed earlier. Once introduced into a cell, however, plasmids have specific advantages compared to viral vectors, including: (1) no potential to be infectious;

38

Mark J. Cooper

(2) levels of gene expression per cell equivalent to other viral vectors that persist as extrachromosomal elements (see Table 1); (3) lack of immunogenicity (allowing for multiple treatments) [141]; (4) lack of toxicity following intravenous injection [142]; (5) low probability of integration during transient periods of expression, thereby reducing potential for insertional mutagenesis; (6) easy coupling to liposome or receptor-mediated gene delivery systems; and (7) long-term stability, requiring no special preparation or storage requirements. Modifications in vector design, including tissue-specific promoters, inducible promoters, and elements enabling the plasmid to replicate extrachromosomally in tumor cells, further enhance the safety of plasmid vectors and significantly augment the level of expression observed in transiently transfected tumor cells.

A. Tissue-Specific Promoters The cytomegalovirus (CMV) immediate-early promoter is often utilized in gene therapy studies due to its high level of activity in diverse tissue types [203,204]. Although it is desirable to express target genes at high levels in tumor cells, transcriptionally active promoters, such as CMV, will also direct high-level expression in unintentionally transfected normal cells following in vivo gene transfer. To approach current limitations in the ability to specifically target a tumor cell for gene transfer, tissue-specific promoters can be employed that limit expression of the therapeutic gene to tumor cells and normal cells of a specific lineage. Many tissue-specific promoters have been developed [205,206], and a short list includes the insulin promoter (β islet cells of the pancreas) [207], elastase promoter (acinar cells of the pancreas) [208], whey acidic protein promoter (breast) [209], tyrosinase promoter (melanocytes) [210], tyrosine hydroxylase promoter (sympathetic nervous system) [211], neurofilament protein promoter (brain neurons) [212], glial fibrillary acidic protein promoter (brain astrocytes) [213], Ren-2 promoter (kidney) [214], collagen promoter (connective tissues) [215], α-actin promoter (muscle) [216], von Willebrand factor promoter (endothelial cells) [217], α-fetoprotein promoter (hepatoma) [218], albumin promoter (liver) [218], surfactant promoter (lung) [219], CEA promoter (gastrointestinal tract, tumors of colon, breast, lung) [220], uroplakin II promoter (bladder) [221], T-cell receptor promoter (T lymphocytes) [222], immunoglobulin heavy-chain promoter (B lymphocytes) [223], prostatic-specific antigen promoter (prostate) [224], and protamine promoter (testes) [225]. Tissue-specific promoters have been utilized in gene therapy studies to evaluate tumor-specific killing mediated by expression of the herpes simplex thymidine kinase gene followed by exposure to ganciclovir. For example, use of the albumin and α-fetoprotein promoter in retroviral constructs encoding HSV-TK specifically killed hepatoma cell lines but had marginal activity in other tumor cells derived from breast,

colon, or skin [218]. In other studies, Vile and Hart recently reported use of plasmid DNA encoding HSV-TK transcriptionally regulated by the murine tyrosinase promoter to treat B16 melanoma tumors growing as subcutaneous explants in syngeneic mice [136]. Established tumors, approximately 4 mm in diameter, were directly injected with 20 μg of calcium-phosphate-precipitated plasmid DNA, and 2 days later mice were administered daily injections of intraperitoneal ganciclovir for 5 days. A statistically significant reduction in tumor size was observed compared to animals not receiving ganciclovir. No local toxicity was observed in the tissues adjacent to the tumor explant, as expected based on the tissue specificity of the tyrosinase promoter. In similar studies, the CEA promoter also has been utilized to control transcription of HSV-TK [220]. CEA-expressing lung cancer cell lines were highly sensitive to ganciclovir in vitro and in vivo following gene transfer of these constructs, whereas non-CEA-expressing lung cell lines were resistant to ganciclovir following gene transfer. Another opportunity to specifically target tumor cells for gene expression is to utilize promoter elements that become activated in chemotherapy-resistant tumor cells. Based on the observation that the metallothionein promoter becomes activated in cisplatin-resistant ovarian carcinoma cells, plasmid DNA encoding the HSV-TK gene transcriptionally controlled by the metallothionein promoter has been introduced into cisplatin-sensitive and -resistant ovarian carcinoma cell lines followed by treatment with ganciclovir [226]. No cytotoxicity was apparent in cisplatin-sensitive, parental 1A9 ovarian carcinoma cells, whereas a cisplatin-resistant subclone was efficiently killed by this treatment. These results suggest a specific approach for gene therapy of cisplatinresistant ovarian carcinoma cells and underscore the potential of using tumor-specific promoter elements.

B. Inducible Promoters In addition to using tissue-specific promoters to minimize target gene expression in unintentionally transfected cells, the timing and duration of gene expression also can be modulated by employing inducible promoters that can be externally controlled. Several inducible systems have been developed, and a few appear to be appropriate for use in clinical gene therapy trials due to lack of apparent toxicity and demonstrated effectiveness in vivo. For example, a tetracycline-controlled expression system has been developed by Gossen and Bujard [227]. A novel hybrid transcriptional transactivation protein was constructed by ligating the ligand and DNA binding domains of the bacterial tetracycline repressor gene to the C-terminal region of the herpes virus VP16 transcriptional regulator protein containing its transactivation domain. In conjunction with reporter genes containing a heptad repeat of the consensus binding domain of the tetracycline repressor upstream of a minimal core


element of the cytomegalovirus immediate-early promoter, tetracycline-controlled expression has been demonstrated in vitro and in vivo in transgenic mice [227–229]. The hybrid transcriptional transactivator binds to the tet operon in the absence of tetracycline, whereas tetracycline efficiently dissociates the transcription factor from its binding site. Hence, efficient reporter gene expression was observed in the absence of tetracycline, whereas transcription is virtually eliminated in the presence of 0.1–1 μg/mL of tetracycline, a concentration readily attainable in humans. This system has also been used to transiently express target genes following direct in vivo gene transfer of these plasmid constructs in rat myocardium [230]. More recently, a tetracycline-on system has been developed utilizing specific point mutations in the tetracycline repressor component of the hybrid transcriptional transactivator [231]. In other studies, tetracycline-controlled transciptional repressors have been constructed by linking the KRAB transciptional repressor downstream from the DNA binding domain of the tetracycline repressor [232]. O’Malley and colleagues have also described a novel, regulated transcriptional activator that consists of a truncated ligand binding domain of the human progesterone receptor (which binds tightly to the synthetic progesterone antagonist RU486 but binds very poorly to progesterone), the DNA binding domain of the yeast transcriptional activator GAL4, and a C-terminal fragment of the herpes simplex VP16 transcriptional regulator protein [233]. In conjunction with a target gene containing four copies of the consensus GAL4 binding site, gene expression was activated only in the presence of RU486, and regulation was achieved both in vitro and in vivo [233,234]. A similar gene switch has been developed by Delort and Capecchi that utilizes different domains of the progesterone receptor and GAL4 binding protein [235]. Wang et al. also have developed an inducible repressor system by substituting the KRAB transcriptional repressor domain for the VP16 transactivation domain [236]. The specificity of these inducible systems is dependent upon the presence of the GAL4 consensus sequence upstream of the target gene of interest. Because GAL4-activated genes are not currently known to be present in the human genome, induction of gene expression in vivo is predicted to solely activate the therapeutic target gene. In addition, the presence of endogenous progesterone receptors in tumor cells would not be expected to interfere with this expression system. Other inducible transcriptional activation systems have been developed to control gene expression. These include the Drosphila ecdysone receptor gene switch and the rapamycincontrolled transactivation system [237,238]. The latter system utilizes two transcription factor fusion proteins that share a high-affinity binding site for rapamycin. The first element consists of the rapamycin binding protein, FKBP12, fused to the ZFHD1 DNA binding protein. The second element consists of a rapamycin binding protein, FRB, fused to the carboxy terminal portion of the NF-κB transcriptional acti-

39

vator protein. In the presence of rapamycin, these two fusion proteins bind to one another and reconstitute an active transcription factor for target reporter genes placed upstream of a minimal CMV promoter region carrying 12 binding sites for ZFHD1. Highly efficient and specific rapamycin-controlled gene expression has been demonstrated both in vitro and in in vivo preclinical models [238]. Several issues need to be addressed when considering any of these systems for cancer gene therapy. Although predicted to specifically inactivate or activate the transcription of target genes downstream from their respective consensus binding sequences in the presence of drug, further experimental testing is required to confirm that endogenous cellular genes, such as tumor suppressor genes and proto-oncogenes, are not unexpectedly regulated by these hybrid transcriptional repressors and transactivators, respectively. In addition, these hybrid transcriptional control proteins may very well generate antigenic peptide sequences derived from the bacterial tetracycline repressor, the yeast Gal4 protein, and the herpes simplex virus VP16 protein. An immune response may therefore be generated against tumor and normal cells following in vivo gene transfer. Although the toxicity of this immune response may be minimal, it may conceivably limit the duration of target gene expression in tumor cells following repetitive treatments. Another example of an inducible promoter system utilizes transcriptional control elements that become active following radiation-induced injury. As developed by Weichselbaum and colleagues, the radiation-responsive consensus sequence from the early growth response (EGR-1) gene promoter was ligated upstream from a gene known to significantly enhance radiation injury, TNF-α [239]. This plasmid construct was electroporated into a hematopoietic cell line, HL525, known to be deficient in radiation-induced expression of TNF-α. These gene-modified HL525 cells were injected into established radiation-resistant human squamous carcinoma xenografts in nude mice. Following radiation exposure to the tumor explant, the squamous carcinomas regressed and most of the animals were apparently cured. In contrast, control animals bearing squamous tumor explants that received radiation therapy alone, radiation plus HL525 cells transfected with the neomycin resistance gene, or TNF-α transfected HL525 cells without radiation all developed progressive tumor growth. In other studies, a quartad repeat of specific transcriptional elements within the EGR-1 gene has been combined with the CMV immediate-early core promoter to produce a radiation-responsive chimeric promoter [240]. These studies demonstrate the ability to induce gene expression in vivo by focused application of radiation and gene therapy in specific areas known to be involved by tumor. Several groups have developed inducible transcriptional regulators that are activated by endogenous metabolic conditions or physical stimuli that can be directly applied to the

40

Mark J. Cooper

tumor. Because tumor masses often have hypoxic regions [241,242], several groups have developed tumor-activated transcriptional regulators that are stimulated by low oxygen content [243,244]. A number of genes are upregulated by hypoxia, including erythropoietin, vascular endothelial growth factor, and some glycolytic enzymes, and specific hypoxia response elements (HREs) have been identified in 5 or 3 flanking regions [245]. Hypoxia-induced transcriptional regulators have been developed using five copies of HREs in conjunction with minimal promoters from E1b or CMV [244,245]. Using the CMV chimeric system, hypoxic conditions induced over a 500-fold increase in gene expression, achieving a level of gene expression comparable to the complete CMV promoter [245]. Using an alternative approach, transcriptional regulators have been developed based on activation of endogenous heat shock genes by hyperthermia [246]. Employing a heat shock gene (hsp70) promoter, transgene expression was induced over 10,000-fold in tissue culture cells by a temperature elevation of 39◦ –43◦ C [247]. Moreover, growth of syngeneic melanoma tumor explants was delayed following a local injection of adenoviral vectors encoding IL-12 regulated by hsp70 only in limbs treated with hyperthermia. Finally, cytoplasmic expression systems have been developed that utilize bacteriophage T7 RNA polymerase to control transcription of transgenes regulated by the T7 promoter. As reported by Gao and Huang, co-delivery of purified T7 RNA polymerase and a plasmid containing a T7 promoter upstream of the CAT reporter gene resulted in rapid, highlevel transgene expression that lasted about 30 hours [248]. Of note, this cytoplasmic expression system does not require plasmid DNA to enter the nucleus to be transcribed. Further modification of this T7 expression system, as reported by Gao et al. [249] and Chen et al. [250], utilized a bicistronic plasmid containing a first T7 promoter upstream of the T7 RNA polymerase gene (T7/T7 autogene) and a second T7 promoter upstream of various reporter genes. In this fashion, T7 RNA polymerase protein initiates transgene transcription, and newly synthesized T7 mRNA replenishes and maintains levels of T7 RNA polymerase. In these studies, expression of transfected reporter genes was maintained for up to 5–6 days in vitro. Furthermore, direct injection of a T7/T7 autogene vector system into mouse liver, muscle, brain, and connective tissue generated up to 200-fold higher levels of luciferase reporter activity than nuclear gene expression vectors [251]. In subsequent studies, T7 autogene bicistronic vectors encoding HSV-TK have been introduced into established human 143B osteosarcoma xenografts in nude mice, and tumor regression was observed in animals receiving intraperitoneal doses of ganciclovir [252]. More recently, the requirement for co-delivery of T7 RNA polymerase protein was bypassed by designing a plasmid vector incorporating the CMV and T7 promoters upstream of the T7 RNA polymerase gene [253]. In this system, initial CMV-based transcription of T7 RNA polymerase results in further enhancement of T7 polymerase

expression by autoregulating the T7 promoter, and high levels of reporter gene expression were maintained for at least 7 days post gene transfer.

C. Replicating Plasmid Vectors: Episomes Expression of genes encoded by plasmids is generally transient, unless specific modifications are made to enable the plasmid to efficiently integrate into genomic DNA or to replicate in human cells. In dividing tumor cells, plasmidmediated gene expression falls to very low levels by several days after gene transfer. This decline in gene expression is mediated by several factors, including a logarithmic decline in the percentage of transfected cells during replication of the target population (as the plasmid does not replicate in human cells) [254], potential loss of the transgene by nuclease destruction or by partitioning to non-nuclear compartments, and promoter inactivation by cytokines, chromatin remodeling, or methylation [255–262]. One approach to maintain plasmid copy numbers in transfected tumor cells is to incorporate sequences from human DNA that enable the plasmid to replicate extrachromosomally. Although sequence-specific human DNA origins have been difficult to clone, Calos and colleagues have identified DNA fragments that replicate semiconservatively during S phase of the cell cycle when incorporated into plasmid vectors [263,264]. These vectors replicate once per cell cycle, and the plasmid copy number per cell is therefore dependent upon the initial transfection conditions. In these studies, the size of the DNA fragment is an important factor in conferring replication competence, with random human DNA fragments over 10–15 kb in length having significant activity [265]. Similar sizes of randomly chosen yeast DNA also are replication competent in human 293 cells [266], and large fragments of bacterial DNA have detectable although minor activity [263]. Plasmids containing these DNA fragments will replicate for several months in human cells if the vector additionally includes a portion of the Epstein–Barr virus (EBV) DNA origin (including a tandem array of repeated sequences) and if the transfected cells express the EBV early gene product, EBNA-1. EBNA-1 binds to these tandem repeat sequences and retains plasmid DNA in the nucleus of dividing cells, thereby conferring stable maintenance of the episomal plasmid [267]. In short-term assays, however, these DNA fragments alone enable plasmids to replicate transiently in human cells over several generations, although the copy number of these vectors is low [263]. The expression characteristics of plasmids containing such autonomously replicating human sequences and the potential role of these vectors for cancer gene therapy are currently undefined. In other studies, human artificial chromosomes have been assembled in vivo by transfecting cells with specific fragments of telomeric and centromeric DNA [268–273]. These separate DNA fragments recombine within the cell to form


mini-chromosomes approximately 6–10 Mb in size, and stable vertical transfer of these extrachromosomal elements has been demonstrated over multiple generations in vitro. These properties make them well suited for introduction into human stem cells, including ex vivo gene transfer into hematopoietic progenitor cells. However, the ability to isolate large quantities of homogeneous, unrearranged artificial chromosomes, transfect them into human cells, and then achieve transfer into the nucleus remains to be demonstrated. Another approach to increase both the peak level and duration of gene expression mediated by plasmid vectors is to include sequences from DNA viruses that enable the plasmid to replicate in human cells. Two elements are required: (1) a viral DNA origin of replication and (2) a viral early gene product. The viral DNA origin alone is not functional in human cells. During the life cycle of DNA viruses, including Epstein–Barr virus and BK virus, an early gene product is synthesized that directly binds to the viral DNA origin [274,275]. This protein/DNA complex is recognized by the infected human cell as a functional DNA origin, and the virus is able to replicate its DNA. In a similar fashion, plasmids encoding a viral DNA origin and its corresponding early gene product can replicate in human cells. Replicating episomal plasmid vectors have two predicted advantages compared to standard plasmid vectors for cancer gene therapy applications: (1) high-level gene expression due to vector amplification and (2) maintenance of gene expression in transiently transfected cells due to efficient vertical transfer of the episome during tumor cell division. These principles are summarized in Table 3 and illustrated in Fig. 2. Plasmid vectors that replicate in human cells have been constructed from several viruses, including Epstein–Barr virus, BK virus, human papillomavirus, and SV40 [274–277]. For example, EBV episomes replicate in lymphoid cells, achieving a steady-state copy number of approximately 10– 50 copies [274]. These plasmids can be stably maintained in cells for many months, and EBV-based vectors containing over 200-kbp inserts have been characterized [278–281]. Constructs derived from BK virus replicate in a wide range of cell types [275,283–284], and stable bladder cell transfectants have been characterized that have approximately 150 copies per cell [284]. In these studies, gene expression was proportional to the episomal plasmid copy number. Additionally, gene expression was maintained in a population of unselected, transiently transfected cells for at least one week TABLE 3 Features of Standard Plasmid and Replication-Competent Episomal Vectors Peak level of gene expression

Sustained expression in dividing tumor cells

Standard plasmid

Low

No

Replication-competent episome

High

Yes

Expression vector

41

following gene transfer, whereas nonreplicating plasmidbased gene expression fell exponentially at a rate predicted by the doubling time of these cells. In studies by Thierry et al., mice receiving intravenous injections of liposome complexes of BK virus episomes generated transgene expression in multiple tissues up to 3 months post-injection; in contrast, a short duration of expression was observed in animals dosed with nonreplicating plasmids [285]. In summary, the predicted advantages of high-level, maintained gene expression of replicating episomal vectors compared to standard plasmids have been observed using BK virus episomes. A key distinction among the multiple types of episomal plasmids derived from DNA viruses is their ability to replicate once or multiple times per cell cycle. Some episomal plasmids, including those derived from EBV and BK virus [274,284], replicate once per cell cycle. In this circumstance, the plasmid copy number per cell will never be higher than the level achieved on the day of gene transfer. In circumstances where in vivo gene transfer is desired, this replication feature significantly limits transgene expression, as only one or several plasmids per transfected cell nucleus likely can be attained. In contrast, episomes derived from SV40 virus replicate multiple times per cell cycle [286]. In this scenario, transgene expression is not limited by the initial gene transfer efficiency, and episomal replication can generate high-level plasmid copy numbers in dividing cells, thereby optimizing transgene expression. Despite the clear advantages of replicating plasmid vectors, a significant obstacle to their development is the transformation properties associated with suitable viral early genes that possess replication transactivator function. For example, the Epstein–Barr virus replication transactivator, EBNA-1, has transformation properties in transgenic mice [287]. In addition, papovavirus early gene products, including the large T antigens from BK virus and SV40 virus, have transformation properties thought to be primarily mediated by binding to host tumor suppressor gene products, including p53, RB, and RB-related proteins, such as p107 and p130 [288–291]. To develop replicating episomal vectors for human gene therapy, our laboratory has recently developed a safetymodified, SV40, large T antigen (107/402-T) that lacks detectable binding to human tumor suppressor gene products yet preserves replication competence (Fig. 3A) [286]. This large T antigen mutant has specific point mutations in codons 107 and 402 and lacks detectable binding to p53, RB, and p107 proteins (Fig. 3B and C, Table 4). Episomal vectors incorporating the 107/402-T replicon amplify in a wide range of human and simian cell lines but not in dog or rodent cells (Table 5). In addition to gene transfer in vitro, we have observed that 107/402-T-based episomal vectors replicate in human tumor cells following direct in vivo gene transfer into human tumor xenografts in nude mice. An example of the replication activity of 107/402-T episomes in human hepatoma (Hep G2) and bladder (HT-1376) cell lines is shown in Fig. 3D and E. Based on transient gene transfer

42

Mark J. Cooper

FIGURE 2 (A) Replicating episomal plasmids yield high levels of target gene expression due to vector amplification. Depicted are multiple copies of an episomal plasmid in the nucleus of a transfected cell which have accumulated because of vector replication. The increased copy number of the expression vector produces high levels of target gene mRNA and consequently high levels of target gene protein. (B) High-level gene expression is maintained in transiently transfected tumor cells due to efficient vertical transfer of the episome (•) as these cells divide.

efficiencies and content of genomic DNA per cell, 107/402-T episomes achieve peak copy numbers of approximately 1400 in HT-1376 cells, and 25,000 in Hep G2 cells (Table 5). As a consequence of vector amplification, we observe significantly enhanced levels (>100-fold) of reporter gene ex-

pression when comparing this replicating episomal expression system to analogous, nonreplicating expression vectors (Fig. 4). When transferred into log-phase tumor cells, high levels of reporter gene expression was maintained for at least 1–2 weeks due to efficient vertical transfer of the


TABLE 4 Binding of Wild-Type and Mutant SV40 Large T Antigens to RB, p107, and p53 Tumor Suppressor Gene Products Tumor suppressor gene

Ta

107-T

402-T

107/402-T

RB

100

0.03

67

0.07

p107

100

0

79

0

p53

100

36.2

0

0

a Shown

is the percentage of binding of T antigen mutants compared to wild-type T antigen. Source: From Cooper, M. J. et al. (1997). Proc. Natl. Acad. Sci. USA 94, 6450–6455. With permission.

replicating episomal expression vectors. In ongoing studies, we have developed an externally controlled replicon switch that employs a novel fusion gene consisting of 107/402-T and a portion of the human progesterone receptor [292].

43

Vector amplification occurs only in the presence of RU486, an FDA-approved synthetic progesterone antagonist. Administration of RU486 to the cancer patient on the day of gene transfer may be sufficient to permit a short burst of vector amplification, thereby boosting levels of transgene expression.

VI. FUTURE DIRECTIONS Clinical cancer gene therapies have only recently been initiated, and results are currently very preliminary. At present, the optimal delivery system, expression vector, and target genes for a given tumor type are entirely unknown. Success of this modality will ultimately depend upon the ability to express the therapeutic gene of interest at high levels, and being able to target the tumor cell for gene delivery will minimize toxicities. Incorporation of tissue-specific and inducible

FIGURE 3 107/402-T lacks binding to human tumor suppressor genes and is replication competent. (A) Point mutations in replication-competent, safety-modified, SV40, large T antigen mutants. Highlighted are domains of T antigen that bind to RB, p53, and the SV40 DNA origin. The codon 107 mutation substitutes lysine for glutamic acid, and the codon 402 mutation substitutes glutamic acid for aspartic acid [286]. (B and C) Co-immunoprecipitation analysis of binding of wild-type and mutant T antigens to human tumor suppressor gene products. 2 × 105 dpm of in vitro translated T antigens were mixed with CV-1 extracts overproducing human RB protein and anti-RB monoclonal antibody G3-245 (B, lanes 3–6), p53 and anti-p53 monoclonal antibody 1801 (B, lanes 7–10), and pl07 and anti-p107 monoclonal antibody SD9 (C, lanes 3–6). As controls, wild-type T antigen is immunoprecipitated with either anti-chromogranin A monoclonal antibody LKH210 (lane 1) or anti-T-antigen monoclonal antibody 416 (lane 2). (D) 107/402-T is replication competent. Hep G2 hepatoma cells (D) were transfected with wild-type and mutant T antigen expression vectors, and total cellular DNA was harvested 2 days post transfection. DNA samples were sequentially digested with ApaI to linearize vector DNA, and then DpnI to distinguish amplified DNA from the input DNA used to transfect these cells. Because human cells lack adenine methylase activity, newly replicated DNA is resistant to digestion by DpnI. Hence, presence of unit-length, linearized plasmid DNA, as indicated by the arrow, demonstrates newly replicated episome. Hybridization probe: pRC/CMV.107/402-T. (E) To evaluate amplification of a cotransfected plasmid in concert with T antigen episomes, HT-1376 bladder carcinoma cells were transfected with T antigen expression vectors and a reporter replication plasmid containing the SV40 DNA origin, pSV2CAT. DNA harvested from cells 4 days post gene transfer was sequentially digested with BamHI to linearize pSV2CAT and then with DpnI. Hybridization probe: BamHI-HindIII CAT fragment. CMV, pRC/CMV transfectants (no T antigen); DC, DpnI digestion control consisting of 5 μg of genomic DNA and 2 ng of either pRC/CMV.107/402-T (D, lane 9) or pSV2CAT (E, lane 9). (From Cooper, M. J. et al. (1997). Proc. Natl. Acad. Sci. USA 94, 6450–6455. With permission.)

44

Mark J. Cooper

TABLE 5 Replication Activity of 107/402-T Based Episomes in Human and Animal Cell Lines Species

Cell line

Human

HT-1376 5637 MCF-7 T98G SW480 Hs68 Hep G2 NCI-H69 NCI-H82 NCI-H146 RAJI

Type

Copy number/cella

Bladder Bladder Breast Brain Colon Fibroblast Hepatoma Lung Lung Lung Lymphoma

1400 100,000 8600 25,000 78 82 25,000 9000 1200 2200 7000 11,000

Simian

CV-1

Kidney

Dog

MDCK-2 D17

Kidney Osteosarcoma

300

80

None

None

13/11

47.1

100

50

Complete

Note: pfu, plaque forming units; LN, lymph node; SC, subcutaneous; IL, interleukin; BCDT, BCNU + cicplatin + DTIC + tamoxifen; IFN, interferon; DCV, DTIC + cisplatin + vinblastine.

maximal antibody titers, we continued to be successful at achieving recombinant gene expression, measured both as vaccinia-encoded GM-CSF (V-GM-CSF) (RT-PCR using primers designed to specifically identify viral encoded GMCSF) and viral thymidine kinase gene (V-TK) mRNA expression (Fig. 6).

F. Intravesical Vaccinia in Patients with Bladder Cancer As the first step of our planned expansion of this strategy to the localized treatment of bladder cancer (our preclinical data demonstrated significant infection/transfection of the orthotopically growing murine bladder tumor MB49 following intravesical administration of recombinant vaccinia [117]), we have completed a phase I study of intravesical vaccinia vector in patients with advanced transitional cell carcinoma. As with our phase I of vector alone in melanoma [113], we used the vaccinia in a dose-escalation study, with each patient receiving three intravesical doses over a 2-week period. Given safety concerns, this study focused on patients with invasive transitional cell carcinoma scheduled for cystectomy the day following the third dose. Table 2 summarizes patient characteristics, doses employed, and toxicity. As noted in our prior clinical trials, patients developed high titers of antivaccinia antibody, although maximal titers were measured after cystectomy in patients given the shortened

course of therapy (data not shown). Also as noted earlier, treatment was associated with a significant degree of inflammation (Fig. 7; see also color insert) and recruitment of activated T lymphocytes (CD3+ , CD45RO+ ) (Fig. 8; see also color insert), as well as dendritic cells (Factor XIIIa+ ) that we feel will enhance prospects for the induction of immunity to tumor.

V. FUTURE DIRECTIONS We have demonstrated (1) that vaccinia virus can be effectively used to infect/transfect tumor cells in vivo in both preclinical melanoma and bladder systems, (2) that the vaccinia virus vector and recombinants can be given safely and with continued infectivity in patients despite preexisting or developing immunity to vaccinia, and (3) that vaccinia recombinants expressing the genes for a panel of cytokines effectively induce infected cells to produce high levels of biologically active cytokines. We have translated our preclinical results into the clinical use of a GM-CSF-encoding vaccinia recombinant given intralesionally to patients with melanoma and have seen encouraging clinical responses (long-term remission in two of seven patients and rejection of uninjected lesions in four of the seven; (see Table 1). Our initial decision to focus on in situ cytokine modulation in an attempt to enhance immune recognition was

216

Edmund C. Lattime et al.

FIGURE 4 Resolution of dermal metastases following intralesional injection of recombinant vaccinia–granulocyte– macrophage colony-stimulating factor (GM-CSF). Patient 3 is a 32-year-old female with extensive dermal metastases of the left thigh before treatment (A), on day 81 (B), and on day 600, 150 days following cessation of treatment (C). Regression was accompanied by gross (D) and histologic (E) evidence of inflammation, including significant T-cell (CD3+ ) involvement (F). (From Mastrangelo, M. J. et al., Cancer Gene Ther., 6(5):409–422, 1999. With permission.) (See color insert.)

based on our early laboratory studies demonstrating that in both melanoma and bladder tumors the tumor-host environment was rich in IL-10, an immune suppressive cytokine [54,129,130]. Having made this clinical observation, we have focused on demonstrating the feasibility of modulating the tumor-host environment with immune-enhancing cytokines

described in this chapter. While doing this, our laboratory studies focused on validating IL-10 and other suppressive molecules as targets for modulation [62]. We are now focused on the production and cloning of a new series of recombinants with the ability not only to instill positive mediators but also to neutralize the negative.

In Situ Immune Modulation Using Recombinant Vaccinia Virus Vectors: Preclinical Studies to Clinical Implementation

FIGURE 5 Regression of uninjected lesions accompanied by T-cell infiltration. A representative uninjected distant regressing lesion prior to (A) and following (B) patient treatment demonstrated T-cell (CD8) infiltration (C). (From Mastrangelo, M. J. et al., Cancer Gene Ther., 6(5):409–422, 1999. With permission.) (See color insert.)

FIGURE 6 High titers of antivaccinia antibody fail to prevent local infection/transfection following injection with vaccinia–GM-CSF recombinant. RT-PCR of mRNA from melanoma biopsies for vaccinia, thymidine kinase (V-TK), vacciniaencoded human GM-CSF (V-GMCSF), human GM-CSF (GMCSF) and actin control (β-actin). Biopsies from injected lesions (lanes 1–3) and an uninjected lesion (lane 4). Lane 1, biopsy 18 hours following the last of a series of multiple injections; lanes 2 and 3, biopsies 18 hours following a single injection; lane 4, uninjected lesion. All biopsies were taken from patient 3 at week 31. From J. Clin. Invest. 105(8); 1031, 2000. With permission.)

217

218


TABLE 2 Intravesical Vaccinia Vector Prior to Cystectomy in Patients with Muscle-Invasive Bladder Cancer Patient

Age/Sex

Dose per treatment (×106 pfu)

Toxicity

Bladder inflammation

1 2 3 4

57/F 36/M 64/M 52/M

1, 5, 10 10, 25, 100 25, 100, 100 25, 100, 100

Mild dysuria Mild dysuria Mild dysuria Mild dysuria

Slight Significant Significant Significant

VI. CONCLUSIONS In summary, we have developed an approach to immunologically based gene therapy logically designed from the requirements to generate a productive cellular immune response. We have translated the approach to clinical trials, with positive antitumor activity seen in a number of patients. As outlined in this review, numerous strategies have been hypothesized and tested in both preclinical and clinical settings with this goal as an endpoint. It is our hypothesis that in situ tumor transfection with cytokine genes will provide a logical extension of the vaccine strategies that have been previously studied. By incorporating genes selected based on their known contribution to the generation of systemic immune responses, we anticipate the ability to optimize the generation of an antitumor response. In addition to this logical in vivo vaccine design, this methodology will allow the generation of a single reagent in a bottle that will be of use

in any tumor type provided it is accessible to injection. This will preclude the need to have sufficient autologous tumor for harvest and subsequent vaccine production and will overcome the significant limitation of the in vitro transfectants for tumor transfection and selection in the lab. As noted above, the use of the patient’s own tumor as a source of antigens in our system optimizes the generation of a T-cell response and has significant advantages over allogeneic vaccine strategies that rely on shared antigens restricted by common MHC antigens.

Acknowledgments The authors wish to thank Arvin Yang and Faryal Mahmud for assistance in developing the graphics and text of this chapter. Our research is supported by ACS grants IM742 and EDT-78842; USPHS grants CA-42908, CA-55322, CA-69253, CA-74543; and the Nat Pincus Trust.

FIGURE 7 Intravesical administration of vaccinia vector in patients with invasive bladder cancer. H&E stained section of bladder from patient 2, taken at time of cystectomy 24 hours after the third of three intravesical instillations of vaccinia, shows widespread inflammation (A) and infection of the urothelium (B). (See color insert.)

In Situ Immune Modulation Using Recombinant Vaccinia Virus Vectors: Preclinical Studies to Clinical Implementation

219

FIGURE 8 Immunohistochemical staining of bladder from patient 2. Pretreatment biopsies (A, C, E) and post-treatment cystectomy sections (B, D, F) were stained for CD3 (A, B), CD45RO (C, D), and Factor XIIIa (dendritic cells) (E, F). (From Mastrangelo, M. J. et al., J. Clin. Invest., 105(8):1031, 2000. With permission.) (See color insert.)

References 1. Thomas, L. (1959). In Cellular and Humoral Aspects of the Hypersensitive States (H. S. Lawrence, ed.), pp. 529–532. Hoeber-Harper, New York. 2. Burnet, F. M. (1970). The concept of immunological surveillance. Prog. Exp. Tumor. Res. 13, 1. 3. Stutman, O. (1975). Immunodepression and malignancy. Adv. Cancer Res. 22, 261–422.

4. Prehn, R. T., and Main, J. M. (1957). Immunity to methylcholanthreneinduced sarcomas. J. Natl. Cancer Inst. 18, 769–778. 5. Bodurtha, A. J., Berkelhammer, J., Kim, Y. H., Laucius, J. F., and Mastrangelo, M. J. (1976). A clinical, histologic, and immunologic study of a case of metastatic malignant melanoma undergoing spontaneous remission. Cancer 37, 735–742. 6. Spontaneous Remission: An Annotated Bibliography, Institute of Noetic Sciences, Savsalito, CA, pp. 1–710. 7. Bornstein, R. S., Mastrangelo, M. J., Sulit, H., Chee, D., Yarbro, J. W.,

220

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21. 22. 23. 24.


Prehn, L. M., and Prehn, R. T. (1973). Immunotherapy of melanoma with intralesional BCG. Natl. Cancer Inst. Monogr. 39, 213–220. Laucius, J. F., Bodurtha, A. J., Mastrangelo, M. J., and Creech, R. H. (1974). Bacillus Calmette-Guerin in the treatment of neoplastic disease. J. Reticuloendothelial Soc. 16, 347–373. Lamm, D. L., Thor, D. E., Harris, S. C., Reyna, J. A., Stogdill, V. D., and Radwin, H. M. (1980). Bacillus Calmette-Guerin immunotherapy of superficial bladder cancer. J. Urol. 124, 38–42. Mastrangelo, M. J., Maguire, H. C. J., Lattime, E. C., and Berd, D. (1995). Whole cell vaccines, in Biological Therapy of Cancer, 2nd ed. (V. T. DaVita, S. Hellman, and S. A. Rosenberg, eds.), pp. 648–658. Lippincott, Philadelphia, PA. Mastrangelo, M. J., Sato, T., Lattime, E. C., Maguire, Jr., H. C., and Berd, D. (1998). Cellular vaccine therapies for cancer, in Biological and Hormonal Therapies of Cancer (K. A. Foon and H. B. Muss, eds.), pp. 35–50. Kluwer, Boston. Hu, X., Chakraborty, N. G., Sporn, J. R., Kurtzman, S. H., Ergin, M. T., and Mukherji, B. (1996). Enhancement of cytolytic T lymphocyte precursor frequency in melanoma patients following immunization with MAGE-1 peptide loaded antigen presenting cell-based vaccine. Cancer Res. 56, 2479–2483. Mule, J. J. (2000). Tumor vaccine strategies that employ dendritic cells and tumor lysates: experimental and clinical studies. Immunol. Invest. 29, 127–129. Geiger, J., Hutchinson, R., Hohenkirk, L., McKenna, E., Chang, A., and Mule, J. (2000). Treatment of solid tumours in children with tumourlysate-pulsed dendritic cells. Lancet 356, 1163–1165. Holtl, L., Rieser, C., Papesh, C., Ramoner, R., Herold, M., Klocker, H., Radmayr, C., Stenzl, A., Bartsch, G., and Thurnher, M. (1999). Cellular and humoral immune responses in patients with metastatic renal cell carcinoma after vaccination with antigen pulsed dendritic cells. J. Urol. 161, 777–782. Kugler, A., Stuhler, G., Walden, P., Zoller, G., Zobywalski, A., Brossart, P., Trefzer, U., Ullrich, S., Muller, C. A., Becker, V., Gross, A. J., Hemmerlein, B., Kanz, L., Muller, G. A., and Ringert, R. H. (2000). Regression of human metastatic renal cell carcinoma after vaccination with tumor cell-dendritic cell hybrids. Nat. Med. 6, 332–336. Ciborowski, P., and Finn, O. J. (1995). Recombinant epithelial cell mucin (MUC-1) expressed in baculovirus resembles antigenically tumor associated mucin, target for immunotherapy. Biomed. Peptides Proteins Nucleic Acids 1, 193–198. Bei, R., Kantor, J., Kashmiri, S. V., Abrams, S., and Schlom, J. (1994). Enhanced immune responses and anti-tumor activity by baculovirus recombinant carcinoembryonic antigen (CEA) in mice primed with the recombinant vaccinia CEA. J. Immunother. Emphasis Tumor Immunol. 16, 275–282. Livingston, P. O., Natoli, E. J., Jones Calves, M., Stockert, E., Oettgen, H. F., and Old, L. J. (1987). Vaccines containing purified GM2 ganglioside elicit GM2 antibodies in melanoma patients. Proc. Natl. Acad. Sci. USA 84, 2911–2915. Maloney, D. G., Liles, T. M., Czerwinski, D. K., Waldichuk, C., Rosenberg, J., Grillo-Lopez, A., and Levy, R. (1994). Phase I clinical trial using escalating single-dose infusion of chimeric anti-CD20 monoclonal antibody (IDEC-C2B8) in patients with recurrent B-cell lymphoma. Blood 84, 2457–2466. Vitetta, E., and Ghetie, V. (1994). Immunotoxins in the therapy of cancer: from bench to clinic. Pharmacol. Ther. 63, 209–234. Pai, L. H., and Pastan, I. (1994). Immunotoxins and recombinant toxins for cancer treatment. Important Adv. Oncol. 3–19. Green, M. C., Murray, J. L., and Hortobagyi, G. N. (2000). Monoclonal antibody therapy for solid tumors. Cancer Treat. Rev. 26, 269–86. White, C. A., Weaver, R. L., and Grillo-Lopez, A. J. (2001). Antibodytargeted immunotherapy for treatment of malignancy. Annu. Rev. Med. 52, 125–145.

25. Schwartz, R. H. (1992). Costimulation of T lymphocytes: the role of CD28, CTLA-4, and B7/BB1 in interleukin-2 production and immunotherapy. Cell 71, 1065–1068. 26. Disis, M. L., Smith, J. W., Murphy, A. E., Chen, W., and Cheever, M. A. (1995). In vitro generation of human cytolytic T-cells specific for peptides derived from the HER-2/neu protooncogene protein. Cancer Res. 54, 1071–1076. 27. Peace, D. J., Smith, J. W., Chen, W., You, S. G., Cosand, W. L., Blake, J., and Cheever, M. A. (1994). Lysis of ras oncogene-transformed cells by specific cytotoxic T lymphocytes elicited by primary in vitro immunization with mutated Ras peptide. J. Exp. Med. 179, 473– 479. 28. Peoples, G. E., Goedegebuure, P. S., Smith, R., Linehan, D. C., Yoshino, I., and Eberlein, T. J. (1995). Breast and ovarian cancer-specific cytotoxic T lymphocytes recognize the same HER2/neu-derived peptide. Proc. Natl. Acad. Sci. USA 92, 432– 436. 29. Van Elsas, A., Nijman, H. W., Van der Minne, C. E., Mourer, J. S., Kast, W. M., Melief, C. J., and Schreir, P. I. (1995). Induction and characterization of cytotoxic T-lymphocytes recognizing a mutated p21ras peptide presented by HLA-A*0201. Int. J. Cancer 61, 389– 396. 30. Storkus, W. J., Zeh, H. J., Maeurer, M. J., Salter, R. D., and Lotze, M. T. (1993). Identification of human melanoma peptides recognized by class I restricted tumor infiltrating T lymphocytes, J. Immunol. 151, 3719–3727. 31. Slingluff, C. L., Hunt, D. F., and Engelhard, V. H. (1994). Direct analysis of tumor-associated peptide antigens. Curr. Opin. Immunol. 6, 733–740. 32. Cox, A. L., Skipper, J., Chen, Y., Henderson, R. A., Darrow, T. L., Shabanowitz, J., Engelhard, V. H., Hunt, D. F., and Slingluff, C. L. (1994). Identification of a peptide recognized by five melanomaspecific human cytotoxic T cell lines. Science 264, 716–719. 33. Correale, P., Walmsley, K., Nieroda, C., Zaremba, S., Zhu, M., Schlom, J., and Tsang, K. Y. (1997). In vitro generation of human cytotoxic T lymphocytes specific for peptides derived from prostate-specific antigen. J. Natl. Cancer Inst. 89, 293–300. 34. Eder, J. P., Kantoff, P. W., Roper, K., Xu, G. X., Bubley, G. J., Boyden, J., Gritz, L., Mazzara, G., Oh, W. K., Arlen, P., Tsang, K. Y., Panicali, D., Schlom, J., and Kufe, D. W. (2000). A phase I trial of a recombinant vaccinia virus expressing prostate-specific antigen in advanced prostate cancer. Clin. Cancer Res. 6, 1632–1638. 35. Klein, J., and Sato, A. (2000). The HLA system [second of two parts]. N. Engl. J. Med. 343, 782–786. 36. Klein, J., and Sato, A. (2000). The HLA system [first of two parts]. N. Engl. J. Med. 343, 702–709. 37. Linsley, P. S., and Ledbetter, J. A. (1993). The role of the CD28 receptor during T cell responses to antigen. Ann. Rev. Immunol. 11, 191–212. 38. Lenschow, D. J., and Bluestone, J. A. (1993). T cell co-stimulation and in vivo tolerance. Curr. Opin. Immunol. 5, 747–752. 39. Gimmi, C. D., Freeman, G. J., Gribben, J. G., Gray, G., and Nadler, L. M. (1993). Human T-cell clonal anergy is induced by antigen presentation in the absence of B7 costimulation. Proc. Natl. Acad. Sci. USA 90, 6586–6590. 40. Janeway, C. A., and Travers, P. (1995). Immunobiology: The Immune System in Health and Disease, Garland Publishing, New York. 41. Finke, J. H., Rayman, P., Alexander, J., Edinger, M., Tubbs, R. R., Connelly, R., Pontes, E., and Bukowski, R. (1990). Characterization of the cytolytic activity of CD4+ and CD8+ tumor-infiltrating lymphocytes in human renal cell carcinoma. Cancer Res. 50, 2363–2370. 42. Li, W. Y., Lusheng, S., Kanbour, A., Herberman, R. B., and Whiteside, T. L. (1989). Lymphocytes infiltrating human ovarian tumors: synergy between tumor necrosis factor α and interleukin 2 in the generation of CD8+ effectors from tumor-infiltrating lymphocytes. Cancer Res. 49, 5979–5985.

In Situ Immune Modulation Using Recombinant Vaccinia Virus Vectors: Preclinical Studies to Clinical Implementation 43. Schoof, D. D., Jung, S.-E., and Eberlein, T. J. (1989). Human tumorinfiltrating lymphocyte (TIL) cytotoxicity facilitated by anti-T-cell receptor antibody. Int. J. Cancer 44, 219–224. 44. Balch, C. M., Riley, L. B., Bae, Y. J., Salmeron, M. A., Platsoucas, C. D., Von Eschenbach, A., and Itoh, K. (1990). Patterns of human tumor-infiltrating lymphocytes in 120 human cancers. Arch. Surg. 125, 200–205. 45. Haas, G. P., Solomon, D., and Rosenberg, S. A. (1990). Tumorinfiltrating lymphocytes from nonrenal urological malignancies. Cancer Immunol. Immunother. 30, 342–350. 46. Disis, M. L., and Cheever, M. A. (1998). HER-2/neu oncogenic protein: issues in vaccine development. Crit. Rev. Immunol. 18, 37–45. 47. Disis, M. L., and Cheever, M. A. (1996). Oncogenic proteins as tumor antigens. Curr. Opin. Immunol. 8, 637–42. 48. Zitvogel, L., Tahara, H., Robbins, P. D., Storkus, W. J., Clarke, M. R., Nalesnik, M. A., and Lotze, M. T. (1995). Cancer immunotherapy of established tumors with IL-12: effective delivery by genetically engineered fibroblasts. J. Immunol. 155, 1393–1403. 49. Kurane, S., Arca, M. T., Aruga, A., Krinock, R. A., Krauss, J. C., and Chang, A. E. (1997). Cytokines as an adjuvant to tumor vaccines: efficacy of local methods of delivery. Ann. Surg. Oncol. 4, 579–585. 50. Yamamura, M., Utemura, K., Deans, R. J., Weinberg, K., Rea, T. H., Bloom, B. R., and Modlin, R. L. (1991). Defining protective responses to pathogens: cytokine profiles in leprosy lesions. Science 254, 277– 279. 51. Scott, P., Pearce, E., Cheever, A. W., Coffman, R. L., and Sher, A. (1989). Role of cytokines and CD4+ T-cell subsets in the regulation of parasite immunity and disease [review]. Immunol. Rev. 112, 161–182. 52. Townsend, S. E., and Allison, J. P. (1993). Tumor rejection after direct costimulation of CD8+ T cells by B7-transfected melanoma cells. Science 259, 368–370. 53. Chen, L., Ashe, S., Brady, W. A., Hellstrom, I., Hellstrom, K. E., Ledbetter, J. A., McGowan, P., and Linsley, P. S. (1992). Costimulation of anti-tumor immunity by B7 counterreceptor for the T lymphocyte molecules CD28 and CTLA-4. Cell 71, 1093–1102. 54. Lattime, E. C., Mastrangelo, M. J., Bagasra, O., Li, W., and Berd, D. (1995). Expression of cytokine mRNA in human melanoma tissues. Cancer Immunol. Immunother. 41, 151–156. 55. Kruger-Krasagakes, S., Krasagakis, K., Garbe, C., Schmitt, E., Huls, C., Blankenstein, T., and Diamantstein, T. (1994). Expression of interleukin 10 in human melanoma. Br. J. Cancer 70, 1182–1185. 56. Chen, Q., Daniel, V., Maher, D. W., and Hersey, P. (1994). Production of IL-10 by melanoma cells: examination of its role in immunosuppression mediated by melanoma. Int. J. Cancer 56, 755–760. 57. Lattime, E. C., McCue, P. A., Keeley, F. X., Li, W., and Gomella, L. G. (1995). Expression of IL10 mRNA in biopsies of superficial and invasive TCC of the human bladder. Proc. Am. Assoc. Cancer Res. 36, 462. 58. Gastl, G. A., Abrams, J. S., Nanus, D. M., Oosterkamp, R., Silver, J., Liu, F., Chen, M., Albino, A. P., and Bander, N. H. (1993). Interleukin10 production by human carcinoma cell lines and its relationship to interleukin-6 expression. Int. J. Cancer 55, 96–101. 59. Pisa, P., Halapi, E., Pisa, E. K., Gerdin, E., Hising, C., Bucht, A., Gerdin, B., and Kiessling, R. (1992). Selective expression of interleukin 10, interferon γ , and granulocyte–macrophage colony-stimulating factor in ovarian cancer biopsies. Proc. Natl. Acad. Sci. USA 89, 7708–7712. 60. McAveney, K. M., Gomella, L. G., and Lattime, E. C. (1994). Induction of TH1 and TH2 associated cytokine mRNA in mouse bladder following intravesical growth of the murine bladder tumor MB49 and BCG immunotherapy. Clin. Immunol. Immunother. 39, 401–406. 61. Gorelik, L., Prokhorova, A., and Mokyr, M. B. (1994). Low-dose melphalan-induced shift in the production of a Th2-type cytokine to a Th1-type cytokine in mice bearing a large MOPC-315 tumor. Clin. Immunol. Immunother. 39, 117–126.

221

62. Halak, B. K., Maguire, Jr., H. C., and Lattime, E. C. (1999). Tumorinduced interleukin-10 inhibits type 1 immune responses directed at a tumor antigen as well as a non-tumor antigen present at the tumor site. Cancer Res. 59, 911–917. 63. Monken, C. E., Gomella, L. G., Li, W., Fink, E., and Lattime, E. C. (1996). IL10 is produced by human transitional cell carcinoma lines immortalized by retroviral transfection with human papilloma virus E6/E7 genes. Proc. Am. Assoc. Cancer Res. 37, 451. 64. Maguire, Jr., H. C., Ketcha, K. A., Halak, B. K., Holmes, K. L., and Lattime, E. C. (1997). Tumor-induced IL10 production in-vivo suppresses the development of delayed type hypersensitivity (DTH) to tumor associated antigens. Proc. Am. Assoc. Cancer Res. 38, 358. 65. Halak, B. K., Maguire, Jr., H. C., and Lattime, E. C. (1998). Tumorassociated IL-10 inhibits immune responses directed at a tumor specific antigen as well as a non-tumor antigen present at the tumor site. Proc. Am. Assoc. Cancer Res. 39, 652. 66. Ding, L., Linsley, P. S., Huang, L. Y., Germain, R. N., and Shevach, E. M. (1993). IL-10 inhibits macrophage costimulatory activity by selectively inhibiting the up-regulation of B7 expression. J. Immunol. 151, 1224–1234. 67. Steinbrink, K., Jonuleit, H., Muller, G., Schuler, G., Knop, J., and Enk, A. H. (1999). Interleukin-10-treated human dendritic cells induce a melanoma-antigen-specific anergy in CD8(+) T cells resulting in a failure to lyse tumor cells. Blood 93, 1634–1642. 68. Qin, Z., Noffz, G., Mohaupt, M., and Blankenstein, T. (1997). Interleukin-10 prevents dendritic cell accumulation and vaccination with granulocyte–macrophage colony-stimulating factor genemodified tumor cells. J. Immunol. 159, 770 –776. 69. De Smedt, T., Van Mechelen, M., De Becker, G., Urbain, J., Leo, O., and Moser, M. Effect of interleukin-10 on dendritic cell maturation and function. Eur. J. Immunol. 27, 1229–1235. 70. Fearon, E. R., Pardoll, D. M., Itaya, T., Golumbek, P., Levitsky, H. I., Simons, J. W., Karasuyama, H., Vogelstein, B., and Frost, P. (1990). Interleukin-2 production by tumor cells bypasses T helper function in the generation of an antitumor response. Cell. 60, 397–403. 71. Asher, A. L., Mulé, J. J., Kasid, A., Restifo, N. P., Salo, J. C., Reichert, C. M., Jaffe, G., Fendly, B., Kriegler, M., and Rosenberg, S. A. (1991). Murine tumor cells transduced with the gene for tumor necrosis factor-α: evidence for paracrine immune effects of tumor necrosis factor against tumors. J. Immunol. 146, 3227–3234. 72. Watanabe, Y., Kuribayashi, K., Miyatake, J., Nishihara, K., Nakayama, E., Taniyama, T., and Sakata, T. (1989). Exogenous expression of mouse interferon-gamma cDNA in mouse neuroblastoma C1300 cells results in reduced tumorigenicity by augmented anti-tumor immunity. Proc. Natl. Acad. Sci. USA 86, 9456–9460. 73. Tepper, R. I., Pattengale, P. K., and Leder, P. (1989). Murine interleukin4 displays potent anti-tumor activity in vivo. Cell 57, 503–512. 74. Connor, J., Bannerji, R., Saito, S., Heston, W., Fair, W., and Gilboa, E. (1993). Regression of bladder tumors in mice treated with interleukin 2 gene-modified tumor cells. J. Exp. Med. 177, 1127–1134. 75. Saito, S., Bannerji, R., Gansbacher, B., Rosenthal, F. M., Romanenko, P., Heston, W. D. W., Fair, W. R., and Gilboa, E. (1994). Immunotherapy of bladder cancer with cytokine gene-modified tumor vaccines. Cancer Res. 54, 3516–3520. 76. Dranoff, G., Jaffee, E., Lazenby, A., Golumbek, P., Levitsky, H., Brose, K., Jackson, V., Hamada, H., Pardoll, D. M., and Mulligan, R. C. (1993). Vaccination with irradiated tumor cells engineered to secrete murine granulocyte–macrophage colony–stimulating factor stimulates potent, specific, and long lasting anti-tumor immunity. Proc. Natl. Acad. Sci. USA 90, 3539–3543. 77. Simons, J. W., Mikhak, B., Chang, J. F., DeMarzo, A. M., Carducci, M. A., Lim, M., Weber, C. E., Baccala, A. A., Goemann, M. A., Clift, S. M., Ando, D. G., Levitsky, H. I., Cohen, L. K., Sanda, M. G., Mulligan, R. C., Partin, A. W., Carter, H. B., Piantadosi, S., Marshall, F. F., and

222

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88. 89. 90. 91.

92.


Nelson, W. G. (1999). Induction of immunity to prostate cancer antigens: results of a clinical trial of vaccination with irradiated autologous prostate tumor cells engineered to secrete granulocyte–macrophage colony-stimulating factor using ex vivo gene transfer. Cancer Res. 59, 5160–5168. Simons, J. W., Jaffee, E. M., Weber, C. E., Levitsky, H. I., Nelson, W. G., Carducci, M. A., Lazenby, A. J., Cohen, L. K., Finn, C. C., Clift, S. M., Hauda, K. M., Beck, L. A., Leiferman, K. M., Owens, Jr., A. H., Piantadosi, S., Dranoff, G., Mulligan, R. C., Pardoll, D. M., and Marshall, F. F. (1997). Bioactivity of autologous irradiated renal cell carcinoma vaccines generated by ex vivo granulocyte–macrophage colony-stimulating factor gene transfer. Cancer Res. 57, 1537– 1546. Nelson, W. G., Simons, J. W., Mikhak, B., Chang, J. F., DeMarzo, A. M., Carducci, M. A., Kim, M., Weber, C. E., Baccala, A. A., Goeman, M. A., Clift, S. M., Ando, D. G., Levitsky, H. I., Cohen, L. K., Sanda, M. G., Mulligan, R. C., Partin, A. W., Carter, H. B., Piantadosi, S., and Marshall, F. F. (2000). Cancer cells engineered to secrete granulocyte–macrophage colony-stimulating factor using ex vivo gene transfer as vaccines for the treatment of genitourinary malignancies. Cancer Chemother. Pharmacol. 46, S67–S72. Sigal, R. K., Lieberman, M. D., Reynolds, J. V., Williams, N., Ziegler, M. M., and Daly, J. M. (1990). Tumor immunization: improved results after vaccine modified with recombinant interferon gamma. Arch. Surg. 125, 308–312. Golumbek, P. T., Lazenby, A. J., Levitsky, H. I., Jaffee, L. M., Karasuyama, H., Baker, M., and Pardoll, D. M. (1991). Treatment of established renal cancer by tumor cells engineered to secrete interleukin4. Science 254, 713–716. Lattime, E. C., McCue, P. A., Ross, R. P., Baltish, M. A., and Gomella, L. (1992). T cells bearing τ/δ receptor in human transitional cell carcinoma of the bladder. Proc. Am. Assoc. Cancer Res. 33, 334. Perussia, B., Chan, S. H., D’Andrea, A., Tsuji, K., Santoli, D., Pospisil, M., Young, D., Wolf, S. F., and Trinchieri, G. (1992). Natural killer cell stimulatory factor or interleukin-12 has differential effects on the proliferation of TCRαβ+, TCRτ δ+ T lymphocytes and NK cells. J. Immunol. 149, 3495–3502. Basker, S., Glimcher, L., Nabavi, N., Jones, R. T., and OstrandRosenberg, S. (1995). Major histocompatibility complex class II+ B71+ tumor cells are potent vaccines for stimulating tumor rejection in tumor bearing mice. J. Exp. Med. 181, 619–629. Hodge, J. W., Rad, A. N., Grosenbach, D. W., Sabzevari, H., Yafal, A. G., Gritz, L., and Schlom, J. (2000). Enhanced activation of T cells by dendritic cells engineered to hyperexpress a triad of costimulatory molecules. J. Natl. Cancer Inst. 92, 1228–1239. Baldick, C. J. J., Keck, J. G., and Moss, B. (1992). Mutational analysis of the core, spacer, and initiator regions of vaccinia virus intermediateclass promoters. J. Virol. 66, 4710–4719. Cochran, M. A., Puckett, C., and Moss, B. (1985). In vitro mutagenesis of the promoter region for a vaccinia virus gene: evidence for tandem early and late regulatory signals. J. Virol. 54, 30–37. Fenner, F., Henderson, D. A., and Anita, I. (1988). Smallpox and Its Eradication, World Health Organization, Geneva. Williams, N. R., and Cooper, B. M. (1993). Counselling of workers handling vaccinia virus. Occup. Med. 43, 125–127. Smith, G. L., and Moss, B. (1983). Infectious poxvirus vectors have capacity for at least 25,000 base pairs of foreign DNA. Gene 25, 21–28. Earl, P. L., Moss, B., and Doms, R. W. (1991). Folding, interaction with GRP78-BiP, assembly, and transport of human immunodefficiency virus type 1 envelope protein. J. Virol. 65, 2047–2055. Scharenberg, A., Lin, S., Cuenod, B., Yamamura, H., and King, F. (1995). Reconstitution of interactions between tyrosine kinases and the high affinity IgE receptor which are controlled by recepter clustering. EMBO J. 14, 3385–3394.

93. Feng, Y., Broder, C. C., Kennedy, P. E., and Berger, M. (1996). HIV-1 entry cofactor: functional cDNA cloning of a seven transmembrane G protein-coupled receptor. Science 272, 872–877. 94. Doranz, B. J., Rucker, Yi, Y., Smyth, R. J., Samson, M., Peiper, S. C., Parmentier, M., Collman, R. G., and Domzig, W. (1996). A dual tropic primary HIV-1 isolate that uses fusin and the beta chemokine receptors CKR-5, CKR-3, and CKR-2b as fusion cofactors. Cell 85, 1149–1158. 95. Kapuiah, G., Woodhams, C. E., Blanden, R. V., and Ramshaw, I. A. (1991). Immunobiology of infection with recombinant vaccinia virus encoding murine IL2. J. Immunol. 147, 4327–4332. 96. Moss, B. (1996). Genetically engineered poxviruses for recombinant gene expression, vaccination, and safety. Proc. Natl. Acad. Sci. USA 93, 11341–11348. 97. Restifo, N. P. (1996). The new vaccines: building viruses that elicit antitumor immunity. Curr. Opin. Immunol. 8, 658–663. 98. Marshall, J. L., Hoyer, R. J., Toomey, M. A., Faraguna, K., Chang, P., Richmond, E., Pedicano, J. E., Gehan, E., Peck, R. A., Arlen, P., Tsang, K. Y., and Schlom, J. (2000). Phase I study in advanced cancer patients of a diversified prime-and-boost vaccination protocol using recombinant vaccinia virus and recombinant nonreplicating avipox virus to elicit anti-carcinoembryonic antigen immune responses. J. Clin. Oncol. 18, 3964–3973. 99. Conry, R. M., Khazaeli, M. B., Saleh, M. N., Allen, K. O., Barlow, D. L., Moore, S. E., Craig, D., Arani, R. B., Schlom, J., and LoBuglio, A. F. (1999). Phase I trial of a recombinant vaccinia virus encoding carcinoembryonic antigen in metastatic adenocarcinoma: comparison of intradermal versus subcutaneous administration. Clin. Cancer Res. 5, 2330–2337. 100. Cochran, M. S., Mackett, M., and Moss, B. (1985). Eukaryotic transient expression system dependent on transcription factors and regulatory DNA sequences of vaccinia virus. Proc. Natl. Acad. Sci. USA 82, 19–23. 101. Fuerst, T. R., Niles, E. G., Studier, F. W., and Moss, B. (1986). Eukaryotic transient-expression system based on recombinant vaccinia virus that synthesizes bacteriophage T7 RNA polymerase. Proc. Natl. Acad. Sci. USA 83, 8122–8126. 102. Rodriguez, D., Zhou, Y., Durbin, R. K., Jimenez, V., McAllister, W. T., and Esteban, M. (1990). Regulated expression of nuclear genes by T3 RNA polymerase and lac repressor using recombinant vaccinia virus vectors. J. Virol. 64, 4851–4857. 103. Perkus, M. E., Limbach, K., and Paoletti, E. (1989). Cloning and expression of foreign genes in vaccinia virus using a host range selection system. J. Virol. 63, 3829–3836. 104. Coupar, B. E. H., Andrew, M. E., and Boyle, D. B. (1988). A general method for the construction of recombinant vaccinia viruses expressing multiple foreign genes. Gene 68, 1–10. 105. Davison, A. J., and Moss, B. (1989). Structure of vaccinia virus early promoters. J. Mol. Biol. 210, 749–769. 106. Davison, A. J., and Moss, B. (1989). Structure of vaccinia virus late promoters. J. Mol. Biol. 210, 771–784. 107. Davison, A. J., and Moss, B. (1990). New vaccinia virus recombination plasmids incorporating a synthetic late promoter for high level expression of foreign proteins. Nucleic Acids Res. 18, 4285– 4286. 108. Merchlinsky, M., and Moss, B. (1992). Introduction of foreign DNA into the vaccinia virus genome by in vitro ligation: recombinationindependent selectable cloning vectors. Virology 190, 522–526. 109. Sutter, G., and Moss, B. (1992). Nonreplicating vaccinia vector efficiently expresses recombinant genes. Proc. Natl. Acad. Sci. USA 89, 10847–10851. 110. Carroll, M. W., and Moss, B. (1997). Host range and cytopathogenicity of the highly attenuated MVA strain of vacinia virus: propagation and generation of recombinant viruses in a nonhuman mammalian cell line. Virology 238, 198–211.

In Situ Immune Modulation Using Recombinant Vaccinia Virus Vectors: Preclinical Studies to Clinical Implementation 111. Paoletti, E. (1996). Application of pox virus vectors to vaccination: an update. Proc. Natl. Acad. Sci. USA 93, 11349–11353. 112. Tsung, K., Yim, J. H., Marti, W., Buller, M. L., and Norton, J. A. (1996). Gene expression and cytopathic effect of vaccinia virus inactivated by psoralen and long-wave UV light. J. Virol. 70, 165–171. 113. Mastrangelo, M. J., Maguire, Jr., H. C., McCue, P. A., Lee, S. S., Alexander, A., Nazarian, L. N., Eisenlohr, L. C., Nathan, F. E., Berd, D., and Lattime, E. C. (1995). A pilot study demonstrating the feasibility of using intratumoral vaccinia injections as a vector for gene transfer. Vaccine Res. 4, 55–69. 114. Mastrangelo, M. J., Maguire, Jr., H. C., Eisenlohr, L. C., Laughlin, C. E., Monken, C. E., McCue, P. A., Kovatich, A. J., and Lattime, E. C. (1999). Intratumoral recombinant GM-CSF-encoding virus as gene therapy in patients with cutaneous melanoma. Cancer Gene Ther. 6, 409–422. 115. Vanderplasschen, A., Hillinshead, M., and Smith, G. L. (1997). Antibodies against vaccinia virus do not neutralize extracellular enveloped virus but prevent virus release from infected cells and comet formation. J. Gen. Virol. 78, 2041–2048. 116. Lee, S. S., Eisenlohr, L. C., McCue, P. A., Mastrangelo, M. J., and Lattime, E. C. (1993). Intravesical gene therapy: vaccinia virus recombinants transfect murine bladder tumors and urothelium. Proc. Am. Assoc. Cancer Res. 34, 337. 117. Lee, S. S., Eisenlohr, L. C., McCue, P. A., Mastrangelo, M. J., and Lattime, E. C. (1994). Intravesical gene therapy: in-vivo gene transfer using vaccinia vectors. Cancer Res. 54, 3325–3328. 118. Lattime, E. C., Maguire, Jr., H. C., McCue, P. A., Eisenlohr, L. C., Berd, D., Lee, S. S., and Mastrangelo, M. J. (1994). Infection of human melanoma cells by intratumoral vaccinia, J. Invest. Dermatol. 102, 568. 119. Gomella, L. G., Mastrangelo, M. J., Eisenlohr, L. C., McCue, P. A., Lee, S. S., and Lattime, E. C. (1995). Localized gene therapy for prostate cancer: strategies for intraprostatic cytokine gene transfection using vaccinia virus vectors. J. Urol. 153, 308A. 120. Lattime, E., Eisenlohr, L., Gomella, L., and Mastrangelo, M. (1999). The use of vaccinia virus vectors for immunotherapy via in-situ tumor transfection, in Gene Therapy of Cancer: Translational Approaches from Preclinical Studies to Clinical Implementation (E. Lattime and S. Gerson, eds.), pp. 125–137. Academic Press, San Diego, CA.

223

121. Lee, S. S., Eisenlohr, L. C., McCue, P. A., Mastrangelo, M. J., Fink, E., and Lattime, E. C. (1995). In-vivo gene therapy of murine tumors using recombinant vaccinia virus encoding GM-CSF. Proc. Am. Assoc. Cancer Res. 36, 248. 122. Lee, S. S., Eisenlohr, L. C., McCue, P. A., Mastrangelo, M. J., and Lattime, E. C. (1994). Vaccinia virus vector mediated cytokine gene transfer for in vivo tumor immunotherapy. Proc. Am. Assoc. Cancer Res. 35, 514. 123. Ramshaw, I., Ruby, J., Ramsay, A., Ada, G., and Karupiah, G. (1992). Expression of cytokines by recombinant vaccinia viruses: a model for studying cytokines in virus infections in-vivo, Immunol. Rev. 127, 157– 182. 124. Mackett, M., Smith, G. L., and Moss, B. (1984). General method for production and selection of infectious vaccinia virus recombinants expressing foreign genes. J. Virol. 49, 857–864. 125. Elkins, K. L., Ennist, D. L., Winegar, R. K., and Weir, J. P. (1994). In-vivo delivery of interleukin-4 by a recombinant vaccinia prevents tumor development in mice. Hum. Gene Ther. 5, 809–820. 126. Whitman, E. D., Tsung, K., Paxson, J., and Norton, J. (1994). In vitro and in vivo kinetics of recombinant vaccinia virus cancer-gene therapy. Surgery 116, 183–188. 127. Qin, H., and Chatterjee, S. K. (1996). Recombinant vaccinia expressing interleukin-2 for cancer gene therapy. Cancer Gene Ther. 3, 163–167. 128. Lattime, E. C., Maguire, H. C. J., McCue, P. A., Eisenlohr, L. C., Berd, D., Lee, S. S., and Mastrangelo, M. J. (1994). Gene therapy using vaccinia vectors: repeated intratumoral injections result in tumor infection in the presence of anti-vaccinia immunity. Proc. Am. Soc. Clin. Oncol. 13, 397. 129. Lattime, E. C., Mastrangelo, M. J., and Berd, D. (1994). Human metastatic melanoma lesions and cell lines express mRNA for IL-10. Proc. Am. Assoc. Cancer Res. 35, 489. 130. Lattime, E. C., McCue, P. A., Keeley, F. X., Li, W., Baltish, M. A., and Gomella, L. G. Biopsies of superficial and invasive TCC of the bladder express mRNA for the immunosuppressive cytokine IL10. J. Urol. 153, 488a. 131. Chakrabarti, S., Brechling, K., and Moss, B. (1985). Vaccinia virus expression vector: coexpression of β-galactosidase provides visual screening of recombinant virus plaques. Mol. Cell. Biol. 5, 3403–3409.


C

H

A

P

T

E

R

13 The Use of Particle-Mediated Gene Transfer for Immunotherapy of Cancer MARK R. ALBERTINI

DAVID M. KING

ALEXANDER L. RAKHMILEVICH

The University of Wisconsin Comprehensive Cancer Center Madison, Wisconsin 53792



I. Introduction 225 II. Background 225 A. B. C. D.

interest; (3) transfect resting, nondividing cells, irrespective of cell lineage; and (4) simultaneously deliver multiple candidate therapeutic genes into one cell or neighboring cells. The latter feature of PMGT can be beneficial for genetic immunization strategies when a gene for a tumor-associated antigen is employed in combination with cytokine genes. Exciting preclinical results suggest potential clinical strategies to stimulate in vivo antitumor T-cell immunity. Several recently completed clinical studies have demonstrated the safety of ex vivo and in vivo PMGT. Rigorous clinical evaluation of PMGT, with careful immunological monitoring, is needed to determine the potential clinical importance of this technology.

Uses of PMGT 225 Technical Aspects 226 In Vivo Applications 226 Vector Considerations 228

III. Recent Advances

228

A. In Vitro Modification of Human Cells 228 B. Antitumor Efficacy in Murine Models 231 C. Antitumor Activity of a Canine Tumor Vaccine

233

IV. Issues Regarding Evaluation in Clinical Trials 234 A. General Considerations 234 B. Skin Penetration Considerations

V. Recent Clinical Trials

234

234

A. Ex Vivo PMGT 234 B. In Vivo PMGT 235

VI. Potential Novel Uses and Future Directions

II. BACKGROUND

235

Acknowledgments 235 References 235

A. Uses of PMGT Particle-mediated gene transfer provides a physical means for intracellular delivery of biologically active molecules. With the use of high-voltage electric discharge, PMGT was originally utilized for genetic transformation of various plants [1,2]. Because of its physical nature, this method displays properties distinct from characteristic properties of chemical and biological gene transfer agents [3]. During the past decade, PMGT technology has been utilized for in vivo transfection of various mammalian somatic tissues, including skin, liver, pancreas, muscle, spleen, and other organs [4,5]. In addition, PMGT has been utilized for ex vivo transfection of brain, mammary, and leukocyte primary cultures or tissue explants [6–9] as well as a wide range of in vitro cell lines [4,8–10]. A potential advantage of PMGT is its applicability to cells in vivo, which is the primary focus of this review.

I. INTRODUCTION The particle-mediated gene transfer (PMGT) technology is a relatively new approach for mammalian gene transfer. This method of gene delivery uses a burst of helium to accelerate DNA-coated gold particles into target cells. Resulting transgene expression levels are often significantly higher than those achieved by other direct DNA delivery methods. Moreover, it takes only several seconds to complete a single treatment by PMGT. Significant advantages of PMGT include the ability to (1) physically confer gene expression by nonviral means; (2) direct the particle delivery to the anatomic site of


225


226

Mark R. Albertini, David M. King and Alexander L. Rakhmilevich

B. Technical Aspects To achieve PMGT, microscopic gold particles are coated with plasmid DNA containing the gene(s) of interest and are accelerated by a motive force to sufficient velocities to penetrate the target cells and provide intracellular delivery of the transgene [11]. The amount of DNA required for PMGT is relatively low, and an amount of 5000 copies of cDNA delivered per cell was found to be optimal for commonly used reporter genes [4,5]. The motive force for PMGT was originally generated by high-voltage electric discharge [1,2]. Subsequently, a helium shock wave was produced by a handheld version of the original device (formerly the Accell gene delivery device; currently known as the Dermal PowderJectXR device, PowderJect Vaccines, Inc., Madison, WI; also commercially available for research purposes from Bio-Rad under the name Helios). DNA/gold particles are loaded in a small Teflon tube that acts as a cartridge; the research handheld helium -pulse transfection device can hold 12 cartridges in a revolving cylinder (Fig. 1A). Each gene transfer can be performed in less than 5 seconds, thus making PMGT an efficient method for repeated, multiple gene deliveries. The device designed for clinical applications accepts prefilled, single-use, disposable nozzles that can be manufactured under good manufacturing practice (GMP) procedures. The design of the clinical device utilized in a recent clinical trial at the University of Wisconsin is shown in Fig. 1B. Parameters for PMGT into skin that require consideration include the composition, size, and shape of the particles utilized for gene transfer. Particles made from dense materials, including gold, tungsten, iridium, and platinum, are all capable of effectively delivering DNA via PMGT. However, gold particles are most commonly chosen for two reasons: Elemental gold is chemically inert with no significant cytotoxic effects [12], and, owing to the common use of gold in the electronics industry, uniformly sized gold particles are commercially available. A wide range of gold particle sizes have been evaluated for PMGT; for mammalian skin tissues, the 2–3-μm gold particle size is frequently utilized [4,5,13]. Gold particles are readily available in different forms (e.g., as round particles, crystals, or even aggregates), for gene transfer into epidermal cells, crystal and spherical gold particles were found to be similarly effective in gene delivery. The DNA loading rate per particle, particle loading rate per target surface area, and the physical acceleration rate for particle penetration into skin are additional important parameters for PMGT into skin. More than 5000 copies of 5–10-kb plasmid DNA can be effectively coated onto a single 1-3-μm gold particle with a Ca2+ /spermidine or poly(ethylene glucol) (PEG) formulation in precipitated form [11]. With a predetermined gold particle loading rate of 0.1 mg/cm2 , approximately 1–2 gold particles (1–3 μm) per cell can be delivered via random distribution into epidermis containing stratified epithelial cells on the order of 15 μm in

diameter [13]. Excessive particle loading rates could cause trauma to transfected tissues. On the other hand, too low a particle load could result in low gene transfer efficiency. For transfecting murine skin tissues, a pressure of 300–500 psi for the helium pulse device has been found to confer high levels of transgene expression [13]. Another important feature of PMGT is the much reduced restriction on the size of the DNA vectors. Plasmid DNA, genomic DNA (∼23 kb), and reporter genes cloned in lambda phage genomic libraries (∼44 kb) can all be effectively delivered into mammalian cells by PMGT [6]. This capability offers new opportunities for transferring multiple genes, largesize genomic DNA sequences, or multiple tandem genes into mammalian somatic tissues. In addition, cotransfection of multiple genes on different plasmids has also been shown to be efficiently achieved by using the PMGT method [14–16]. Furthermore, RNA molecules can be similarly delivered as DNA vectors by PMGT [17]. The major current disadvantages of PMGT in vivo are the limited transfection efficiency for certain tissue systems (particularly if permanent gene transfer to the target cells is necessary) and the depth of tissue that can be accessed. Although transient transfection efficiencies from 3% to approximately 50% are possible in vitro [11], efficiencies for stable (i.e., integrative) gene transfer in vivo are apparently low and have not been clearly established in various transfected somatic tissues. Long-term transgene expression following PMGT has been observed in muscle and dermis, but these tissues seem to be the exception rather than the rule [13]. More general longterm gene expression may be possible through a combination of PMGT and replicating or actively integrating vector systems, but for the present it appears that the technique is most suitable in applications where short- to medium-term transgene expression is sufficient or desirable, such as DNA vaccine applications. The PMGT method at present also cannot deliver genes systemically to cell fractions scattered in large, three-dimensional tissues such as liver or brain, as can certain other gene transfer systems when administered through the circulatory system [18].

C. In Vivo Applications Among the applications of PMGT technology, transfection of skin tissues of live animals has resulted in some of the most interesting findings. High levels of transgene expression were first demonstrated by in vivo PMGT of skin epidermal tissues in rodents [5,19]. These results were highly reproducible for various large animals, including turkeys [20], rabbits [21], dogs [22], pigs [23], horses [24], and rhesus monkeys [25]. Safe and effective skin transfection by PMGT has also been recently demonstrated in humans [26]. Efficient delivery and expression of transgenes in skin tissues have been extended to several reporter genes [4,6], candidate tumor-antigen genes [27–30], cytokine genes [31,32], viral

The Use of Particle-Mediated Gene Transfer for Immunotherapy of Cancer

227

FIGURE 1 Design and operation of the PMGT device. (A) Diagram of the research hand-held helium pulse transfection device. (Artwork by Dr. K. Barton. From Rakhmilevich, A. and Yang, N., in Methods in Molecular Medicine. Vol. 35. Gene Therapy: Methods and Protocols, W. Walther and U. Stein, eds., Humana Press, Totowa, NJ, 2000. With permission.) (B) Diagram of the clinical device utilized in a recent clinical trial at the University of Wisconsin. (Artwork provided by PowderJect Vaccines, Inc., Madison, WI. With permission.)

antigen genes [25,33–36], and bacterial antigen genes [37,38], demonstrating the wide-ranging applicability of this gene transfer strategy. PMGT into skin tissues has been used for developing genetic immunization approaches [25,27– 29,34], gene therapy of subcutaneous tumors [31,32,39],

wound healing [40], delivery of RNA as transgenes or immunogens [17], analysis of transcriptional promoters and other regulatory sequences in gene expression vectors [5,41], and the study of various cell types in transgene expression and migration following DNA immunization [42,43].

228


Histological studies show that cDNA expression vectors can be introduced into and expressed in different cell layers of epidermal or dermal tissues by adjusting the ballistic variables. These include the motive force for particle acceleration, gold particle size, pretreatment and manipulations of epidermal and dermal tissues prior to transfection, and the DNA formulations for coating gold particles. In general, topical skin gene delivery results in high-level transient expression almost exclusively in the epidermal cell layers [4,5,39] which may vary considerably in different animals (e.g., 3–4 cell layers and 10–30 cell layers for mouse and pig skin, respectively). Treatment of dermal tissues via exposure of the underside of a skin flap can result in long-term transgene expression, though at a reduced level [5,31]. Because easily accessible epidermis is competent for eliciting both a humoral and a cell-mediated immune response and is efficient in the synthesis and secretion of transgenic proteins, it is an attractive gene transfer target for vaccination approaches.

D. Vector Considerations Several vector design strategies to enhance the antigenicity of target cells following PMGT include maximizing transgene expression and the use of multiple vectors or vectors encoding several transgenes. Additional strategies include the enhancement of proteosome-dependent protein degradation, production of secreted cytokine to enhance the immune response, and use of immunostimulatory sequences. The use of a strong promoter is important to optimize transgene expression in mammalian cells. Viral promoters from cytomegalovirus (CMV) and simian virus 40 are commonly used. In addition, incorporation of polyadenylation sequences can be used for stabilization of mRNA transcripts [44]. Plasmid vectors can be used to modify an immune response depending on whether the expressed protein is retained in the cell or is membrane bound or secreted. Incorporation of DNA sequences into the vector can direct intracellular processing to target major histocompatibility complex (MHC) class I or class II pathways [44,45]. N-terminal ubiquitination signals target expressed proteins to proteosomes for rapid cytoplasmic degradation and presentation via the class I pathway. The E3 leader sequence from adenovirus can be used to facilitate transport of antigens directly into the endoplasmic reticulum for class I presentation. Both these methods appear to increase cytotoxic T lymphocyte (CTL) activity following vaccination in vivo [42,46]. If the specific antigen for CTL activation is known, its gene can be incorporated as a “minigene” for antigen expression through the class I pathway. Epitope-specific CTL activation can be seen with both PMGT and intramuscular delivery of plasmid [47]. Tumors are heterogenous and express many different antigens. Thus, potential benefit may be found following vaccination with a plasmid encoding several antigenic peptides. Expression of two or more genes or “minigenes” in a cell

following transfection can be undertaken with dicistronic or multicistronic vectors with internal ribosome entry sites. This multigene strategy can also combine cytokine genes to enhance the activation of CTLs [48]. Cytokine delivery at the site of plasmid vaccination can be used to influence the predominate expression of either a Th1 or a Th2 response [49,50]. Several in vivo vaccine studies with PMGT describe a predominately Th2 response following vaccination [34,51]. Another strategy for gene delivery involves vectors containing minimal amounts of DNA. These vectors incorporate a conditional origin of replication that can only replicate in limited hosts and eliminate antibiotic resistance markers. Reducing the amount of bacterial sequences and severely limiting the ability of the plasmid to replicate provide another vector strategy for nonviral gene therapy [45,52]. Plasmid vectors can also contribute to the immunogenicity of the transgene by utilizing immunostimulatory sequences (ISSs) of DNA in the vector. Cytosine–phosphate–guanosine (CpG) motifs, found in increased numbers in bacterial DNA, can act as potent immunostimulatory molecules [53]. CpGoligodeoxynucleotide sequences have been shown to activate and mobilize dendritic cells (DCs), inducing the increased production of interleukin-12 (IL-12) that may account for the shift toward Th1-predominant immune responses [54]. The effects of ISSs appear to be less apparent following PMGT than after direct DNA injection (intramuscular or intradermal).

III. RECENT ADVANCES A. In Vitro Modification of Human Cells The stimulation of selective recognition and destruction of tumor cells by components of the immune system is a central goal of tumor immunotherapy. Because the T cell is felt to be an important component in the immune response against cancer, several strategies to activate T cells are being considered as candidate vaccine approaches. T-cell stimulation and activation by antigen require interaction of a T cell with a tumor cell or antigen-presenting cell (APC) that can provide appropriate MHC presentation of the antigen as well as necessary costimulation of the T cell [55–58] (Fig. 2). The use of PMGT to directly modify tumor cells or APC with one or more transgenes presents an opportunity to rapidly evaluate promising strategies to activate immune-mediated destruction of human cancer. One of these strategies is to make a tumor cell express costimulatory molecules, and this strategy has been shown to enhance the immunogenicity of tumor cells in murine studies [59]. 1. Modification of Melanoma Cells Human melanoma cells with no detectable baseline surface expression of the B7-1 costimulatory molecule can


229

FIGURE 2 The requirements for T-cell activation by tumor cells include T-cell recognition of tumor antigen that is presented by the appropriate MHC molecule and the delivery of an additional costimulatory signal to the T cell. T-cell activation for tumor cell lysis requires each of these components to be present (A), and the lack of any of these components in the stimulating cell could result in a failure of the tumor cell to activate a T-cell response (B). (From Albertini, M. R. and Sondel, P. M., in Clinical Oncology, 2nd ed., Abeloff, M. D. et al., eds., Saunders, Philadelphia, PA, 2000, pp. 214–241. With permission.)

have 8–31% of cells become B7-1 positive with no selection procedure after in vitro PMGT with human cDNA for B7-1 [16]. To evaluate whether the antigenicity of B7-1-expressing melanoma cells would vary with the level of B7-1 expression, melanoma cell lines with different levels of stable B7-1 expression were obtained [60]. This was accomplished by initially transfecting cells by PMGT with a B7-1-neo cDNA vector and proceeding with subsequent selection in media containing G418. Cells were then sorted on a FACStar Plus flow cytometer by brightness of B7-1 staining (Fig. 3), and cell lines that maintained different levels of stable B7-1 expression (85, 62, 26, and 14%) were obtained.

Functional studies determined the antigenicity of the human melanoma cells with transient (me15-B7, M-21-B7) or stable (me15-B7-neo, M-21-B7-neo) expression of B7-1. Allogeneic normal donor peripheral blood mononuclear cells (PBMCs) secreted greater amounts of granulocyte– macrophage colony-stimulating factor (GM-CSF) when incubated with B7-1-transfected melanoma cells than did PBMCs incubated with unmodified melanoma cells. Similarly, cell-mediated cytotoxicity against unmodified melanoma cells was greater in PBMCs cultured for 5 days with B7-1-transfected cells in comparison with PBMCs cultured with unmodified melanoma cells, and the level of

230


TABLE 1 GP-100 Expression by Transfected B-cell Lines Transgeneb

HLA-A phenotypec

GP-100 expressiond

721 poly

Neo/B7-gp 100

1,2

10%

.221 CL.5

Neo/B7-gp 100

-,-

60%

.221(A1)CL.8

Hygro/gp 100

1,-

90%

.221(A1)CL.9

Hygro/gp 100

1,-

60%

.221(A1)N-CL.5

Hygro-vector

1,-

0%

.221(A2)CL.10

Hygro/gp 100

2,-

70%

.221(A2)N-CL.29

Hygro-vector

2,-

0%

B-cell Linesa

a B-cell line variants of LCL-721 (721, .221, .221(A1), and .221(A2)) were provided by Dr. Robert DeMars for these experiments. The HLA class 1 negative 221 cells were derived from the parental 721 cells by gamma ray irradiation followed by immunoselection with appropriate anlisera or monoclonal antibodies. Transfer of cDNA for HLA-A1 or HLA-A2 into .221 cells was performed for the .221(A1) and .221(A2) cells, respectively. b The indicated transgene was stably transfected into each B-cell line. c The HLA-A phenotype is shown for each of the B-cell lines. d The GP-100 expression of each cell line was determined by flow cytometry.

cytotoxicity was proportional to the level of B7-1 expression on the stimulating cells [60]. Thus, PMGT of cDNA for B7-1 into human melanoma cells increased expression of functional B7-1 and enhanced the antigenicity of the genemodified cells proportional to the level of B7-1 expression by the modified tumor cells. As melanoma cells with enhanced expression of both HLA and B7-1 molecules may better stimulate T-cell immunity than unmodified melanoma cells, additional experiments have determined that enhanced expression for each of these molecules could be obtained within the same melanoma cell following PMGT (Fig. 4). The expression of two genes in the same cell after PMGT of separate genetic constructs provides an efficient means to evaluate distinct molecular modifications to stimulate T-cell immunity to human melanoma [16]. 2. Modification of B-Cell Lines

FIGURE 3 Flow cytometric analysis of B7-1 expression by M-21 cells. The B7-1 expression is shown for nontransfected M-21 cells (panel A), as well as for unsorted M-21–B7 stable transfectants (Panel B) and for M-21–B7 (bright) cells sorted to contain M-21 cells with high B7-1 expression (panel C). (From McCarthy, D. et al., Cancer Immunol. Immunother., 49, 85–93, 2000. With permission.

As our initial experiments determined conditions to enhance human leukocyte antigen (HLA) expression as well as conditions to enhance or induce B7-1 expression by melanoma cells, subsequent experiments determined the ability to transfect cells with the gene for the melanomaassociated antigen gp100. We proceeded to use PMGT to gene-modify B-cell lines with defined HLA phenotypes1 to have stable expression of gp100 (Table 1; unpublished results). Because these gp100-expressing B-cell lines are expected to present epitopes of gp100 protein in the context of different HLA molecules, they will be used as target cells in upcoming functional experiments to determine the 1 The

B-cell lines with various HLA phenotypes were obtained from Dr. Robert DeMars.


231

FIGURE 4 Expression of both B7-1 and HLA-A2 following dual delivery of the cDNAs for B7-1 and HLA-A2. Fluorescence intensity of Mel 117 melanoma cells following particle-mediated gene transfer with gold beads coated with (A) β-galactosidase DNA, (B) human B7-1 DNA, (C) human HLA-A2 DNA, or (D) the combination of human B7-1 and human HLA-A2 DNA. Cells were dually stained with anti-BB1 plus goat anti-mouse (GAM) FITC plus anti-HLA-A2 antibody and then analyzed on a FACScan flow cytometer. The percentage of cells in each quadrant is reported. (From Albertini, M. R. et al., Cancer Gene Ther., 3(3), 192–201, 1996. With permission.).

HLA restriction of T cells following stimulation with gp100expressing cells [61].

been ulilized as a means for DNA vaccination with cDNA for tumor-associated molecules and with cDNA for various cytokines [11].

B. Antitumor Efficacy in Murine Models Particle-mediated gene transfer has been used to generate murine tumor vaccines by genetically modifying tumor cells in vitro or ex vivo, primarily with cytokine genes. In addition, in vivo gene delivery using PMGT into skin has

1. DNA Vaccination This approach involves the delivery into the skin of genes encoding tumor-associated antigens (TAAs) in an attempt to induce antitumor immune responses. Using PMGT,

232


successful vaccinations against tumors have been achieved with the genes encoding model TAAs such as β-galactosidase [27,30] or ovalbumin [30] and natural TAAs such as mutant p53 oncoprotein [28,62], carcinoembryonic antigen [29], or human papillomavirus E7 antigen [63]. In a study by Ross et al., vaccination by PMGT was found to be more effective than vaccination with a peptide even in combination with an adjuvant [30]. In several studies in both mice and monkeys, PMGT required 100–500-fold less DNA than intramuscular or intradermal DNA inoculations to induce comparable or higher levels of cellular or antibody responses [64–69]. In an attempt to elucidate the mechanism of induction of immune response by PMGT into the skin, trafficking of cutaneous DCs carrying DNA/gold beads to draining lymph nodes has been demonstrated [42]. These cutaneous DCs directly transfected with DNA-encoded antigen via PMGT have been shown to play a predominant role in antigen presentation to CD8+ T cells [64]. The results suggest that enhanced antigen presentation by bone-marrow-derived [43] DCs to T lymphocytes is responsible for induction of antitumor immunity in cutaneous gene immunization experiments. PMGT for co-delivery of adjuvant cytokine genes with tumor-antigen genes has been successfully used to further augment antigen presentation and induce enhanced antitumor immunity in mouse models. This was achieved by co-delivering plasmids encoding such cytokines as IL-12 [62,65], IFN-γ [70], and GM-CSF [29,70]. We have recently established a model of gene vaccination against melanoma using PMGT for gp100 gene delivery in the skin of mice [71]. Before evaluating the antitumor efficacy of gp100 DNA vaccination, it was necessary to determine whether B16 melanoma transfection with the human gp100 (hgp100) cDNA expression plasmid results in production of transgenic hgp100 protein. Murine B16 cells were transfected with hgp100 cDNA, selected in vitro for positive clones (B16–gp100), and then analyzed for hgp100 expression using flow cytometry. The control wild-type B16 cells had a lowlevel, positive expression of a molecule cross-reactive with hgp100 (and recognized by HMB-45 mAb against hgp100), reflecting the 80% homology of murine gp100 (mgp100) with hgp100. However, B16-gp100 cells showed a much greater fluorescence intensity with the HMB-45 mAb, documenting the greater expression of gp100 induced by transfection with the hgp100 DNA. Specificity of B16 tumor staining by HMB-45 mAb was confirmed in separate studies, where a nonmelanoma murine cell line, 4T1 adenocarcinoma, did not stain with HMB-45 mAb. In a second approach, we looked at the expression of hgp100 in B16–gp100 cells by using reverse transcription–polymerse chain reaction (RT–PCR) analysis. RT-PCR using the human-specific primer set for the human gp100 gene showed that only the human gp100-transfected cell line, B16–gp100, resulted in an amplification product.

The hgp100 DNA vaccination, either alone or in combination with GM-CSF DNA, also resulted in transgene transcription in the skin tissues 24 hours post transfection. The gp100 DNA vaccination was performed by PMGT into the skin of mice with hgp100-encoding plasmid DNA, alone or in combination with mGM-CSF-encoding plasmid DNA. This resulted in a total delivery of 2.5 μg of each plasmid. Seven days later, vaccinated or naive mice were challenged intradermally with 5×104 B16–gp100 cells, and survival of mice was followed. The gp100 gene vaccination resulted in substantial protection against B16–gp100 tumors, with 40% of mice remaining tumor-free for at least 2 months. Importantly, co-delivery of GM-CSF cDNA with hgp100 cDNA resulted in complete tumor protection in all vaccinated mice (Fig. 5). The ability of GM-CSF gene codelivery to enhance the effect of gp100 DNA vaccination was consistently reproducible in 12 subsequent experiments, although the degree of protection varied. We next investigated whether T cells were responsible for the protection induced by gp100–GM-CSF gene vaccination. Our results demonstrated that in vivo depletion of T cells with anti-CD4 and anti-CD8 mAbs abrogated the protection induced by the gp100–GM-CSF DNA vaccine, indicating that this protection is T-cell mediated [71]. In addition, vaccinated mice that remained tumor-free for more than a month after tumor challenge were partially or completely immune to a secondary tumor challenge. The experiments described above demonstrate that gp100 gene vaccination of naive mice, especially in combination

FIGURE 5 GM-CSF cotransfection enhances the effect of gp100 gene vaccination against B16–gp100 melanoma. Skin of C97BL/6 mice was transfected at four sites with hgp100 cDNA (2.5 μg/mouse) or hgp100 cDNA in combination with GM-CSF cDNA (2.5 μg/mouse of each) using PMGT. Expression of transgenic GM-CSF in the skin was confirmed by ELISA (not shown). Seven days following vaccination, mice were injected intradermally with 5 × 104 B16–gp100 tumor cells, and the survival of mice was followed. Each group included five mice.


with GM-CSF, can result in induction of a T-cell-mediated immune response that is capable of protecting mice from a subsequent tumor challenge. To allow the experimental conditions to better simulate the clinical situation, we investigated the therapeutic effect of gp100–GM-CSF gene therapy in mice bearing established tumors. The results demonstrated that the gp100–GM-CSF gene combination induced suppression of tumor growth when compared to control mice ( p < 0.05 starting from day 17 of tumor growth). The effect of gp100–GM-CSF nucleic acid immunization against established tumors was further confirmed by the extended survival of mice (Fig. 6). The survival was calculated based on the number of days it took for the tumor diameter to reach 15 mm, at which time the mice were sacrificed. Thus, untreated tumor-bearing mice and mice treated with the empty vector–GM-CSF cDNA survived for 26.11 ± 0.96 and 25.5 ± 1.42 days, respectively, whereas mice treated with gp100–GM-CSF cDNA survived for 37.5 ± 2.72 days ( p < 0.005). This experiment was repeated two times, with the treatment started on day 4 or on day 7 post tumor cell implantation, and similar results were obtained [71]. 2. Cytokine Gene Therapy A second approach using PMGT involves the treatment of already established subcutaneous tumors with cytokine genes. We have reported a powerful T-cell-mediated tumor regression following skin transfection with IL-12 gene in several mouse tumor models [39]. Moreover, the localized IL-12 gene delivery into the skin overlaying the immunogenic tumor has resulted in a systemic effect against visceral metastases [39] and a distant solid tumor [32]. Histological analysis showed that IL-12 transgene expression was readily detectable in the epidermis, although the DNA-coated gold beads did not reach the implanted tumor. These data suggest that a gradual, continuous release of small doses of transgenic IL-12 protein by transfected epidermal cells in the vicinity of a tumor can effectively result in activation of systemic antitumor immunity. This gene therapy approach using PMGT was later shown to be applicable and effective not only against cutaneous tumors, but also against tumors in visceral organs such as liver [72]. In contrast to immunogenic tumors, IL-12 gene therapy of poorly immunogenic tumors via PMGT may elicit different antitumor mechanisms. Thus, IL-12 gene transfer into the skin surrounding and overlaying a poorly immunogenic 4T1 mammary adenocarcinoma has resulted in a significant reduction of spontaneous lung metastases while having no effect on the growth of the primary intradermal tumor [73]. In contrast to the above-described antitumor effect of IL-12 gene therapy with immunogenic tumors, this antimetastatic effect was not mediated by T cells, but involved natural killer

233

FIGURE 6 Suppression of the growth of established tumors by the gp100–GM-CSF DNA vaccine. C97BL/6 mice were injected intradermally with 5×104 B16–gp100 cells. On days 7, 10, 13, and 17, the skin of tumorbearing mice was transfected at two sites of the abdominal area by PMGT with hgp100 cDNA in combination with GM-CSF cDNA (625 ng of each plasmid per transfection site). Control mice were untreated or received empty vector DNA plus GM-CSF DNA. The survival of mice was determined (mice were sacrificed on the day the tumor diameter reached 15 mm). Each group consisted of five mice.

(NK) cells and IFN-γ . We have also shown that co-delivery of additional gene(s) with IL-12 cDNA, a task that can be easily achieved by PMGT, may enhance the antitumor and antimetastatic activity of IL-12 gene therapy [74].

C. Antitumor Activity of a Canine Tumor Vaccine Hogge et al. [75] have utilized the dog as a valid translational model for human cancers. A pilot study of seven research dogs demonstrated that a tumor vaccine composed of irradiated canine melanoma cells transfected ex vivo by PMGT with cDNA for human GM-CSF was safe, nontoxic, and well tolerated [75]. Vaccine-site biopsies demonstrated production of human GM-CSF at the vaccine sites, and histological analysis revealed an influx of neutrophils and macrophages. Dogs with spontaneous tumors were evaluated in an additional study. These dogs had tumors surgically excised and processed with mechanical and enzymatic digestion. The tumor cells were irradiated, received PMGT with cDNA for human GM-CSF, and were injected intradermally as a vaccine for the dog. Biopsy samples revealed biologically active human GM-CSF protein, and neutrophil and macrophage infiltration was seen at the vaccine sites. Objective antitumor responses were described in 3 of the 16 dogs in this trial [22].

234


IV. ISSUES REGARDING EVALUATION IN CLINICAL TRIALS A. General Considerations A primary issue for further clinical development of PMGT will be the ability to achieve clinically relevant levels of transgene expression following both ex vivo and in vivo PMGT. Identification of informative in vivo biomarkers of a successful immune response following PMGT will be essential for this technology, as well as other vaccine approaches, to move forward.

B. Skin Penetration Considerations The ability to transfect a sufficient number of Langerhans cells, as well as other APCs in the skin, will be a key accomplishment for in vivo PMGT. The epidermis of human skin is made up of five layers: (1) stratum germinativum (stratum basale), the germinal layer of the epidermis where cell division occurs; (2) stratum spinosum (prickle cell layer), which contains cells in the process of growth and early keratin synthesis; (3) stratum granulosum (granular layer), characterized by the presence of granules within the cells which contribute to the process of keratinization; (4) stratum lucidum, found only in thick skin, and (5) stratum corneum, which consists of dead cell remnants flattened and fused and composed primarily of the fibrous protein keratin. The epidermis ranges in thickness from 0.075–0.15 mm in thin skin and 0.4–0.6 mm in thick skin. Keratinocytes are the predominate cell type within the epidermis. Another cell type found is the Langerhans cell. Langerhans cells, mainly found in the stratum germinativum, are from a monocyte lineage and act as cutaneous DCs [76]. Keratinocytes and deeper myocytes can also act as APCs though they lack costimulatory molecules. Preliminary observations from PMGT in humans reveal beads present in the epidermis and papillary dermis with a depth of penetration ranging between 0.1 and 0.4 mm [77]. The largest concentration of gold particles was found in the stratum corneum (greater than 90%), the outermost layer of the epidermis [77]. The optimal number of beads needed in the stratum germinativum for an effective immune response is not known. Again, Langerhans cells, which act as professional APCs, are located in the lower layers of the epidermis layer and not in the stratum corneum. In order to increase the penetration of gold particles into Langerhans cells and into viable keratinocytes, a decrease or thinning of the stratum corneum may be needed. Multiple different established techniques exist for decreasing the stratum corneum [78,79], including mechanical removal utilizing cellophane adhesive tape [80,81] and chemical peels. The primary use of chemical peels is the treatment of photo-aged skin and precancerous lesions and to reduce wrinkles. Agents used in chemical peel include salicylic acid

[82,83], glycolic acid [84,85], and retinoic acid [86, 87]. All of these provide mild to moderate removal of the stratum corneum and result in increased growth of the stratum germativum. Additionally, treatment of skin with retinoic acid appears to increase the number of Langerhans cells at the treatment site [88]. These skin pretreatment strategies could be investigated in clinical trials evaluating PMGT.

V. RECENT CLINICAL TRIALS A. Ex Vivo PMGT The potential advantage of cancer immunotherapy is the ability of the immune system to selectively recognize and attack tumor cells throughout the body. It is important to note, in this regard, that many human cancers express TAAs that can be recognized by the immune system [89]. Immunization using whole tumor cell vaccine, TAA peptides, or TAA genes is a promising therapeutic strategy for cancers [90]. However, vaccination with tumor cells alone, without adjuvants, has rarely achieved a substantial antitumor effect in either experimental models or clinical practice due to insufficient antigen presentation or costimulation [89]. Possibilities for clinical cancer immunotherapy are suggested by the exciting findings that some nonimmunogenic tumors, as defined based on conventional vaccination methods, may be genetically modified to allow them to induce immune responses that can be detected following certain experimental approaches [90,91]. PMGT has been successfully used to transfect tumor cells with costimulatory genes [16] or cytokine genes. Transfection of weakly immunogenic tumor cells with GM-CSF DNA, followed by vaccination of mice with these cells, resulted in the generation of an immune response that was able to control growth of parental tumors in the mouse B16 melanoma model [92]. Whereas viral methods of gene delivery require establishing tumor cell cultures from patients’ samples, PMGT can be used with tumor cells freshly harvested from a patient [92]. Based on these findings, the first cancer clinical study using PMGT was conducted utilizing the method of ex vivo transfection of tumor cells with GM-CSF DNA [93]. Preliminary results from this trial (led by Dr. David Mahvi at the University of Wisconsin Comprehensive Cancer Center) demonstrate that this technique of gene transfer is safe and feasible. No treatment-related toxicity was seen in patients receiving this treatment. The entire procedure of autologous tumor biopsy and vaccination by PMGT of irradiated tumor cells could be accomplished in less than 6 hours (D. Mahvi, pers. comm.). Other investigators recently confirmed the efficacy of PMGT for transfecting human tumors using primary renal carcinoma cell lines. These cells were obtained from patient tumor tissues and produced high levels of GM-CSF following transfection with GM-CSF DNA using PMGT [94]. A separate approach of transfecting DC with TAA genes via


PMGT for the induction of immune response has also been recently described [95].

B. In Vivo PMGT Multiple modifications of APCs may be required to activate effective in vivo anti-melanoma T-cell immunity. We recently conducted a phase I study of in vivo PMGT of cDNAs for gp100 and GM-CSF into uninvolved skin of melanoma patients to evaluate clinical toxicity, transgene expression, and immunological activation with this treatment [77]. Three treatment groups of six patients were originally planned for this trial. Treatment group I (gp100 alone) has been completed, and subsequent treatment groups were planned to receive PMGT with cDNA for GM-CSF administered either 3 days prior to cDNA for gp100 or on the day of gp100 administration. In treatment group I, PMGT was administered with helium at 500 psi using the PowderJect XR-1 device. Our initial treatment group of six patients received gp100 PMGT (0.25 μg DNA and 250 μg gold/treatment) at two (three patients) or four (three patients) separate vaccine sites each 3-week cycle of treatment. Vaccine site reactions consisted of localized erythema that resolved within 1–2 weeks and a brief tingling sensation at the vaccine site. Vaccine site biopsies 2 and 4 days after PMGT revealed beads primarily in the epidermis, but also in the papillary dermis, and with a depth of bead penetration ranging from 0.1–0.4 mm. The greatest bead concentration was in the stratum corneum, and a perivascular lymphoid infiltrate was present in the dermis. Gold bead localization and gp100 transgene expression were detected in large cells of the epidermis, most likely Langerhans cells. One of three patients in dose level 1 and two of the three patients in dose level 2 had stable disease following two 3-week cycles of treatment. Immunological assays to determine the magnitude and specificity of T-cell reactivity and antibody responses are in progress. Our preliminary conclusions are that PMGT of cDNA for gp100 has minimal toxicity and can achieve gp100 transgene expression in normal human skin. Additional clinical evaluation is needed to determine the immune activation that can be achieved with PMGT of gold beads carrying cDNA for gp100, either alone or in combination with gold beads carrying cDNA for GM-CSF.

VI. POTENTIAL NOVEL USES AND FUTURE DIRECTIONS Various cytokine genes, including IL-2, IL-4, IL-12, IFN-γ , and GM-CSF, have been effective in mediating either T-cell-dependent or inflammatory responses which lead to tumor regression [96,97]. Importantly, the synergistic antimelanoma effects of melanoma peptide vaccines and systemic administration of cytokines such as GM-CSF have been

235

reported [98]. In our murine experiments, we have established a model of gene vaccination against melanoma using PMGT for gene delivery in the skin. The results obtained so far show that coadministration of human gp100 and murine GM-CSF plasmid DNAs into mice led to a much greater vaccination effect than delivery of the gp100 cDNA vaccine alone. Moreover, this DNA combination was found effective in suppression of melanoma growth and extended survival of mice bearing established B16 melanoma that genetically expressed human gp100 [71]. Further plans for investigation are based on the hypothesis that selective activation of immunologic recognition mechanisms against melanoma, followed by activation of the effector mechanisms downstream from the immune recognition event, might provide preferential destruction of the melanoma cells in vivo. The use of PMGT to directly modify APCs with one or more transgenes presents an opportunity to translate successful in vitro findings into treatment approaches for patients in the clinic. Exciting preclinical results have demonstrated anti-melanoma efficacy following vaccination with cDNAs for gp100 and GM-CSF by PMGT. Rigorous clinical evaluation of PMGT, with careful immunological monitoring, is required to determine the potential clinical importance of this technology. In addition, the evaluation of combination immunotherapy strategies in preclinical models will provide a foundation for subsequent clinical trials to enhance antitumor T-cell immunity. Primary issues for further clinical development of PMGT will be the ability to achieve clinically relevant levels of transgene expression following both ex vivo and in vivo PMGT. Identification of informative in vivo biomarkers of a successful immune response following PMGT will be essential for this technology, as well as other vaccine approaches, to more forward.

Acknowledgments The authors thank Sandy Keller for preparation of this manuscript and PowderJect Vaccines, Inc., for providing a dermal PowderJect-XR device for this research. We thank Drs. Paul Sondel, David Mahvi, Jacquelyn Hank, KyungMann Kim, and Ning-Sun Yang for stimulating discussions about PMGT. Our research investigating PMGT was supported by the National Institutes of Health (R29-CA68466; U01-CA61498), the Jay Van Sloan Memorial from the Steven C. Leuthold Family Foundation, gifts to the University of Wisconsin Comprehensive Cancer Center, and the University of Wisconsin General Clinical Research Center (M01 RR03186).

References 1. McCabe, D., Swain, W., Martinell, B., and Christou, P. (1988). Stable transformation of soybean (hylcine max) by particle acceleration. Bio/Technology 6, 923–926.

236


2. Christou, P., McCabe, D., Martinell, B., and Swain, W. (1990). Soybean genetic engineering-commercial production of transgenic plants. Trends Biotechnol. 8, 145–151. 3. Heiser, W. (1994). Gene transfer into mammalian cells by particle bombardment. Analyt. Biochem. 217, 185–196. 4. Yang, N., Burkholder, J., Martinell, B., and McCabe, D. (1990). In vivo and in vitro gene transfer to mammalian somatic cells by particle bombardment. Proc. Natl. Acad. Sci. USA 87, 9568–9572. 5. Cheng, L., Ziegelhoffer, P., and Yang, N. (1993). In vivo promoter activity and transgene expression in mammalian somatic tissues evaluated by using particle bombardment. Proc. Natl. Acad. Sci. USA 90, 4455– 4459. 6. Yang, N.-S., and Ziegelhoffer, P., (1994). The particle bombardment system for mammalian gene transfer, in Yang N-S and Christou P, eds. Particle Bombardment Technology for Gene Transfer (N.-S. Yang and Christou, p., eds.), pp. 117–141. Oxford University Press, New York. 7. Jiao, S., Cheng, L., Wolff, J., and Yang, N.-S. (1993). Particle bombardment-mediated gene transfer and expression in rat brain tissues. J. Immunol. 159, 11–14. 8. Thompson, T., Gould, M., Burkholder, J., and Yang, N.-S. (1993). Transient promoter activity in prmary rat mammary epithelial cells evaluated using particle bombardment gene transfer. In Vitro Cell Dev. Biol. 29A, 165–170. 9. Burkholder, J., Decker, J., and Yang, N.-S. (1993). Transgene expression in lymphocyte and macrophage primary cultures after particle bombardment. J. Immunol. Meth. 165, 149–156. 10. Ye, Z., Qiu, P., Burkholder, J., Turner, J., Culp, J., Roberts, T., Shahidi, N., and Yang, N. (1998). Cytokine transgene expression and promoter usage in primary CD34+ cells using particle-mediated gene delivery. Hum. Gene Ther. 9, 2197–2205. 11. Rakhmilevich, A., and Yang, N. (2000). In vivo particle-mediated gene transfer for cancer therapy, in Methods in Molecular Medicine. Vol. 35. Gene Therapy: Methods and Protocols (W. Walther and U. Stein, eds.), Humana Press, Totowa, NJ. 12. Merchant, B. (1998). Gold, the noble metal and the paradoxes of its toxicology. Biologicals 26, 49–59. 13. Yang, N.-S., Burkholder, J., McCabe, D., Neumann, V., and Fuller, D. (1997). Particle-mediated gene delivery in vivo and in vitro. Curr. Protocols Hum. Genet. Suppl. 12, 12.6.1–12.6.14. 14. Chen, R. and Ruggle, S. (1993). In situ cDNA polymerase chain reaction: a novel technique for detecting mRNA expression. Am. J. Pathol. 143, 1527–1534. 15. Williams, R., Johnston, S., Riedy, M., DeVit, M., McElligot, S., and Sanford, J. (1991). Introduction of foreign genes into tissues of living mice by DNA-coated microprojectiles. Proc. Natl. Acad. Sci. USA 88, 2726–2730. 16. Albertini, M. R., Emler, C. A., Schell, K., Tans, K. J., King, D. M., and Sheehy, M. J. (1996). Dual expression of human leukocyte antigen molecules and the B7-1 costimulatory molecule (CD80) on human melanoma cells after particle-mediated gene transfer. Cancer Gene Ther. 3(3), 192–201. 17. Qiu, P., Ziegelhoffer, P., Sun, J., and Yang, N.-S. (1996). Gene gun delivery of mRNA in situ results in efficient transgene expression and immunization. Gene Ther. 3, 262–268. 18. Perales, J., Perkol, T., Beegen, H., Ratnoff, O., and Hanson, R. (1994). Gene transfer in vivo: sustained expression and regulation of genes introduced into the liver by receptor-targeted uptake. Proc. Natl. Acad. Sci. USA 91, 4086–4090. 19. Tang, D., DeVit, M., and Johnston, S. (1992). Genetic immunization is a simple method for eliciting an immune response. Nature 356, 152– 154. 20. Vanrompay, D., Cox, E., Vandenbussche, F., Volckaert, G., and Goddeeris, B. (1999). Protection of turkeys against Chlamydia psittaci challenge by gene gun-based DNA immunizations. Vaccine 17, 2628– 2635.

21. Han, R., Cladel, N., Reed, C., Peng, X., and Christensen, N. (1999). Protection of rabbits from viral challenge by gene gun-based intracutaneous vaccination with a combination of cottontail rabbit papillomavirus E1, E2, E6, and E7 genes. J. Virol. 73, 7039–7043. 22. Hogge, G., Burkholder, F. J., Culp, J., Dubielzig, R., Albertini, M., Keller, E., Yang, N.-S., and MacEwen, E. (1998). Development of human granulocyte macrophage-colony stimulating factor transfected tumor cell vaccines for the treatment of spontaneous canine cancer. Hum. Gene Ther. 9, 1851–1861. 23. Macklin, M., McCabe, D., Mcgregor, M., Neumann, V., Meyer, T., Callan, R., Hinshaw, V., and Swain, W. (1998). Immunization of pigs with a particle-mediated DNA vaccine to influenza A virus protects against challenge with homologous virus. J. Virol. 72, 1491–1496. 24. Lunn, D., Soboll, G., Schram, B., Quass, J., McGregor, M., Drape, R., Macklin, M., McCabe, D., Swain, W., and Olsen, C. (1999). Antibody responses to DNA vaccination of horses using the influenza virus hemagglutinin gene. Vaccine 17, 2245–2258. 25. Fuller, D., Corb, M., Barnett, S., Steimer, K., and Haynes, J. (1997). Enhancement of immunodeficiency virus-specific immune responses in DNA-immunized rhesus macaques. Vaccine 15, 924–926. 26. Tacket, C., Roy, M., Widera, G., Swain, W., Broome, S., and Edelman, R. (1999). Phase I safety and immune response studies of a DNA vaccine encoding hepatitis B surface antigen delivered by a gene delivery device. Vaccine 17, 2826–2829. 27. Irvine, K. R., Rao, J. B., Rosenberg, S. A., and Restifo, N. P. (1996). Cytokine enhancement of DNA immunization leads to effective treatment of established pulmonary metastases. J. Immunol. 156(1), 238– 245. 28. Ciernik, F., Berzofsky, J., and Carbone, D. (1996). Induction of cytoxic T lymphocytes and antitumor immunity with DNA vaccines expressing single T cell epitopes. J. Immunol. 156, 2369–2375. 29. Conry, R., Widera, G., LoBuglio, A. et al. (1996). Selected strategies to augment polynucleotide immunization. Gene Ther. 3, 67–74. 30. Ross, H., Weber, L., Wang, S. et al. (1997). Priming for T-cell-mediated rejection of established tumors by cutaneous DNA immunization. Clin. Cancer Res. 3, 2191–2196. 31. Sun, W., Burkholder, J., Decker, J., Turner, J., Lu, X., Pugh, T., Ershler, W., and Yang, N. (1995). In vivo cytokine gene transfer by particle bombardment reduces tumor growth in mice. Proc. Natl. Acad. Sci. USA 92, 2889–2893. 32. Rakhmilevich, A., Janssen, K., Turner, J., Culp, J., and Yang, N.-S. (1997). Cytokine gene therapy of cancer using gene gun technology: superior antitumor activity of IL-12. Hum. Gene Ther. 8, 1303–1311. 33. Chen, H., Pan, C., Liau, M., Jou, R., Tsai, C., Wu, J., Lin, Y., and Tao, M. (1999). Screening of protective antigens of Japanese encephalitis virus by DNA immunization: a comparative study with conventional viral vaccines. J. Virol. 73, 10137–10145. 34. Feltquate, D., Heaney, S., Webster, R., and Robinson, H. (1997). Different T helper cell types and antibody isotypes generated by saline and gene gun DNA immunization. J. Immunol. 163, 4510–4518. 35. Fomsgaard, A., Nielsen, H., Kirkby, N., Bryder, K., Corbet, S., Nielsen, C., Hinkula, J., and Buus, S. (1999). Induction of cytotoxic T-cell responses by gene gun DNA vaccination with minigenes encoding influenza A virus HA and NP CTL-epitopes. Vaccine 18, 681–691. 36. Hanke, T., Neumann, V., Blanchard, T., Sweeney, P., Hill, A., Smith, G., and McMichael, A. (1999). Effective induction of HIV-specific CTL by multi-epitope using gene gun in combined vaccination regime. Vaccine 17, 589–596. 37. Belperron, A., Feltquate, D., Fox, B., Horii, T., and Bzik, D. (1999). Immune responses induced by gene gun or intramuscular injection of DNA vaccines that express immunogenic regions of the serine repeat antigen from Plasmodium falciparum. Infect. Immun. 67, 5163–5169. 38. Fensterle, J., Grode, L., Hess, J., and Kaufmann, S. (1999). Effective DNA vaccination against listeriosis by prime/boost innoculation with the gene gun. J. Immunol. 163, 4510–4518.

The Use of Particle-Mediated Gene Transfer for Immunotherapy of Cancer 39. Rakhmilevich, A., Turner, J., Ford, M., McCabe, D., Sun, W., Sondel, P., Grota, K., and Yang, N. (1996). Gene gun-mediated skin transfection with interleukin 12 gene results in regression of established primary and metastatic murine tumors. Proc. Natl. Acad. Sci. USA 93, 6291–6296. 40. Andree, C., Swain, W., Page, C., Macklin, M., Slama, J., Hatzis, D., and Eriksson, E. (1994). In vivo transfer and expression of an EGF gene accelerated wound repair. Proc. Natl. Acad. Sci. USA 91, 12188–12192. 41. Rajagopalan, L., Burkholder, J., Turner, J., Culp, J., Yang, N.-S., and Malter, J. (1995). Targeted mutagenesis of GM-CSF cDNA increases transgenic mRNA stability and protein expression in normal cells. Blood 86, 2551–2558. 42. Condon, C., Watkins, S., Celluzzi, M., Thompson, K., and Falo, L. (1996). DNA-bsed immunization by in vivo transfection of dendritic cells. Natu. Med. 2(10), 1122–1128. 43. Iwasaki, A., Torres, C., Ohashi, P., Robinson, H., and Barber, B. (1997). The dominant role of bone marrow-derived cells in CTL induction following plasmid DNA immunization at different sites. J. Immunol. 159, 11–14. 44. Gurunathan, S., Klinman, D., and Seder, R. (2000). DNA vaccines: immunology, application, and optimization. Annu. Rev. Immunol. 18, 927–974. 45. Soubrier, F., Cameron, B., Manse, B., Somarriba, S., Dubertret, C., Jaslin, G., Jung, G., LeCaer, C., Dang, D., Mouvault, J., Scherman, D., Mayaux, J., and Crouzet, J. (1999). pCOR: a new design of plasmid vectors for nonviral gene therapy. Gene Ther. 6, 1482–1488. 46. Wu, Y., and Kipps, T. (1997). Deoxyribonucleic acid vaccines encoding antigens with rapid proteasome-dependent degradation are highly efficient inducer of cytolotic T lymphocytes. J. Immunol. 159, 6037– 6043. 47. Iwasaki, A., DelaCruz, C., Young, A., and Barber, B. (1999). Epitopespecific cytoxic T lymphocyte induction by minigene DNA immunization. Vaccine 17, 2081–2088. 48. Thomson, S., Sherritt, M., Medveczky, J., Elliot, S., Moss, D., Fernando, G., Brown, L., and Suhrbier, A. (1998). Delivery of multiple CD8 cytotoxic T cell epitopes by DNA vaccination. J. Immunol. 160, 1717– 1723. 49. O’Garra, A. (1998). Cytokines induce the development of functionally heterogeneous T helper cell subsets. Immunity 8, 275–283. 50. Murphy, K., Ouyang, W., Farrar, J., Yang, J., Ranganath, S., Asnagli, H., Afkarian, M., and Murphy, T. (2000). Signaling and transcription in T helper development. Annu. Rev. Immunol. 18, 451–494. 51. Torres, C., Iwasaki, A., Barber, B., and Robinson, H. (1997). Differential dependence on target site tissue for gene gun and intramuscular DNA immunizations. J. Immunol. 158, 4529–4532. 52. Darquet, A.-M., Rangara, R., Kreiss, P., Schwartz, B., Naimi, S., Delaere, P., Crouzet, J., and Scherman, D. (1999). Minicircle: an improved DNA molecule for in vitro and in vivo gene transfer. Gene Ther. 6, 209–218. 53. Sato, Y., Roman, M., Tighe, H., Lee, D., Corr, M., Nguyen, M.-D., Silverman, G., Lotz, M., Carson, D., and Raz, E. (1996). Immunostimulatory DNA sequences necessary for effective intradermal gene immunization. Science 273, 352–354. 54. Jakob, T., Walker, P., Krieg, A., Udey, M., and Vogel, J. (1998). Activation of cutaneous dendritic cells by CpG-containing oligodexoynucleotides: a role for dendritic cells in the augmentation of Th1 responses by immunostimulatory DNA. J. Immunol. 161, 3042–3049. 55. Jenkins, M. K., and Johnson, J. G. (1993). Molecules involved in T-cell costimulation. Curr. Opin. Immunol. 5(3), 361–367. 56. Bluestone, J. A. (1995). New perspectives of CD28-B7-mediated T cell costimulation. Immunity 2(6), 555–559. 57. Linsley, P. S., and Ledbetter, J. A. (1993). The role of the CD28 receptor during T cell responses to antigen. Annu. Rev. Immunol. 11, 191– 212. 58. Lenschow, D. J., Walunas, T. L., and Bluestone, J. A. (1996). CD28/B7 system of T cell costimulation. Annu. Rev. Immunol. 14, 233–258.

237

59. Townsend, S., and Allison, J. (1993). Tumor rejection after direct costimulation of CD8+ cells by B7-transfected melanoma cells. Science 259, 368–370. 60. McCarthy, D., Glowacki, N., Schell, K., Emler, C., and Albertini, M. (2000). Antigenicity of human melanoma cells transfected to express the B7-1 co-stimulatory molecule (CD80) varies with the level of B7-1 expression. Cancer Immunol. Immunother. 49, 85–93. 61. King, D., Ye, Z.-Q., Roberts, T., Glowacki, N., and Albertini, M. (2000). Cytokine release by thiguanine-resistant T cell clones from melanoma patients. Proc. Am. Assoc. Cancer. Res. 41, 697. 62. Tuting, T., Gambotto, A., Robbins, P., Storkus, J., and Deleo, A. (1999). Co-delivery of T helper 1-biasing cytokine genes enhances the efficacy of gene gun immunization of mice: studies with the model tumor antigen beta-galactosidase and the BALB/c Meth A p53 tumor-specific antigen. Gene Ther. 6, 629–636. 63. Chen, C., Ji, H., Suh, K., Choti, M., Pardoll, D., and Wu, T. (1999). Gene gun-mediated DNA vaccination induces antitumor immunity against human papillomavirus type 16 E7-expressing murine tumor metastases in the liver and lungs. Gene Ther. 6, 1972–1981. 64. Porgador, A., Irvine, K., Iwasaki, A., Barber, B., Restifo, N., and Germain, R. (1998). Predominant role for directly transfected dendritic cells in antigen presentation to CD8+ T cells after gene gun immunization. J. Exp. Med. 188, 1075–1082. 65. Tan, J., Yang, N., Turner, J., Niu, G., Maassab, H., Sun, J., Herlocher, M., Chang, A., and Yu, H. (1999). Interleukin-12 cDNA skin transfection potentiates human Yu papillomavirus E6 DNA vaccineinduced antitumor immune response. Cancer Gene Ther. 6, 331– 339. 66. Leitner, W. W., Seguin, M. C., Ballou, W. R., Seitz, J. P., Schulz, A. M., Sheehy, M. J., and Lyon, J. A. (1997). Immune responses induced by intramuscular or gene gun injection of protective deoxyribonucleic acid vaccines that express the circumsporozoite protein from Plasmodium berghei malaria parasites. J. Immunol. 159, 6112–6119. 67. Fynan, E. F., Webster, R. G., Fuller, D. H., Haynes, J. R., Santoro, J. C., and Robinson, H. L. (1993). DNA vaccines: protective immunizations by parenteral, mucosal, and gene-gun inoculations. Proc. Natl. Acad. Sci. USA 90, 11478–11482. 68. McCluskie, M. J., Millan, C. L. B., Gramzinski, R. A., Robinson, H. L., Santoro, J. C., Fuller, J. T., Widera, G., Haynes, J. R., Purcell, R. H., and Davis, H. L. (1999). Route and method of delivery of DNA vaccine influence immune responses in mice and non-human primates. Mol. Med. 5, 287–300. 69. Pertmer, T. M., Eisenbraun, M. D., McCabe, D., Prayaga, S. K., Fuller, D. H., and Haynes, J. R. (1995). Gene gun-based nucleic acid immunization: elicitation of humoral and cytotoxic T lymphocyte responses following epidermal delivery of nanogram quantities of DNA. Vaccine 13(15), 1427–1430. 70. Charo, J., Ciupitu, A., LeChevalier De Preville, A., Trivedi, P., Klein, G., Hinkula, J., and Kiessling, R. (1999). A long-term memory obtained by genetic immunization results in full protection from a mammary adenocarcinoma expressing an EBV gene. J. Immunol. 163, 5913– 5919. 71. Rakhmilevich, A., Imboden, M., Hao, Z., Macklin, M., Roberts, T., Wright, K., Albertini, M., Yang, N.-S., and Sondel, P. (2001). Effective particle-mediated vaccination against mouse melanoma by co-administration of plasmid DNA encoding gp100 and granulocytemacrophage colony-stimulating factor. Clin. Cancer Res. 7, 952–961. 72. Weber, S., Shi, F., Heise, C., Warner, T., and Mahvi, D. (1999). Interleukin-12 gene transfer results in CD8-dependent regression of murine CT26 liver tumors. Ann. Surg. Oncol. 6, 186–194. 73. Rakhmilevich, A., Janssen, K., Hao, Z., Sondel, P., and Yang, N.-S. (2000). Interleukin 12 gene therapy of a weakly immunogenic mouse mammary carcinoma results in reduction of spontaneous lung metastases via a T cell-independent mechanism. Cancer Gene Ther. 7, 826– 838.

238


74. Oshikawa, K., Shi, F., Rakhmilevich, A., Sondel, P., Mahvi, D., and Yang, N.-S. (1999). Synergistic inhibition of tumor growth in a murine mammary adenocarcinoma model by combinational gene therapy using interleukin-12, pro-interleukin-18 and IL-1β-converting enzyme cDNA. Proc. Natl. Acad. Sci. USA 96, 13351–13356. 75. Hogge, G., Burkholder, J., Culp, J., Dubielzig, R., Albertini, M., Yang, N.-S., and MacEwen, E. (1999). Preclinical development of hGM-CSF transfected melanoma cell vaccine using established canine cell lines and normal canines. Cancer Gene Ther. 6, 26–36. 76. Wheater, P., Burkitt, H., and Daniels, V. (1987). Functional Histology, Churchill-Livingstone, London, pp. 130–141. 77. Albertini, M., King, D., Mahvi, D., Warner, T., Glowacki, N., Roberts, T., Schalch, H., Kim, K., Schiller, J., Rakhmilevich, A., Yang, N., Roy, M., Swain, W., Hank, J., and Sondel, P. (2000). Phase I trial of immunization using particle-mediated transfer of genes for gp100 and GM-CSF into uninvolved skin of patients with melanoma. Proc. Am. Soc. Clin. Oncol. 19, 467a. 78. Moy, L., Peace, S., and Moy, R. (1996). Comparison of the effect of various chemical peeling agents in a mini-pig model. Dermatol. Surg. 22, 429–432. 79. Tse, Y., Ostad, A., Lee, H.-S., Levine, V., Koenig, K., Kamino, H., and Ashinoff, R. (1996). A clinical and histologic evaluation of two medium-depth peels. Dermatol. Surg. 22, 781–786. 80. Tanaka, M., Zhen, Y., and Tagami, H. (1997). Normal recovery of the stratum corneum barrier function following damage induced by tape stripping in patients with atopic dermatitis. Br. J. Derm. 136, 966– 967. 81. Marttin, E., Neelissen-Subnel, M., De Haan, F., and Bodde, H. (1996). A critical comparison of methods to quantify stratum corneum removed by tape stripping. Skin Pharmacol. 9, 69–77. 82. Kligman, D., and Kligman, A. (1998). Salicyclic acid peels for the treatment of photoaging. Dermatol. Surg. 24, 325–328. 83. Loden, M., Bostrom, P., and Kneczke, M. (1995). Distribution and keratolytic effect of salicyclic acid and urea in human skin. Skin Pharmacol. 8, 173–178. 84. Piaquadio, D., Dobry, M., Hunt, S., Andree, C., Grove, G., and Hollenbach, K. (1996). Short contact 70% glycolic acid peels as a treatment for photodamaged skin. Dermatol. Surg. 22, 449–452. 85. Newman, N., Newman, A., Moy, L., Babapour, R., Harris, A., and Moy, R. (1996). Clinical improvement of photoaged skin with 50% glycolic acid. Dermatol. Surg. 22, 455–460. 86. Creidi, P., and Humbert, P. (1999). Clinical use of topical retinaldehyde on photoaged skin. Dermatology 199 (suppl. 1), 49–52.

87. Sachsenberg-Studer, E. (1999). Tolerance of topical retinaldehyde in humans. Dermatology 199 (suppl. 1), 61–63. 88. Meunier, L., Bohjanen, K., Voorhees, J., and Cooper, K. (1994). Retinoic acid upregulates human Langerhans cell antigen presentation and surface expression of HLA-DR and CD11c, a B2 integrin critically involved in T-cell activation. J. Inv. Dermatol. 103, 775–779. 89. Pardoll, D. M. (1998). Cancer vaccines. Nat. Med. 4(5, suppl.), 525–531. 90. Roth, J., and Cristiano, R. (1997). Gene therapy for cancer: what have we done and where are we going? J. Natl. Cancer Inst. 89, 21–39. 91. Colombo, M., Modelsti, A., Parmiani, G., and Forni, G. (1992). Local cytokine availability elicits tumor rejection and systemic immunity through granulocyte-T-lymphocyte cross-talk. Cancer Res. 52, 4853–4857. 92. Mahvi, D. M., Burkholder, J. K., Turner, J., Culp, J., Malter, J. S., Sondel, P. M., and Yang, N.-S. (1996). Particle-mediated gene transfer of granulocyte–macrophage colony-stimulating factor cDNA to tumor cells: implications for a clinically relevant tumor vaccine. Hum. Gene Ther. 7(13), 1535–1543. 93. Mahvi, D. M., Sondel, P. M., Yang, N.-S., Albertini, M. R., Schiller, J. H., Hank, J., Heiner, J., Gan, J., Swain, W., and Logrono, R. (1997). Phase I/IB study of immunization with autologous tumor cells transfected with the GM-CSF gene by particle-mediated transfer in patients with melanoma or sarcoma. Hum. Gene Ther. 8(7), 875–891. 94. Seigne, J., Turner, J., Diaz, J., Hackney, J., Pow-Sang, J., Helal, M., Lockhart, J., and Yu, H. (1999). A feasibility study of gene gun mediated immunotherapy for renal cell carcinoma. J. Urol. 162, 1259–1263. 95. Tuting, T., and Albers, A. (2000). Walther, W., and Stein, U., Eds. Particle-mediated gene transfer into dendritic cells: a novel strategy for the induction of immune responses against tumor antigens, in Gene Therapy of Cancer: Methods and Protocols, Vol. 35, pp. 27–47, Totow, N.J.: Humana Press. 96. Tahara, H., Zitvogel, L., Storkus, W., Zeh, H., McKinney, T., Schreiber, R., Bubler, U., Robbins, P., and Lotze, M. (1995). Effective eradication of established murine tumors with IL-12 gene therapy using a polycistronic retroviral vector. J. Immunol. 154, 6466–6474. 97. Cavallo, F., Di Pierro, F., Giovarelli, M., Gulino, A., Vacca, A., Stoppacciaro, A., Forni, M., Modesti, A., and Forni, G. (1993). Protective and curative potential of vaccination with interleukin-2 gene-transfected cells from a spontaneous mouse mammary adenocarcinoma. Cancer Res. 53, 5067–5070. 98. Jager, E., Ringhoffer, M., Dienes, H., Arand, M., Karbach, J., Jager, D., Ilsemann, Hagerdorn, M., Oesch, F., and Knuth, A. (1996). Granulocyte– macrophage-colony-stimulating factor enhances immune responses to melanoma-associated peptides in vivo. Int. J. Cancer 67, 54–62.

P A R T IId

GENETICALLY MODIFIED EFFECTOR CELLS FOR IMMUNE BASED IMMUNOTHERAPY


C

H

A

P

T

E

R

14 Applications of Gene Transfer in the Adoptive Immunotherapy of Cancer KEVIN T. McDONAGH

ALFRED E. CHANG

Department of Internal Medicine Division of Hematology/Oncology Comprehensive Cancer Center University of Michigan Ann Arbor, Michigan 48109

Department of Surgery Division of Surgical Oncology Comprehensive Cancer Center University of Michigan Ann Arbor, Michigan 48109

I. Introduction 241 II. Use of Gene-Modified Tumors to Generate Antitumor-Reactive T Cells 242

The feasibility of adoptive immunotherapy for cancer is predicated on two fundamental observations derived from animal models. The first is that tumor cells express antigens that are qualitatively or quantitatively different from normal cells and can elicit an immune response within the syngeneic host. The second is that the immune rejection of established tumors can be mediated by the adoptive transfer of appropriately sensitized lymphoid cells. In 1943, Gross [1] was the first to recognize that inbred mice could be immunized against a tumor that was developed in a mouse of the same inbred strain, thus documenting the existence of tumor-associated antigens. Over the years, it has become apparent that individual tumors vary greatly in the nature of their “immunogenicity.” The immunogenicity of a tumor has a direct influence on the ability to develop cellular antitumor immune responses. The transfer of immunity to a naive host by the use of cells was first described by Landsteiner and Chase [2] in 1942. They reported that hypersensitivity to simple compounds could be transferred to normal rats by the transfer of peritoneal exudate cells of sensitized donor animals. In 1954, Billingham et al. [3] documented the ability to transfer skin allograft immunity to a normal murine host by the use of regional lymph node cells from animals that had rejected primary skin allografts. These investigators developed the term adoptive immunotherapy to describe the acquisition of immunity in a normal subject as a result of the transference, not of preformed antibody, but of immunologically competent cells. In 1955, Mitchison [4] was the first to report about the adoptive immunotherapy of tumors in a rodent model. In this study, the adoptive transfer of lymph node cells from mice

A. TILs Derived from Gene-Modified Tumors 242 B. Lymph Node Cells Sensitized with Gene-Modified Tumor Vaccines 243

III. Genetic Manipulation of T Cells to Enhance Antitumor Reactivity 246 A. Genetic Transduction of T Cells 247 B. TIL-Marking Studies 247 C. T Cells as Delivery Vehicles for Immunoregulatory Molecules 248 D. Genetic Reprogramming of Effector Cells with Tumor-Antigen-Specific Receptors 248

IV. Genetic Modulation of Dendritic Cells V. Summary 251

250

References 251

I. INTRODUCTION This chapter focuses on the applications of gene transfer to adoptive immunotherapy of malignancy. This form of cellular therapy refers to the infusion of tumor-reactive immune cells into the tumor-bearing host to mediate, directly or indirectly, regression of established tumor. This review is divided into three areas, each one involving different methods of genetic transfer to generate immune cells into the tumor-bearing host for subsequent adoptive transfer: (1) the use of gene-modified tumor to serve as immunogens to generate effector T cells, (2) genetic manipulation of T cells to enhance antitumor reactivity, and (3) genetic modulation of dendritic cells (DCs).


241


242

Kevin T. McDonagh and Alfred E. Chang

that rejected tumor allografts conferred accelerated rejection of the same tumor allografts in naive hosts. However, more germane to clinical therapy is the ability to transfer immunity to autologous tumors (i.e., syngeneic tumors in inbred rodent strains) using lymphocytes. Borberg et al. [5] were the first to clearly show that the infusion of syngeneic immune cells from hyperimmunized donor animals was capable of mediating the regression of established tumor. During the ensuing years, several other investigators have documented the ability to successfully treat established syngeneic tumors by the adoptive transfer of immune “effector” cells [6,7]. One of the major obstacles in extrapolating the concepts developed in the animal models to the clinical treatment of cancers was the inability to generate adequate numbers of immune cells for therapy. Methods to grow or expand lymphoid cells while retaining their immunologic reactivity were limited; however, in 1976, the discovery of interleukin-2 (IL-2) as a T-cell growth factor made it possible to culture activated T cells in large quantities [8]. In 1981, Cheever et al. [9] showed that tumor-reactive T cells could be expanded in IL-2 and still maintain their therapeutic efficacy in adoptive immunotherapy of tumors in mice. The first successful clinical application of cellular therapy in humans was reported by Rosenberg et al. [10] in 1985. These investigators generated large quantities of IL-2activated peripheral blood lymphocytes (approximately 1–2 ×1011 cells/patient), which were infused along with the concomitant administration of IL-2. These lymphokine-activated killer (LAK) cells were nonspecifically cytolytic to tumor cells by in vitro measurements. In these early clinical trials, significant tumor burdens regressed in a subset of patients, and the feasibility of generating large numbers of cells for clinical therapy was realized. Based on subsequent animal studies [11,12], tumor infiltrating lymphocyte (TILs) were found to be an alternative population of cells that were more potent than LAK cells in mediating tumor regression. Rosenberg and co-workers were the first to report significant clinical responses using TILs in the treatment of patients with advanced melanoma [13,14]. Despite the progress in this field, cellular therapy is still in its infancy. The isolation and expansion of potent immune effector cells derived from the tumor-bearing host remains a formidable task. The use of genetic approaches to alter the host-tumor immune response offers potential opportunities to develop more potent cellular reagents (Table 1).

II. USE OF GENE-MODIFIED TUMORS TO GENERATE ANTITUMOR-REACTIVE T CELLS Genetic modification of tumors to secrete or express immunomodulatory peptides has been found to significantly alter the biology of the tumor cells when these are inoculated into the syngeneic host. A majority of these studies have

TABLE 1 Genetic Approaches to Adoptive Immunotherapy Generation of effector cells using gene-modified tumors TILs isolated from gene-modified tumors Lymph node cells draining gene-modified tumors Gene modification of T cells Marking studies of effector cells T cells as delivery vehicles for immunoregulatory molecules Reengineering T cells with tumor-antigen-specific receptors Antigen-specific T-cell receptors Chimeric T-cell receptor Genetic modification of dendritic cells Genes encoding tumor-associated antigens Genes encoding cytokines/chemokines

involved the use of various cytokine genes introduced into tumor cells. Many of the observed changes are related to an enhanced cellular immune response to tumor-associated antigens expressed on the parental tumor. In selected animal models, regression of established tumors by the inoculation of genetically modified tumor cells that secrete IL-4 or interferon-gamma (IFN-γ ) administered as a tumor vaccination has been observed [15–17]. These studies have served as a rationale to initiate vaccination trials in humans for the therapy of cancers. Based on these early animal studies, investigators have used gene-modified tumors to develop cellular reagents for adoptive immunotherapy. The generation of TILs and vaccine-primed lymph node cells using gene-modified tumors as immunogens is described in this section.

A. TILs Derived from Gene-Modified Tumors Tumor-infiltrating lymphocytes represent lymphoid cells derived from tumors that are disaggregated ex vivo and cultured in IL-2 [11]. A significant impediment to generating therapeutic TILs resides in the inherent immunogenicity of the tumor from which the TILs are derived. In animal models using poorly immunogenic tumors, the therapeutic efficacy of the TILs is limited [11]. These observations suggest that TILs represent a heterogeneous population of cells and that a significant portion of these cells are not appropriately sensitized within “poorly” or nonimmunogenic tumors. Because human tumors are postulated to be nonimmunogenic based on their spontaneous origins and ability to escape the host immune system, methods to enhance the isolation of therapeutic TILs from these tumors could have significant clinical applications. Tumors genetically engineered to produce certain cytokines have been found to contain TILs with enhanced in vitro and in vivo antitumor reactivity. Using the poorly immunogenic methylcholanthrene (MCA)-induced 101 murine

Applications of Gene Transfer in the Adoptive Immunotherapy of Cancer

fibrosarcoma, Restifo et al. [18] showed that therapeutic TILs could be generated if these tumor cells were genetically engineered to secrete IFN-γ . Transduction of MCA 101 tumor cells to secrete IFN-γ resulted in upregulated expression of major histocompatibility complex (MHC) class I molecules. TILs derived from IFN-γ -secreting tumors mediated regression of established parental tumor metastases compared with TILs derived from wild-type tumor, which were ineffective. In additional studies, they found that TILs from the transduced tumors were capable of presenting viral antigen to sensitized T cells, whereas TILs from the parental tumor could not. These studies showed that tumors can be genetically altered with the IFN-γ gene to become “nonprofessional” antigen-presenting cells and that such tumors appeared to be a more reliable source for therapeutically effective TILs. In a model of established lung metastases, Marincola et al. [19] evaluated the immunologic response of the host to an inoculation of syngeneic tumor modified to secrete tumor necrosis factor alpha (TNF-α). TNF-α can stimulate T-cell proliferation and cytotoxic T lymphocyte (CTL) activity. In these studies, the poorly immunogenic MCA 102 sarcoma was modified to secrete TNF-α. The tumorigenicity of the TNF-α-secreting tumor cells was not different from that of wild-type tumor in normal animals. However, TNF-αsecreting tumor cells inoculated subcutaneously regressed in the presence of established wild-type pulmonary metastases, in contrast to inocula of wild-type tumor cells, which grew progressively. In this setting, TILs that were generated from TNF-α-secreting tumors mediated significant antitumor reactivity in adoptive transfer studies compared with wild-type tumor cells, which could not. These observations are important in documenting that therapeutic TILs can be generated from poorly immunogenic tumors modified to secrete TNFα; moreover, this was accomplished in hosts bearing significant tumor burden. IL-7 is another cytokine that has been shown to enhance recovery of TILs from tumors. When the gene for murine IL-7 was retrovirally transferred into an immunogenic murine fibrosarcoma, the tumorigenicity of the tumor was significantly diminished [20]. Moreover, by flow cytometry the IL-7-secreting tumor showed a greater than fivefold increase in infiltrating T cells compared with wild-type tumor. These infiltrating cells were primarily CD8+ cells that mediated enhanced in vitro cytotoxicity against wild-type tumor compared with T cells isolated from control tumors transfected with a neomycin-resistance gene. These studies indicated that IL-7 secretion by the gene-modified tumors can promote the recruitment of induction by cytolytic T cells to the site of that same tumor. A novel variation to alter TIL reactivity has been developed at the University of Michigan. In earlier animal studies, tumors treated by the in vivo transfer of an allogeneic MHC class I gene complexed with liposomes resulted in expres-

243

FIGURE 1 In vitro cytolytic activity of TILs derived from a melanoma nodule before (pretreatment) and after (posttreatment) intralesional HLAB7/liposome inoculations given three times every 2 weeks. Antitumor reactivity was assessed against autologous and allogeneic melanoma targets at various effector/target (E/T) ratios. Tumor-specific TIL cytolytic reactivity was enhanced after HLA-B7 inoculation.

sion of the class I molecules by tumor cells [21]. This also resulted in tumor regression and induction of T cells that were reactive not only to transfected tumor cells but also to unmodified cancer cells. We have also been able to achieve gene expression in advanced melanoma patients treated by the in vivo inoculation of tumor with DNA/liposome complexes containing a foreign MHC class I gene [22]. Based upon these initial observations, we are currently evaluating the immune reactivity of TILs derived from tumors modified by direct in vivo gene transfer utilizing an allogeneic MHC class I gene, HLA-B7, in patients with stage IV melanoma. We have confirmed, in in vitro assays, an enhanced reactivity of TIL derived from patients inoculated with the foreign class I gene (Fig. 1). We postulate that the allogeneic response induced by the expression of HLA-B7 results in elaboration of other cytokines that enhance TIL reactivity to tumor-associated antigens. The advantage of this approach is that it does not employ a viral vector and does not require the establishment of a cultured tumor line to accomplish gene transfer.

B. Lymph Node Cells Sensitized with Gene-Modified Tumor Vaccines Our laboratory has had a long-standing interest in the use of tumor-draining lymph nodes (TDLNs) or vaccine-primed lymph nodes (VPLN) as a source of T cells for adoptive immunotherapy in murine models and in human trials [23–31]. We have shown that lymphoid cells derived from TDLNs or VPLNs by themselves do not possess antitumor reactivity but require secondary activation in vitro to gain functional antitumor activity. We have called these TDLN or VPLN pre-effector cells. Secondary in vitro activation may be accomplished by coculture with irradiated tumor and IL-2

244


[23,24,26] or the sequential activation with anti-CD3 monoclonal antibody (mAb) and expansion in IL-2 [25,27]. Systemic micrometastases from weakly immunogenic murine tumors such as MCA 205 can be successfully treated using TDLN cells derived from either method of secondary activation. Poorly immunogenic tumors such as the B16–BL6 melanoma, however, fail to sensitize the TDLN cells to become pre-effector cells. To overcome this problem, we have previously shown that the admixture of the B16–BL6 tumor with the potent immunologic adjuvant Corynebacterium parvum primes draining lymph nodes to develop pre-effector cells [28]. These cells, upon secondary in vitro activation, mediated regression of established metastases in murine models. This reinforces the premise that poorly immunogenic tumors can be genetically modified to be more immunogenic. Another method that we have explored to sensitize T cells within the draining lymph node has included the transfection of tumor with an allogeneic MHC class I gene [29]. The B16–BL6 melanoma (H-2b ) was transfected in vivo with an allogeneic MHC class I (H-2d ) gene using lipofection techniques. Cells from the TDLN were sequentially cultured in an anti-CD3 mAb and IL-2. When adoptively transferred into animals with wild-type pulmonary metastases, these activated TDLN cells showed significant antitumor activity compared to parental TDLN cells, which had minimal therapeutic effect. We have found that sensitization of TDLN using tumor cells that have been genetically modified to secrete cytokines has been useful in generating T cells reactive to poorly immunogenic tumors. B16–BL6 tumor that has been transfected with the murine IL-4 gene was used to sensitize the regional TDLN. The pre-effector cells sensitized by the IL-4-secreting tumor were effective in mediating regression of preexisting lung micrometastases in a tumor-bearing animal. The in vivo antitumor reactivity of these lymphocytes was comparable to that of lymphocytes sensitized by wild-type tumor admixed with C. parvum. Another cytokine that we found enhanced the sensitization of the TDLN cells was granulocyte– macrophage colony-stimulating factor (GM-CSF) [30]. GM-CSF is a potent stimulator of macrophages and DCs, which are important antigen-presenting cells involved in the induction of immune responses (See Chapters 7 and 10). Using the B16–BL6 tumor, GM-CSF-secreting tumors were associated with a significant influx of tissue macrophages within the tumor and TDLNs. Activated TDLN cells from mice inoculated with GM-CSF-secreting tumors mediated significant regression of established tumor in adoptive immunotherapy compared with parental TDLN cells, which had no activity. More important, TDLN cells primed with GM-CSF-secreting tumors were more effective in adoptive immunotherapy than were those sensitized by parental tumor admixed with C. parvum or tumor cells transduced to secrete other cytokines (e.g., IL-2, IFN-γ , IL-4) (Fig. 2) [31].

FIGURE 2 Comparison of the adjuvant effect of various cytokines (IL-2, IFN-γ , IL-4, and GM-CSF) elaborated at the site of tumor inoculation for priming pre-effector cells in the tumor-draining lymph nodes. C. parvum was also included as a bacterial adjuvant admixed with tumor cells. The antitumor reactivity of the tumor-draining lymph node cells was assessed in adoptive immunotherapy experiments. Each dot represents a group of animals from separate experiments and the percent reduction of pulmonary metastases recorded for each group. GM-CSF was the most potent adjuvant compared to the other agents studied.

Based upon the above preclinical observations, we conducted a clinical study in patients with advanced melanoma to evaluate the immunobiological effects of retrovirally transduced autologous tumor cells given as a vaccine to prime draining lymph nodes (Fig. 3) [32]. Patients were inoculated with both wild-type (WT) and GM-CSF gene-transduced tumor cells in different extremities. Approximately 7 days later, VPLNs were removed. There was an increased infiltration of DCs in the GM-CSF-secreting vaccine sites compared with the WT vaccine sites (Fig. 4 [see also color insert], Table 2).

FIGURE 3 Schema of a clinical protocol being performed at the University of Michigan. Melanoma patients with stage IV disease have tumor harvested for transduction with a retrovirus encoding GM-CSF. The genemodified tumor cells are inoculated intradermally in the thigh, and the vaccine-primed lymph nodes are harvested one week later for ex vivo activation and expansion. These cells are subsequently transferred back to the patient intravenously along with the concomitant administration of IL-2.

FIGURE 4 Skin vaccine sites sampled from a patient 8 days after injection with either wild-type (A) or granulocyte– macrophage colony-stimulating factor (GM-CSF)-transfected (B to J) autologous melanoma cells. (A) Routine light microscopic appearance of skin injected with mock-transfected melanoma cells with sparse superficial and deep perivascular lymphocyte infiltrate. (B to D) Routine light microscopic appearance demonstrating dense mononuclear cell infiltrate beginning in upper papillary dermis and extending into deep reticular dermis. This inflammatory infiltrate included lymphocyte, neutrophils, and eosinophils. (E) Presence of melanoma cells confirmed by vimentin immunoreactivity. Note that the malignant cells have enlarged nuclei with vesicular chromatic and prominent nucleoli (arrows). (F) Occasional MAC387-positive macrophages are present near melanoma cells (arrows). (G and H) Extensive increase in Factor XIIIapositive dermal dendrocytes with elongated cytoplasmic processes in mid- and deep dermis closely associated with melanoma tumor cells (arrows). (I and J) Numerous CD45RO-positive “memory” T cells are present throughout the dermis admixed with melanoma tumor cells (arrows). (See color insert.)

245

246


TABLE 2 Summary of Immunohistochemical Analysis of Vaccine Sites Scoring of infiltrateb Yield of VPLN (×108 )

Patient no.

Vaccine sitea

Overall

DCs

PMNs

Mφ

1

GM-CSF

4+

4+

2+

0–1+

7

2

GM-CSF WT

5+ 1+

5+ 1+

2+ 1+

0 0

8 2

3

GM-CSF WT

3+ 1+

1+ 1+

1–2+ 1+

0–1+ 0

8 2

4

GM-CSF WT

5+ 3+

2–3+ 1+

1–2+ 2+

0 0

6 1.5

5

GM-CSF WT

4+ 2+

3+ 1–2+

1–2+ 1–2+

0–1+ 0–1+

0.4 0.2

a GM-CSF-transduced (GM-CSF) or nontransduced wild-type (WT) tumor cells were irradiated and inoculated intradermally into opposite thighs. Approximately 7 to 8 days later, the vaccine sites were harvested at the time of VPLN excision. b The infiltrate was scored on a 1- to 5-point scale with 1+ being mild, 3+ moderate, and 5+ severe. The overall infiltrate was scored on the basis of hematoxylin and eosin staining. DCs, dendritic cells; PMNs, polymorphonuclear cells; Mφ, macrophages).

This resulted in a greater number of cells harvested from the GM-CSF VPLNs compared with the WT VPLNs at a time when serum levels of GM-CSF were not detectable. Four patients proceeded to have the adoptive transfer of GM-CSF VPLN cells secondarily activated and expanded ex vivo with anti-CD3 mAb and IL-2. One of these patients has had a durable complete remission of metastatic tumor. In additional animal models, we have utilized the gene gun technology to deliver GM-CSF plasmids to sites of tumor in order to generate tumor-reactive T cells in TDLNs [33]. This procedure provides an alternative nonviral method for introducing cytokine genes into tumor cells to generate effector T cells for adoptive immunotherapy. (For further discussion of gene gun technology, see Chapter 13). Other investigators have utilized similar models to evaluate the sensitization of TDLN cells. Shiloni et al. [34] used the poorly immunogenic MCA 102 sarcoma that was modified with the gene encoding for IFN-γ and reported an increased expression of MHC class I molecules on transduced tumor cells. These cells were inoculated in the flanks of animals to induce an immune response in the TDLNs, which were excised several days later. After secondary in vitro activation, these TDLN cells were adoptive transferred into animals with systemic micrometastases and mediated enhanced antitumor efficacy compared with lymph node cells draining wild-type tumor. In summary, genetic modification of tumor cells to secrete cytokines or express immunomodulatory proteins has shown promise in enhancing the antitumor reactivity of TILs and TDLN cells in animal models. These observations have established the rationale for evaluating these approaches in clinical studies, which our laboratory is currently pursuing.

III. GENETIC MANIPULATION OF T CELLS TO ENHANCE ANTITUMOR REACTIVITY The adoptive transfer of LAK cells, TILs, or TDLN cells in combination with the systemic administration of IL-2 has resulted in the regression of several types of tumors in both murine models and human clinical trials [27,35–38]. Specific limitations in adoptive immunotherapy were recognized early in its use. First, the antitumor activity of adoptive immunotherapy is directly proportional to the administered cell dose, yet the frequency of antigen-reactive T cells in cancer patients is postulated to be extremely low (280 days posttransplant.

M.D. Anderson Cancer Center; Houston, TX

[47]

TABLE 4 (Continued) Patient population Adult AML or ALL

Completed vs. ongoing

Protocol

CD34 selection

Results vs. desired outcomes

Institution

Ref.

Completed

5–19% of BM obtained during second clinical remission (AML) or first clinical remission (ALL) was exposed to the G1N retroviral vector supernatant for 4 hours, and reinfused with untransduced marrow following ablative therapy.

No

Patient had BM and PB + for neo by PCR at 1 year. Neither of the 2 relapsed patients (AML) had neo gene marked leukemic blasts.

Indiana University School of Medicine; Indianapolis, IN

[48]

Pediatric AML in first clinical remission

Ongoing

One third of BM exposed to LNL6, a second third of BM exposed to G1N for 6 hours without cytokine support. Each set of BM would then be exposed to a separate ex vivo purging regimen.

No

With relapse, assess the presence of a neo-specific marker in the leukemic blasts to determine the efficacy of the two purging regimens.

St Jude’s Children’s Hospital; Memphis, TN

[49,55]

Breast cancer or malignant lymphoma

Ongoing

A portion of MPB to be transduced via a 5-day retroviral vector (LN) supernatant exposure in media with IL-1, IL-3, IL-6, and SCF; transduced and untransduced MPB to be reinfused after ablative therapy.

Yes

Study is designed to determine the ability of MPB to contribute to long-term hematopoiesis.

Fred Hutchinson Cancer Center; Seattle, WA

[34]

Metastatic breast cancer or lymphoma

Ongoing

One third of harvested BM and one half of any 2 MPB harvests will be enriched for CD34 cells, preincubated for 42 hours in media with IL-3 and IL-6, and then incubated for 6 hours in LNL6 or G1Na retroviral vector supernatant.

Yes

Differential gene marking of BM and MPB is used to detemine the relative contributions of either to hematopoeitic reconstitution.

USC Comprehensive Cancer Center; Los Angeles, CA

[50]

Multiple myeloma

Ongoing

10–30% of BM will be placed in LTMC and exposed to GTk1. SvNa. 7 (LN derivative with herpes simplex thymidine kinase gene) retroviral vector supernatant 3 times in 21 days.

No

Study is designed to determine the contribution to relapse of the gene-marked autograft.

University of Toronto; Toronto, Canada

[53]

Patients receiving ABMT for leukemias or solid tumors

Ongoing

Two thirds of BM will undergo CD34+ cell selection. One half of the selected population will be exposed to LNL6 and the other half to G1Na for 6 hours. The transduced cells will then be randomized to either cytokine exposure (IL-3, IL-6, and SCF for 5 to 7 days) or no cytokine exposure.

Yes

Study is designed to investigate the ex vivo exposure of CD34+ BM cells to growth factors and the time to hematopoietic engraftment in patients receiving ablative therapy.

St Jude’s Children’s Hospital; Memphis, TN

[51]

Relapsed follicular nonHodgkin’s lymphoma

Ongoing

A portion of the MPB and BM was exposed to either LNL6 or G1Na retroviral vector supernatant without exposure to cytokines or stroma; both transduced and nontransduced MPB and BM were reinfused into patients after ablative therapy.

Yes

2 patients are currently enrolled in the study: At day 45 both patients have neo+ BM cells by PCR (1.5 and 4.3%). The goal of the study is to determine the contribution of contaminated BM or MPB to relapse.

M.D. Anderson Cancer Center, Houston, TX

[54]

Adult recurrent germ cell tumors

Ongoing

CD34+ cells are maintained ex vivo with SCF and IL-6 then cultured with MDR-1 vector on plates coated with CH-296 for 4 hours on days 3 and 4.

Yes

Gene transfer seen in: 25% at oneyear in BM. 5.6% at 1 month and 0.5% at 9 months in circulating mature cells.

Indiana University School of Medicine; Indianapolis, IN

[71]

Note: ABMT, autologous bone marrow transplant; ALL, acute lymphoblastic leukemia; AML, acute monocytic leukemia; CML, chronic monocytic leukemia; SCF, stem cell factor; MDR, multiple drug resistance.

266

Edsel U. Kim, Lee G. Wilke and James J. Mul´ e

a small population of functioning HSCs that could reconstitute their hematopoeitic system despite the myeloablative therapy. Of the completed studies, it can be concluded that human HSCs can be “marked” with a retroviral vector containing a neomycin-resistant gene [43,54–57]. Brenner et al. revealed that five of five evaluable pediatric patients with acute monocytic leukemia (AML) or neuroblastoma had from 0 to approximately 15% of cells in their BM that were G418 resistant in CFU assay at 1 year posttransplant. These investigators thus concluded that harvested autologous BM does contribute to hematopoietic recovery after ablative chemotherapy [54]. Cornetta et al. also demonstrated that in an adult leukemic patient, treated with a BMT in which a portion of the HSC had been exposed to a retroviral vector with the neomycin-resistant gene, both BM and PB cells contained the transgene 1 year posttransplant [56]. Deisseroth et al. found that one of their adult chronic myelogenous leukemia (CML) patients, who had received a PBSCT, had cells in the peripheral blood that were positive via reverse transcription–polymerase chain reaction (RT-PCR) for the neomycin gene, 280 days posttransplant. This finding led to the conclusion that reinfused, retrovirally exposed, mobilized peripheral blood, as well as BM, could contribute to the patient’s hematopoietic recovery [57]. Dunbar et al. reported the long-term presence of a marker gene in the circulating granulocytes and T and B cells (0.01–1% of cells) of two patients with myeloma and one patient with breast cancer 18 months after a combined transplant with marked bone marrow and mobilized peripheral blood [4,43]. By using two different vectors with the neomycin gene, the investigators were able to discriminate the source of the engrafted cells after the transplant. Using a transduction protocol similar to that of Brenner et al., in which the HSCs were exposed to vector supernatant for 6 hours without cytokine support, Dunbar and collegues were unable to demonstrate gene marking in adult patients with breast cancer or myeloma who received a BMT or PBSCT [58]. Further study into the conditions necessary for optimal HSC gene transduction for individual cancers at different stages will need to be performed to resolve the discrepancies in results from different laboratories. Despite the differences in outcomes, it has been possible to use the technology of gene marking to address a clinically relevant question about the restoration of human hematopoiesis [59]. As shown in Table 4, several ongoing trials seek to further address the question of the source of hematopoeitic recovery after myeloablation [39,60,61]. A second question of interest to those investigators focusing on the biology of BMT and PBSCT in the cancer setting is that of the origin of relapse after transplantation for leukemia. Brenner et al. and Deisseroth et al. both discovered that, in some of their patients who relapsed from pediatric AML, pediatric neuroblastoma, and adult CML, the leukemic blasts contained the neomycin gene. The authors thus concluded that the reinfused autologous BM or MPB was a contribu-

tor to the cancer recurrence [55,57,59,61]. Cornetta et al. reported the relapse of two of four patients with AML who were treated with BMT, in which a portion of the graft was exposed to a retroviral vector with the neomycin resistant gene. However, neither patient had evidence of vector-marked leukemic blasts [56]. Clinical studies are ongoing to further clarify this question concerning the source of relapse in patients with varied forms of cancer at different clinical stages, including multiple myeloma and non-Hodgkin’s lymphoma [62,63]. A third area of interest is the role that “purging” of the BMTs and/or PBSCTs with cytokines or chemotherapeutic agents prior to reinfusion into the patient will play in preventing disease recurrence. Brenner et al. have initiated studies using two distinct retroviral vectors with the neomycinresistant gene in pediatric AML patients to address this question [64]. As improvements in vector design and HSC transduction occur, more clinical studies can be undertaken to further elucidate the biology of BMTs and PBSCTs and those ex vivo and in vivo therapies that will benefit the patient with cancer.

C. Drug Resistance Genes in Hematopoietic Stem Cells In addition to its role in the study of BMTs and PBSCTs, gene therapy with HSCs has been introduced as a means to potentially improve existing transplant protocols. The dose of chemotherapeutic agents used to treat malignancies is limited to some extent by myelosuppression. If a patient’s bone marrow were tolerant to the antineoplastic drugs, increasing doses of the agents could be used to improve disease eradication. The multiple drug resistance gene-1 (MDR1) encodes a 170-kDa P-glycoprotein which functions as an adenosine-triphosphate-dependant efflux pump for lipophilic compounds [65]. Those chemotherapeutic agents that are “removed” by cells with the MDR1 pump include the anthracyclines, the vinca alkaloids, the epipodophyllotoxins, actinomycin D, and TaxolR [65]. As previously noted, murine HSCs have been transduced with the human MDR1 gene via a retroviral vector, and selection with TaxolR has been successful [35]. The ability to select for MDR1-containing cells in humans would conceivably permit dose escalation of the antineoplastic agents as well as decreased intervals between chemotherapy administration due to improved hematopoietic recovery [65]. Several human clinical trials, with primarily breast and/or ovarian cancer patients, are currently ongoing to answer the question of whether human HSCs from BM and/or MPB can be transduced without toxicity to the HSCs and in adequate quantities to permit growth selection with TaxolR [66–69]. Preliminary results from one group indicated that the CD34+ cells from the BM and MPB of ovarian and breast cancer patients, respectively, could be transduced with the MDR1 gene by exposing the cells to the retroviral supernatant on a stromal monolayer with cytokine supplementation

Update on the Use of Genetically Modified Hematopoietic Stem Cells for Cancer Therapy

in the culture media. The transgene was identified in the BM of five of eight evaluable patients using a solution DNA PCR assay [70]. A more recent trial in patients with recurrent germ cell tumors has shown the highest level of transgene expression using the fibronectin fragment CH-296 in the transduction of HSCs with a retroviral vector containing the MDR1 gene [71]. Approximately 25% of the clonogenic progenitors were transduced in the bone marrow at 1 year with approximately 0.9% at 9 months in the peripheral blood. While the level of transduction was high, expression of the full-length MDR1 cDNA ranged from 10%–50% in the first month. Much of this loss could be attributed to cryptic splice sites that were found in the MDR1 gene, resulting in a nonfunctional product. While the levels of transduction from this trial are the highest reported to date, the subtherapeutic levels of functional expression show that more progress is needed before this technology can enter the clinical arena. As results from the ongoing trials with MDR1-transduced HSCs become available and indicate the feasibility of BM selection, a second use for the drug-resistance genes may arise. Retroviral constructs containing the MDR1 gene and a second therapeutic gene could be introduced into the HSCs and those cells with the transgene selected for using TaxolR [65]. This strategy would then increase the frequency of genemodified cells with a potentially therapeutic antitumor gene (see later discussion). Without the selection process, the antitumor gene might only be present in a small number of HSCs due to low transduction efficiency with current retroviral vectors. With selection, however, the number of cells with the transgene DNA would increase, thereby improving the therapeutic potential after the BMT and/or PBSCT.

D. Chimeric Receptor Genes in Hematopoietic Stem Cells “Purging” autologous marrow or peripheral blood prior to reinfusion after myeloablative therapy represents one means to address the problem of disease recurrence. Another means of approaching this problem in the cancer patient might be to provide the patient receiving a gene-modified HSC transplant with an immune system that is biased to recognize autologous tumor. Eshhar et al. developed a chimeric single-chain receptor consisting of the Fv domain (scFv) of an antibody linked with the γ or ζ chains, the signal-transducing subunits of the immunoglobulin receptor and the T-cell receptor (TCR) [72]. The scFv domains, which are a combination of the heavy and light variable regions (VH and VL ) of the antibody, appear to possess the specificity and affinity of the intact Fab fragment [72]. Originally, a chimeric gene with scFv from an antitrinitophenyl (TNP) antibody was found to be expressed as a functional surface receptor in a murine cytolytic T-cell line and to specifically secrete IL-2 upon exposure to the TNP antigen [72]. T-cell activation via this

267

receptor is not major histocompatibility (MHC) restricted because of antibody recognition, as opposed to peptide recognition with the TCR complex. Studies utilizing the chimeric receptor were expanded to investigate antitumor antibodies. Stancoviski et al. introduced a chimeric gene with an antiHER2/neu antibody into a murine Cytolytic T-lymphocyte (CTL) line and demonstrated specific antigen recognition and lysis of cells overexpressing Neu/HER2 [73]. Moritz et al. [74] similarly utilized an anti-HER2/neu scFv linked to the ζ chain of the TCR complex and transduced a murine CTL line. Target cells that overexpress HER2/neu were lysed in vitro by the transduced CTLs, and the growth of HER2/neu-transformed NIH3T3 cells in nude mice was retarded, though not prevented, by CTLs transduced by the chimeric receptor and subsequently injected into the mice [74]. Moving beyond the stable cell lines, Hwu et al. transduced CD8+ human CTLs with a chimeric receptor designed to recognize a defined human ovarian carcinoma antigen [75]. Again, the transduced cells recognized target cells with the ovarian cancer antigen and secreted granulocyte– macrophage-colony stimulating factor (GM-CSF) upon incubation with the specific antigen [75]. The next step toward applying this technology to the therapy of human cancer involves transduction of human HSCs with a chimeric receptor. Figure 1 depicts a model therapeutic strategy for patients receiving a chimeric receptor-genemodified HSC transplant. The progeny lymphocytes and myeloid cells (monocytes and neutrophils) from the transduced HSCs would potentially have the ability to recognize and kill in a non-MHC-restricted fashion those cells expressing the specific tumor antigen incorporated into the chimeric gene. Because the HSCs are self-renewing and pluripotent, they would offer the cancer patient receiving a BMT and/or PBSCT a long-term immune system with antitumor function. We have recently introduced into human CD34+ cells, isolated from MPB, the gene encoding for a chimeric receptor to HER2/neu [76]. Following selection, under G418, 81% of the colonies derived from the transduced HSCs (11%) contained the transgene. We are now planning a phase I clinical trial of chimeric receptor-gene-modified HSCs into advanced breast cancer patients receiving PBSCT. Because of the concern that recognition of “normal” levels of HER2/neu on normal tissues, as opposed to “overexpressed” levels on tumor, might lead to toxicity, vector constructs will also contain suicide cytosine deaminase (CD) or thymidine kinase (TK) genes to eliminate expressing effector progeny cells if adverse targeting occurs. Another strategy based on redirecting the immune response to recognize tumor-associated antigens after a BMT or PBSCT is the introduction of genes encoding “classic” TCRs. Unlike the chimeric receptor approach, this strategy would be limited to progeny T cells only, would be MHC restricted, and would be more specific to tumor cells (as opposed to normal tissue). Recently, TCRs for a cytotoxic CTL-defined peptide

268


FIGURE 1 HSC gene transfer: chimeric TCR approach.

expressed by HLA-A2 on melanoma has been cloned [77]. The prohibitive limitation of this approach, however, may be the difficulty in successfully expressing two independent genes that encode the separate alpha and beta chains of the TCRs in HSC target cells and their T-lymphoid progeny.

V. CONCLUSIONS Genetic modification of human HSCs for cancer therapy remains in its infancy but already has shown great potential as a tool to enhance the disease-free survival of patients undergoing a BMT and/or PBSCT for bloodborne or solid tumors. Areas of ongoing and future focus include novel vectors and improvement on existing retroviral vectors for HSC transduction, ex vivo HSC proliferation and the cytokines necessary to support survival of the self-renewing pluripotent HSCs, and in vivo models to study transduced HSCs. As human clinical trials are completed, more information concerning the biology of BMTs and/or PBSCTs, the feasibility of HSC transduction with multidrug resistance genes, and their ability to provide bone marrow protection will become available. These data will provide information on the safety of this technology and offer a stepping stone toward the use of dual drugresistance and therapeutic genes, including those encoding receptor molecules that recognize tumor-associated antigens.

References 1. Baum, C. M., Weissman, I. L., Tsukamoto, A. S. et al. (1992). Isolation of a candidate human hematopoietic stem-cell population. Proc. Natl. Acad. Sci. USA 89(7), 2804–2808.

2. Morrison, S. J., Uchida, N., and Weissman, I. L. (1995). The biology of hematopoietic stem cells. Annu. Rev. Cell. Dev. Biol. 11, 35–71. 3. Roth, J. A., and Cristiano, R. J. Gene therapy for cancer: what have we done and where are we going? J. Natl. Cancer Inst., (1997). 89(1), 21–39. 4. Dunbar, C. E. (1996). Gene transfer to hematopoietic stem cells: implications for gene therapy of human disease. Annu. Rev. Med. 47, 11–20. 5. Brenner, M. K. (1996). Gene transfer to hematopoietic cells. N. Engl. J. Med. 335(5), 337–339. 6. Kerr, W. G., and Mule, J. J. (1994). Gene therapy: current status and future prospects. J. Leukoc. Biol. 56(2), 210–214. 7. Bagnis, C., Chabannon, C., and Mannoni, P. (1996). Gene transfer into haemopoietic cells: a challenge for gene therapy. European Concerted Action Workshop, Marseille, 18–19 October, 1995. Gene Ther. 3(4), 362–364. 8. Schwarzenberger, P., Spence, S. E., Gooya, J. M. et al. (1996). Targeted gene transfer to human hematopoietic progenitor cell lines through the c-kit receptor. Blood 87(2), 472–478. 9. Fisher-Adams, G., Wong, Jr., K. K., Podsakoff, G. et al. (1996). Integration of adeno-associated virus vectors in CD34+ human hematopoietic progenitor cells after transduction. Blood 88(2), 492–504. 10. Geraghty, P. J., and Chang, A. E. (1995). Basic principles associated with gene therapy of cancer. Surg. Oncol. 4(3), 125–137. 11. Adam, M. A., and Miller, A. D. (1988). Identification of a signal in a murine retrovirus that is sufficient for packaging of nonretroviral RNA into virions. J. Virol. 62(10), 3802–3806. 12. Rollins, S. A., Birks, C. W., Setter, E. et al. (1996). Retroviral vector producer cell killing in human serum is mediated by natural antibody and complement: strategies for evading the humoral immune response. Hum. Gene Ther. 7(5), 619–626. 13. Donahue, R. E., Kessler, S. W., Bodine, D. et al. (1992). Helper virus induced T cell lymphoma in nonhuman primates after retroviral mediated gene transfer. J. Exp. Med. 176(4), 1125–1135. 14. Purcell, D. F., Broscius, C. M., Vanin, E. F. et al. (1996). An array of murine leukemia virus-related elements is transmitted and expressed in a primate recipient of retroviral gene transfer. J. Virol. 70(2), 887–897. 15. Kohn, D. B. (1997). Gene therapy for haematopoietic and lymphoid disorders. Clin. Exp. Immunol. 107(suppl. 1), 54–57.

Update on the Use of Genetically Modified Hematopoietic Stem Cells for Cancer Therapy 16. Mastrangelo, M. J., Berd, D., Nathan, F. E. et al. (1996). Gene therapy for human cancer: an essay for clinicians. Semin. Oncol. 23(1), 4–21. 17. Medin, J. A., and Karlsson, S. (1997). Viral vectors for gene therapy of hematopoietic cells. Immunotechnology 3(1), 3–19. 18. Dick, J. E., Kamel-Reid, S., Murdoch, B. et al. (1991). Gene transfer into normal human hematopoietic cells using in vitro and in vivo assays. Blood 78(3), 624–634. 19. Cassel, A., Cottler-Fox, M., Doren, S. et al. (1993). Retroviral-mediated gene transfer into CD34-enriched human peripheral blood stem cells. Exp. Hematol. 21(4), 585–591. 20. Moritz, T., Keller, D. C., and Williams, D. A. (1993). Human cord blood cells as targets for gene transfer: potential use in genetic therapies of severe combined immunodeficiency disease. J. Exp. Med. 178(2), 529– 536. 21. Lu, M., Maruyama, M., Zhang, N. et al. (1994). High efficiency retroviral-mediated gene transduction into CD34+ cells purified from peripheral blood of breast cancer patients primed with chemotherapy and granulocyte–macrophage colony-stimulating factor. Hum. Gene Ther. 5(2), 203–208. 22. Akkina, R. K., Rosenblatt, J. D., Campbell, A. G. et al. (1994). Modeling human lymphoid precursor cell gene therapy in the SCID-hu mouse. Blood 84(5), 1393–1398. 23. Champseix, C., Marechal, V., Khazaal, I. et al. (1996). A cell surface marker gene transferred with a retroviral vector into CD34+ cord blood cells is expressed by their T-cell progeny in the SCID- hu thymus. Blood 88(1), 107–113. 24. Conneally, E., Bardy, P., Eaves, C. J. et al. (1996). Rapid and efficient selection of human hematopoietic cells expressing murine heat-stable antigen as an indicator of retroviral-mediated gene transfer. Blood 87(2), 456–464. 25. Medin, J. A., Migita, M., Pawliuk, R. et al. (1996). A bicistronic therapeutic retroviral vector enables sorting of transduced CD34+ cells and corrects the enzyme deficiency in cells from Gaucher patients. Blood 87(5), 1754–1762. 26. Nolta, J. A., Dao, M. A., Wells, S. et al. (1996). Transduction of pluripotent human hematopoietic stem cells demonstrated by clonal analysis after engraftment in immune-deficient mice. Proc. Natl. Acad. Sci. USA 93(6), 2414–2419. 27. Bauer, Jr., T. R., and Hickstein, D. D. (1997). Transduction of human hematopoietic cells and cell lines using a retroviral vector containing a modified murine CD4 reporter gene. Hum. Gene Ther. 8(3), 243– 252. 28. Emerson, S. G., (1996). Ex vivo expansion of hematopoietic precursors, progenitors, and stem cells: the next generation of cellular therapeutics. Blood 87(8), 3082–3088. 29. Arca, M. J., Mule, J. J., and Chang, A. E. (1996). Genetic approaches to adoptive cellular therapy of malignancy. Semin. Oncol. 23(1), 108–117. 30. Herrmann, F. (1995). Cancer gene therapy: principles, problems, and perspectives. J. Mol. Med. 73(4), 157–63. 31. Kafri, T., van Praag, H., Ouyang, L. et al. (1999). A packaging cell line for lentivirus vectors. J. Virol. 73(1), 576–584. 32. Miyoshi, H., Smith, K. A., Mosier, D. E. et al. (1999). Transduction of human CD34+ cells that mediate long-term engraftment of NOD/SCID mice by HIV vectors. Science 283(5402), 682–686. 33. Miyoshi, H., Blomer, U., Takahashi, M. et al. (1998). Development of a self-inactivating lentivirus vector. J. Virol. 72(10), 8150–8157. 34. Wilson, J. M., Danos, O., Grossman, M. et al. (1990). Expression of human adenosine deaminase in mice reconstituted with retrovirustransduced hematopoietic stem cells. Proc. Natl. Acad. Sci. USA 87(1), 439–443. 35. Sorrentino, B. P., Brandt, S. J., Bodine, D. et al. (1992). Selection of drugresistant bone marrow cells in vivo after retroviral transfer of human MDR1. Science 257(5066), 99–103. 36. Bodine, D. M., McDonagh, K. T., Seidel, N. E. et al. (1991). Survival and retrovirus infection of murine hematopoietic stem cells in vitro: effects of 5-FU and method of infection. Exp. Hematol. 19(3), 206–2012.

269

37. Fraser, C. C., Eaves, C. J., Szilvassy, S. J. et al. (1990). Expansion in vitro of retrovirally marked totipotent hematopoietic stem cells. Blood 76(6), 1071–1076. 38. Sorrentino, B. P., McDonagh, K. T., Woods, D. et al. (1995). Expression of retroviral vectors containing the human multidrug resistance 1 cDNA in hematopoietic cells of transplanted mice. Blood 86(2), 491– 501. 39. Schuening, F., Miller, A. D., Torok-Storb, B. et al. (1994). Study on contribution of genetically marked peripheral blood repopulating cells to hematopoietic reconstitution after transplantation. Hum. Gene Ther. 5(12), 1523–1534. 40. Moritz, T., Patel, V. P., and Williams, D. A. (1994). Bone marrow extracellular matrix molecules improve gene transfer into human hematopoietic cells via retroviral vectors. J. Clin. Invest. 93(4), 1451–1457. 41. Traycoff, C. M., Srour, E. F., Dutt, P. et al. (1997). The 30/35 kDa chymotryptic fragment of fibronectin enhances retroviral-mediated gene transfer in purified chronic myelogenous leukemia bone marrow progenitors. Leukemia 11(1), 159–167. 42. Hanenberg, H., Xiao, X. L., Dilloo, D. et al. (1996). Colocalization of retrovirus and target cells on specific fibronectin fragments increases genetic transduction of mammalian cells. Nat. Med. 2(8), 876–882. 43. Dunbar, C. E., Cottler-Fox, M., O’Shaughnessy, J. A. et al. (1995). Retrovirally marked CD34-enriched peripheral blood and bone marrow cells contribute to long-term engraftment after autologous transplantation. Blood 85(11), 3048–3057. 44. Bordignon, C., Notarangelo, L. D., Nobili, N. et al. (1995). Gene therapy in peripheral blood lymphocytes and bone marrow for ADA- immunodeficient patients. Science 270(5235), 470–475. 45. Lu, M. Zhang, N., Maruyama, M. et al. (1996). Retrovirus-mediated gene expression in hematopoietic cells correlates inversely with growth factor stimulation. Hum. Gene Ther. 7(18), 2263–2271. 46. Leonard, J., May, C., Gallardo, H. et al. (1996). Retroviral transduction of human hematopoietic progenitor cells using a vector encoding a cell surface marker (LNGFR) to optimize transgene expression and characterize transduced cell populations. Blood 88, 433a. 47. Ward, M., Pioli, P., Ayello, J. et al. (1996). Retroviral transfer and expression of the human multiple drug resistance (MDR) gene in peripheral blood progenitor cells. Clin. Cancer Res. 2(5), 873–876. 48. Carter, R. F., Abrams-Ogg, A. C., Dick, J. E. et al. (1992). Autologous transplantation of canine long-term marrow culture cells genetically marked by retroviral vectors. Blood 79(2), 356–364. 49. Bodine, D. M., Moritz, T., Donahue, R. E. et al. (1993). Long-term in vivo expression of a murine adenosine deaminase gene in rhesus monkey hematopoietic cells of multiple lineages after retroviral mediated gene transfer into CD34+ bone marrow cells. Blood 82(7), 1975– 1980. 50. Bienzle, D., Abrams-Ogg, A. C., Kruth, S. A. et al. (1994). Gene transfer into hematopoietic stem cells: long-term maintenance of in vitro activated progenitors without marrow ablation. Proc. Natl. Acad. Sci. USA 91(1), 350–354. 51. Donahue, R. E., Byrne, E. R., Thomas, T. E. et al. (1996). Transplantation and gene transfer of the human glucocerebrosidase gene into immunoselected primate CD34+Thy-1+ cells. Blood 88(11), 4166–4172. 52. Bodine, D. M., McDonagh, K. T., Brandt, S. J. et al. (1990). Development of a high-titer retrovirus producer cell line capable of gene transfer into rhesus monkey hematopoietic stem cells. Proc. Natl. Acad. Sci. USA 87(10), 3738–3742. 53. Cavazzana-Calvo, M., Hacein-Bey, S., de Saint Basile, G. et al. (2000). Gene therapy of human severe combined immunodeficiency (SCID)-X1 disease [see comments]. Science 288(5466), 669–672. 54. Brenner, M. K., Rill, D. R., Holladay, M. S. et al. (1993). Gene marking to determine whether autologous marrow infusion restores long-term haemopoiesis in cancer patients. Lancet 342(8880), 1134–1137. 55. Rill, D. R., Santana, V. M., Roberts, W. M. et al. (1994). Direct demonstration that autologous bone marrow transplantation for solid tumors can return a multiplicity of tumorigenic cells. Blood 84(2), 380–383.

270


56. Cornetta, K., Srour, E. F., Moore, A. et al. (1996). Retroviral gene transfer in autologous bone marrow transplantation for adult acute leukemia. Hum. Gene Ther. 7(11), 1323–1329. 57. Deisseroth, A. B., Zu, Z., Claxton, D. et al. (1994). Genetic marking shows that Ph+ cells present in autologous transplants of chronic myelogenous leukemia (CML) contribute to relapse after autologous bone marrow in CML. Blood 83(10), 3068–3076. 58. Emmons, R. V., Doren, S., Zujewski, J. et al. (1997). Retroviral gene transduction of adult peripheral blood or marrow-derived CD34+ cells for six hours without growth factors or on autologous stroma does not improve marking efficiency assessed in vivo. Blood 89(11), 4040– 4046. 59. Heslop, H. E., Rooney, C. M., Rill, D. R. et al. (1996). Use of gene marking in bone marrow transplantation. Cancer Detect. Prev. 20(2), 108–113. 60. Douer, D., Levine, A., Anderson, W. F. et al. (1996). High-dose chemotherapy and autologous bone marrow plus peripheral blood stem cell transplantation for patients with lymphoma or metastatic breast cancer: use of marker genes to investigate hematopoietic reconstitution in adults. Hum. Gene Ther. 7(5), 669–684. 61. Brenner, M. K., Rill, D. R., Moen, R. C. et al. (1993). Gene-marking to trace origin of relapse after autologous bone-marrow transplantation. Lancet 341(8837), 85–86. 62. Stewart, A. K., Dube, I. D., Kamel-Reid, S. et al. (1995). A phase I study of autologous bone marrow transplantation with stem cell gene marking in multiple myeloma. Hum. Gene Ther. 6(1), 107–119. 63. Bachier, C., Giles, R., Ellerson, D. et al. (1996). Retroviral gene marking in relapsed follicular non-Hodgkin’s lumphoma (FNHL) to determine the origin of relapse following autologous bone marrow (ABMT) and peripheral stem cell transplant (PSCT) [meeting abstract]. Proc. Annu. Meet. Am. Assoc. Cancer Res. 37, A1398. 64. Brenner, M., Krance, R., Heslop, H. E. et al. (1994). Assessment of the efficacy of purging by using gene marked autologous marrow transplantation for children with AML in first complete remission. Hum. Gene Ther. 5(4), 481–499. 65. Koc, O. N., Allay, J. A., Lee, K. et al. (1996). Transfer of drug resistance genes into hematopoietic progenitors to improve chemotherapy tolerance. Semin. Oncol. 23(1), 46–65. 66. Hesdorffer, C., Antman, K., Bank, A. et al. (1994). Human MDR gene transfer in patients with advanced cancer. Hum. Gene Ther. 5(9), 1151– 1160. 67. Deisseroth, A. B., Kavanagh, J., and Champlin, R. (1994). Use of safety-modified retroviruses to introduce chemotherapy resistance se-

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

quences into normal hematopoietic cells for chemoprotection during the therapy of ovarian cancer: a pilot trial. Hum. Gene Ther. 5(12), 1507–1522. Deisseroth, A. B., Holmes, F., Hortobagyi, G. et al. (1996). Use of safetymodified retroviruses to introduce chemotherapy resistance sequences into normal hematopoietic cells for chemoprotection during the therapy of breast cancer: a pilot trial. Hum. Gene Ther. 7(3), 401–416. O’Shaughnessy, J. A., Cowan, K. H., Nienhuis, A. W. et al. (1994). Retroviral mediated transfer of the human multidrug resistance gene (MDR-1) into hematopoietic stem cells during autologous transplantation after intensive chemotherapy for metastatic breast cancer. Hum. Gene Ther. 5(7), 891–911. Hanania, E. G., Giles, R. E., Kavanagh, J. et al. (1996). Results of MDR-1 vector modification trial indicate that granulocyte/macrophage colony-forming unit cells do not contribute to posttransplant hematopoietic recovery following intensive systemic therapy [published erratum appears in Proc. Natl. Acad. Sci. USA 94(10), 5495,1997]. Proc. Natl. Acad. Sci. USA 93(26), 15346–15351. Abonour, R., Williams, D. A., Einhorn, L. et al. (2000). Efficient retrovirus-mediated transfer of the multidrug resistance 1 gene into autologous human long-term repopulating hematopoietic stem cells. Nat. Med. 6(6), 652–658. Eshhar, Z., Waks, T., Gross, G. et al. (1993). Specific activation and targeting of cytotoxic lymphocytes through chimeric single chains consisting of antibody-binding domains and the gamma or zeta subunits of the immunoglobulin and T-cell receptors. Proc. Natl. Acad. Sci. USA 90(2), 720–724. Stancovski, I., Schindler, D. G., Waks, T. et al. (1993). Targeting of T lymphocytes to Neu/HER2-expressing cells using chimeric single chain Fv receptors. J. Immunol. 151(11), 6577–6582. Moritz, D., Wels, W., Mattern, J. et al. (1994). Cytotoxic T lymphocytes with a grafted recognition specificity for ERBB2-expressing tumor cells. Proc. Natl. Acad. Sci. USA 91(10), 4318–4322. Hwu, P., Shafer, G. E., Treisman, J. et al. (1993). Lysis of ovarian cancer cells by human lymphocytes redirected with a chimeric gene composed of an antibody variable region and the Fc receptor gamma chain. J. Exp. Med. 178(1), 361–366. Wilke, L., Reynolds, C., McDonough, K. et al. (1997). Engineering immune effector cells to target the HER-2/neu breast cancer antigen (meeting abstract). SSO Annual Cancer Symposium, 51, P4 Cole, D. J., Weil, D. P., Shilyansky, J. et al. (1995). Characterization of the functional specificity of a cloned T-cell receptor heterodimer recognizing the MART-1 melanoma antigen. Cancer Res. 55(4), 748–752.

P A R T III

ONCOGENE-TARGETED GENE THERAPY


C

H

A

P

T

E

R

16 Clinical Applications of Tumor-Suppressor Gene Therapy

I. II. III. IV.

RAYMOND D. MENG

WAFIK S. EL-DEIRY

Laboratory of Molecular Oncology and Cell Cycle Regulation Howard Hughes Medical Institute Departments of Medicine and Genetics Cancer Center and The Institute for Human Gene Therapy University of Pennsylvania School of Medicine Philadelphia, Pennsylvania 19104


Introduction 273 p53 273 BRCA1 275 ONYX-015 Adenoviruses

tosis makes the p53 tumor suppressor an important target for replacement. 275

II. p53

A. Introduction 275 B. Gene Therapy 275 C. Clinical Trials 276

V. Summary and Future Work

Currently, of the tumor suppressors being considered for gene replacement in cancer, only the p53 gene is being extensively tested in clinical trials (see Table 1). Much of the pioneering work with p53 gene therapy was conducted by Jack Roth and colleagues at the M.D. Anderson Cancer Center. They reported the results of the first clinical trials using p53 delivered by a retrovirus vector to treat patients with non-small-cell lung cancer who had failed other treatments [28]. The virus was administered intratumorally and caused no toxic side effects up to 5 months later. Wild-type p53 was detected in lung biopsies by in situ hybridization and PCR amplification, and apoptosis (as determined by the TUNEL assay) was increased in posttreatment biopsy samples. Of the nine patients in the study, three showed tumor growth stabilization, and three showed slight tumor regression. It was also reported by the same group that tumor growth inhibition was enhanced when p53 gene therapy was combined with systemic cisplatin [26]. In another phase I study of patients with non-smallcell lung cancers, p53-expressing adenovirus (Ad-p53) treatments only stabilized two of 15 patients, with one patient being stable at more than 6 months following his treatments [32]. A comprehensive phase I study using Ad-p53 for recurrent head and neck cancers was also recently reported [3].

276

References 277

I. INTRODUCTION Multiple human tumors show mutations or deletions in tumor suppressor genes, which control cellular growth by regulating the cell cycle or by inducing cellular apoptosis. In particular, it has been shown that the p53 tumor suppressor gene is mutated or deleted in over half of all human malignancies (reviewed by Levine [see Ref. 83 in Chapter 18]). Hence, one strategy in cancer gene therapy has focused on the replacement or overexpression of tumor suppressors genes. Over the past decade, extensive research has been conducted in vitro with multiple tumor suppressors, including p53, p21WAF1/CIP1 , p16, Rb, p14ARF , p27, E2F-1, BRCA1, VHL, and FHIT. Overexpression of tumor suppressors in selective cancer cell lines has resulted in either growth suppression or apoptosis. Currently, clinical trials for cancer gene therapy are being conducted with select tumor suppressor genes, most notably the p53 gene. Its dual role in both cell-cycle arrest and in apop-


273


274

Raymond D. Meng and Wafik S. El-Deiry

TABLE 1 Selected p53 Gene Therapy Clinical Trials in 2000 Cancer Bladder cancer

Mode of treatment

Site of trial

Results

Adenovirus

Houston, TX

Phase I planned

Adenovirus

Hamburg, Germany; New Brunswick, New Jersey; San Diego, CA

Phase I completed; intravesical injection more effective than intratumoral injection; both are safe [31]

Colorectal cancer

Adenovirus

Houston, TX

Phase I planned

Glioblastoma

Adenovirus

Houston, TX

Phase I planned

Head–neck, squamous cell carcinoma

Adenovirus; multiple injections

Houston, TX

Phase I completed; 2 of 17 nonresectable had partial regression; 40% of resectable were disease-free for 6 months; Phase II in progress

Head–neck squamous cell cancer, mutant p53

Adenovirus; intratumoral

Pittsburgh, PA; London, England

Phase I planned; 6–18 patients

Liver cancer, primary and metastatic; mutant p53

Adenovirus; hepatic artery infusions

San Francisco, CA; Philadelphia, PA


Liver cancer

Adenovirus

Houston, TX

Phase I planned

Malignant ascites

Adenovirus

Houston, TX

Phase I planned

Non-small-cell lung cancer

Adenovirus and retrovirus; intratumoral

Houston, TX

Phase I completed, Ad-p53 slowed growth at high doses; phase II in progress.

Non-small-cell lung cancer, mutant p53

Adenovirus; intratumoral

Mainz, Germany; Basel, Switzerland


Ovarian cancer

Adenovirus; combined with cisplatin

Houston, TX

Phase I planned

Ovarian, fallopian tube, or peritoneal cancer; mutant p53

Adenovirus; intraperitoneal

Karolinksa, Sweden; Iowa City, IA


Prostate cancer

Adenovirus

Houston, TX

Phase I planned; intraprostatic injections in 30 patients

Source: Data compiled from Roth and Cristiano [39], National Cancer Institute PDQ Clinical Trial Database, and Genetic Engineering News, June 15, 1997.

Patients with surgically nonresectable and previously irradiated head and neck cancers were given cycles of treatment consisting of three injections of Ad-p53 per week for 2 weeks followed by 2 weeks of rest. In 18 patients, only two patients experienced a decrease in tumor size of more than 50%, although this remission lasted almost 8 months. Six patients, however, had no change in tumor size. The major side effects encountered in this study were pain at the injection site and fever and headache with the higher doses of Ad-p53 used. Following treatment, the authors were able to detect antibodies to the adenovirus in the blood and could detect p53 in the blood and urine [3]. In addition, a phase I trial using Ad-p53 to treat invasive bladder cancers was recently completed [31]. Patients with locally invasive bladder cancers were administered Ad-p53 either by cystoscopically aided guidance or by intravesical treatment. Major side effects of the treatment consisted of

urethral discomfort or abdominal pain. In five of six patients given intravesical administration of Ad-p53, PCR could detect p53 expression 3 days later; however, in the patients with intratumoral injections of Ad-p53, no p53 protein could be detected at that time. Another study examined the utility of using computed tomography (CT) of the chest to document the effectiveness of Ad-p53 gene therapy for non-small-cell lung cancers [24]. From 33 tumors administered Ad-p53, CT scans showed no change in 20 tumors, a decrease in size in six tumors, and an increase in size in seven tumors. A biopsy from the lung tumors revealed that in two of the six tumors that had decreased in size no evidence of malignant cells could be found; in contrast, in the seven tumors that increased in size, the biopsies were all positive for malignant cells. Unfortunately, no significant features from the CT scan, such as the detection of necrosis or of decreased attenuation, were specific enough to document the efficacy of

Clinical Applications of Tumor-Suppressor Gene Therapy

Ad-p53 treatments for lung tumors. Phase I trials of Ad-p53 have also been planned for metastatic liver tumors, whereby Ad-p53 will be infused through the hepatic artery [10]; for locally advanced prostate cancers, whereby Ad-p53 will be administered directly by intraprostatic injections [33]; and for non-small-cell lung cancers [37].

III. BRCA1 One clinical trial is currently being conducted at Vanderbilt University to examine the feasibility of BRCA1-mediated gene therapy for the treatment of ovarian cancers. Unlike the p53 clinical trials, this study utilized a retrovirus vector to deliver BRCA1. Following injection into patients, the retroviral BRCA1 vector was stable, generated minimal antibody responses against itself, and produced subtle decreases in tumor size [34]. The clinical trials then progressed to a phase II level, but the results were unfortunately not very successful. Retroviral BRCA1 produced no decreases in tumor size and also generated strong antibody responses [34]. The authors felt that the retroviral vector was more immunogenic because it was being packaged in mouse cells; therefore, a new retroviral BRCA1 vector was developed that could be packaged in human 293 cells [35]. Currently, further trials utilizing retrovirus BRCA1 are being conducted.

275

with wild-type p53, although the kinetics of replication are much slower and the total viral load much less than that observed in cells with mutant p53 [27]. This finding was also reported by another group that observed that ONYX replication was threefold slower when a cell line with a temperaturesensitive p53 mutant had wild-type p53 [12]. It has been hypothesized that two modes of cell death may be induced by adenoviruses: a rapid p53-dependent death observed with wild-type adenoviruses that requires E1B 55-KDa protein binding to p53, and a slower p53-independent cell death observed with ONYX [4]. In addition, the presence or absence of p14ARF may be important for efficient apoptosis following infection with ONYX-015. Infection with ONYX–015 induced cell death in mesothelioma cell lines lacking p14ARF , whereas cell lines with wild-type p14ARF were resistant [36]. Transfection of wild-type p14ARF rendered the previously ONYX-015-sensitive cells more resistant to the virus [36]. Despite this controversy over the exact cellular mechanisms underlying the apoptosis induced by ONYX-015, the virus does eliminate tumor cells effectively. It was shown that mixing as few as 5% of tumor cells infected with ONYX-015 along with wild-type tumor cells was enough to inhibit tumor cell growth in vivo for 8 weeks in nude mice [16]. Recent studies have combined ONYX-015 with cisplatin and 5-fluorouracil in head and neck cancer patients [20] and have examined the ability of a mutant E1A adenovirus to lyse tumor cells [14,15].

IV. ONYX-015 ADENOVIRUSES

B. Gene Therapy

A. Introduction

Multiple studies have now shown that ONYX-015 can infect and replicate in a wide range of tumor cells with either mutant p53 or wild-type p53, although to different degrees. The therapeutic efficacy of ONYX-015 has been enhanced by combining the virus with either chemotherapy or radiotherapy. For example, in human lung cancer cells, treatment with ONYX-015 and with paclitaxel or cisplatin enhanced the degree of apoptosis [38]. Similarly, the combination of ONYX-015 and radiotherapy caused an additive increase in the degree of cell death, although high doses of γ -irradiation (approximately 20 Gy) began to inhibit ONYX-015 replication [5,27]. Several groups have attempted to improve the efficacy of the ONYX-015 virus by modifying its delivery. First, the schedule of administration appears to be important, as multiple dosings of ONYX-015 decrease tumor burden to a greater extent than a single intratumoral dose [16]. Second, a single intratumoral injection of ONYX-015 was more effective at inhibiting growth when given in a large volume (100 μL) as opposed to a small volume (40 μL) [16]. Finally, the efficiency of ONYX-015 can be improved almost threefold simply by changing the injection carrier. Clinical trials have shown that following an intratumoral injection of ONYX015, the majority of the virus will remain in the center of the

Intratumoral delivery of adenoviruses that selectively replicate in p53-deficient tumor cells have been designed that may take full advantage of introducing a small amount of virus in some tumor cells followed by propagation and toxicity only within the tumor mass [1,17] (reviewed by Heise and Kirn [13] and Hermiston [18]). The adenovirus E1B gene region, which binds to and inactivates wild-type p53 in host cells, was deleted from an adenovirus backbone [1]. Consequently, because it can no longer inactivate p53, this vector can only replicate in p53-null cells, effectively targeting this virus to cells with p53 mutations, which are usually cancer cells. It was shown that normal cells are highly resistant to the E1B-deleted adenovirus [17]. Recent studies, however, suggest that the replication of ONYX-015 may also occur in cell lines with wild-type p53 rather than being limited solely to cells with mutant p53 [8,9,11,17,29]. For example, ONYX-015 was able to replicate in and lyse lung cancer cell lines with mutant p53, as expected, but also in cells with wild-type p53 [38]. The apoptosis induced by ONYX-015, however, was ten times less efficient in cell lines with wild-type p53 compared to those with mutant p53 [38]. The virus can replicate in cells

276


tumor, causing an unequal distribution of necrosis, mostly in the center and not in the periphery. One group hypothesized that the intratumoral spread of the virus could be improved by injecting in a carrier with lidocaine, which would vasodilate the tumor, or with hyaluronidase, which degrades tissue [23]. The most efficient tumor spread of ONYX-015 was observed with lidocaine, which was almost threefold greater than injection in PBS alone. Finally, the ONYX-015 virus itself is being modified to improve its efficacy. First, ONYX-015/CD contains the cytosine deaminase gene, which is a prodrug that converts 5-fluorocytosine to 5-flurouracil [19]. Infection of human pancreatic tumor cells xenografted on nude mice with ONYX-015/CD significantly inhibited tumor growth [19]. Other adenoviruses are being created that also have deletions within the E1B region. The E1B region contains primarily two adenoviral proteins: 55 kDa, which is deleted in ONYX015, and 19 kDa, which is hypothesized to delay apoptosis in cells infected with wild-type adenovirus. An adenovirus was created that only had the 19-kDa protein in the E1B region deleted (Ad-337). Infection of human lung cancer cells with Ad-337 induced apoptosis more extensively than wild-type adenovirus [30]. Finally, viruses with deletions in the E1A region rather than the E1B region, like ONYX-015, have also been created. ONYX-838 contains a partial deletion in the E1A region, which usually binds the Rb protein [21]. Infection of human tumor cell lines in vitro or in vivo with ONYX838 caused greater apoptosis than that observed with wildtype adenovirus and with fewer toxic side effects [14,15,21]. Intravenous injection of ONYX-838 was especially efficient with decreasing lung and lymph node metastases in an orthotopic mouse model of human breast cancer [14,15]. The ONYX-838 virus could still be detected 2 months after a single intratumoral or intravascular injection. Another E1A-deleted adenovirus (Ad-24) was created in which amino acids 120–127, which bind Rb, were deleted. Infection of glioma cells in vitro or in vivo with Ad-24 caused greater cell death than wild-type adenovirus but had no effect on normal lung fibroblasts [6]. Resistance to Ad-24 was correlated with the presence of endogenous wild-type Rb protein. In another approach, an E1-deleted adenovirus was modified to allow it to replicate but only on a limited basis [2]. An E1-deleted adenovirus was modified by the addition of an exogenous plasmid containing the E1 region to allow one round of infection. This adenovirus was then used to infect HeLa tumor cells xenografted on nude mice, which resulted in growth suppression.

C. Clinical Trials The results of a phase I trial using ONYX-015 to treat recurrent head and neck cancers that had failed traditional

chemotherapeutic or radiotherapeutic regimens was recently reported [7]. Intratumoral injections of ONYX-015 given to 22 patients produced no significant side effects, with the most common symptom being low-grade fever. Blood samples taken immediately following the injection of virus and up to 29 days later did not reveal ONYX-015, as determined by polymerase chain reaction (PCR) analysis. This suggested that ONYX-015 virus was not being shed into the blood or that it did not persist in the blood. Furthermore, tissue samples taken from the injection site and from the oropharynx of the patients did not reveal the adenoviral hexon protein, as determined by direct fluorescence. In terms of efficacy, three of the 22 patients experienced a decrease in tumor size of more than 50%, and two saw decreases of 25%. Interestingly, of these five patients who experienced a response, four of them had head and neck tumors with mutant p53. Despite these responses, eventually all of the patients in the study developed progression of the original tumor, new tumors, or metastases. In examining a possible reason for failure, the authors showed that 21 of the 22 patients had also developed increasing antibody levels to the adenovirus [7]. Another phase I study examined the utility of ONYX-015 in refractory cancers [25]. Phase II studies examining the efficacy of ONYX-015 for other head and neck cancers are currently being conducted. A phase II trial recently examined the efficacy of ONYX-015 in treating recurrent head and neck cell cancers in combination with the chemotherapeutic agents cisplatin and 5-fluorouracil [20,21]. Out of 30 patients in the study, almost 60% experienced some degree of tumor size decrease, with 30% of the patients experiencing a complete response. Follow-up 5 months later revealed no tumor recurrence in these patients with a complete response. A phase IIB trial with ONYX-015 is also being conducted for hepatic or billiary tumors [22]. ONYX-015 is being administered by CT-guided intratumoral injections in ten patients with locally advanced or metastatic hepatic or billiary tumors.

V. SUMMARY AND FUTURE WORK The use of tumor suppressors in gene therapy represents an important strategy in the war on cancer. p53 gene therapy remains the most important tumor suppressor strategy being developed, and its combination with chemotherapy or radiotherapy may prove to be even more beneficial. Currently, only p53 gene therapy has progressed to clinical trials. However, p53 may not represent the ideal choice for gene therapy in all cancers. In tumor cells that overexpress MDM2 or have HPV16 E6, other tumor suppressors such as p21 may be more desirable targets of gene therapy because they can bypass the inactivation of p53. In addition to p53 and p21, other tumor suppressors that have been studied for gene replacement,

Clinical Applications of Tumor-Suppressor Gene Therapy

include p16, Rb, p27, p14, PTEN, BRCA1, VHL, and FHIT. Although significant progress in gene therapy for cancer has been made within the past decade, several problems still need to be resolved. First, an efficient vector needs to be designed that can effect prolonged high expression of the transduced gene while only targeting cancer cells. Second, further criteria need to be established for determining which tumor suppressor to employ for gene therapy. The use of tumor suppressors represents a potentially important anticancer treatment that needs to be further investigated.

References 1. Bischoff, J. R., Kirn, D. H., Williams, A., Heise, C., Horn, S., Muna, M., Ng, L., Nye, J. A., Samspon-Johannes, A., Fattaey, A., and McCormick, F. (1996). An adenovirus mutant that replicates selectively in p53deficient human tumor cells. Science 274, 373–376. 2. Clarke, M. F., Qian, D., Han, J., Nunez, G., and Wicha, M. (1999). Targeting the programmed cell death pathway for cancer treatment. Cancer Gene Ther. 6(suppl.), S1. 3. Clayman, G. L., El-Naggar, A. K., Lippman, S. M., Henderson, Y. C., Frederick, M., Merritt, J. A., Zumstein, L. A., Timmons, T. M., Liu, T. J., Ginsberg, L., Roth, J. A., Hong, W. K., Bruso, P., and Goepfert, H. (1998). Adenovirus-mediated p53 gene transfer in patients with advanced recurrent head and neck squamous cell carcinoma. J. Clin. Oncol. 16, 2221–2232. 4. Dix, B. R., O’Carroll, S. J., Myers, C. J., Edwards, S. J., and Braithwaite, A. W. (2000). Efficient induction of cell death by adenoviruses requires binding of E1B55k and p53. Cancer Res. 60, 2666–2672. 5. Duque, P. M., Alonso, C., Sanchez-Prieto, R., Lleonart, M., Martinez, C., Conzalez de Buitrago, G., Cano, A., Quintanilla, M., and Ramon y Cajal, S. (1999). Adenovirus lacking the 19-kDa and 55-kDa E1B genes exerts a marked cytotoxic effect in human malignant cells. Cancer Gene Ther. 6, 554–563. 6. Fueyo, J., Gomez-Manzano, C., Alemany, R., Lee, P. S. Y., McDonnell, T. J., Mitlianga, P., Shi, Y.-X., Levin, V. A., Yung, W. K. A., and Kyritsis, A. P. (2000). A mutant oncolytic adenovirus targeting the Rb pathway produces anti-glioma effect in vivo. Oncogene 19, 2–12. 7. Ganley, I., Kirn, D., Eckhardt, S. G., Rodriguez, G. I., Soutar, D. S., Otto, R., Robertson, A. G., Park, O., Gulley, M. L., Heise, C., Von Hoff, D. D., and Kaye, S. B. (2000). A phase I study of ONYX015, an E1B attenuated adenovirus, administered intratumorally to patients with recurrent head and neck cancer. Clin. Cancer Res. 6, 798– 806. 8. Goodrum, F. D., and Ornelles, D. A. (1998). p53 status does not determine outcome of E1B 55-kilodalton mutant adenovirus lytic infection. J. Virol. 72, 9479–9490. 9. Goodrum, F. D., and Ornelles, D. A. (1997). The early region 1B 55-kilodalton oncoprotein of adenovirus relieves growth restrictions imposed on viral replication by the cell cycle. J. Virol. 71, 548– 561. 10. Habib, N. A., Hodgson, H. J., Lemoine, N., and Pignatelli, M. (1999). A phase I/II study of hepatic artery infusion with wtp53-CMV-Ad in metastatic malignant liver tumours. Hum. Gene Ther. 10, 2019– 2034. 11. Hall, A. R., Dix, B. R., O’Carroll, S. J., and Braithwaite, A. W. (1998). p53-dependent cell death/apoptosis is required for a productive adenoviral infection. Nat. Med. 4, 1068–1072. 12. Harada, J., and Berk, A. (1999). p53-independent, and -dependent requirements for E1B-55K in adenovirus type 5 replication. J. Virol. 73, 5333–5344.

277

13. Heise, C., and Kirn, D. H. (2000). Replication-selective adenoviruses as oncolytic agents. J. Clin. Invest. 105, 847–851. 14. Heise, C. C., Hermiston, T., Brooks, G., Samspon-Johannes, A., Trown, P., and Kirn, D. H. (2000). An adenovirus E1A mutant, ONYX-838, that demonstrates potent and selective systemic and local anti-tumoral efficacy. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 350. 15. Heise, C., Hermiston, T., Johnson, L., Brooks, G., Samspon-Johannes, A., Williams, A., Hawkins, L., and Kirn, D. (2000). An adenovirus E1A mutant that demonstrates potent and selective systemic anti-tumoral efficacy. Nat. Med. 6, 1134–1139. 16. Heise, C. C., Williams, A., Olesch, J., and Kirn, D. H. (1999). Efficacy of a replication-competent adenovirus (ONYX-015) following intratumoral injection: intratumoral spread and distribution effects. Cancer Gene Ther. 6, 499–504. 17. Heise, C., Sampson-Johannes, A., Williams, A., McCormick, F., von Hoff, D. D., and Kirn, D. H. (1997). ONYX-015, an E1B gene attenuated adenovirus, causes tumor-specific cytolysis and antitumoral efficacy that can be augmented by standard chemotherapeutic agents. Nat. Med. 3, 639–645. 18. Hermiston, T. (2000). Gene delivery from replication-selective viruses: arming guided missiles in the war against cancer. J. Clin. Invest. 105, 1169–1172. 19. Hermiston, T., Hawkins, L., Nye, J., Hatfield, J. M., Lemmon, M., Johnson, L., Trown, P., Kirn, D., and Heise, C. (1999). Superior efficacy of a selectively replicating adenovirus, ONYX-015/CD, expressing the cytosine deaminase gene, and development of new replicating adenoviral gene delivery systems. Cancer Gene Ther. 6(suppl.), S13. 20. Khuri, F. R., Nemunaitis, J., Ganly, I., Arseneau, J., Tannock, I. F., Romel, L., Gore, M., Ironside, J., MacDougall, R. H., Heise, C., Randlev, B., Gillenwater, A. M., Bruso, P., Kaye, S. B., Hong, W. K., and Kirn, D. H. (2000). A controlled trial of intratumoral ONYX-015, a selectively-replicating adenovirus, in combination with cisplatin and 5-fluorouracil in patients with recurrent head and neck cancer. Nat. Med. 6, 879–885. 21. Kirn, D., Nemunaitis, J., Heise, C., and Hermiston, T. (1999). Selectively-replicating oncolytic adenoviruses for the treatment of cancer: clinical development of ONYX-015 (p53-targeted) and preclinical studies with ONYX-838 (RB pathway-targeted). Cancer Gene Ther. 6(suppl.), S10. 22. Makower, D., Rozenblit, A., Edelman, M., Haugenlicht, L., Kaufman, H., Haynes, H., Zwiebel, J., and Wadler, S. (1999). Phase IIB trial of ONYX-015 therapy in hepatobiliary tumors. Cancer Gene Ther. 6(suppl.), S16. 23. Morley, S. E., Brown, R., and Kaye, S. (2000). Improvement in distribution of ONYX-015 adenovirus within tumor xenografts by altering the carrier medium. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 350–351. 24. Munden, R. F., Truong, M. T., Swisher, S., and Roth, J. A. (1999). CT evaluation of nonsmall cell lung cancer undergoing gene therapy with adenoviral p53. Radiology 213P, 173. 25. Nemunaitis, J., Cunningham, C., Edelman, G., Berman, B., and Kirn, D. (1999). Phase I dose escalation trial of intravenous infusion of ONYX-015 in patients with refractory cancer. Cancer Gene Ther. 6(suppl.), S16–S17. 26. Nguyen, D. M., Spitz, F. R., Yen, N., Cristiano, R. J., and Roth, J. A. (1996). Gene therapy for lung cancer: enhancement of tumor suppression by a combination of sequential systemic cisplatin and adenovirusmediated p53 gene transfer. J. Thoracic Cardiovascular Surg. 112, 1372–1376. 27. Rogulski, K. R., Freytag, S. O., Zhang, K., Gilbert, J. D., Paielli, D. L., Kim, J. H., Heise, C. C., and Kirn, D. H. (2000). In vivo antitumor activity of ONYX-015 is influenced by p53 status and is augmented by radiotherapy. Cancer Res. 60, 1193–1196. 28. Roth, J. A., Nguyen, D., Lawrence, D. D., Kemp, B. L., Carrasco, C. H., Ferson, D. Z., Hong, W. K., Komaki, R., Lee, J. J., Nesbitt, J. C., Pisters,

278

29.

30.

31.

32.

33.


K. M., Putnam, J. B., Schea, R., Shin, D. M., Walsh, G. L., Dolormente, M. M., Han, C. I., Martin, F. D., Yen, N., Xu, K., Stephens, L. C., McDonnell, T. J., Mukhopadhyay, T., and Cai, D. (1996). Retrovirusmediated wild-type p53 gene transfer to tumors of patients with lung cancer. Nat. Med. 2, 974–975. Rothmann, T., Hengstermann, A., Whitaker, N. J., Scheffner, M., and zur Hansen, H. (1998). Replication of ONYX-015, a potential anticancer adenovirus, is independent of p53 status in tumor cells. J. Virol. 72, 9470–9478. Sauthoff, H., Heitner, S., Rom, W. N., and Hay, J. G. (1999). Deletion of the E1B-19-kDa gene enhances the tumoricidal effect of a replicating adenoviral vector. Cancer Gene Ther. 6(suppl.), S10. Schuler, M., Kuball, J., Leibner, J., Atkins, D., Wen, S. F., Engler, H., Meinhardt, P., Uhlenbusch, R., Horowitz, J. A., Hutchins, B., Maneval, D. C., Storkel, S., Thuroff, J. W., and Huber, C. (1999). A phase I study of adenovirus-mediated wild-type p53 gene transfer in patients with invasive bladder cancer. Cancer Gene Ther. 6(suppl.), S2. Schuler, M., Rochlitz, C., Horowitz, J. A., Schlegel, J., Perruchoud, A. P., Kommoss, F., Bolliger, C. T., Kauczor, H. U., Dalquen, P., Fritz, M. A., Swanson, S., Herrmann, R., and Huber, C. (1998). A phase I study of adenovirus-mediated wild-type p53 gene transfer in patients with advanced non-small cell lung cancer. Hum. Gene Ther. 9, 2075–2082. Sweeney, P., and Pisters, L. L. (2000). Ad5CMVp53 gene therapy for locally advanced prostate cancer—where do we stand? World J. Urol. 18, 121–124.

34. Tait, D. L., Obermiller, P. S., Hatmaker, A. R., Redlin-Frazier, S., and Holt, J. T. (1999). Ovarian cancer BRCA1 gene therapy: phase I and II trial differences in immune response and vector stability. Clin. Cancer Res. 5, 1708–1714. 35. Tait, D. L., Obermiller, P. S., Jensen, R. A., and Holt, J. T. (1999). Ovarian cancer gene therapy with a BRCA1 retroviral vector. Cancer Gene Ther. 6(suppl.), S1. 36. Yang, C.-T, You, L., Song, J., Kirn, D. H., McCormick, F., and Jabions, D. M. (1999). Adenoviral therapy of human mesotheliomas. Cancer Gene Ther. 6(suppl.), S17. 37. Yen, N., Ioannides, C. G., Xu, K., Swisher, S. G., Lawrence, D. D., Kemp, B. L., El-Naggar, A. K., Cristiano, R. J., Fang, B., Glisson, B. S., Hong, W. K., Khuri, F. R., Kurie, J. M., Lee, J. J., Lee, J. S., Merritt, J. A., Mukhopadhyay, T., Nesbitt, J. C., Nguyen, D., Perez-Soler, R., Pisters, K. M. W., Putnam, J. B., Jr., Schrump, D. S., Shin, D. M., and Roth, J. A. (2000). Cellular and humoral immune responses to adenovirus and p53 protein antigens in patients following intratumoral injection of an adenovirus vector expressing wild-type p53 (Ad-p53). Cancer Gene Ther. 7, 530–536. 38. You, L., Yang, C.-T., and Jablons, D. M. (2000). ONYX-015 works synergistically with chemotherapy in lung cancer cell lines and primary cultures freshly made from lung cancer patients. Cancer Res. 60, 1009–1013. 39. Roth, J. A., and Cristiano, R. J. (1997). Gene therapy for cancer: what have we done and where are we going? J. Natl. Cancer Inst. 89, 21–39.

C

H

A

P

T

E

R

17 Cancer Gene Therapy with Tumor Suppressor Genes Involved in Cell-Cycle Control RAYMOND D. MENG

WAFIK S. EL-DEIRY



I. Introduction II. p21WAF1/CIP1

279 280

VI. p27Kip1

A. Introduction 280 B. Gene Therapy with p21 280 C. Gene Therapy with a p21 Mutant Deficient in PCNA Interaction 281 D. Gene Therapy with p21 as an Alternative to p53 281 E. Combination of Ad-p21 and Chemotherapy or Radiotherapy 283 F. p21 Overexpression and Induction of Senescence 283

III. p16INK4

284

A. Introduction 284 B. Gene Therapy with p16

IV. Rb

284

A. Introduction 285 B. Gene Therapy with Rb 285 C. Gene Therapy for Rb-Resistant Tumors

V. p14ARF

VII. E2F-1 VIII. PTEN

286

286

A. Introduction 286 B. Gene Therapy with p14ARF

286

BRCA1 288 VHL 289 FHIT 289 Apoptosis-Inducing Genes

288

289

Introduction 289 Gene Therapy with bax 290 Gene Therapy with Fas 290 Gene Therapy with the TRAIL Receptors

XIII. Conclusions

291

291

References 291

I. INTRODUCTION

to induce both cell cycle arrest and/or apoptosis. In certain conditions, however, p53 replacement may not be the ideal strategy. Consequently, other tumor suppressors have been studied for possible cancer gene therapy (see Fig. 1), including p21WAF1/CIP1 , p16, Rb, p14ARF , p27, E2F-1, BRCA1, VHL, and FHIT. The following sections will discuss the background of each gene and its current role in gene therapy for cancer.

Tumorigenesis results from genetic alterations, including the mutation or deletion of genes involved in growth inhibition. These tumor suppressor genes, therefore, have become a prime focus in cancer gene replacement. The p53 tumor suppressor has been studied extensively both in the laboratory and in clinical trials because of its dual ability


287

288

A. Introduction 288 B. Gene Therapy with PTEN

A. B. C. D.

286

287

A. Introduction 287 B. Gene Therapy with E2F-1

IX. X. XI. XII.

285

286

A. Introduction 286 B. Gene Therapy with p27Kip1 286 C. Gene Therapy with a Modified Form of p27

279


280


FIGURE 1 Tumor suppressors that are targets for gene therapy of Cancer. This review discusses tumor suppressors involved in cell-cycle regulation that have been studied as potential targets for gene replacement in the treatment of cancer. Loss of these tumor suppressors, most notably p53, results in tumor development and progression. p53 mediates the cellular response to DNA damage, resulting in growth arrest or in apoptosis. p21 is a main effector of p53 that mediates growth arrest and is a CDKI, along with p16 and p27, which help to regulate G1 transition. Rb helps to mediate cell-cycle progression from G1 to S phase. In addition, the tumor suppressors BRCA1 (involved in breast cancer), VHL (involved in Von Hippel-Landau familial disease) FHIT (involved in chromosomal breakages), and PTEN (involved in cell attachment) also suppress growth through novel mechanisms. Likewise, the apoptosis induced by p53 is based on the activation of select targets. The recently cloned novel TRAIL target KILLER/DR5 and the Fas family of death receptors can be activated by p53 and can induce apoptosis through initiation of a proteolytic caspase cascade. Two other p53-mediated targets involved in apoptosis, bax and the p53-induced genes or (PIGs), initiate cell death through reactive oxygen species.

II. p21WAF1/CIP1

B. Gene Therapy with p21

A. Introduction

Most of the studies on p21 gene therapy have used adenovirus vectors. It has been shown that p21-expressing adenovirus (Ad-p21) can infect a variety of cancer cell types and can produce readily detectable p21 protein within 24 hours after infection (see Table 1) [205]. The induction of p21 was comparable to, if not greater than, that induced by Ad-p53 [38,80,104]. Infection with Ad-p21 was able to inhibit tumor growth both in vitro and in vivo, causing cell-cycle arrest at G0/G1 and altering tumor morphology. Retroviral infection with p21 also caused growth suppression in mice [17]. Infection of normal tissues in vivo with Ad-p21 produced no adverse effects [76,205]. The exogenously transferred p21 was also shown to be functional, as histone H1 kinase assays showed that CDC2 activity and CDK2 activity were both decreased after Ad-p21 infection of glioma cells [20]. The growth inhibition induced by Ad-p21 was found to be greater than that produced by Ad-p53 in endometrial cancer

p21WAF1/CIP1 was originally cloned as a transcriptionally activated target of p53 [40] that was found to be a potent universal inhibitor of cyclin-dependent kinases (CDKs) [57,193]. It was shown that following DNA damage, p21 is required for p53-mediated G1 arrest [33]. p21 also associates with PCNA, which results in inhibition of DNA polymerase δ processivity in vitro [44]. The negative growth regulatory effects of p21 are also observed during differentiation [208]. Because p21 is an important downstream target of p53 and because it helps to mediate the growth suppressive effects of p53, its effectiveness as an anticancer treatment in gene therapy replacement has been studied. In addition, p21 has growth regulatory effects independent of p53 (reviewed by El-Deiry [39]), as it has been shown that expression of p21 effectively inhibits cancer cell growth in vivo [189].

Cancer Gene Therapy with Tumor Suppressor Genes Involved in Cell-Cycle Control

TABLE 1 Infection of Selected Cell Lines by Ad-p21 Name of cell line

p53a

MOIb

In vitro

Breast

MCF-7 MDA-MB-231 SKBr3

wt mut mut

10 10 20

x x x

[80] [80] [128]

Bladder

RT4 UMUC3

wt mut

10 10

x x

[54] [54]

Cervical

C33A HeLA

wt wt

N.A.c N.A.

x x

[171] [171]

Cell type

In vivo

Ref.

Choriocarcinoma

JEG3

wt

100

x

[104]

Colon

DLD-1 HCT116 LoVo SW480

mut wt wt mut

50 20 50 20

x x x x

[79] [104] [79] [128]

Endometrial

SPEC-2

mut

50

x

[132]

Esophageal

TE-1 TE-3

mut wt

100 100

x x

[76] [76]

Glioblastoma

U-87 MG U-373 MG

wt mut

100 10

x x

MDA-686-LN MDA 886 Tu-138 Tu-177

wt wt mut mut

100 100 100 100

x x x x

[27] [27] [27] [27]

H-358 H460 H1299

null wt null

10 20 50

x x x

[80] [128] [79]

Melanoma

7336 A875

wt wt

100 100

x x

[104] [104]

Prostate

LNCaP

wt

20

x

[51]

Renal cell carcinoma

293

N.A.

200

x

[205]

Head and neck

Lung

x

[50] [20]

a p53 mutations are being examined because p21 mutations are very rare in human cancers. b MOI indicates >50% transduction. c N.A., not available.

cells [132]. Although cell death was noted in some tumors after Ad-p21 infection, no evidence of massive apoptosis was observed [80,104,128,205]. In contrast, apoptosis was induced by Ad-p21 overexpression in endometrial cancer cells [132], cervical cancer cells [171], and esophageal cancer cells [76]. Infection with Ad-p21 has also been reported to block apoptosis through inhibition of caspases [194].

C. Gene Therapy with a p21 Mutant Deficient in PCNA Interaction It has been shown that the N-terminal (cyclin- and CDKinteracting and inhibitory) domain of p21 is sufficient for cancer cell growth inhibition [128]. p21-341 is a p21 mutant in which a premature stop codon has been inserted at nucleotide 341 to delete the C-terminal domain that binds to proliferating cell nuclear antigen (PCNA). Transfection of

281

p21-341 into human colon cancer cells inhibits their growth more than that of wild-type p21 [128]. In addition, loss of the PCNA-interacting region of p21 contributes to a repair defect [97]. A strategy using p21 lacking the PCNAinteracting domain would be expected to inhibit growth but not stimulate DNA repair due to the absence of interaction with PCNA [97,128]. Therefore, an adenovirus containing p21-341 was constructed to evaluate whether it can play a role in growth suppression. It was shown that the p21-341 adenovirus (Ad-p21-341) can infect various human cancer cells in vitro like Ad-p21 and produced a significant suppression of DNA synthesis comparable to that of Ad-p53 and independent of p53 status. Furthermore, some DNA fragmentation was observed in lung and colon cancer cells following infection with Ad-p21-341. Interestingly, it has been reported that the N-terminal domain of p21 may suppress growth by a different mechanism than the C-terminal domain, as the N-terminal domain also seemed to inhibit E2F-1 activity [139]. Therefore, Ad-p21-341 may be an effective candidate for gene replacement in cancer.

D. Gene Therapy with p21 as an Alternative to p53 In evaluating the potential for p21 gene therapy, most groups have compared it to Ad-p53. In contrast to Ad-p53, Ad-p21 causes little or no apoptosis following the infection of many cell lines, including head and neck cancer [26], prostate cancer [51], lung cancer [80], gliomas [50], and melanomas [104]. However, p21 is a potent suppressor of cancer cell growth; thus, Ad-p21 may be an important alternative to Adp53 in situations where p53 is inactivated (see Fig. 2). For example, in some cell lines, p53 is nonfunctional because it is targeted for degradation by the human papillomavirus (HPV) type 16 or 18 E6 protein or because it is bound to inactivating cellular proteins such as the human homolog of the mouse double minute-2 (MDM2) oncoprotein. In these situations, where overexpression of these proteins may inactivate transduced exogenous p53, p21 may represent a more promising approach.

1. HPV16 E6 Inactivates p53 Infection by HPV type 16 or 18 has been correlated with a greatly increased risk of cervical cancer worldwide [175]. It was discovered that the E6 protein of HPV targets human p53 for degradation through ubiquitin-mediated proteolysis [145]. In the presence of E6, a cellular protein called E6-associated protein (E6AP) binds to p53 and functions as an E3 ubiquitin ligase in mediating the degradation of p53. It has also recently been proposed that the HPV E6 gene can also downregulate p53 function by targeting CBP/p300, a transcriptional co-activator of p53 [210].

282


FIGURE 2 Mechanisms of resistance to p53 gene therapy. Although delivery of wild-type p53 by a virus vector suppresses the growth of and induces apoptosis in many human cancer cell lines, some cells are resistant to p53 gene therapy. An important cause of nonresponsiveness to Ad-p53 is target cell resistance to adenovirus infection, although the mechanism for this resistance is currently unknown. In other cell lines, overexpression of MDM2 or SV40 T antigen, two proteins that bind to and inactivate p53, can decrease the effectiveness of exogenously transduced p53. The expression of HPV16 E6 causes enhanced degradation of p53. In these cases, gene replacement with other tumor suppressors, such as p21, may bypass this resistance.

In studies on the role of p53 in chemosensitivity, ovarian cancer cells that stably express HPV16 E6 protein were engineered, leading to endogenous p53 degradation [190]. Infection of these ovarian cancer cells with Ad-p53 produced only slight inhibition of DNA synthesis [128]. In contrast, infection of E6-overexpressing cells with Ad-p21-341 produced a significant suppression of DNA synthesis, which occurred at a much lower multiplicity of infection (MOI) as well. Infection with Ad-p21-341 also caused some DNA fragmentation, indicative of apoptosis. Thus, in HPV-associated cancers in which E6 may be overexpressed, the inactivation of p53 may be bypassed by Ad-p21. 2. MDM2 Overexpression Inactivates p53 p53 can also be inactivated by binding to MDM2, which targets p53 for degradation by MDM2. The MDM2 oncogene is a target for transcriptional activation by p53 [9], but upon binding MDM2 conceals the transactivation domain of p53 and inhibits p53-dependent transcriptional activation [108]. The importance of MDM2 in development was shown when its targeted disruption in mice led to embryonic lethality [75,113]. Thus, although p53 activates MDM2, it is MDM2 that downregulates p53 in a feedback loop that

inhibits p53 function in both growth arrest and apoptosis. Recently, in addition to inactivating p53, MDM2 was also shown to promote the rapid degradation of p53 [58,87]. Therefore, in tumors where MDM2 is elevated, exogenous p53 gene replacement may not lead to optimal growth suppression. To test this hypothesis, Ad-p53 was used to infect several human cancer cell lines that have high expression levels of the MDM2 protein [104]. In comparison to cell lines with low levels of MDM2, the tumor cell lines with elevated MDM2 were resistant to the growth inhibitory effects of Ad-p53. Although cancer cells that overexpress MDM2 were still readily infected by Ad-p53 and induced high expression of p53 protein as determined by western immunoblotting, their rate of DNA synthesis was only slightly decreased as compared to mock-infected or Ad-LacZ-infected cells, and they displayed a blunted induction of p21. Because the inhibitory effect of the exogenous p53 was blunted in these cell lines, Ad-p21 was tested to determine if it could bypass this MDM2-mediated inhibition of p53 because p21 is a downstream target of p53. Infection of MDM2-overexpressing cells with Ad-p21, however, resulted in a strong inhibition of cell-cycle progression and of cellular viability. Similar results were obtained with Ad-p21-341, suggesting that the


cyclin/CDK-interacting domain is sufficient for bypassing p53 resistance in MDM2-overexpressing cells. Furthermore, persistence of the hyperphosphorylated form of the Rb protein correlated with tumor resistance to Ad-p53 infection, suggesting that the phosphorylation state of Rb may be a good indicator of p53-mediated growth inhibition. Therefore, Ad-p21 may be able to effectively bypass MDM2-mediated inactivation of p53 in cancer therapy.

3. SV40 Large T Antigen Inactivates p53 The p53 tumor suppressor was originally discovered in 1979 as a 53-kDa simian virus 40 (SV40) large T antigen (Tag)-associated protein [92]. SV40 Tag is known to bind to and inactivate several tumor suppressor genes, including p53. It was reported that over 60% of human mesotheliomas express SV40-like sequences [16]. In several mesothelioma samples, it was shown that p53 coexpressed with Tag, and Tag coprecipitated with p53, suggesting that these sequences may bind to and inactivate p53 [15]. It was also shown that these SV40-like sequences from mesotheliomas can bind to the Rb family members as well [31]. Therefore, in some mesotheliomas that are refractory to standard therapy, experimental treatment with p21 gene therapy may be a useful substitute for p53.

4. Hepatitis B Virus and Cytoplasmic Retention of p53 The hepatitis B virus X protein is negatively regulated by p53, and the X protein, in turn, inhibits p53 function [177]. The binding of X protein to p53 has been reported to exclude p53 from the nucleus [172]. Thus, in hepatitis-virusdependent liver cancer, p53 may be dysfunctional due to exclusion from the nucleus by X protein [166]. This would be another situation where p21 or other tumor suppressors may be more suitable than p53. 5. Overexpression of TRAIL Decoy Receptor TRUNDD A recent study suggests that p53-mediated apoptosis may be inhibited by overexpression of an anti-apoptotic TRAIL receptor, TRUNDD (also known as DcR2) [102]. It was previously shown that p53 overexpression can induce the pro-apoptotic TRAIL receptor KILLER/DR5 [188] (reviewed by Wu et al. [185]). Recent reports now suggest that Ad-p53 can upregulate the two anti-apoptotic TRAIL receptors, TRUNDD [102] and TRID (or DcR1) [151]. TRUNDD is similar to KILLER/DR5 except that it lacks the cytoplasmic apoptotic-signaling region, the “death domain”; consequently, TRUNDD is hypothesized to attenuate cell death by binding the apoptotic ligand TRAIL and preventing it from binding KILLER/DR5 [115,116,152].

283

E. Combination of Ad-p21 and Chemotherapy or Radiotherapy Because p21 helps to induce cell-cycle arrest following DNA damage, loss of p21 in cancer cells might cause a deficiency in DNA repair, leading to chemosensitivity or to radiosensitivity. p21−/−cells have defective repair of damaged DNA in vitro and are more sensitive to ultraviolet radiation or to chemotherapeutic agents than cells with wild-type p21 [97,176]. When wild-type p53 in a human tumor is not correlated with radiosensitivity, p21 may be involved, and gene therapy with p21 may further enhance such sensitivity. For example, a retrovirus construct containing p21 was used to infect a rat glioma cell line [64]. In addition to tumor growth suppression, the introduction of p21 but not p53 rendered these cells more radiosensitive. In a colony formation assay following 8-Gy exposure, the number of p21-infected cells was decreased 93% compared to the controls and the p53-infected cells. Also, in some tumor cells p21 may help to mediate chemosensitivity. It has been shown that overexpression of p21 in a human sarcoma cell that lacked both p53 and Rb resulted in enhanced chemosensitivity [91]. However, in a human colon cancer cell line with disruption of both alleles of p21, chemosensitivity to DNA crosslinking agents appears to be enhanced [42,97,176]. The schedule of p21 administration may be important, as well, in determining synergy with chemotherapy, as infection with Ad-p21 prior to treatment with etoposide in osteosarcoma cells actually protected the cells from the cytotoxic effects of etoposide [127].

F. p21 Overexpression and Induction of Senescence Another application for p21 gene therapy may be in the induction of terminal differentiation or of senescence. The expression of p21 was originally found to be elevated in senescent cells, and overexpression of p21 can induce premature senescence [111]. Some groups have hypothesized that the induction of senescence by p21 overexpression may be a mechanism to inhibit the growth of tumor cell lines, which usually have mechanisms to escape senescence. Ad-p21 infection of several human cancer cell lines, including lung, osteosarcoma, and colon, induced growth inhibition, induced morphological features characteristic of senescence, and decreased telomerase activity [79]. These senescent features included an enlarged and flattened cytoplasmic shape, an increased cytoplasmic-to-nuclear ratio, a decreased cell density, and an increase in a β-galactosidase activity characteristically associated with senescence [79]. The induction of senescence was also hypothesized to be a mechanism whereby normal cells may evade the effects of p21 overexpression. Infection of esophageal cancer cells with Ad-p21 eventually resulted in the onset of apoptosis;

284


however, normal epithelial keratinocytes underwent terminal differentiation following Ad-p21, withdrew from the cell cycle, and escaped cell death [76].

III. p16

INK4

A. Introduction p16INK4/CDKN2 is a tumor suppressor gene that encodes a specific inhibitor of cyclin D-CDK4 and CDK6. By controlling the activity of CDK4, p16 helps to control the phosphorylation of Rb at late G1 [148]. p16 has been termed a tumor suppressor because it is frequently mutated or homozygously deleted in several types of cancers. p16 is also a major target for hypermethylation, leading to its inactivation in many cancers [61]. Homozygous p16 deletions have been found in over 50% of gliomas [110], but mutations in p16 are also found in other tumors, including esophageal, pancreatic, and non-small-cell lung cancer; lymphomas; and familial melanomas (reviewed by Foulkes et al. [45] and Pinyol et al. [124]).

TABLE 2 Infection of Selected Cell Lines by Ad-p16 Cell type

p16

MOIa

In vitro

In vivo

Ref.

Breast

MCF7

null

300

x

[143]

Cervical

C33A

wt

50

x

[143]

Colon

Lovo

null

30

x

[143]

Glioma

U-87 MG U-251 MG

null null

125 125

x x

[46] [46]

Head and neck

JHU012 JHU022 Tu-138 Tu-177

methc wt wt wt

5 5 100 100

x x x x

[90] [90] [60] [60]

Hepatocellular

HuH7

null

50

x

HL60 Jurkat K562

null null null

N.A.b N.A. N.A.

x x x

H322 H460

null null

50 50

x x

Ovarian

OVCA420 SKOV3

wt mut

100 100

x x

[183] [183]

Pancreas

MIAPaCa-2 NP-9

null mut

30 25

x x

[83] [18]

Prostate

DU145 LNCaP PC-3 PPC-1 TSU

mut null null null null

200 200 200 200 200

x x x x x

[4] [4] [4] [4] [163]

Leukemia

Lung

B. Gene Therapy with p16 A p16-expressing adenovirus (Ad-p16) was first used to infect non-small-cell lung cancer lines that had homozygous deletions of p16 (see Table 2) [72]. These cell lines were readily infected by Ad-p16 in vitro and in vivo and expressed p16 at high levels; their growth rates were inhibited up to 90%; and cell-cycle arrest occurred at G0 /G1 . In contrast, infection of a normal mammary epithelial cell line with Ad-p16 did not cause growth inhibition or cellcycle arrest. Ad-p16 has also been used to successfully infect and inhibit tumor growth up to 80% in several malignant glioma cells in vitro, with either wild-type p16 or homozygous p16 deletions [46]. Although Ad-p16 inhibited the growth of esophageal squamous cell cancers in vitro, no effect was observed for esophageal adenocarcinoma because these cells were poorly infected [146]. The utility of Ad-p16 for tumor inhibition has now been extended to other cell lines, including head and neck squamous cell carcinomas [137], prostate cancer [163] (reviewed by Allay et al. [4]), ovarian cancer cells [183], and pancreatic carcinoma lines [83]. Overexpression of p16 has also been reported to induce senescence. In prostate cancer cells, Ad-p16 induced growth arrest and cellular senescence only in those lines that expressed wild-type Rb; tumor lines containing mutant Rb were growth-inhibited by Ad-p16 but did not undergo senescence [162]. Because retroviruses infect hematopoietic cells more efficiently than adenoviruses, a p16 retrovirus was created and used to infect several leukemia cell lines in vitro.

Name of cell line

x

[143] [131] [131] [131]

x

x

[72] [72]

a MOI

indicates >50% transduction. not available. c Cell line has wild-type p16, but it is functionally inactivated by promoter hypermethylation. b N.A.,

Strong growth inhibition was observed in three lines with homozygous deletions of p16, but no inhibition was observed for a leukemia cell line with mutant p16 [131]. In contrast, another study reported that an ovarian cancer cell line with coexpression of both wild-type Rb and wild-type p16 was resistant to Ad-p16 [170]. Interestingly, a recent report suggests that intratumoral delivery of Ad-p16 to colon cancers in mice may decrease the incidence of liver metastases [184]. Finally, the feasibility of liposomally delivered p16 has recently been studied. Squamous cell carcinoma lines were transfected with liposomes containing p16, and cell-cycle arrest was subsequently observed with sustained p16 expression [99]. The efficacy of p16 gene therapy has been compared to p21 and p53 in causing growth inhibition in prostate cancer cells [51]. At comparable titers, p53 inhibited prostate cancer cell growth in vitro more significantly than either p21 or p16, which were comparable to each other. It was also shown that Ad-p53 induced a higher percentage of apoptosis among infected cells. In an in vivo model of prostate cancer in nude mice, all three viruses could inhibit tumor growth when initial


tumor size was less than 200 mm3 ; however, only Ad-p53 was effective in larger sized tumors. In terms of chemosensitivity, the results with Ad-p16 are mixed. One group reported that overexpression of p16 in an IPTG-p16-inducible cell line made them more resistant to methotrexate, vinblastine, and cisplatin [164]. In contrast, transfection of p16 into a glioma with a homozygous deletion of p16 did not increase its chemosensitivity to nitrogen mustards [55]. Recent studies now suggest that the schedule of Ad-p16 administration in concert with chemotherapy may explain the efficacy of Ad-p16. It was reported that infection of human osteosarcoma cells with Ad-p16 before or during treatment with etoposide protected the cells from chemotherapy-induced cell death, whereas infection with Ad-p16 after etoposide treatment had no such protective effect [127]. It was hypothesized that pretreatment with Ad-p16 decreases chemosensitivity because the infected cells undergo G1 arrest; hence, there are fewer cells in S phase, which is the cellular phase targeted by etoposide or gemcitabine. Another report suggests that Ad-p16 infection several days after treatment of human prostate cancer cells with cisplatin will actually increase chemosensitivity compared to either treatment alone [2]. Therefore, the timing of Ad-p16 infection and the induction of growth arrest appear to be important in regard to any possible synergy with chemotherapy. Unlike p53 or even p21, the role or p16 in radiosensitivity remains controversial. One group reported that transfection of p16 into two human malignant melanoma cell lines, one of which had a homozygous deletion of p16, increased the radiosensitivity of both cell lines [100]. In contrast, the levels of p16 in bladder cancer cells were not altered following irradiation [135], and the presence of p16 was not correlated with radiosensitivity nor with the induction of p53 in several tumor cell lines [174]. p16, however, has been combined with Ad-p53, which induces apoptosis, to infect a panel of cancer cells [143]. It was shown that infection of tumor cell lines in vitro with both viruses induced apoptosis, whereas neither virus alone, at the MOIs used, caused apoptosis. Such a strategy may prove useful for tumors that have mutations in different tumor suppressors, such as gliomas, which often have deletions of p16 and mutations in p53. Thus, the combination of p16 and p53 may induce apoptotic cell death in cancer cells, although p16 alone has not been shown to induce apoptosis in any system. It is not entirely clear that, in vivo, the combination of p53 and p16 offers anything that cannot be achieved by p53 alone, if used at a sufficiently high MOI. It has not been shown, for example, that the combination of p53 and p16 is either more tumor specific or less toxic to normal cells. Ad-p16 has also been combined with Ad-p21 to enhance growth inhibition [107]. To improve upon the efficacy of wild-type p16 for gene therapy, a fusion p16 construct was recently created that tar-

285

geted glioma cells [1]. To wild-type p16 were added 500 base pairs of antisense uPAR (urokinase-type plasminogen activator receptor), which is a gene involved in malignant glioma invasion. The entire construct was then encoded in an adenovirus. Infection of glioma cells with this adenovirus containing antisense uPAR and p16 inhibited the ability of glioma cells to metastasize in Matrigel and in spheroid assays, presumably by decreasing the levels of uPAR while increasing the expression of p16 [1].

IV. Rb A. Introduction The retinoblastoma, or Rb, gene plays a role in the progression of the cell into the S phase. It was the first tumor suppressor identified and has been shown to be a nuclear phosphoprotein that regulates cell cycle progression by binding to several transcription factors needed for DNA synthesis, most notably the E2F family of transcription factors [180] (reviewed by Harbour and Dean [56]). The function of Rb is dependent upon its phosphorylation state. If Rb is hypophosphorylated, it can bind to and inactivate E2F, halting progression through the cell cycle. Hyperphosphorylation of Rb, however, releases it from binding to E2F, and the cell enters S phase. It has been hypothesized that Rb functions as a “guardian” at the R point, the point at which the cell commits itself in G1 to progress to the M phase [180]. Although Rb was originally identified as being deleted in the rare disease retinoblastoma, it can be found mutated in many other tumors, including osteosarcoma, breast cancers, hepatocellular carcinoma, and bladder cancer.

B. Gene Therapy with Rb The initial studies on the role of Rb as a tumor suppressor focused on the replacement of Rb into various Rb-defective human cell lines in vitro, which suppressed the tumorigenicity of the cell lines (reviewed in Xu [195]). An adenovirus construct encoding the full-length wild-type Rb was created and was used to infect several Rb−/− cell lines, including a non-small-cell lung carcinoma, bladder cancer, breast cancer, and an osteosarcoma [196]. Following infection, the cell lines expressed high levels of exogenous Rb proteins, mostly in the hypophosphorylated or unphosphorylated forms, as determined by immunocytochemistry and by western blot analyses. Infection of established bladder tumors in mice with the Rb adenovirus (Ad-Rb) slightly decreased the rate of growth of the tumors. Ad-Rb was also used to treat spontaneous pituitary melanotroph tumors in Rb ± mice in vivo [136]. Gene replacement with Ad-Rb decreased the growth of the tumors and increased the survival of the mice, compared to untreated

286


controls. Ad-Rb was recently shown to infect a wide range of human cancer cell lines [32].

C. Gene Therapy for Rb-Resistant Tumors

susceptible to Ad-p14-induced apoptosis than the HCT116 cell line genetically engineered with a homozygous deletion of p53 [202]. However, recent studies suggest that ARF may also have growth suppressive effects that are p53 independent but MDM2 dependent [179].

Some tumor cells, however, even after replacement of the Rb gene, remain resistant to growth inhibition [65,209], suggesting that the Rb pathway may be inactivated in these cells [195]. Two approaches have been used to circumvent Rb-resistant tumors with Rb gene therapy. First, an N-terminal truncated Rb mutant has been reported to cause enhanced tumor suppression in vitro compared to the fulllength wild-type Rb [198]. When this Rb mutant was expressed in an adenovirus and used to infect human bladder cancer cells in an in vivo mouse model, complete growth inhibition of the tumors was observed, and tumor regression was even noted in 50% of the tumors [196]. Second, another approach to Rb-resistant tumors focuses on using the other Rb family members, specifically p130. A retrovirus encoding p130 was used to infect a lung tumor cell line and caused growth inhibition both in vitro and in vivo in a nude mouse model [25].

p27 is a universal cyclin-dependent kinase inhibitor that was first identified as a downstream effector of TGF-β and contact inhibition [84]. It belongs to the same family of CDK inhibitors as p21WAF1/CIP1 and can also arrest cells in G1 [126]. p27Kip1 is believed to be a tumor suppressor because it maps to a chromosomal site often deleted in leukemias, because it functions as a CDK inhibitor like p21, and because it can be found mutated in some tumors [161]. Recent efforts have uncovered frequent loss of p27 protein expression in colon, breast, and lung cancer through increased ubiquitinmediated proteolysis of p27 in cancer [41,95,168].

V. p14ARF

B. Gene Therapy with p27Kip1

A. Introduction The INK4A/ARF locus on chromosome 9p21 encodes two alternative transcripts, which have been identified as the tumor suppressors p16INK4a and p14ARF (reviewed by Chin et al. [23]). It has been calculated that mutations in p16/ARF are the second most common mutations in human cancers, after p53 perturbations [53]. For example, it has been reported that INK4A/ARF mutations occur in over 70% of mesotheliomas [22]. ARF (alternative reading frame) is the β transcript of the INK4A/ARF locus, utilizing exons 1β, 2, and 3 [37,130]. ARF is hypothesized to stabilize the p53 tumor suppressor protein by degrading MDM2, which targets p53 for ubiquitinmediated degradation [125,208]. Overexpression of ARF induces growth inhibition through both a reported G1 and G2 arrest and eventual apoptosis [130].

B. Gene Therapy with p14ARF Because of the role of ARF as a possible tumor suppressor and its frequency of deletions in human cancers, one group has studied the feasibility of using ARF in cancer gene therapy studies. An adenovirus encoding p14 (Ad-p14) was used to infect a panel of mesothelioma cell lines (with mutant ARF and p16) and caused both G1 growth arrest and eventual apoptosis [202]. Ad-p14 infection induced both p53 and p21 protein and dephosphorylated Rb. Interestingly, the apoptosis induced by Ad-p14 seemed to be dependent upon p53 as the HCT116 cancer cell line with wild-type p53 was more

VI. p27Kip1 A. Introduction Kip1

To study its possible role in growth suppression, two groups constructed adenoviruses encoding p27Kip1 [20,28]. The p27 adenovirus (Ad-p27) efficiently infected a glioma and a squamous cell carcinoma cell line in vitro, producing profound growth suppression that was caused by cell-cycle arrest at G0 /G1 [20]. Tumor growth was also inhibited in an in vivo glioma model [20]. Growth of a human breast cancer xenograft in nude mice was also inhibited by Ad-p27 [81]. In a human chloangiocarcinoma cell line, Ad-p27 infection induced apoptosis in vitro, with upregulation of Fas ligand but not Fas ligand mRNA [199]. Another study compared the efficacy of Ad-p27 to another tumor suppressor, Ad-p21. When compared to Ad-p21 following infection of a breast cancer cell line, Ad-p27 produced greater cytotoxicity and caused G1 /S arrest and decreased CDK2 activity at a lower MOI, suggesting that in some breast cancer cell lines p27 may be a better gene therapy agent than p21 [28].

C. Gene Therapy with a Modified Form of p27 Another group has attempted to improve the growth inhibition induced by p27 by modifying the tumor suppressor itself. p27 is usually degraded by ubiquitination following phosphorylation on residue 187T, which releases p27 from a complex with cyclin E; subsequently, a mutant p27 was created in which 187T was deleted [118]. Human lung cancer cell lines were then infected with Ad-p27 or the adenovirusencoding mutant p27 (Ad-p27-mut). Ad-p27-mut induced

287


a stronger G1 arrest than Ad-p27 based on the percentage of cells in S phase as determined by fluorescence-activated cell sorting (FACS) analysis (control adenovirus was 42%, Ad-p27 was 22%, and Ad-p27-mut was 10%). Finally, another group created a chimeric tumor suppressor between p27 and p16 and encoded it in an adenovirus construct (Ad-p27/p16) [120]. In prostate, colon, pancreatic, and lung tumor cells, Ad-p27/p16 caused a stronger growth inhibition than either Ad-p27 or Ad-p16 alone [120]. This growth inhibition was independent of endogenous p53 or Rb status of the tumor cells. Interestingly, human cancer cell lines that were resistant to Ad-p16 or Ad-p27 were effectively induced to undergo apoptosis by Ad-p27/p16 [120].

VII. E2F-1 A. Introduction E2F-1 belongs to the E2F family of transcription factors that regulate the expression of genes involved in cell-cycle progression, especially at the G1 to S phase boundary (reviewed by Adams and Kaelin [3]). When complexed to the DP family of transcription factors [88], E2F-1 binds to DNA and activates the transcription of numerous genes that are involved in DNA synthesis and cell proliferation. The activity of E2F-1, in turn, is regulated by Rb, as binding of hypophosphorylated Rb to E2F-1 prevents E2F-1 from activating its target genes [19]. When E2F-1 is overexpressed in the absence of growth factors, it can drive quiescent cells into S phase [74]. Consequently, E2F-1 has been proposed to be an oncogene. When its regulation is disrupted, E2F-1 overexpression can cause uncontrolled growth, leading to tumor development [197]. In cooperation with ras, E2F-1 can transform rat embryonic fibroblasts [73,159]. In studies using Rb± and E2F-1 −/− mice, tumor development was delayed compared to growth in Rb± mice [200]. Although it can function as an oncogene, E2F-1 has also been postulated to be a tumor suppressor because it can induce apoptosis. Mice with a homozygous deletion of E2F-1 have a defect in thymocyte apoptosis, show increased cellular proliferation, and eventually develop numerous tumors [43,201]. Overexpression of E2F-1 in quiescent fibroblasts can cause apoptosis [74,129]. The apoptosis induced by E2F-1 overexpression does not seem to depend on the ability of E2F-1 to transactivate target genes or to induce DNA synthesis but rather on its ability to bind to DNA [63,123]. Whether the apoptosis induced by E2F-1 is dependent on the tumor suppressor p53, however, is unclear. In some cell lines, this apoptosis appears to be p53 dependent [191], whereas in other cells it seems to be independent of both p53 and Rb, although it can be inhibited by overexpression of Rb [63]. A recent study showed that apoptosis mediated by E2F-1 overexpression induces p53 protein, and both the induction of

apoptosis and accumulation of p53 can be blocked by expression of MDM2 [86]. Furthermore, DNA damage was reported to elevate E2F-1 expression independent of p53 status, suggesting a role for E2F-1 upregulation in apoptosis [103]. Recent studies have identified the p53 family member p73 as a transcriptional target of E2F-1 and as a potential mediator of E2F-1-dependent apoptosis [69,93].

B. Gene Therapy with E2F-1 Because of its possible role in the induction of apoptosis, E2F-1 overexpression in tumor cell lines was studied using an adenovirus that encodes full-length wild-type E2F-1 (Ad-E2F-1). Infection of human breast and ovarian cancer cell lines with Ad-E2F-1 produced a strong apoptotic response in vitro and in vivo in a nude mouse model (see Table 3) [67]. The induction of apoptosis was also observed following Ad-E2F-1 infection in human head and neck cancer cells [94], melanoma cells [34], glioma cells [49], and esophageal cancer cells [203]. The effect was related to a functional overexpression of E2F-1 as E2F-1 infection activated a fusion Rb–CAT reporter with E2F-1 binding sites [94]. The expression of other cell-cycle or apoptotic proteins was then examined following E2F-1 overexpression. In melanoma cells that underwent apoptosis following Ad-E2F-1 infection, the levels of two pro-apoptotic proteins, Bax and Bak, were not altered, although the expression of Bcl-Xl and Mcl-1, two anti-apoptotic proteins, decreased [34]. Similarly, Ad-E2F-1 infection of esophageal cancer cells decreased the protein levels of Bcl-2, Mcl-1, and Bcl-XL [203]. However, in esophageal cancer cells that were resistant to Ad-E2F-1mediated apoptosis, the levels of these anti-apoptotic proteins were not altered [203]. In contrast, in glioma cells infected with Ad-E2F-1, the expression of the anti-apoptotic Bcl-2 was not altered, and, in fact, an increase in the protein levels of p21 and p27 was observed [49]. The apoptosis induced by Ad-E2F-1 was independent of p53 in human breast, ovarian, and head and neck cancer cell lines, as these lines readily underwent apoptosis despite having endogenous mutant or null p53 alleles [67,94]. Ad-E2F-1 infection of esophageal cancer cells caused apoptosis but did TABLE 3 Infection of Selected Cell Lines by Ad-E2F-1 Cell type

Name of cell line

p53

MOIa

In vitro

Esophageal

Yes-4 Yes-6

mut mut

100 100

x x

Head and neck

Tu-138 Tu-167

mut mut

100 100

x x

Melanoma

SK-MEL-2 SK-MEL-28

mut wt

100 100

x x

a MOI

indicates >50% transduction.

In vivo

Ref. [203] [203]

x

[94] [94] [34] [34]

288


not induce the expression of p53 protein [203]. Consequently, Ad-E2F-1 may be a method to eradicate tumor cells that are normally resistant to p53-mediated gene therapy. For example, tumor cell lines that overexpress MDM2 are resistant to infection with Ad-p53 but not with Ad-E2F-1, which induces a strong apoptotic response [204]. These studies, however, do not abrogate a contribution from p53 in the induction of apoptosis by E2F-1. In fact, p53-mediated apoptosis may further contribute to E2F-1-induced cell death. In human esophageal cancer cells infected with Ad-E2F-1, the expression of ARF was increased while the levels of MDM2, which inhibits p53, were decreased [70]. Thus, one group studied the feasibility of combined Ad-p53 and Ad-E2F-1 infection of esophageal cancer cells. Cells were first infected with Ad-p53 and then with Ad-E2F-1, and it was shown that this serial infection strategy was more efficient than simultaneous infection with both adenoviruses in inducing apoptosis. Presumably, infection with Ad-p53 first induces apoptosis but also enhances the levels of MDM2, which helps to downregulate the expression of p53; however, infection with Ad-E2F-1, in turn, downregulates MDM2 [70].

TABLE 4 Infection of Selected Cell Lines by Ad-PTEN Cell type

A. Introduction PTEN (also cloned as MMAC1 or TEP1) was originally identified to have homology to protein tyrosine phosphatases and was localized to human chromosome 10q23.3. The gene was found to be mutated in sporadic breast cancers and glioblastomas and in genetic diseases such as Cowden’s syndrome, a hereditary predisposition to breast cancers. Functionally, PTEN dephosphorylated phosphatidylinositols phosphorylated by phosphatidylinositol 3 -kinase (PI3K). It was also reported that PTEN shared homology with a structural protein tensin, creating a role for PTEN in integrin signaling (reviewed by Tamura et al. [167]). Stable transfection of PTEN in glioblastoma or breast cancer cell lines lacking endogenous wild-type PTEN caused growth inhibition and eventual apoptosis [96,169,182]. This mechanism of growth inhibition has been linked to the Rb protein [117]. It has also been reported that overexpression of PTEN in selected glioblastoma cell lines can enhance their radiosensitivity but not chemosensitivity [182].

B. Gene Therapy with PTEN Several groups have examined the utility of PTEN overexpression for gene therapy by constructing adenoviruses that encode wild-type PTEN (Ad-PTEN). Multiple cancer cell lines have been shown to be growth inhibited in vitro by Ad-PTEN, especially brain tumor cells (see Table 4). Infection of glioblastoma cell lines with Ad-PTEN caused growth

MOIa

In In vitro vivo

Ref.

Endometrial

3H12 HEC1-A KLE RL95-2

N.A.b N.A. N.A. N.A.

x x x x

Glioblastoma

LN-18 LN229 U87MG

N.A. N.A. N.A.

x x x

Ovarian

Caov-3 ES-2 MCAS MDAH 2774 OV-1063 OCAR-3 SKOV3 SW626 TYK-nu

100 100 100 100 100 100 100 100 100

x x x x x x x x x

[106] [106] [106] [106] [106] [106] [106] [106] [106]

Prostate

LNCap

N.A.

x

[29]

a MOI b N.A.,

VIII. PTEN

Name of cell line

x

x x

[142] [142] [142] [142] [182] [11,182] [11,21,169]

indicates >50% transduction. not available.

inhibition and anoikis and also disrupted Akt-mediated signaling [30]. Another group showed that the mechanism of growth inhibition was related to the induction of p27Kip1 to cyclin E, decreased CDK2 kinase activity, and decreased Rb protein [21]. In U87MG glioblastoma cells, Ad-PTEN induced anoikis, which was enhanced by TGF-β [11]. In ovarian cancer cell lines, Ad-PTEN caused G1 arrest and apoptosis [106]. Another group showed that the induction of apoptosis in endometrial cancer cell lines with Ad-PTEN was independent of endogenous PTEN status (either wild-type or mutant) [142]. Finally, the efficacy of Ad-PTEN has been compared to that of Ad-p53. In prostate cancer cells that lack endogenous PTEN, Ad-PTEN inhibited growth more effectively than Ad-p53 but did not induce the same degree of apoptosis [29]. The apoptosis induced by Ad-PTEN, however, was blocked by overexpression of the anti-apoptotic protein Bcl-2, which did not have an effect in cells infected with Ad-p53 [29]. In contrast, in PTEN-mutant glioblastoma xenografts on nude mice, tumor growth was inhibited to the same degree by either Ad-PTEN or Ad-p53 [11].

IX. BRCA1 BRCA1, or breast cancer susceptibility gene 1, was originally identified as a tumor suppressor lost in familial breast or ovarian cancers [47,105]. BRCA1, however, does not seem to play an important role in the development of primary breast tumors. Interestingly, it has been reported that overexpression of BRCA1 can induce p53 expression both by transcriptional


coactivation of p53 and by upregulation and stabilization of p14ARF [160]. The function of BRCA1 is currently unknown, although several hypotheses exist. It has been shown that an adenovirus encoding wild-type BRCA1 (Ad-BRCA1) induces growth arrest following infection of human colon, lung, and breast cancer cell lines by dephosphorylating retinoblastoma protein and decreasing cyclin-dependent kinase activity [98]. In another study, infection of human breast or osteosarcoma cell lines with Ad-BRCA1 induced apoptosis independent of endogenous p53 status [149]. Concomitantly, it was observed that BRCA1 overexpression elevated the protein levels of GADD45, Fas, FasL, and p21 [149]. These results suggest that BRCA1 overexpression may help inhibit tumor growth through the activation of proteins involved in either DNA damage or cell-cycle regulation.

X. VHL Von Hippel-Landau (VHL) loss is associated with cancers, including renal cell carcinomas [48] and central nervous system hemangioblastomas. Initially, VHL was shown to be able to inhibit transcription [35]. Because renal cell tumors and hemangioblastomas are highly vascular in nature, it was hypothesized that VHL may be involved in angiogenesis. In fact, it was recently reported that VHL may play a role in the proteolysis of hypoxia-inducible factors [101]. Because of its hypothesized tumor suppressor role, the utility of an adenovirus encoding VHL (Ad-VHL) for gene therapy of tumors was studied. Infection of renal cell carcinoma and breast cell carcinoma lines with Ad-VHL produced a G1 cellcycle arrest, resulting in growth inhibition but not apoptosis [82]. It was also shown that VHL overexpression increased the protein levels of p27, suggesting that p27 may be involved in VHL-mediated cell cycle arrest [82]. The tumor suppressive ability of Ad-p27 was then compared to other tumor suppressors, such as Ad-p53, Ad-p21, Ad-p27, and Ad-p16, in an in vivo xenograft model. Although Ad-p53 caused the strongest growth suppression, Ad-VHL was observed to cause the next highest inhibition of growth [173]. Further research may examine how VHL-mediated growth suppression can be applied to tumor growth inhibition, perhaps specifically focusing upon renal cell or breast cell carcinomas.

XI. FHIT The fragile histidine triad (FHIT ) gene, localized on human chromosome 3p14.2, was found to be mutated in early lung cancers and in renal cell carcinomas, suggesting a possible role as a tumor suppressor in select cancers (reviewed by Huebner et al. [66]). It has also been reported that mutation of FHIT occurs in esophageal squamous cell

289

carcinomas [153], head and neck cancers [52], and endometrial carcinomas [147]. The protein product of FHIT was later shown to function as a dinucleoside triphosphate hydrolase [10]. Overexpression of FHIT in lung tumor cells was then shown to cause apoptosis in vitro [144]. In vivo, xenografts of renal cell carcinomas with stable expression of FHIT had delayed tumor formation compared to cells without expression of FHIT [181]. Because of the tumor-suppressive effects of FHIT, several groups have engineered viral constructs that encode FHIT. Infection of human lung or head and neck cancer cells with an adenovirus that expresses FHIT (Ad-FHIT) suppressed tumor growth both in vitro and in vivo [71]. Another group inhibited the growth of pancreatic cancer cell lines following infection with an adeno-associated adenovirus encoding FHIT [36]. This tumor-suppressive effect of FHIT, however, is not always observed in all cell lines. Infection of human cervical cancer cell lines with a retrovirus encoding FHIT did not delay anchorageindependent growth [186]. The ability of FHIT to suppress growth of tumors may be limited to specific cell lineages.

XII. APOPTOSIS-INDUCING GENES A. Introduction Because many tumor suppressors are currently known to function in the mediation of apoptosis, with p53 being a prime example, another strategy for cancer gene therapy has been to directly introduce pro-apoptotic genes into tumor cell lines. For example, rather than overexpressing p53, which presumably activates downstream targets involved in apoptosis such as bax or Fas, it may be more advantageous to overexpress these apoptosis-inducing genes directly. Two groups of apoptosis-inducing genes have been extensively studied: the pro-apoptotic members of the bcl-2 gene family (including bax) and the Fas death receptor members. One of the difficulties in using apoptosis-inducing genes, however, has been in the construction of viral vectors to encode them (reviewed by Bruder et al. [13]). Because the apoptosisinducing genes cause cell death upon overexpression, production of adenoviruses is difficult because overexpression of the pro-apoptotic genes eliminates the producer cells that package the viral constructs. For example, in creating an adenovirus that encodes FasL, it was shown that the cells usually used to propagate adenoviruses, human 293 cells, were eliminated by the shuttle plasmid that encoded murine FasL, presumably because the FasL was binding the FasR on 293 cells [89]. Several strategies have been employed to circumvent overexpression of the pro-apoptotic genes in viral constructs. In some vectors, the pro-apoptotic gene is placed under a promoter that can be regulated, such as the tetracycline-inducible promoter [139], or the myelin basic promoter [154]. Other groups modified the adenoviral vector to contain Cre-LoxP

290


excision sites to keep the pro-apoptotic gene dormant unless it is released by coinfection with an adenovirus that expresses the Cre recombinase [85,140,157,192], or had the proapoptotic gene under the control of a GAL4 promoter which was only activated by coinfection with an adenovirus containing the GAL4 activator [119]. For example, bax was cloned into an adenovirus vector containing 5 GAL4-binding sites and a TATA box, and, when used for infections, this virus was coinfected with an adenovirus containing the transactivator GAL4/VP-16 [77,78]. Finally, it has been reported that production of adenoviruses containing pro-apoptotic genes is facilitated using the Ad-Easy system [178], which employs Escherichia coli, rather than mammalian cells, as the viral producer cells [59].

B. Gene Therapy with bax bax is a pro-apoptotic member of the bcl-2 family, thought to induce apoptosis through enhancing the mitochondrial release of cytochrome c [114]. Loss of bax in genetically engineered mice resulted in increased tumor incidence, suggesting that bax may have a tumor suppressor role in vivo [206]. Frameshift mutations have also been reported in various cancer cell lines, including colon cancer and hematopoietic cancers [12,133]. Furthermore, bax is one of the target genes observed to be upregulated following Ad-p53 infection [8,121,134], suggesting that it may be an important mediator of p53-induced apoptosis. Overexpression of bax in human cancer cell lines has been reported to induce apoptosis independent of endogenous p53 status and to enhance the chemosensitivity or radiosensitivity of select tumor lines [85,141]. Several groups, therefore, have successfully developed adenoviruses that encode bax (Ad-Bax). Ad-Bax infection of human lung cancer cell lines, both in vitro and in vivo, caused apoptosis that was independent of endogenous p53 status [77,78]. Interestingly, lung cancer cell lines resistant to Ad-p53 were susceptible to apoptosis following Ad-Bax. In nude mice, Ad-Bax infection was initially not hepatotoxic, as determined by measuring the levels of the liverassociated enzymes ALT and AST. Using a binary vector system, another group regulated the expression of bax following Ad-Bax infection of prostate cancer cell lines and reported extensive apoptosis, which furthermore, was not inhibited by overexpression of the anti-apoptotic bcl-2 [62]. Infection with Ad-Bax was found to enhance the radiosensitivity of human ovarian cancer cell lines to radiotherapy through an increased induction of apoptosis in vitro or in vivo [6]. The gene therapy potential of Ad-Bax was recently combined with an adenovirus that expresses caspase-8 (Ad-Caspase-8) [156]. The caspases are cysteine proteases activated during apoptosis that are involved in the actual cleavage of intracellular protein targets. The caspases act in a

pyramid cascade, whereby activation of caspase-8 induces the activity of other caspases. Ad-p53 has been observed to elevate the expression of various caspases, including caspase-8 [8,150]. Both Ad-Bax and Ad-Caspase-8 were encoded in adenoviruses containing loxP sequences; in order for the genes to be expressed, they had to be coinfected with an adenovirus carrying the Cre recombinase, which excises the genes [156]. In two glioblastoma cell lines, the combination of Ad-Bax and Ad-Caspase-8 induced more cell death (83%) than either Ad-Bax (49%) or Ad-Caspase-8 (55%) alone. The authors suggest that the two adenoviruses may work through separate pathways, as infection with Ad-Bax did not induce the expression of caspase-8 protein, and, likewise, infection with Ad-Caspase-8 did not induce Bax protein [156]. The other pro-apoptotic members of the bcl-2 family, related to bax, have also been tested as potential targets for cancer gene therapy. First, Bak has been shown to be upregulated following Ad-p53 infection in human lung cancer cells [8,121]. Bak was then cloned in an adenoviral vector containing GAL4 binding sites (Ad-Bak) [119]. Ad-Bak was then activated by coinfection with an adenovirus expressing a GAL4/GV16 fusion protein. Infection of several human cancer cell lines in vitro with Ad-Bax caused 40–60% apoptosis, which was induced independent of endogenous p53 or bcl-2 status. Infection with Ad-Bak in vivo also decreased tumor growth [119]. Finally, another group has created an adenovirus that encodes bcl-xs (Ad-Bcl-xs ) [24], another proapoptotic bcl-2 family member. Ad-Bcl-xs induced apoptosis in various tumor lines, including breast, colon, and stomach cancer cells, but not in normal cells. The virus, however, had only limited penetration into the tumors.

C. Gene Therapy with Fas The Fas death receptor initiates apoptosis following binding of Fas ligand [165] (reviewed by Nagata and Golstein, [109] and Peter and Krammer [122]). In addition, Fas has been observed to be upregulated following Ad-p53 infection, suggesting it may function as a mediator of p53-induced apoptosis [134]. Currently, several groups have shown that infection of tumor cell lines with an adenovirus encoding Fas death receptor (Ad-Fas) produces a rapid and extensive apoptosis [7,154,158]. Other studies have examined the ability of an adenovirus encoding Fas ligand (Ad-FasL) to induce apoptosis. Prostate cancer cells, previously found to be resistant to apoptosis induced by Fas antibody, were infected with AdFasL in vitro and found to undergo extensive cell death [68]. Infection with Ad-FasL also caused extensive apoptosis in breast and brain cancer cells [112]. Similar results were also reported in vivo following injection of Ad-FasL into prostate cancer xenografts on nude mice [68]. It has been suggested that the apoptosis observed in vivo following Ad-FasL may


also depend upon a bystander effect, as only 25–50% but not 100% of brain tumor cells were infected with Ad-FasL [112]. To enhance the apoptotic effect of Ad-FasL, it has been combined with an adenovirus that expresses caspase-3 (Ad-Caspase-3) [155]. Coinfection of glioma cells with Ad-FasL and Ad-Caspase-3 synergistically enhances the degree of apoptosis, compared to infection with other adenovirus alone [155].

D. Gene Therapy with the TRAIL Receptors Currently, the use of other death receptors in cancer gene therapy is being explored. An adenovirus was recently constructed that contained the cytoplasmic domain of DR4 (Ad-DR4-CD) [194]. Infection of human breast, lung, and colon cancer cells with Ad-DR4-CD caused extensive apoptosis independent of cellular p53 status. Normal fibroblasts, however, are resistant to the apoptotic effects of Ad-DR4-CD.

XIII. CONCLUSIONS Although most tumor suppressor replacement strategies have focused upon the p53 tumor suppressor gene, other genes have been studied. The p53 target, p21WAF1/CIP1, is involved in cell-cycle inhibition and has been studied as an alternative to p53, especially in situations where p53 may not be entirely effective. The cell-cycle inhibitors p16, Rb, p14ARF, and p27 may prove effective in specific tumors with mutations in those genes. Similarly, the tumor suppressors BRCA1, VHL, and FHIT may also be tumor-specific replacement strategies. Finally, the E2F-1 tumor suppressor may also induce apoptosis, when expressed at high levels, and may also be able to play a role in suppressing growth. Likewise, genes that specifically induce apoptosis, such as bax, Fas, or the TRAIL receptors, may prove important in cancer gene therapy. Cancer gene therapy strategies utilizing either tumor suppressor genes or apoptosis-inducing genes may eventually comprise clinical trials.

References 1. Adachi, Y., Liu, T., Mohanam, S., Mohan, P. M., Rajan, M. K., Gokasian, Z. L., Kyritsis, A. P., Sawaya, R., and Rao, J. S. (2000). Adenovirus-mediated delivery of a biscistronic construct containing antisense uPAR and sense p16 gene sequences suppress glioma invasion. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 387–388. 2. Adams, C. D., Steiner, M. S., and Allay, J. A. (2000). Adenovirus p16 synergizes with cisplatin to inhibit human prostate xenograft growth. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 122. 3. Adams, P. D., and Kaelin, Jr., W. G. (1995). Transcriptional control by E2F. Semin. Cancer Biol. 6, 99–108.

291

4. Allay, J. A., Steiner, M. S., Zhang, Y., Reed, C. P., Cockroft, J., and Lu, Y. (2000). Adenovirus p16 gene therapy for prostate cancer. World J. Urol. 18, 111–120. 5. Aoki, K., Akyurek, L. M., San, H., Leung, K., Parmacek, M. S., Nabel, E. G., and Nabel, G. J. (2000). Restricted expression of an adenoviral vector encoding Fas ligand (CD95L) enhances safety for cancer gene therapy. Mol. Ther. 1, 555–565. 6. Arafat, W. O., Gomez-Navarro, J., Xiang, J., Barnes, M. N., Mahasrenshti, P., Alvarez, R. D., Siegal, G. P., Badib, A. L., Buschsbaum, D., Curiel, D. T., and Stackhouse, M. A. (2000). An adenovirus encoding proapoptotic bax induces apoptosis and enhances the radiation effect in human ovarian cancer. Mol. Ther. 1, 545–554. 7. Arai, H., Gordon, D., Nabel, E. G., and Nabel, G. J. (1997). Gene transfer of Fas ligand induces tumor regression in vivo. Proc. Natl. Acad. Sci. USA 25, 13862–13867. 8. Atencio, I., and Demers, G. W. (1999). Apoptosis in a hepatocellular carcinoma cell line in response to adenovirus-mediated transfer of wild-type p53 gene occurs by regulation of bcl-2 family members and activation of caspases. Cancer Gene Ther. 6(suppl.), S5. 9. Barak, Y., Juven, T., Haffner, R., and Oren, M. (1993). MDM2 expression is induced by wild-type p53 activity. EMBO J. 12, 461–468. 10. Barnes, L. D., Garrison, P. N., Siprashvili, Z., Guranowski, A., Robinson, A.K., Ingram, S.W., Croce, C.M., Ohta, M., and Huebner, F. (1996). FHIT, a putative tumor suppressor in humans, is a dinucleoside 5 , 5 -P-1,P-3-triphosphate hydrolase. Biochemistry 35, 11529– 11535. 11. Bookstein, R, Cheney, W., Neuteboom, S., Levy, A., Anderson, S., Johnson, D., Vaillancourt, M., Ramachandra, M., and Steck, P. (1999). Prospects for MMAC1/PTEN gene therapy of cancer. Cancer Gene Ther. 6(suppl.), S1. 12. Brimmell, M., Mendiola, R., Mangion, J., and Packham, G. (1998). Bax frameshift mutations in cell lines derived from human haemopoietic malignancies are associated with resistance to apoptosis and microsatellite instability. Oncogene 16, 1803–1812. 13. Bruder, J. T., Appiah, A., Kirkman, W. M., Chen, P., Tian, J., Reddy, D., Brough, D. E., Lizonova, A., and Kovesdi, I. (2000). Improved production of adenovirus vectors expressing apoptotic transgenes. Hum. Gene Ther. 11, 139–149. 14. Cairns, P., Polascik, T. J., Eby, Y., and Sidransky, D. (1995). High frequency of homozygous deletion at p16/CDKN2 in primary human tumors. Nat. Genet. 11, 210–212. 15. Carbone, M., Rizzo, P., Grimley, P. M., Procopio, A., Mew, D. J. Y., Shridhar, V., de Bartolomeis, A., Esposito, V., Giuliano, M. T., Steinberg, S. M., Levine, A. S., Giordano, A., and Pass, H.I. (1997). Simian virus-40 large T antigen binds p53 in human mesotheliomas. Nat. Med. 3, 908–912. 16. Carbone, M., Pass, H. I., Rizzo, P., Marinetti, M., Di Muzio, M., Mew, D. J., Levine, A. S., and Procopio, A. (1994). Simian virus 40-like DNA sequences in human pleural mesothelioma. Oncogene 9, 1781–1790. 17. Cardinali, M., Jakus, J., Shah, S., Ensley, J. F., Robbins, K. C., and Yeudall, W. A. (1998). p21(WAF1/CIP1) retards the growth of human squamous cell carcinomas in vivo. Oral Oncol. 34, 211–218. 18. Cascallo, M., Mercade, E., Capella, G., Lluis, F., Fillat, C., Gomez-Foix, A. M., and Mazo, A. (1999). Genetic background determines the response to adenovirus-mediated wild-type p53 expression in pancreatic tumor cells. Cancer Gene Ther. 6, 428–436. 19. Chellappan, S. P., Hiebert, S., Mudryj, M., Horowitz, J. M., and Nevins, J. R. (1991). The E2F transcription factor is a cellular target for the RB protein. Cell 65, 1053–1061. 20. Chen, J., Willingham, T., Shuford, M., Bruce, D., Rushing, E., Smith, Y., and Nisen, P.D. (1996). Effects of ectopic overexpression of p21WAF1/CIP1 on aneuploidy and the malignant phenotype of human brain tumor cells. Oncogene 13, 1395–1403.

292


21. Cheney, I. W., Neuteboom, S. T., Vaillancourt, M. T., Ramachandra, M., and Bookstein, R. (1999). Adenovirus-mediated gene transfer of MMAC1/PTEN to glioblastoma cells inhibits S phase entry by the recruitment of p27Kip1 into cyclin E/CDK2 complexes. Cancer Res. 59, 2318–2323. 22. Cheng, J. Q., Jhanwar, S. C., Klein, W. M., Bell, D. , Lee, W. C., Altomare, D. A., Nobori, T., Olopade, O. I., Buckler, A. J., and Testa, J. R. (1994). p16 alterations and deletion mapping of 9p21–p22 in malignant mesothelioma. Cancer Res. 54, 5547–5551. 23. Chin, L., Pomerantz, J., and DePinho, R. A. (1998). The INK4a/ARF tumor suppressor: one gene—two products—two pathways. Trends Biochem. Sci. 23, 291–296. 24. Clarke, M. F., Qian, D., Han, J., Nunez, G., and Wicha, M. (1999). Targeting the programmed cell death pathway for cancer treatment. Cancer Gene Ther. 6(suppl.), S1. 25. Claudio, P. P., Howard, C. M., Pacilio, C., Cinti, C., Romano, G., Minimo, C., Maraldi, N. M., Minna, J. D., Gelbert, L., Leoncini, L., Tosi, G.M., Hicheli, P., Caputi, M., Giordano, G. G., and Giordano, A. (2000). Mutations in the retinoblastoma-related gene RB2/p130 in lung tumors and suppression of tumor growth in vivo by retrovirus-mediated gene transfer. Cancer Res. 60, 372–382. 26. Clayman, G. L., Liu, T.-J., Overholt, S. M., Mobley, S. R., Wang, M., Janot, F., and Goepfert, H. (1996). Gene therapy for head and neck cancer: comparing the tumor suppressor gene p53 and a cell cycle regulator WAF1/CIP1 (p21). Arch. Otolaryngol. Head Neck Surg. 122, 489–493. 27. Clayman G. L., El-Naggar, A. K., Roth, J. A., Zhang, W.-W., Goepfert, H., Taylor, D. L., and Liu, T.-J. (1995). In vivo molecular therapy with p53 adenovirus for microscopic residual head and neck squamous carcinoma. Cancer Res. 55, 1–6. 28. Craig, C., Wersto, R., Kim, M., Ohri, E., Li, Z., Katayose, D., Lee, S. J., Trepel, J., Cowan, K., and Seth, P. (1997). A recombinant adenovirus expressing p27Kip1 induces cell cycle arrest and loss of cyclin-Cdk activity in human breast cancer cells. Oncogene 14, 2283– 2289. 29. Davies, M. A., Koul, D., Dhesi, H., Berman, R., McDonnell, T. J., McConkey, D., Yung, W. K., and Steck, P. A. (1999). Regulation of Akt/PKB activity, cellular growth, and apoptosis in prostate cancer cells by MMAC/PTEN. Cancer Res. 59, 2551–2556. 30. Davies, M. A., Lu, Y., Sano, T., Fang, X., Tang, P., LaPushin, R., Koul, D. Bookstein, R., Stokoe, D., Yung, W. K., Mills, G. B., and Steck, P. A. (1998). Adenoviral transgene expression of MMAC/PTEN in human glioma cells inhibits Akt activation and induces anoikis. Cancer Res. 58, 5285–5290. 31. DeLuca, A., Baldi, A., Esposito, V., Howard, C. M., Bagella, L., Rizzo, P., Caputi, M., Pass, H. I., Giordano, G. G., Baldi, F., Carbone, M., and Giordano, A. (1997). The retinoblastoma gene family pRb/p105, p107, pRb2/p130 and simian virus-40 large T-antigen in human mesotheliomas. Nat. Med. 3, 839–840. 32. Demers, G. W., Harris, M. P., Wen, S. F., Engler, H., Nielsen, L. L., and Maneval, D.C. (1998). A recombinant adenoviral vector expressing full-length human retinoblastoma susceptibility gene inhibits human tumor cell growth. Cancer Gene Ther. 5, 207–214. 33. Deng, C., Zhang, P., Harper, J. W., Elledge, S. J., and Leder, P. (1995). Mice lacking p21CIP1/WAF1 undergo normal development, but are defective in G1 checkpoint control. Cell 82, 675–684. 34. Dong, Y.-B., Yang, H.-L., Elliott, M. J., Liu, T.-J., Stilwell, A., Atienza, Jr., C., and McMasters, K. M. (1999). Adenovirus-mediated E2F-1 gene transfer efficiently induces apoptosis in melanoma cells. Cancer 86, 2021–2033. 35. Duan D. R., Pause, A., Burgess, W. H., Aso, T., Chen, D. Y. T., Garrett, K. P., Conaway, R. C., Conaway, J. W., Linehan, W. M., and Klausner, R. D. (1995). Inhibition of transcription elongation by the VHL tumorsuppressor protein. Science 269, 1402–1406.

36. Dumon, K. R., Chakrani, F., During, M. J., Rosato, E. F., Williams, N. N., and Croce, C. C. (2000). Inhibition of tumor growth and induction of apoptosis by adeno-associated virus-mediated fragile histidine triad (FHIT) gene overexpression in pancreatic cancer cell lines. Gastroenterology 118 (suppl.), 3271. 37. Duro, D., Bernard, O., Della Valle, V., Berger, R., and Larsen, C. J. (1995). A new type of p16INK4/MTS1 gene transcript expressed in B-cell malignancies. Oncogene 11, 21–29. 38. Eastham, J. A., Hall, S. J., Sehgal, I., Wang, J., Timme, T. L., Yang, G., Connell-Crowley, L., Elledge, S.J., Zhang, W.-W., Harper, J. W., and Thompson, T. C. (1995). In vivo gene therapy with p53 or p21 adenovirus for prostate cancer. Cancer Res. 55, 5151–5155. 39. El-Deiry, W. S. (1997). Role of oncogenes in resistance and killing by cancer therapeutic agents. Curr. Opin. Oncol. 9, 79–87. 40. El-Deiry, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Parsons, R., Trent, J. M., Lin, D., Mercer, W. E., Kinzler, K. W., and Vogelstein, B. (1993). WAF1, a potential mediator of p53 tumor suppression. Cell 75, 817–825. 41. Esposito, V., Baldi, A., De Luca, A., Groger, A. M., Loda, M., Giordano, G. G., Caputi, M., Baldi, F., Pagano, M., and Giordano, A. (1997). Prognostic role of the cyclin-dependent kinase inhibitor p27 in non-small-cell lung cancer. Cancer Res. 57, 3381–3385. 42. Fan, S., Chang, J. K., Smith, M. L., Duba, D., Fornace, Jr., A. J., and O’Connor, P.M. (1997). Cells lacking CIP1/WAF1 genes exhibit preferential sensitivity to cisplatin and nitrogen mustard. Oncogene 14, 2127–2136. 43. Field, S. J., Tsai, F.-Y., Kuo, F., Zubiaga, A. M., Kaelin, W. G., Livingston, D. M., Orkin, S. H., and Greenberg, M. E. (1996). E2F-1 functions in mice to promote apoptosis and suppress proliferation. Cell 85, 549–561. 44. Flores-Rozas, H., Kelman, Z., Dean, F. B., Pan, Z. Q., Harper, J. W., Elledge, S. J., O’Donnell, M., and Hurwitz, J. (1994). CDK-interacting protein 1 directly binds with proliferating cell nuclear antigen and inhibits DNA replication catalyzed by the DNA polymerase delta holoenzyme. Proc. Natl. Acad. Sci. USA 91, 8655–8659. 45. Foulkes, W. D., Flanders, T. Y., Pollock, P. M., and Hayward, N. K. (1997). The CDKN2A (p16) gene and human cancer. Mol. Med. 3, 5–20. 46. Fueyo, J., Gomez-Manzano, C., Yung, W. K. A., Clayman, G. L., Liu, T.-J., Bruner, J., Levin, V. A., and Kyritsis, A. P. (1996). Adenovirusmediated p16/CDKN2 gene transfer induces growth arrest and modifies the transformed phenotype of glioma cells. Oncogene 12, 103–110. 47. Futreal, P. A., Liu, Q. Y., Shattuckeidens, D., Cochran, C., Harshman, K., Tavtigian, S., Bennett, L. M., Haugenstrano, A., Swensen, J., Miki, Y., Eddington, K., McClure, M., Frye, C., Weaverfeldhaus, J., Ding, W., Gholami, Z., Soderkvist, P., Terry, L., Jhanwar, S., Berchuck, A., Iglehart, J. D., Marks, J., Ballinger, D.G., Barrett, J. C., Skolnick, M. H., Kamb, A., and Wiseman, R. (1994). BRCA1 mutations in primary breast and ovarian carcinomas. Science 266, 120–122. 48. Gnarra, J. R., Tory, K., Weng, Y., Schmidt, L., Wei, M. H., Li, H., Latif, F., Liu, S., Chen, F., Duh, F. M., Lubensky, I., Duan, D.R., Florence, C., Pozzatti, R., Walther, M. M., Bander, N. H., Grossman, H. B., Brauch, H., Pomer, S., Brooks, J. D., Isaacs, W. B., Lerman, M. I., Zbar, B., and Linehan, W.M. (1994). Mutations of the VHL tumor-suppressor gene in renal-carcinoma. Nat. Genet. 7, 85–90. 49. Gomez-Manzano, C., Mitlianga, P., Fueyo, J., Liu, T.-J., Kyritsis, A. P., and Yung, W. K. A. (2000). Transfer of E2F-1 results in upregulation of Bcl-2, p21, and p27: action–reaction? Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 351. 50. Gomez-Manzano, C., Fueyo, J., Kyritsis, A. P., McDonnell, T. J., Steck, P. A., Levin, V. A., and Yung, W. K. A. (1997). Characterization of p53 and p21 functional interactions in glioma cells en route to apoptosis. J. Natl. Cancer Inst. 89, 1036–1044.

Cancer Gene Therapy with Tumor Suppressor Genes Involved in Cell-Cycle Control 51. Gotoh, A., Kao, C., Ko, S.-C., Hamada, K., Liu, T. J., and Chung, L. W. K. (1997). Cytotoxic effects of recombinant adenovirus p53 and cell cycle regulator genes (p21WAF1/CIP1 and p16CDKN4 ) in human prostate cancers. J. Urol. 158, 636–641. 52. Gotte, K., Hadaczek, P., Coy, J. F., Wirtz, H. W., Riedel, F., Neubauer, J., and Hormann, K. (2000). FHIT expression is absent or reduced in a subset of primary head and neck cancer. Anticancer Res. 20, 1057– 1060. 53. Haber, D. A. (1997). Splicing into senescence: the curious case of p16 and p19ARF. Cell 91, 555–558. 54. Hall, M. C., Li, Y., Pong, R.-Y., Ely, B., Sagalowsky, A. I., and Hsieh, J.-T. (2000). The growth inhibitory effect of p21 adenovirus on human bladder cancer cells. J. Urol. 163, 1033–1038. 55. Hama, S., Sadatomo, T., Yoshioka, H., Kurisu, K., Tahara, E., Naruse, I., Heike, Y., and Saijo, N. (1997). Transformation of human glioma cell lines with the p16 gene inhibits cell proliferation. Anticancer Res. 17, 1933–1938. 56. Harbour, J. W., and Dean, D. C. (2000). Rb function in cell-cycle regulation and apoptosis. Nat. Cell Biol. 2, E65–E67. 57. Harper, J. W., Adami, G. R., Wei, N., Keyomarsi, K., and Elledge, S. J. (1993). The p21 CDK-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell 75, 805–816. 58. Haupt, Y., Maya, R., Kazaz, A., and Oren, M. (1997). MDM2 promotes the rapid degradation of p53. Nature 387, 296–299. 59. He, T. C., Zhou, S., Da Costa, L. T., Yu, J., Kinzler, K. W., and Vogelstein, B. (1998). A simplified system for generating recombinant adenoviruses. Proc. Natl. Acad. Sci. USA 95, 2509–2514. 60. Henderson, Y. C., Breau, R. L., Liu, T.-J., and Clayman, G. L. (2000). Telomerase activity in head and neck tumors after introduction of wild-type p53, p21, p16, and E2F-1 genes by means of recombinant adenovirus. Head Neck 22, 347–354. 61. Herman, J. G., Jen, J., Merlo, A., and Baylin, S. B. (1996). Hypermethylation-associated inactivation indicates a tumor suppressor role for p16NK4B . Cancer Res. 56, 722–727. 62. Honda, T., Kagawa, S., Higuchi, M., Spurgers, K. B., Gjertsen, B. T., Bruckheimer, E. M., Brisbay, S. M., Roth, J. A., Fang, B., and McDonnell, T. J. (2000). A recombinant adenovirus expressing wild-type bax induces apoptosis in prostate cancer cells. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 50. 63. Hsieh, J.-K., Fredersdorf, S., Kouzarides, T., Martin, K., and Lu, X. (1997). E2F-1-induced apoptosis requires DNA binding but not transactivation and is inhibited by the retinoblastoma protein through direct interaction. Genes Dev. 11, 1840–1852. 64. Hsiao, M., Tse, V., Carmel, J., Costanzi, E., Strauss, B., Haas, M., and Silverberg, G. D. (1997). Functional expression of human p21(WAF1/CIP1) gene in rat glioma cells suppresses tumor growth in vivo and induces radiosensitivity. Biochem. Biophys. Res. Comm. 233, 329–335. 65. Huang, H.-J., Yee, J. K., Shew, J. Y., Chen, P. L., Bookstein, R., Friedmann, T., Lee, E. Y., and Lee, W. H. (1988). Suppression of the neoplastic phenotype by replacement of the retinoblastoma gene product in human cancer cells. Science 242, 1563–1566. 66. Huebner, K., Garrison, P. N., Barnes, L. D., and Croce, C. M. (1998). The role of the FHIT/FRA3B locus in cancer. Ann. Rev. Genet. 32, 7–31. 67. Hunt, K. K., Deng, J., Liu, T.-J., Wilson-Heiner, M., Swisher, S. G., Clayman, G., and Hung, M.-C. (1997). Adenovirus-mediated overexpression of the transcription factor E2F-1 induces apoptosis in human breast and ovarian carcinoma cell lines and does not require p53. Cancer Res. 57. 68. Hyer, M. L., Rubinchik, S., Dong, J.-Y., Voelkel-Johnson, C., Gunhan, M., and Norris, J. S. (2000). Gene therapy of prostate cancer utilizing a FasL adenoviral vector. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 50–51.

293

69. Irwin, M., Marin, M. C., Phillips, A. C., Seelan, R. S., Smith, D. I., Liu, W., Flores, E. R., Tsai, K. Y., Jacks, T., Vousden, K. H., and Kaelin, W. G. (2000). Role for the p53 homologue p73 in E2F-1–induced apoptosis. Nature 407, 645–648. 70. Itoshima, T., Fujiwara, T., Kataoka, M., Kodowaki, Y., Fukazawa, Shao, J., Waku, T., Tanaka, N., and Kodama, M. (2000). Induction of apoptosis in human esophageal cancer cells by sequential p53 and E2F-1 gene transfer: involvement of p53 accumulation via ARF-mediated MDM2 downregulation. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 352. 71. Ji, L., Fang, B., Yen, N., Fong, K., Minna, J. D., and Roth, J. A. (1999). Induction of apoptosis and inhibition of tumorigenicity and tumor growth by adenovirus vector-mediated fragile histidine triad (FHIT) gene overexpression. Cancer Res. 59, 3333–3339. 72. Jin, X., Nguyen, D., Zhang, W.-W., Kyritsis, A. P., and Roth, J. A. (1995). Cell cycle arrest and inhibition of tumor cell proliferation by the p16INK4 gene mediated by an adenovirus vector. Cancer Res. 55, 3250–3253. 73. Johnson, D. G., Cress, W. D., Jakoi, L., and Nevins, J. R. (1994). Oncogenic capacity of the E2F1 gene. Proc. Natl. Acad. Sci. USA 91, 12823–12827. 74. Johnson, D. G., Schwarz, J. K., Cress, W. D., and Nevins, J. R. (1993). Expression of E2F-1 induces quiescent cells to enter S-phase. Nature 365, 349–352. 75. Jones, S. N., Roe, A. E., Donehower, L. A., and Bradley, A. (1995). Rescue of embryonic lethality in MDM2-deficient mice by absence of p53. Nature 378, 206–208. 76. Kadowaki, Y., Fujiwara, T., Fukazawa, T., Shao, J., Yasuda, T., Itoshima, T., Kagawa, S., Hudson, L. G., Roth, J. A., and Tanaka, N. (1999). Induction of differentiation-dependent apoptosis in human esophageal squamous cell carcinoma by adenovirus-mediated p21 sdi1 gene transfer. Clin. Cancer Res. 5, 4233–4241. 77. Kagawa, S., Gu, J., Swisher, S. G., Ji, L., Roth, J. A., Lai, D., Stephens, L. C., and Fang, B. (2000). Antitumor effect of adenovirus-mediated bax gene transfer on p53-sensitive and p53-resistant cancer lines. Cancer Res. 60, 1157–1161. 78. Kagawa, S., Pearson, S. A., Ji, L., Xu, K., McDonnell, T. J., Swisher, S., Roth, J. A., and Fang, B. (2000). A binary adenoviral vector for expressing high levels of the proapoptotic gene bax. Gene Ther. 7, 75–79 79. Kagawa, S., Fujiwara, T., Kadowaki, Y., Fukazawa, T., Sok-Joo, R., Roth, J.A., and Tanaka, N. (1999). Overexpression of the p21 SDI1 gene induces senescence-like state in human cancer cells: implication for senescence-directed molecular therapy for cancer. Cell Death Differ. 6, 765–772. 80. Katayose, D., Wersto, R., Cowan, K. H., and Seth, P. (1995). Effects of a recombinant adenovirus expressing WAF1/CIP1 on cell growth, cell cycle, and apoptosis. Cell Growth Differ. 6, 1207–1212. 81. Kim, J., Hwang, E. S., Kim, J. S., You, E.-H., Lee, S. H., and Lee, J.-H. (1999). Intraperitoneal gene therapy with adenoviral-mediated p53 tumor suppressor gene for ovarian cancer model in nude mouse. Cancer Gene Ther. 6, 172–178. 82. Kim, M., Katayose, Y., Li, Q., Rakkar, A. N., Li, Z., Hwang, S. G., Katayose, D., Trepel, J., Cowan, K.H., and Seth, P. (1999). Cancer gene therapy with a recombinant adenovirus expressing a Von HippelLandau tumor suppressor gene. Cancer Gene Ther. 6(suppl.), S1–S2. 83. Kobayashi, S., Shirasawa, H., Sashiyama, H., Kawahira, H., Kaneko, K., Asano, T., and Ochiai, T. (1999). p16INK4A expression adenoviral vector to suppress pancreas cancer cell proliferation. Clin. Cancer Res. 5, 4182–4185. 84. Koff, A., Ohtsuki, E., Roberts, J., and Massague, J. (1993). Negative regulator of G1 in mammal cells: inhibition of cyclin E-dependent kinase by TGF-β. Science 257, 1689–1694. 85. Komatsu, K., Suzuki, S., Shimosegawa, T., Miyazaki, J., and Toyota, T. (2000). Cre-loxP-mediated bax gene activation reduces growth rate

294

86.

87. 88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

100.

101.

102.


and increases sensitivity to chemotherapeutic agents in human gastric cancer cells. Cancer Gene Ther. 7, 885–892. Kowalik, T. F., DeGregori, J., Leone, G., Jakoi, L., and Nevins, J. R. (1998). E2F1-specific induction of apoptosis and p53 accumulation, which is blocked by MDM2. Cell Growth Differ. 9, 113–118. Kubbutat, M. H. G., Jones, S. N., and Vousden, K. H. (1997). Regulation of p53 stability by MDM2. Nature 387, 299–303. La Thangue, N. B. (1994). DP and E2F proteins: components of a heterodimeric transcription factor implicated in cell cycle control. Curr. Opin. Cell Biol. 6, 443–450. Larregina, A. T., Morelli, A. E., Dewey, R. A., Castro, M. G., Fontana, A., and Lowenstein, P. R. (1998). FasL induces Fas/Apo1-mediated apoptosis in human embryonic kidney 293 cells routinely used to generate E1-deleted adenoviral vectors. Gene Ther. 5, 563–568. Li, D., Ling, D., Freimuth, P., and O’Malley, Jr., B. W. (1999). Variability of adenovirus receptor density influences gene transfer efficiency and therapeutic response in head and neck cancer. Clin. Cancer Res. 5, 4175–4181. Li, W. W., Fan, J., Hochhauser, D., and Bertino, J. R. (1997). Overexpression of p21waf1 leads to increased inhibition of E2F-1 phosphorylation and sensitivity to anticancer drugs in retinoblastoma-negative human sarcoma cells. Cancer Res. 57, 2193–2199. Linzer, D. I. H., and Levine, A. J. (1979). Characterization of a 54K dalton cellular SV40 tumor antigen present in SV40-transformed cells and uninfected embryonal carcinoma cells. Cell 17, 43–52. Lissy, N. A., Davis, P. K., Irwin, M., Kaelin, W. G., and Dowdy, S. F. (2000). A common E2F-1 and p73 pathway mediates cell death induced by TCR activation. Nature 407, 642–645. Liu, T.-J., Wang, M., Breau, R. L., Henderson, Y., El-Naggar, A. K., Steck, K. D., Sicard, M. W., and Clayman, G. L. (1999). Apoptosis induction by E2F-1 via adenoviral-mediated gene transfer results in growth suppression of head and neck squamous cell carcinoma cell lines. Cancer Gene Ther. 6, 163–171. Loda, M., Cukor, B., Tam, S. W., Lavin, P., Fiorentino, M., Draetta, G. F., Jessup, J. M., and Pagano, M. (1997). Increased proteasomedependent degradation of the cyclin-dependent kinase inhibitor p27 in aggressive colorectal carcinomas. Nat. Med. 3, 152–154. Lu, Y., Lin, Y. Z., LaPushin, R., Cuevas, B., Fang, X., Yu, S. X., Davies, M. A., Khan, H., Furui, T., Mao, M., Zinner, R., Hung, M. C., Steck, P., Siminovitch, K., and Mills, G. B. (1999). The PTEN/MMAC1/TEP tumor suppressor gene decreases cell growth and induces apoptosis and anoikis in breast cancer cells. Oncogene 18, 7034–7045. McDonald, III, E. R., Wu, G. S., Waldman, T., and El-Deiry, W. S. (1996). Repair defect in p21WAF1/CIP1 –/– human cancer cells. Cancer Res. 56, 2250–2255. MacLachlan, T. K., Somasundaram, K., Sgagias, M., Shifman, Y., Muschel, R. J., Cowan K. H., and El-Deiry, W. S. (2000). BRCA1 effects on the cell cycle and the DNA damage response are linked to altered gene expression. J. Biol. Chem. 275, 2777–2785. Marron-Terada, P. G., Park, J. W., and Belinsky, S. A. (2000). Liposome-mediated delivery of the p16INK4A gene in non-small cell lung cancer. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 465–466. Matsumura, Y., Yamagishi, N., Miyakoshi, J., Imamura, S., and Takebe, H. (1997). Increase in radiation sensitivity of human malignant melanoma cells by expression of wild-type p16 gene. Cancer Lett. 115, 91–96. Maxwell, P. H., Wiesener, M. S., Chang, G. W., Clifford, S. C., Vaux, E. C., Cockman, M. E., Wykoff, C. C., Pugh, C. W., Maher, E. R., and Ratcliffe, P.J. (1999). The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399, 271–275. Meng, R. D., McDonald, E. R., III, Sheikh, M. S., Fornace, A. J., and El-Deiry, W. S. (2000). The TRAIL decoy receptor TRUNDD (DcR2, TRAIL-R4) is induced by adenovirus-p53 overexpression and

103.

104.

105.

106.

107.

108.

109. 110.

111.

112.

113.

114.

115.

116.

117.

118.

can delay TRAIL-, p53-, and KILLER/DR5-dependent colon cancer apoptosis. Molec. Ther. 1, 130–144. Meng, R. D., Phillips, P., and El-Deiry, W. S. (1999). p53-independent increase in E2F-1 expression enhances the cytotoxic effects. Intl. J. Oncol. 14, 5–14. Meng, R., Shih, H., Prabhu, N. S., George, D. L., and El-Deiry, W. S. (1997). Bypass of abnormal MDM2 inhibition of p53-dependent growth suppression. Clin. Cancer Res. 4, 251–259. Miki, Y., Swensen, J., Shattuckeidens, D., Futreal, P. A., Harshman, K., Tavtigian, S., Liu, Q. Y., Cochran, C., Bennett, L. M., Ding, W., Bell, R., Rosenthal, J., Hussey, C., Tran, T., McClure, M., Frye, C., Hattier, T., Phelps, R., Haugenstrano, A., Katcher, H., Yakumo, K., Gholami, Z., Shaffer, D., Stone, S., Bayer, S., Wray, C., Bogden, R., Dayananth, P., Ward, J., Tonin, P., Narod, S., Bristow, P. K., Norris, F. H., Helvering, L., Morrison, P., Rosteck, P., Lai, M., Barrett, J. C., Lewis, C., Neuhausen, S., Cannonalbright, L., Goldgar, D., Wiseman, R., Kamb, A., and Skolnick, M. H. (1994). A strong candidate for the breast and ovarian-cancer susceptibility gene BRCA1. Science 266, 66–71. Minaguchi, T., Mori, T., Kanamori, Y., Matsushima, M., Yoshikawa, H., Taketani, Y., and Nakamura, Y. (1999). Growth suppression of human ovarian cancer cells by adenovirus-mediated transfer of the PTEN gene. Cancer Res. 59, 6063–6067. Mobley, S. R., Liu, T. J., Hudson, J. M., and Clayman, G. L. (1998). In vitro growth suppression by adenoviral transduction of p21 and p16 in squamous cell carcinoma of the head and neck: a research model for combination gene therapy. Arch. Otolaryngol. Head Neck Surg. 124, 88–92. Momand, J., Zambetti, G. P., Olson, D. C., George, D., and Levine, A. J. (1992). The MDM-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell 69, 1237– 1245. Nagata, S., and Golstein, P. (1995). The Fas death factor. Science 267, 1449–1456. Nobori, T., Miura, K., Wu, D. J., Lois, A., Takabayashi, K., and Carson, D. A. (1994). Deletions of the cyclin-dependent kinase-4 inhibitor gene in multiple human cancers. Nature 368, 753–756. Noda, A., Ning, Y., Venable, S. F., Pereira, S. D., and Smith, J. R. (1994). Cloning of senescent cell-derived inhibitors of DNA synthesis using an expression screen. Exp. Cell Res. 211, 90–98. Norris, J. S., Hyer, M. L., Rubinchik, S., Wang, D., Ding, R.-Z., Berger, M. S., and Dong, J. Y. (1999). Adenoviral-delivered Fas ligand fusion gene therapy for breast, prostate, and breast cancer. Cancer Gene Ther. 6(suppl.), S3–S4. de Oca Luna, R. M., Wagner, D. S., and Lozano, G. (1995). Rescue of early embryonic lethality in MDM2-deficient mice by deletion of p53. Nature 378, 203–206. Oltavi, Z. N., Milliman, C. L., and Korsmeyer, S. J. (1993). Bcl-2 heterodimerizes in vivo with a conserved homolog, bax, that accelerates programmed cell death. Cell 74, 609–619. Pan, G., Ni, J., Wei, Y.F., Yu, G. L., Gentz, R., and Dixit, V. M. (1997). An antagonist decoy receptor and a death domain-containing receptor for TRAIL. Science 277, 815–818. Pan, G., O’Rourke, K., Chinnaiyan, A. M., Gentz, R., Ebner, R., Ni, J., and Dixit, V. M. (1997). The receptor for the cytotoxic ligand TRAIL. Science 276, 111–113. Paramio, J. M., Navarro, M., Segrelles, C., Gomez-Cassero, E., and Jorcano, J. L. (1999). PTEN tumor suppressor is linked to the cell cycle through the retinoblastoma protein. Oncogene 18, 7462– 7468. Park, K.-H., Lee, J. H., Yoo, C. G., Chung, H. S., Shim, Y. W., Han, S. K., Shim, Y. W., and Lee, C.-T. (2000). An adenovirus expressing mutant p27 showed more potent antitumor response than adenoviruswild p27. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 352.

Cancer Gene Therapy with Tumor Suppressor Genes Involved in Cell-Cycle Control 119. Pataer, A., Fang, B., Yu, R., Kagawa, S., Hunt, K. K., McDonnell, T. J., Roth, J. A., and Swisher, S. G. (2000). Adenoviral bak overexpression mediates caspase-dependent tumor killing. Cancer Res. 60, 788–792. 120. Patel, S. D., Tran, A. C., Lampere, L., Gyuris, J., and McArthur, J. G. (1999). Cancer gene therapy with novel chimeric p27/p16 tumor suppressor genes. Cancer Gene Ther. 6(suppl.), S4. 121. Pearson, A. S., Spitz, F. R., Swisher, S. G., Kataoka, M., Sarkiss, M. G., Meyn, R. E., McDonnell, T. J., Cristiano, R. J., and Roth, J.A. (2000). Up-regulation of the proapoptotic mediators bax and bak after adenovirus-mediated p53 gene transfer in lung cancer cells. Clin. Cancer Res. 6, 887–890. 122. Peter, M. E., and Krammer, P. H. (1998). Mechanisms of CD95 (APO-1/Fas)-mediated apoptosis. Curr. Opin. Immunol. 10, 545–551. 123. Phillips, A. C., Bates, S., Ryan, K. M., Helin, K., and Vousden, K. H. (1997). Induction of DNA synthesis and apoptosis are separable functions of E2F-1. Genes Dev. 11, 1853–1863. 124. Pinyol, M., Hernandez, L., Caz Balbin, M., Jares, P., Fernandez, P. L., Montserrat, E., Cardesa, A., Lopez-Otin, C., and Campo, E. (1997). Deletions and loss of expression of p16INK4a and p21WAF1 genes are associated with aggressive variants of mantle cell lymphomas. Blood 89, 272–280. 125. Pomerantz, J., Schreiber-Agus, N., Liegeois, N. J., Silverman, A., Alland, L., Chin, L., Potes, J., Chen, K., Orlow, I., Lee, H. W., CordonCardo, C., and DePinho, R. A. (1998). The INK4a tumor suppressor gene product, p19ARF , interacts with MDM2 and neutralizes MDM2’s inhibition of p53. Cell 92, 713–723. 126. Ponce-Casteneda, M., Lee, M., Latres, E., Polyak, K., Lacombe, L., Montgomery, K., Matthew, S., Krauter, K., Sheifeld, J., Massague, J. et al. (1995). p27Kip1 : chromosomal mapping to 12p12.2–12p13.1 and absence of mutations in human tumors. Cancer Res. 55, 1211–1214. 127. Prabhu, N. S., Somasundaram, K., Tian, H., Enders, G. H., Satyamoorthy, K., Herlyn, M., and El-Deiry, W. S. (1999) The administration schedule of cyclin-dependent kinase inhibitor gene therapy in etoposide chemotherapy is a major determinant of cytotoxicity. Int. J. Oncol. 15, 209–216. 128. Prabhu, N. S., Blagosklonny, M. V., Zeng, Y.-X., Wu, G. S., Waldman, T., and El-Deiry, W. S. (1996). Suppression of cancer cell growth by adenovirus expressing p21WAF1/CIP1 deficient in PCNA interaction. Clin. Cancer Res. 2, 1221–1229. 129. Qin, X.-Q., Livingston, D. M., Kaelin, W. G., and Adams, P. D. (1994). Deregulated transcription factor E2F-1 expression leads to S-phase entry and p53-mediated apoptosis. Proc. Natl. Acad. Sci. USA 91, 10918– 10922. 130. Quelle, D. E., Zindy, F., Ashmun, R. A., and Sherr, C. J. (1995). Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest. Cell 83, 993–1000. 131. Quesnel, B., Preudhomme, C., Lepelley, P., Hetuin, D., Vanrumbeke, M., Bauters, F., Velu, T., and Fenaux, P. (1996). Transfer of p16INK4A/CDKN2 gene in leukaemic cell lines inhibits cell proliferation. Br. J. Hematol. 95, 291–298. 132. Ramondetta, L., Mills, G. B., Burke, T. W., and Wolf, J. K. (2000). Adenovirus-mediated expression of p53 or p21 in a papillary serous endometrial carcinoma cell line (SPEC-2) results in both growth inhibition and apoptotic cell death: potential application of gene therapy to endometrial cancer. Clin. Cancer Res. 6, 278–284. 133. Rampino, N., Yamamoto, H., Ionov, Y., Li, Y., Sawai, H., Reed, J. C., and Perucho, M. (1997). Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype. Science 275, 967–969. 134. Reiser, M., Neumann, I., Schmiegel, W., Wu, P.-C., and Lau, J. Y. N. (2000). Induction of cell proliferation arrest and apoptosis in hepatoma cells through adenoviral-mediated transfer of p53 gene. J. Hepatol. 32, 771–782.

295

135. Ribeiro, J. C., Hanley, J. R., and Russell, P. J. (1996). Studies of X-irradiated bladder cancer cell lines showing differences in p53 status: absence of a p53-dependent cell cycle checkpoint pathway. Oncogene 13, 1269–1278. 136. Riley, D. J., Nikitin, A. Y., and Lee, W. H. (1996). Adenovirus-mediated retinoblastoma gene therapy suppresses spontaneous pituitary melanotroph tumors in Rb ±/–mice. Nat. Med. 2, 1316–1321. 137. Rocco, J. W., Li, D., Liggett, W. H., Jr., Duan, L., Saunders, Jr., J. K., Sidransky, D., and O’Malley, Jr., B. W. (1998). p16INK4A adenovirus mediated gene therapy for human head and neck squamous cell cancer. Clin. Cancer Res. 4, 1697–1704. 138. Rousseau, D., Cannella, D., Boulaire, J., Fitzgerald, P., Fotedar, A., and Fotedar, R. (1999). Growth inhibition by CDK-cyclin and PCNA binding domains of p21 occurs by distinct mechanisms and is regulated by ubiquitin-proteasome pathway. Oncogene 18, 4313–4325. 139. Rubinchik, S., Ding, R., Qiu, A. J., Zhang, F., and Dong, J. (2000). Adenoviral vector which delivers FasL-GFP fusion protein regulated by the tet-inducible expression system. Gene Ther. 7, 875–885. 140. Sakai, K., Mitani, K., and Miyazaki, J. (1995). Efficient regulation of gene expression by adenovirus vector-mediated delivery of the Cre recombinase. Biochem. Biophys. Res. Commun. 217, 393–401. 141. Sakakura, C., Sweeney, E. A., Shirahama, T., Igarashi, Y., Hakomori, S., Tsujimoto, H., Imanishi, T., Ogaki, M., Ohyama, T., Yamazaki, J., Hagiwara, A., Yamaguchi, T., Sawai, K., and Takahashi, T. (1996). Overexpression of bax sensitizes human breast cancer MCF-7 cells to radiation-induced apoptosis . Int. J. Cancer 67, 101–105. 142. Sakurada, A., Hamada, H., Fukushige, S., Yokoyama, T., Yoshinaga, K., Furukawa, T., Sato, S., Yajima, A, Sato, M., Fujimura, S., and Horii, A. (1999). Adenovirus-mediated delivery of the PTEN gene inhibits cell growth by induction of apoptosis in endometrial cancer. Int. J. Oncol. 15, 1069–1074. 143. Sandig, V., Brand, K., Herwig, S., Lukas, J., Bartek, J., and Strauss, M. (1997). Adenovirally transferred p16INK4/CDKN2 and p53 genes cooperate to induce apoptotic tumor cell death. Nat. Med. 3, 313– 319. 144. Sard, L., Accornero, P., Tornielli, S., Delia, D., Bunone, G., Campiglio, M., Colombo, M.P., Gramegna, M., Croce, C. M., Pierotti, M. A., and Sozzi, G. (1999). The tumor-suppressor gene FHIT is involved in the regulation of apoptosis and in cell cycle control. Proc. Natl. Acad. Sci. USA 96, 8489–8492. 145. Scheffner, M., Werness, B. A., Huibregtse, J. M., Levine, A. J., and Howley, P. M. (1990). The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell 63, 1129–1136. 146. Schrump, D. S., Chen, G. A., Consuli, U., Jin, X., and Roth, J. A. (1996). Inhibition of esophageal cancer proliferation by adenovirally mediated delivery of p16INK4 . Cancer Gene Ther. 3, 357–364. 147. Segawa, T., Sasagawa, T., Saijoh, K., and Inoue, M. (2000). Clinicopathological significance of fragile histidine triad transcription protein expression in endometrial carcinomas. Clin. Cancer Res. 6, 2341–2348. 148. Serrano, M., Hannon, G. J., and Beach, D. (1993). A new regulatory motif in cell-cycle control causing specific inhibition of cyclin D/CDK4. Nature 366, 704–707. 149. Sgagias, M. K., Kim, M., Rikiyama, T., Yun, J. J., and Cowan, K. H. (2000). Induction of apoptosis by adenovirus-mediated overexpression of BRCA1 in T47D human breast cancer cells. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 537. 150. Sheikh, M. S., Fernandes-Salas, E., Chou, J., Brooks, K., Huang, Y., and Fornace, A. J. (2000). p53 mediated apoptosis in a caspase 8-dependent but FADD-independent manner. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 555. 151. Sheikh, M. S., Huang, Y., Fernandez-Salas, E. A., El-Deiry, W. S., Friess, H., Amundson, S., Yin, J., Meltzer, S.J., Holbrook, N. J.,

296

152.

153.

154.

155.

156.

157.

158.

159.

160.

161.

162.

163.

164.

165.

166.

167.


and Fornace, Jr., A. J. (1999). The antiapoptotic decoy receptor TRID/TRAIL-R3 is a p53-regulated DNA damage-inducible gene that is overexpressed in primary tumors of the gastrointestinal tract. Oncogene 18, 4153–4159. Sheridan, J. P., Marsters, S. A., Pitti, R. M., Gurney, A., Skubatch, M., Baldwin, D., Ramakrishnan, L., Gray, C.L., Baker, K., Wood, W. I., Goddard, A. D., Godowski, P., and Ashkenazi, A. (1997). Control of TRAIL-induced apoptosis by a family of signaling and decoy receptors. Science 277, 818–821. Shimada, Y., Sato, F., Watanabe, G., Yamasaki, S., Kato, M., Maeda, M., and Imamura, M. (2000). Loss of fragile histidine triad gene expression is associated with progression of esophageal squamous cell carcinoma, but not with the patient’s prognosis and smoking history. Cancer 89, 5–11. Shinoura, N., Ohashi, M., Yoshida, Y., Kirino, T., Asai, A., Hashimoto, M., and Hamada, H. (2000). Adenovirus-mediated overexpression of fas induces apoptosis of gliomas. Cancer Gene Ther. 7, 224– 232. Shinoura, N., Muramatsu, Y., Yoshida, Y., Asai, A., Kirino, T., and Hamada, H. (2000). Adenovirus-mediated transfer of caspase-3 with Fas ligand induces drastic apoptosis in U-373MG glioma cells. Exp. Cell Res. 256, 423–433. Shinoura, N., Saito, K., Yoshida, Y., Hashimoto, M., Asai, A., Kirino, T., and Hamada, H. (2000). Adenovirus-mediated transfer of bax with caspase-8 controlled by myelin basic protein promoter exerts an enhanced cytotoxic effect in gliomas. Cancer Gene Ther. 7, 739– 748. Shinoura, N., Yoshida, Y., Asai, A., Kirino, T., and Hamada, H. (2000). Adenovirus-mediated transfer of p53 and Fas ligand drastically enhances apoptosis in gliomas. Cancer Gene Ther. 7, 732–738. Shinoura, N., Yoshida, Y., Sadata, A., Hanada, K.I., Yamamoto, S., Kirino, T., Asai, A., and Hamada, H. (1998). Apoptosis by retrovirusand adenovirus-mediated gene transfer of Fas ligand to glioma cells: implications for gene therapy. Hum. Gene Ther. 9, 1983–1993. Singh, C. J., Wong, S. H., and Hong, W. (1994). Overexpression of E2F1 in rat embryo fibroblasts leads to neoplastic transformation. EMBO J. 13, 3329–3338. Somasundaram, K., MacLachlan, T. K., Burns, T. F., Sgagias, M., Cowan, K. H., Weber, B. L., and El-Deiry, W. S. (1999). BRCA1 signals ARF-dependent stabilization and coactivation of p53. Oncogene 18, 6605–6614. Spirin, K. S., Simpson, J. F., Takeuchi, S., Kawamata, N., Miller, C. W., and Koeffler, H.P. (1996). p27/Kip1 mutation found in breast cancer. Cancer Res. 56, 2400–2404. Steiner, M. S., Wang, Y., Zhang, Y., Zhang, X., and Lu, Y. (2000). p16/MTS1/INK4A suppresses prostate cancer by both pRb dependent and independent pathways. Oncogene 19, 1297–1306. Steiner, M. S., Zhang, Y., Farooq, F., Lerner, J., Wang, Y., and Lu, Y. (2000). Adenoviral vector containing wild-type p16 suppresses prostate cancer growth and prolongs survival by inducing cell senescence. Cancer Gene Ther. 7, 360–372. Stone, S., Dayananth, P., and Kamb, A. (1996). Reversible, p16-mediated cell cycle arrest as protection from chemotherapy. Cancer Res. 56, 3199–3202. Suda, T., Takahashi, T., Golstein, P., and Nagata, S. (1993). Molecular cloning and expression of the Fas ligand, a novel member of the tumor necrosis factor family. Cell 75, 1169–1178. Takada, S., Kaneniwa, N., Tsuchida, N., and Koike, K. (1997). Cytoplasmic retention of the p53 tumor suppressor gene product is observed in the hepatitis B virus X gene-transfected cells. Oncogene 15, 1895–1901. Tamura, M., Gu, J., Tran, H., and Yamada, K. M. (1999). PTEN gene and integrin signaling in cancer. J. Natl. Cancer Inst. 91, 1820– 1828.

168. Tan, P., Cady, B., Wanner, M., Worland, P., Cukor, B., Magi-Galluzzi, C., Lavin, P., Draetta, G., Pagano, M., and Loda, M. (1997). The cell cycle inhibitor p27 is an independent prognostic marker in small (T1a,b) invasive breast carcinomas. Cancer Res. 57, 1259–1263. 169. Tian, X. X., Pang, J. C., To, S. S., and Ng, H. K. (1999). Restoration of wild-type PTEN expression leads to apoptosis, induces differentiation, and reduces telomerase activity in human glioma cells. J. Neuropathol. Exp. Neurol. 58, 472–479. 170. Todd, M. C., Sclafani, R. A., and Langan, T. A. (2000). Ovarian cancer cells that coexpress endogenous Rb and p16 are insensitive to overxpression of functional p16 protein. Oncogene 19, 258–264. 171. Tsao, Y. P., Huang, S. J., Chang, J. L., Hsieh, J. T., Pong, R. C., and Chen, S. L. (1999). Adenovirus-mediated p21 (WAF1/SDII/CIP1) gene transfer induces apoptosis in human cervical cancer cell lines. J. Virol. 73, 4983–4990. 172. Ueda, H., Ullrich, S. J., Gangemi, J. D., Kappel, C. A., Ngo, L., Feitelson, M. A., and Jay, G. (1995). Functional inactivation but not structural mutation of p53 causes liver cancer. Nat. Genet. 9, 41–47. 173. Vahanian, N., Reichardt, K., Higginbotham, J., Link, C., and Seth, P. (2000). In vivo study of VHL tumor suppressor gene, its potential role in angiogenesis and cancer gene therapy, Abstract No. 408, 3rd Annual Meeting of the American Society of Gene Therapy. Mol. Ther. 1, 5157. 174. Valenzuela, M. T., Nunez, M. I., Villalobos, M., Siles, E., McMillan, T. J., Pedraza, V., and Ruiz de Almodovar, J. M. (1997). A comparison of p53 and p16 expression in human tumor cells treated with hyperthermia or ionizing radiation. Int. J. Cancer 72, 307–312. 175. Villa, L. L. (1997). Human papillomaviruses and cervical cancer. Adv. Cancer Res. 71, 321–341. 176. Waldman, T., Lengauer, C., Kinzler, K. W., and Vogelstein, B. (1996). Uncoupling of S phase and mitosis induced by anticancer agents in cells lacking p21. Nature 381, 643–644. 177. Wang, X. W., Forrester, K., Yeh, H., Feitelson, M. A., Gu, J., and Harris, C. C. (1994). Hepatitis B virus X protein inhibits p53 sequence-specific DNA binding, transcriptional activity, and association with transcription factor ERCC3. Proc. Natl. Acad. Sci. USA 91, 2230–2234. 178. Watzlik, A., Dufter, C., Jung, M., Opelz, G., and Terness, P. (2000). Fas ligand gene-carrying adeno-5 AdEasy viruses can be efficiently propagated in apoptosis-sensitive human embryonic retinoblast 911 cells. Gene Ther. 7, 70–74. 179. Weber, J. D., Jeffers, J. R., Rehg, J. E., Randle, D. H., Lozano, G., Roussel, M. F., Sherr, C. J., and Zambetti, G.P. (2000). p53independent functions of the p19ARF tumor suppressor. Genes Dev. 14, 2358–2365. 180. Weinberg, R. (1995). The retinoblastoma protein and cell cycle control. Cell 81, 323–330. 181. Werner, N. S., Siprashvili, Z., Fong, L. Y., Marquitan, G., Schroder, J. K., Bardenheuer, W., Seeber, S., Huebner, K., and Schutte, J., and Opalka, B. (2000). Differential susceptibility of renal carcinoma cell lines to tumor suppression by exogenous FHIT expression. Cancer Res. 60, 2780–2785. 182. Wick, W., Furnari, F. B., Naumann, U., Cavenee, W. K., and Weller, M. (1999). PTEN gene transfer in human malignant glioma: sensitization to irradiation and CD95L-induced apoptosis. Oncogene 18, 3936–3943. 183. Wolf, J. K., Kim, T. E., Fightmaster, D., Bodurka, D., Gershenson, D. M., Mills, G., and Wharton, J. T. (1999). Growth suppression of human ovarian cancer cell lines by the introduction of a p16 gene via recombinant adenovirus. Gynecol. Oncol. 73, 27–34. 184. Wolff, G., Schumacher, A., Karawajew, L., Ruppert, V., Arnold, W., Daniel, P., and Doerken, B. (2000). Adenovirus-mediated gene transfer of p16 INK4/CDKN2 to colorectal cancer, Abstract No. 418, 3rd Annual Meeting of the American Society of Gene Therapy. Mol. Ther. 1, 5161.

Cancer Gene Therapy with Tumor Suppressor Genes Involved in Cell-Cycle Control 185. Wu, G. S., Kim, K., and El-Deiry, W. S. (2000). KILLER/DR5, a novel DNA-damage inducible death receptor gene, links the p53-tumor suppressor to caspase activation and apoptotic death. Adv. Exp. Med. Biol. 465, 143–151. 186. Wu, R., Connolly, D. C., Dunn, R. L., and Cho, K. R. (2000). Restored expression of fragile histidine triad protein and tumorigenicity of cervical carcinoma cells. J. Natl. Cancer Inst. 92, 338–344. 187. Wu, G. S., Burns, T. F., McDonald, E. R., III, Meng, R. D., Kao, G., Muschel, R., Yen, T., and El-Deiry, W. S. (1999). Induction of the TRAIL receptor KILLER/DR5 in p53-dependent apoptosis but not growth arrest. Oncogene 18, 6411–6418. 188. Wu, G. S., Burns, T. F., McDonald III, E. R., Jiang, W., Meng, R., Krantz, I. D., Kao, G., Gan, D.-D., Zhou, J.-Y., Muschel, R., Hamilton, S. R., Spinner, N. B., Markowitz, S., Wu, G., and El-Deiry, W. S. (1997). KILLER/DR5 is a DNA damage-inducible p53-regulated death receptor gene. Nat. Genet. 17, 141–143. 189. Wu, H., Wade, M., Krall, L., Grisham, J., Xiong, Y., and Van Dyke, T. (1996). Targeted in vivo expression of the cyclin-dependent kinase inhibitor p21 halts hepatocyte cell-cycle progression, postnatal liver development, and regeneration. Genes Dev. 10, 245–260. 190. Wu, G. S., and El-Deiry, W. S. (1996). Apoptotic death of tumor cells correlates with chemosensitivity, independent of p53 or bcl-2. Clin. Cancer Res. 2, 623–633. 191. Wu, X., and Levine, A. J. (1994). p53 and E2F-1 cooperate to mediate apoptosis. Proc. Natl. Acad. Sci. USA 91, 3602–3606. 192. Xiang, J., Piche, A., Rancourt, C., Gomez-Navarro, J., Siegal, G. P., Alvarez, R. D., and Curiel, D. T. (1999). An inducible recombinant adenoviral vector encoding bax selectively induces apoptosis in ovarian cancer cells. Tumor Targeting 4, 84–91. 193. Xiong, Y., Hannon, G. J., Zhang, H., Casso, D., Kobayashi, R., and Beach, D. (1993). p21 is a universal inhibitor of cyclin kinases. Nature 366, 701–704. 194. Xu, S. Q., and El-Deiry, W. S. (2000). p21(WAF1/CIP1) inhibits initiator caspase cleavage by TRAIL death receptor DR4. Biochem. Biophys. Res. Commun. 269, 179–190. 195. Xu, H.-J. (1997). Strategies for approaching retinoblastoma tumor suppressor gene therapy. Adv. Pharmacol. 40, 369–397. 196. Xu, H.-J., Zhou, Y. L., Seigne, J., Perng, G. S., Mixon, M., Zhang, C. Y., Li, J., Benedict, W. F., and Hu, S. X. (1996). Enhanced tumor suppressor gene therapy via replication-deficient adenovirus vectors expressing an N-terminal truncated retinoblastoma protein. Cancer Res. 56, 2245– 2249. 197. Xu, G., Livingston, D. M., and Krek, W. (1995). Multiple members of the E2F transcription factor family are the products of oncogenes. Proc. Natl. Acad. Sci. USA 92, 1357–1361. 198. Xu, H.-J., Xu, K., Zhou, Y., Li, J., Benedict, W. F., and Hu, S. X. (1994). Enhanced tumor cell growth suppression by an N-terminal truncated retinbolastoma protein. Proc. Natl. Acad. Sci. USA 91, 9837–9841.

297

199. Yamamoto, K., Katayose, Y., Suzuki, M., Unno, M., Endo, E., Takemura, S., Yoshida, H., Cowan, K. H, Seth, P., and Matsuno, S. (2000). Adenovirus expressing cyclin-dependent kinase inhibitor p27Kip1 induces apoptosis on human cholangiocarcinoma cell lines utilizing fas pathway. Proc. Annu. Meet. Am. Assoc. Cancer Res. 41, 90. 200. Yamasaki, L., Bronson, R., Williams, B. O., Dyson, N. J., Harlow, E., and Jacks, T. (1998). Loss of E2F-1 reduces tumorigenesis and extends the lifespan of Rb1(+/−) mice. Nat. Genet. 18, 360–364. 201. Yamasaki, L., Jacks, T., Bronson, R., Goillot, E., Harlow, E., and Dyson, N. (1996). Tumor induction and tissue atrophy in mice lacking E2F-1. Cell 85, 537–548. 202. Yang, C.-T., Liang, Y., Yeh, C.-C., Chang, J. W.-C., Zhang, F., McCormick, F., and Jablons, D.M. (2000). Adenovirus-mediated p14ARF gene transfer in human mesothelioma cells. J. Natl. Cancer Inst. 92, 636–641. 203. Yang, H. L., Dong, Y. B., Elliott, M. J., Liu, T. J., and McMasters, K. M. (2000). Caspase activation and changes in bcl-2 family member protein expression associated with E2F-1-mediated apoptosis in human esophageal cancer cells. Clin. Cancer Res. 6, 1579–1589. 204. Yang, H. L., Dong, Y. B., Elliott, M. J., Liu, T. J., Atienza, Jr., C., Stilwell, A., and McMasters, K. M. (1999). Adenovirus-mediated E2F-1 gene transfer inhibits MDM2 expression and efficiently induces apoptosis in MDM2-overexpressing tumor cells. Clin. Cancer Res. 5, 2242–2250. 205. Yang, Z.-Y., Perkins, N. D., Ohno, T., Nabel, E. G., and Nabel, G. J. (1995). The p21 cyclin-dependent kinase inhibitor suppresses tumorigenicity in vivo. Nat. Med. 1, 1052–1056. 206. Yin, C., Kundson, C. M., Korsmeyer, S. J., and Van Dyke, T. (1997). Bax suppresses tumorigenesis and stimulates apoptosis in vivo. Nature 385, 637–640. 207. Zhang, Y., Xiong, Y., and Yarbrough, W. G. (1998). ARF promotes MDM2 degradation and stabilizes p53: ARF-INK4a locus deletion impairs both the Rb and p53 tumor suppression pathways. Cell 92, 725–734. 208. Zhang, W., Grasso, L., McClain, C. D., Gambel, A. M., Cha, Y., Travali, S., Deisseroth, A. B., and Mercer, W. E. (1995). p53-independent induction of WAF1/CIP1 in human leukemia cells is correlated with growth arrest accompanying monocyte/macrophage differentiation. Cancer Res. 55, 668–674. 209. Zhu, L., van der Heuvel, S., Helin, K., Fattaey, A., Ewen, M., Livingston, D., Dyson, N., and Harlow, E. (1993). Inhibition of cell proliferation by p107, a relative of the retinoblastoma protein. Genes Dev. 7, 1111–1125. 210. Zimmermann, H., Degenkolbe, R., Bernard, H. U., and O’Connor, M. J. (1999). The human papillomavirus type 16 E6 oncoprotein can down-regulate p53 activity by targeting the transcriptional coactivator CBP/p300. J. Virol. 73, 6209–6819.


C

H

A

P

T

E

R

18 Cancer Gene Therapy with the p53 Tumor Suppressor Gene RAYMOND D. MENG

WAFIK S. EL-DEIRY

Laboratory of Molecular Oncology and Cell Cycle Regulation Howard Hughes Medical Institute Departments Medicine and Genetics Cancer Center and The Institute for Human Gene Therapy University of Pennsylvania School of Medicine Philadelphia, Pennsylvania 19104

Laboratory of Molecular Oncology and Cell Cycle Regulation Howard Hughes Medical Institute Departments Medicine and Genetics Cancer Center and The Institute for Human Gene Therapy University of Pennsylvania School of Medicine Philadelphia, Pennsylvania 19104

I. Introduction 299 II. Vectors for Gene Therapy

Before the cell can enter the next phase of the cell cycle, it must pass through a checkpoint that decides if all the previous processes have been completed. The decision to enter the cell cycle is made during the G1 phase by cyclins and their regulatory units, the cyclin-dependent kinases (CDKs) (reviewed by McDonald and El-Deiry [98]). Control of the CDKs is achieved by phosphorylation on different sites of the protein and by the activity of CDK inhibitors, which are composed of two families [105]. The INK4 proteins (consisting of p15, p16, p18, and p19) bind to CDK4 or to CDK6, whereas the CIP/KIP proteins (including p21, p27, and p57) bind to cyclin-CDK complexes. Cell-cycle progression is also affected by environmental stimuli. If a eukaryotic cell is deprived of nutrients or growth factors, the cell responds by activation of a checkpoint leading to cell-cycle arrest until conditions become favorable for cell division (reviewed in Murray and Hunt [106]). In mammalian cells, growth factors and mitogens regulate the expression level of cyclin D1, which is involved in driving forward the transition from the G1 to S phase (reviewed by Sherr [139]). In mammalian cells, the growth-factor-dependent G1 to S checkpoint is regulated by the retinoblastoma (Rb) protein (reviewed by Weinberg [165] and Harbour and Dean [53]). Once a cell has traversed this so-called “restriction point” [120], it is committed to going through the S, G1 , and M phases, leading to two daughter cells. One of the hallmarks of cancer cells is loss of this checkpoint control. Both the viral oncoproteins (reviewed by Nevins [108]) and mutations in human cancer (reviewed by Sherr [139]) appear to target two parallel, yet related, cell-cycle controlling pathways: (1) the

300

A. Liposomes 300 B. Retroviruses 300 C. Adenoviruses 301 D. Adeno-Associated Viruses 301 E. Lentiviral Vectors 301 F. Chimeric Viral Vectors 301 G. Route of Administration 302

III. p53

302

A. Introduction 302 B. Gene Therapy with p53 302 C. Cell-Cycle Arrest and Apoptosis Induced by Ad-p53 304 D. Applications of p53 Gene Therapy 304 E. Combination of Ad-p53 with Other Adenoviruses 307 F. Synthetic p53 Molecules in Gene Therapy 307 G. Gene Therapy with p53 Family Members 308

IV. Conclusions

308

References 308

I. INTRODUCTION Cancer cells accumulate numerous genetic alterations that contribute to tumorigenesis, tumor progression, and chemotherapeutic drug resistance. Most of these alterations affect the regulation of the cell cycle. In normal cells, a balance is achieved between proliferation and cell death by tightly regulating the progression through the cell cycle with cellular checkpoints (reviewed by Hartwell and Kastan [55]).


299


300


p16–cyclin D1–CDK4–Rb pathway, and (2) the ATM–p53– p21 pathway. The p16–cyclin D1–CDK4–Rb pathway helps the cell progress through the G1 phase. Cyclin D1 binds to and positively regulates CDK4, which helps to phosphorylate Rb. In a hypophosphorylated state, Rb is a negative regulator of cell-cycle progression because it binds to and inactivates the E2F transcription factor family. However, when it is phosphorylated by cyclin D1–CDK4, Rb becomes inactive, and the cell can continue into the S phase. p16 is a negative regulator of CDK4, which prevents it from phosphorylating Rb. Abnormalities of each component of this pathway have been identified in human cancers and are usually mutually exclusive (reviewed by Sherr [139]). The detection of DNA damage is governed by the tumor suppressor p53. Multiple genes have been designated tumor suppressors because of their ability to limit cancer cell growth; subsequently, they have a high frequency of mutation or of deletion in human tumors (reviewed by Macleod [92]). Following DNA damage, p53 arrests the cell to allow time for repair, but if the damage is extensive enough p53 initiates programmed cell death, or apoptosis. Loss of these various molecular checkpoints has been found to underlie the development of many tumors because cell-cycle progression becomes deregulated. The accumulation of genetic alterations also contributes to enhanced chemoresistance resulting from the loss of the ability to respond to DNA damage (reviewed by El-Deiry [33]). Of the many alterations that have been identified, changes involving the tumor suppressors, such as p53, are the most common. With loss of growth suppression, progression through the cell cycle remains unchecked, and tumorigenesis results. Therefore, a major strategy in gene therapy for cancer has focused on replacing the tumor suppressors in cancer cells that have been lost through deletion or through mutation (see Fig. 1).

II. VECTORS FOR GENE THERAPY Successful cancer gene therapy depends on successful delivery of the tumor suppressor to the intended target. Several vectors are currently under development for efficacious gene delivery: nonviral systems, such as liposomes, or the more extensively studied viral systems, including retroviruses and adenoviruses.

A. Liposomes Liposomes are a nonviral gene delivery system in which the DNA is directly mixed with liposomes to form a complex, linked by charge interactions (reviewed by Li and Huang [84]). The DNA–liposome complex is then endocytosed by the cell membranes. The advantages of liposomes are that they are relatively convenient to use, they can be produced

FIGURE 1 Tumor suppressors that are targets for gene therapy of cancer. This review discusses tumor suppressors involved in cell-cycle regulation that have been studied as potential targets for gene replacement in the treatment of cancer. Loss of these tumor suppressors, most notably p53, results in tumor development and progression. p53 mediates the cellular response to DNA damage, resulting in growth arrest or in apoptosis. p21 is a main effector of p53 that mediates growth arrest and is a CDKI, along with p16 and p27, which help to regulate G1 transition. Rb helps to mediate cell-cycle progression from G1 to S phase. In addition, the tumor suppressors BRCA1 (involved in breast cancer), VHL (involved in Von Hippel–Landau familial disease), FHIT (involved in chromosomal breakages), and PTEN (involved in cell attachment) also suppress growth through novel mechanisms. Likewise, the apoptosis induced by p53 is based on the activation of select targets. The recently cloned novel TRAIL target KILLER/DR5 and the Fas family of death receptors can be activated by p53 and can induce apoptosis through initiation of a proteolytic caspase cascade. Two other p53-mediated targets involved in apoptosis, bax and the p53-induced genes (PIGs), initiate cell death through reactive oxygen species.

on a large scale, and, most importantly, they do not elicit a strong immune response in vivo. Liposomes, however, do not have high transfection efficiencies, they lack target specificity, and in vivo gene expression is transient because they are rapidly cleared by the reticuloendothelial system. Recent reports also suggest that the unmethylated CpG islands in the DNA plasmid delivered by liposomes may elicit an immune response in mice [84,86,181].

B. Retroviruses Viruses have often been used as vectors for gene therapy because they can transfer genes efficiently and at high expression. The early gene therapy experiments used retroviruses because they can integrate into the chromosomes of the host cell to provide prolonged gene expression and because they can carry large genes, up to 10 kb (reviewed by Palu et al. [119]). Retroviruses, however, have two disadvantages. First, because retroviruses integrate into host chromosomes, they can only efficiently infect cells that are dividing; hence, quiescent cells, such as neurons, are not effectively infected.

Cancer Gene Therapy with the p53 Tumor Suppressor Gene

Second, it is difficult to prepare sufficiently high titers of retroviruses for in vivo gene therapy.

C. Adenoviruses In contrast to retroviruses, adenoviruses infect a wide range of cells, including dividing and nondividing cells, and they can be prepared at extremely high titers, on the order of 1012 plaque-forming units (PFU) per milliliter, which is sufficiently concentrated for in vivo use (reviewed by Hitt et al. [59]). Although they offer high gene expression, adenoviruses do not undergo chromosomal integration. In cancer gene therapy, however, successful eradication of a tumor would obviate the need for prolonged gene expression. In fact, almost 75% of the patients enrolled in gene therapy clinical trials in 1996 were being treated for malignancies, primarily solid tumors [94]. Adenoviruses also do not infect hematopoietic cells with high efficiency (reviewed by Marini et al. [96]). The most significant drawback to adenoviruses, however, is that they elicit a strong host immune response [144,180]. Furthermore, because gene delivery by adenoviruses is transient, multiple treatments may be required, which would further elicit a strong humoral immune response. Consequently, many of the preclinical in vivo experiments with adenoviruses have been conducted in immunodeficient hosts, such as nude mice. Currently, efforts are underway to try to lessen the immune response, either by direct suppression of host immunity [133] or by modification of the adenovirus vector to lessen its inherent immunogenicity [64,68]. Although it may seem intuitive that a heightened immune response may be good in cancer gene therapy, it is less desirable on a practical scale because the immune response helps to eliminate the vector and decreases expression of the transduced gene. Consequently, recent research has focused on improving the gene delivery capability of the adenoviral vectors. First, many groups have now shown that the efficacy of adenoviral gene delivery correlates with the ability of the adenovirus to bind to its receptor on cancer cells which was recently identified as the coxsackie/adenovirus receptor (CAR) [8]. The adenoviral fiber protein binds CAR through a domain in its carboxyl terminus, and another adenoviral protein, the penton base, mediates viral internalization by binding to host-cell integrin proteins αvβ3 and αvβ5 [170]. Subsequently, several groups have shown that the efficiency of adenoviral gene transfer into cancer cell lines depends on the presence of CARs [85] and on the integrins [28,121,151]. It was also shown that the degree of infectivity of head and neck cancer cell lines could be improved by coinfection with an adenovirus expressing CARs, which presumably enhances low CAR levels [85]. Transfection of CARs was also reported to improve the transduction efficiency of an adenovirus encoding the tumor suppressor p21WAF1/CIP1 in bladder cancer cells [115].

301

Finally, several groups have attempted to create adenoviruses that can infect specific targets (reviewed by Miller and Whelan [101] and Wickham [169]). Besides specificity, targeted adenoviruses may also allow systemic rather than intratumoral administration. Although intratumoral injection effectively directs adenovirus to the tumor, it transduces tumor along the needle path rather than the entire tumor, and it does not allow elimination of metastases (reviewed by Wickham [169]). To target selected tumors, the adenovirus backbone has been modified at the transcriptional level by adding tissue-specific promoters. Another strategy adds specific antibody-fragments to the virus vector (reviewed by Bilbao et al. [10]). Recently, tetracycline-responsive recombinant adenoviruses have been developed to try to regulate gene expression following infection [61,183].

D. Adeno-Associated Viruses Adeno-associated viruses (AAVs) are derived from a nonpathogenic and defective human parvovirus that often coexists with adenoviruses (reviewed by Monahan and Samulski [104]). They are relatively safe and nonimmunogenic, provide high titers, and can infect a wide range of cell types, including quiescent cells, hematopoietic cells, and terminally differentiated cells (reviewed by Flotte and Carter [38]). Because the virus can integrate, gene expression induced by AAVs can be long-term, lasting up to 1 year after infection [73,177]. AAVs, however, have several disadvantages. Currently, high-yield AAV production is difficult because of possible contamination with wild-type AAVs or with other helper viruses. Finally, because of the small size of AAVs, only small inserts, up to 5 kb, can be packaged.

E. Lentiviral Vectors Lentiviruses, such as HIV, can infect nondividing and even growth-arrested cells because virus entry is gained using the host-cell nuclear import system (reviewed by Buchschacher and Wong-Staal [18]). Vectors based on lentiviruses are first deleted of their virulence genes [31,187] or made less virulent by decreasing their transcriptional activity [102]. Lentiviral vectors, however, do not have efficient delivery into some terminally differentiated cells, such as skeletal muscle cells or hepatocytes (reviewed by Trono [157]). Currently, because of the shadow of HIV, no clinical trials have been attempted with attenuated lentiviral vectors.

F. Chimeric Viral Vectors Finally, new vectors are being designed that incorporate the advantages of different viral systems. For example, a novel chimeric vector was derived from both adenovirus and retrovirus genes [37]. These vectors were then used to infect human cancer cells, causing them to act as transient producer

302


cells which release retroviruses that can then integrate into nearby cells. Recently, a fusion adenovirus and Epstein–Barr virus (EBV) vector was created, which allows high titer production like adenoviruses but stable expression because of the EBV episome [152].

G. Route of Administration With respect to delivery in vivo, infectivity represents the gateway into the target cells when the virus reaches its destination. However, the ability of a particular agent to reach the tumor is influenced by its route of administration. The most direct route is intratumoral, which provides delivery to the target tissue. Furthermore, the existence of a bystander effect obviates the need to deliver the agent into every cell, and local delivery minimizes toxicity to the normal tissues where the agent is not present. Intratumoral delivery, however, is not always entirely efficient, as the spread of virus can be hindered by a pressure gradient within large tumors, by limited distribution from blood vessels, or by the extracellular matrix (reviewed by Jain [67]). Similarly, delivery of virally expressed genes by intravascular or intracavitary injections also presents barriers. In intravascular administration, instillation into a peripheral vein dilutes the vehicle so only a small portion may ultimately reach the tumor. Intravascular administration also elicits a powerful immune response [122]. Tropism for organs such as the liver (for example, by adenovirus) can be a disadvantage if delivery is intended elsewhere or it can be advantageous if the liver is the target [44]. Even with regional intravascular administration, the virus must traverse the endothelial wall and travel against pressures within an expanding tumor mass. In the case of intracavitary administration (i.e., intrapleural or intraperitoneal), the surface of the tumor mass is coated by virus, but intratumoral delivery within a solid mass represents an important barrier.

III. p53 A. Introduction Among the tumors suppressors being considered for gene replacement in malignancies, p53 has been the focus of many groups for several reasons. First, p53 plays a pivotal role in the fate of a cell following DNA damage. It determines if the damaged cell will undergo growth arrest in order to repair itself [72], or if the cell will undergo programmed cell death or apoptosis because the damage is too extensive [90,91]. Therefore, loss of p53 or mutations in p53, some of which can act in a dominant-negative manner to inhibit residual wildtype p53, significantly contribute to tumor development, tumor progression, and chemotherapeutic resistance (reviewed by Velculescu and El-Deiry [162] and Levine [83]). A survey of the toxicity of hundreds of anticancer drugs toward over 60 human cancer and leukemia cell lines has indicated that

the vast majority of clinically useful drugs are most effective in cells that express wild-type p53 [168]. Second, mutations in p53 are the most common genetic alterations in tumors, being mutated or deleted in over 50% of all human cancers [50]. Germline transmission of a mutant p53 allele predisposes individuals with Li–Fraumeni syndrome to a high risk of cancers [93]. In p53 knockout mice, 75% developed tumors by 6 months of age, and all died by 2 years [30]. Third, loss of p53 results in decreased apoptosis [149] and decreased susceptibility to radiotherapy or chemotherapy [90,91]. With respect to gene replacement therapy, p53 is a potent inducer of cancer cell apoptosis and is effective despite the presence of multiple genetic changes in the cancer cells [5]. Because p53 offers a promising way to regulate the growth of cancers in vivo, much work has been directed at developing this form of gene therapy (reviewed by Nielsen and Maneval [112]; Baselga [6]; Roth et al. [128]). Cancer gene therapy strategies have focused on replacing or even overexpressing wild-type p53 in the hopes that aberrant cell cycle control can once again be tightly regulated. It is important, however, to realize that, despite its strengths, p53 gene therapy has important limitations that must be considered for its clinical development as an anticancer agent.

B. Gene Therapy with p53 The initial p53 gene therapy experiments used retroviruses to deliver the tumor suppressor gene into various cancer cell lines (see Table 1). Two groups introduced wild-type p53 with a retrovirus vector into a non-small-cell lung cancer line and suppressed the growth of the tumor both in vitro [20] and in vivo in a nude mouse model [42]. It was also shown in a phase I clinical trial that retrovirus-transferred p53 can be used to infect human non-small-cell lung cancers by intratumoral injection and that it may effectively limit growth in a small minority of these patients with advanced terminal cancer [131]. Although liposomal delivery of p53 is being studied [186], the majority of p53-directed gene replacement strategies have now shifted toward using an adenovirus vector because the adenovirus can infect numerous cell types and because it can be produced in high titers. It has been previously reported that an adenovirus expressing β-galactosidase (Ad-LacZ) is capable of infecting tumor cells from a wide range of tissues [12]. Furthermore, the ability of the adenovirus to infect these tumor cells was independent of the endogenous p53 status. Some cell lines, however, remain inherently resistant to adenovirus infection. For example, it was reported that two leukemia and two lymphoma cell lines showed less than 0.001% infectivity following infection with an Ad-LacZ at a multiplicity of infection (MOI) of 150 [12]. Other adenovirus-resistant cells include a breast cancer cell line MDA-MB-435 with mutant p53 [113] and a choriocarcinoma cell line JEG3 with wild-type p53 [99]. The explanation for cellular resistance to adenovirus

303

Cancer Gene Therapy with the p53 Tumor Suppressor Gene

TABLE 1 Infection of Selected Cell Lines by Ad-p53 Name of cell line

p53 MOIa

Bladder

UMUC3

mut

20

x

[99]

Breast

MCF7 wt MDA-MB-435 mut SKBr3 mut

10 N.I.b 80% suppression) than angiostatin, endostatin, and neuropilin viruses (20–30% suppression) in both VEGF corneal micropocket assays and tumor growth suppression assays, suggesting the predictive nature of the former assay. These comparative data suggest that soluble VEGF receptors may be particularly amenable for use in systemic antiangiogenic gene therapy, while viral approaches with endostatin and angiostatin will require further optimization.

VIII. ISSUES REGARDING CLINICAL TRANSLATION OF ANTIANGIOGENIC GENE THERAPY A. Safety of Vectors A foremost concern regarding the use of antiangiogenic gene therapy in the oncology clinic is the safety of the viral vectors themselves. Current nonviral vectors, while poten-

tially less dangerous, are unlikely to achieve systemic levels of soluble VEGF receptors sufficient to inhibit significant tumor burdens. On the other hand, while first-generation adenoviruses lacking E1 and E3 can produce extremely robust and therapeutic levels, these viruses have been associated with significant toxicities, likely related to systemic adenoviral infection, as well as cytotoxicity from viral protein production in transduced cells (i.e., liver) [46]. Certainly, these toxicities would be substantially reduced if infection could be accomplished in a remote, although peripheral, site in which the graft could be physically removed or chemically inactivated, such as muscle, skin, or intratumoral. Our data, however, indicate that suppression of bulky preestablished tumors using current adenoviral vectors may require expression levels achievable only by transduction of a substantial tissue mass such as liver. Thus, while current adenoviral vectors may suffice for applications such as the suppression of metastasis by i.m. injection or the treatment of accessible disease by intratumoral injection, the substantial disease often seen in the oncology clinic may require robust but safer vectors delivered by systemic or portal infusion, such as gutless adenoviruses.

B. Safety of Transgene Products Little published information currently exists regarding the safety of long-term VEGF inhibition in adult animals. VEGFA knockout mice exhibit an embryonic lethal phenotype manifest even in heterozygotes, hampering the analysis of loss of function in the adult [16]. The administration of a Flt1 (1–3)–Fc fusion protein to neonatal mice has been reported to produce growth retardation with hepatic and renal dysfunction in a manner decreasing with age, as well as elevations of hematocrit and decreases in platelet counts [26]. Morerover, the Flt1(1–3)–Fc fusion protein has been reported to inhibit endochondral bone formation in 24-day-old mice [27] and corpus luteum development in rats [17]. In adult mice, prolonged administration of anti-Flk1 monoclonal antibodies has not been reported to induce histologic evidence of toxicity in any organs surveyed [53]. On the other hand, angiogenesisdependent processes in the adult include wound healing and menses [19], and the blockade of angiogenesis would seem to be contraindicated in patients with cerebrovascular insufficiency, cardiac ischemia, or peripheral vascular disease. In our experience, although both Ad Flk1-Fc and Ad Flt1 (1–3) potently inhibit tumor growth over a 2-week period, Ad Flt(1–3) mice develope ascites with approximately 30% penetrance and exhibit frequent lethality after 22–28 days, while Ad Flk1-Fc mice are grossly asymptomatic for >1 year (C. J. Kuo, F. Farnebo, and E. Yu, unpublished data). Certainly, more evidence regarding the safety of both transient and prolonged VEGF blockade from phase I trials using small molecule inhibitors or monoclonal antibodies would be desirable prior to proceding with unregulated gene therapy approaches with current vectors.

VEGF-Targeted Antiangiogenic Gene Therapy

C. Context of Translation 1. Control of Local Disease Mutiple contexts can be envisioned in which VEGFdirected gene therapy could be used in the clinic, such as in the treatment of regional disease with local virus administration. Such a strategy could take the form of intratumoral delivery or intratracheal delivery for disease limited to the lung. This might have several safety advantages compared to systemic/intravenous viral administration, including decreased risk of side effects from systemic inhibition of angiogenesis and the ability to remove the graft by surgery or chemical inactivation. The efficacy of local virus delivery, though, has perhaps not been as rigorously demonstrated as possible. While Crystal and colleages [40] have obtained locoregional activity by intratracheal or portal vein infusions of Ad sFlt, these have only been tested against 3-day-old tumor burdens that are not grossly visible and are much smaller than the typical preexisting disease seen in the oncology clinic. Intratumoral injections of self-replicating adenoviruses which selectively multiply in p53-deficient tumors [37] have demonstrated impressive efficacy against head and neck cancer. By analogy to this strategy, it should prove interesting to evaluate the efficacy of intratumoral injections of Ad Flk-Fc or Ad Flt, although it is possible that the nonreplicative nature of firstgeneration adenoviruses will hamper the ability to infect a sufficiently large percentage of the tumor mass to achieve local therapeutic levels. Because of the soluble nature of these VEGF receptors, finite although smaller risks exist for side effects of systemic angiogenesis inhibition from local vector administration. Nevertheless, because of safety factors, local administration for regional control of tumor angiogenesis appears to be a promising initial avenue for clinical translation.

2. Systemic Control of Distant Disease Data from Takayama et al. [63] and from our group have also established the systemic efficacy of adenoviral delivery of soluble VEGF receptors. The intramuscular administration of Ad Flt1-Fc quite effectively inhibits the growth of tumor cells injected at the same time at a remote site. These data indicate that i.m. delivery, which would be anticipated to be safer than i.v. routes, can be efficacious against very small tumor burdens or perhaps in the prevention of metastasis. On the other hand, we have not found intramuscular administration of Ad Flt1(1–3) or Ad Flk1-Fc to be effective against pre-established day 10–14 tumor burdens of >100 mm3 , in contrast to i.v. injection, which produces >80% suppression. Consequently, therapeutic inhibition of distant, bulky, and preestablished disease may require the >2- to 3-log higher circulating levels achieved by transduction of large tissue masses during i.v. injection and accompanying transduction of >50% of hepatocytes.

429

The impressive systemic antiangiogenic and antitumor activity of i.v. administration are certainly accompanied by increased toxicity risks from both virus and transgene, as discussed above. In contrast, although potentially less toxic, intramuscular injections may be effective only in the treatment of very small tumor burdens or in prevention of metastasis. Possibly, reasonable strategies for clinical translation of systemic VEGF receptor treatment by gene therapy would be to use i.m. routes as adjuvant treatment for prevention of metastasis or i.v. routes for administering regulated, less toxic “gutless” adenoviruses against bulky tumor burdens.

3. Combination with Conventional Chemotherapy and Radiotherapy A distinct use of VEGF-directed antiangiogenic gene therapy would be in the combination with conventional modalities such as chemotherapy or radiotherapy. Experimental evidence suggests that anti-KDR antibodies can produce additive-to-synergistic effects with either chemotherapy or radiotherapy against experimental tumors in mice. Our preliminary data in murine tumor models suggest that intravenous treatment with Ad Flk1-Fc in combination with radiotherapy results in impressive gains in tumor suppression and survival relative to either modality alone (F. Farnebo, K. Camphausen, and C. J. Kuo, unpubl. observ.). The limited duration of expression from conventional adenoviruses (i.e., 2–4 weeks) may well be suited to the duration of administration of conventional radiotherapy and chemotherapy. Clinical trials involving conventional radiotherapy or chemotherapy with or without administration of VEGF-directed antiangiogenic gene therapy could therefore be a reasonable option in patient populations in which it would not be ethical to withhold standard treatments.

D. Assessment of Response to Antiangiogenic Gene Therapy 1. Microvessel Density The cytostatic nature of most antiangiogenic agents has several implications for the design of clinical trials, including the need for longer follow-up periods than traditional therapies [19] and the need to measure potential tumor response to treatment before a change in anatomical tumor volume can be detected. Indeed, given these characteristics, the ability to document functional rather than dimensional changes in tumor progression may be quite important during phase I trials in which overt changes in tumor size may only occur after prolonged treatment periods. One biologic endpoint commonly employed for determination of efficacy of antiangiogenic agents is the microvessel density of tumor biopsy samples. Weidner and associates [71] established methodology by which tumor sections are stained with specific antibodies against endothelial

430

Calvin J. Kuo et. al.

antigens (vWF, CD31, CD34), and microvessel “hot spots” are counted by a microscope under 200-fold magnification. This microvessel density of histologic samples of breast cancer has been shown to correlate with patient survival [71], a finding that has been extended to various solid neoplasms (reviewed in Weidner, [70]). Other investigators have been less successful at establishing such a correlation [8,29,41], perhaps reflecting differences in quantitation techniques or from antibody cross-reactivity with other antigens such as on plasma cells (CD31) or perivascular stromal cells (CD34). Microvessel density has been successfully applied to the assessment of biologic response to anti-KDR monoclonal antibodies and dominant-negative KDR, among others [45,53]. In our experience, antiangiogenic gene therapy with Ad Flk1-Fc (Fig. 5; see color insert) or Ad Flt1(1–3) (data not shown) results in decreased microvessel density as measured by vWF staining of tumors from treated versus mockvirus-infected controls. Overall, the use of histological microvessel count as a biologic endpoint to assess efficacy of VEGF-targeted therapy seems reasonable if performed by experienced investigators and well-defined protocols. 2. MRI Imaging of Tumor Blood Flow and Vascularity While technically feasible, the repeated biopsy of a tumor for microvessel density determination may often be impractical because of the invasive nature of the sampling procedure, the requirement for easy tissue accessibility, and sampling errors due to tumor heterogeneity. To circumvent these shortcomings, noninvasive imaging techniques have been developed with potential application for detecting changes in tumor vasculature and blood flow in response to antiangiogenic agents .

Magnetic resonance imaging (MRI) is a cross-sectional imaging technique utilizing strong magnetic fields and multiple radiofrequency pulses to generate an image with outstanding spatial resolution and tissue contrast. Other than for the detection of tumor masses, MRI has recently been used to detect fluid motion such as blood flow. Furthermore, functional changes in various tissues can be evaluated with this method. a. Functional Magnetic Resonance Imaging By measuring the transverse relaxation time (T2) of nuclei activated by a radiofrequency pulse, functional changes in tissues can be measured. Functional MRI (FMRI) was first used to map regions of cortical brain activity [43,49], with activated and nonactivated regions having different imaging properties depending the oxygenation status of local hemoglobin due to differential metabolism. The same principles can be applied in fMRI of tumor tissue. Inhalation of a gas mixture of 95% oxygen and 5% carbon dioxide (carbogen) leads to changes in T2 signal as a result of local vasodilatation and a stimulatory effect on central respiratory regions of the brainstem. Vasodilatation and increased oxygenation of hemoglobin and tumor tissue can be monitored by MRI [1]. b. Dynamic Contrast-Enhanced Magnetic Resonance Imaging In the 1980s, low-molecular-weight gadolinium-based contrast media for MRI were introduced which greatly facilitated the detection of neoplastic lesions due to their hypervascularity. This method, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), is now a well-established adjunct to mammography and ultrasound in breast cancer

FIGURE 5 Reduction in microvessel density in tumors treated with Ad Flk1–Fc. Mice bearing Lewis lung carcinoma tumors of approximately 100 mm3 received intravenous injections of 109 plaque-forming units of the control virus Ad Fc or the anti-angiogenic adenoviruses Ad Flk1–Fc. Tumors were harvested for immunohistochemistry with anti-CD31 antibody after 4 days. The density of CD31 immunostaining is reduced in the tumor treated with Ad Flk1–Fc (right panel) relative to the Ad Fc control virus (left panel). (See color insert.)

VEGF-Targeted Antiangiogenic Gene Therapy

detection [34] and has also been used to detect numerous neoplasms including hepatic [66] and brain [61]. Low-molecular-contrast agents are particularly useful in enhancing MRI detection of microvessels during the early postinjection phase. Unfortunately, gadolinium-based contrast agents exhibit a tendency to leak into the extravascular space, making it almost impossible to assess blood volume in the tumor microvasculature with low molecular contrast agents. This leak is presumably mediated by the vascular permeability activity of intratumoral VEGF, which prior to vessel sprouting induces a strong increase in transendothelial permeability and leakage of large molecules (e.g., fibrin, albumin) into the extravascular space. More importantly, gadolinium-based contrast agents permeate into the extravascular space of normal tissues, making differentiation of benign and malignant lesions problematic [11]. New macromolecular contrast agents with a higher probability of staying intravascular are under investigation in preclinical trials [6]. Nevertheless, by virtue of the permeability-modulating activity of VEGF, the therapeutic efficacy of VEGF antagonism could be indirectly measured by alterations in permeability for MRI contrast agents. For example, a single dose of anti-VEGF antibody decreases transendothelial leakage of a macromolecular contrast agent in a human breast cancer by 98% after 24 hours [6]. Additionally, new macromolecular contrast agents may greatly facilitate measure changes in perfusion and microvessel density in tumor tissues over the course of various antiangiogenic treatments. The sensitivity of DCE-MRI may be increased by specific targeting of the contrast agents to the tumor vasculature. Sipkins et al. [59] coupled paramagnetic liposomes to antibodies directed against the αV β3 integrin expressed in tumor vasculature. Using these paramagnetic liposome–antibody complexes in DCE-MRI, enhancement and distinct localization of tumor angiogenesis were observed in a rabbit carcinoma model with sensitivity exceeding conventional MRI. Certainly, this enhanced in vivo imaging method could be used in the evaluation of VEGF-targeted antiangiogenic therapy. 3. PET Imaging of Tumor Blood Flow and Vascularity Few diagnostic fields in oncology have undergone as comparable an expansion as positron emission tomography (PET). In this imaging modality, radionucleotides with a short half-life are administered either intravenously or via inhalation. The unstable nucleus emits a proton that collides with an electron, creating energy in the form of two gamma rays traveling in opposite directions. Subsequently, sites of accumulated radioactivity can be detected with a camera rotating around the patient. Position emission tomography is well adapted for the measurement of blood flow and volume and tumor metabolism. To assess increased blood flow in areas with disproportionate amounts of microvasculature such as tumors,

431

radioactive oxygen (15 O) in the form of H2 15 O is injected intravenously. H2 15 O has a half-life of approximately 2 minutes and is able to diffuse freely in various tissues. By measuring the rate of delivery to the tissue or organ, the extent of diffusion, and the speed of washout from the tissue, it is possible to calculate blood flow in milliliters per minute per gram of tissue and any alterations after antiangiogenic therapy. Radioactive carbon monoxide (11 CO) can be used for blood volume measurements. Only minute amounts of the radioactive gas are needed for this procedure so that toxic effects are negligible. Radiolabeled red blood cells are detected depending on the blood volume in the specific tissue or organ, any background signals are subtracted by computer to enhance contrast, and the effects of antiangiogenic therapy on blood volume in the tumor tissue are calculated. Another very elegant method used with PET is based on the fact that tumors metabolize glucose but often lack sufficient amounts of enzymes to metabolize the intermediate product glucose-6-PO4 . By radiolabeling glucose with radioactive fluorine (18 F) and injecting this compound intravenously (fluorodeoxyglucose, FDG) an accumulation of FDG-6-PO4 in tumor tissue gives a strong signal in the PET scan. A correlation between high glucose uptake and microvessel density in human gliomas has recently been described [3]. Although FDG is not highly selective for tumor tissue uptake, the pyrimidine analogue (3 -deoxy-3 fluorothymidine, Flt) appears to have superior discrimination [58].

4. Ultrasound Imaging of Tumor Blood Flow and Vascularity Ultrasound is an inexpensive, nonisotopic imaging modality often used in oncology to verify pelvic or testicular masses or to clarify irregularities found by other imaging techniques. The development and improvement of color and power Doppler ultrasound permits real-time assessment of blood flow in various organs and tissues. Tumor vessels often lack multiple smooth muscle cell layers and are therefore easily distensible [9], resulting in increased diastolic flow and a low resistance index. Doppler ultrasound is based on the technique that the Doppler signal can be color encoded, thereby allowing visualizion of not only blood flow but also velocity. A specific combination of speed and direction of blood flow is assigned a color designation. With conventional Doppler ultrasound, blood flow can be detected in vessels with a diameter of approximately 100 μm or more. The resulting ratio of colored pixels with the tumor section to the total number of pixels in that section is defined as the color Doppler vascularity index (CDVI). Chen et al. [10] used transabdominal Doppler ultrasound in patients with colon carcinoma to find a positive

432


correlation between the CDVI and neovascularization to predict distant metastasis and survival. The authors hypothesized that by conventional Doppler ultrasound the CDVI indirectly accounts for intratumoral neovascularization as the number of capillaries is directly proportional to the detectable larger vessels. Additionally, intratumoral Doppler ultrasound signals can be amplified by injecting a perfluorocarbon-based contrast agent intravenously into tumor-bearing mice. These microbubbles increase signal intensity up to 10,000-fold and are small enough to pass freely through capillaries, allowing determination of CDVI based upon the actual microvascular network rather than only the larger supplying and draining vessels (C. Becker and G. Taylor, unpubl.). Further study will be required to evaluate the feasibility of ultrasound determination of vascularity given the need to standardize variables such as the pressure of application of the transducer or the timing of measurement after application of contrast.

IX. CONCLUSION The use of VEGF-directed antiangiogenic gene therapy derives strong mechanistic rationale from abundant experimentation supporting the physiologic role of VEGF in angiogeneisis and the broad-spectrum antiangiogenic and antitumor activities of VEGF-or KDR/Flk-1-targeting monoclonal antibodies and small molecule kinase inhibitors. In its current state, VEGF-directed antiangiogenic gene therapy represents a powerful experimental tool to affirm the therapeutic potential of VEGF inhibition, in which vectors encoding soluble Flk1 or Flt1 ectodomains are easily propagated and conveniently administered. These characteristics have allowed our group and others to rapidly evaluate local and systemic inhibition of VEGF function, to perform comparative analyses, and to begin to assess combinations with conventional chemotherapy and radiotherapy. At the same time, the experimental use of VEGF-directed antiangiogenic gene therapy should provide substantial preclinical information guiding eventual translation into the oncology clinic. Indeed, these viruses should greatly facilitate the study of toxicity of both viral vectors and their transgene products such as soluble VEGF receptors. Additionally, such studies will also allow validation of surrogate endpoints such as microvessel density or imaging correlates of tumor blood flow and vascularity. The inherent safety profile of first-generation adenoviral vectors may restrict their current utility to the adjuvant setting or in the treatment of minimal disease. However, in the future, different vector systems may allow robust and systemic suppression of diffuse bulky disease, fulfilling the promise of a long-term, singleinjection, and economically advantageous antiangiogenic agent.

Acknowledgments We thank George Taylor and Bruce Zetter for allowing us to cite unpublished observations. We are indebted to Cecile Chartier for helpful comments. This work was supported by CaPCURE, the Radley Family Foundation, Deutsche Forschungsgemeinschaft, HHMI, and the National Institutes of Health.

References 1. Abramovitch, R., Frenkiel, D., and Neeman, M. (1998). Analysis of subcutaneous angiogenesis by gradient echo magnetic resonance imaging. Magn. Reson. Med. 39, 813–824. 2. Anderson, W. F. (1998). Human gene therapy. Nature 392, 25–30. 3. Aronen, H. J., Pardo, F. S., Kennedy, D. N., Belliveau, J. W., Packard, S. D., Hsu, D. W., Hochberg, F. H., Fischman, A. J., and Rosen, B. R. (2000). High microvascular blood volume is associated with high glucose uptake and tumor angiogenesis in human gliomas. Clin. Cancer Res. 6, 2189–2200. 4. Asahara, T., Masuda, H., Takahashi, T., Kalka, C., Pastore, C., Silver, M., Kearne, M., Magner, M., and Isner, J. M. (1999). Bone marrow origin of endothelial progenitor cells responsible for postnatal vasculogenesis in physiological and pathological neovascularization. Circ. Res. 85, 221–228. 5. Boehm, T., Folkman, J., Browder, T., and O’Reilly, M. S. (1997). Antiangiogenic therapy of experimental cancer does not induce acquired drug resistance. Nature 390, 404–407. 6. Brasch, R., and Turetschek, K. (2000). MRI characterization of tumors and grading angiogenesis using macromolecular contrast media: status report. Eur. J. Radiol. 34, 148–155. 7. Carmeliet, P., and Collen, D. (2000). Molecular basis of angiogenesis. Role of VEGF and VE-cadherin. Ann. N.Y. Acad. Sci. 902, 249–262. discussion 262–264. 8. Carnochan, P., Briggs, J. C., Westbury, G., and Davies, A. J. (1991). The vascularity of cutaneous melanoma: a quantitative histological study of lesions 0.85–1.25 mm in thickness. Br. J. Cancer 64, 102– 107. 9. Catellino, R. A. (1997). Imaging techniques in cancer management. In Cancer: Principles and Practice of Oncology (V. DeVita, ed.), pp. 633–689. Lippincott-Raven, New York. 10. Chen, C. N., Cheng, Y. M., Liang, J. T., Lee, P. H., Hsieh, F. J., Yuan, R. H., Wang, S. M., Chang, M. F., and Chang, K. J. (2000). Color Doppler vascularity index can predict distant metastasis and survival in colon cancer patients. Cancer Res. 60, 2892–2897. 11. Daldrup, H., Shames, D. M., Wendland, M., Okuhata, Y., Link, T. M., Rosenau, W., Lu, Y., and Brasch, R. C. (1998). Correlation of dynamic contrast-enhanced MR imaging with histologic tumor grade: comparison of macromolecular and small-molecular contrast media. Am. J. Roentgenol. 171, 941–949. 12. Dumont, D. J., Jussila, L., Taipale, J., Lymboussaki, A., Mustonen, T., Pajusola, K., Breitman, M., and Alitalo, K. (1998). Cardiovascular failure in mouse embryos deficient in VEGF receptor-3. Science 282, 946–949. 13. Dvorak, H. F., Brown, L. F., Detmar, M., and Dvorak, A. M. (1995). Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am. J. Pathol. 146, 1029–1039. 14. Eliceiri, B. P., Paul, R., Schwartzberg, P. L., Hood, J. D., Leng, J., and Cheresh, D. A. (1999). Selective requirement for Src kinases during VEGF-induced angiogenesis and vascular permeability. Mol. Cell. 4, 915–924.

VEGF-Targeted Antiangiogenic Gene Therapy 15. Ferrara, N., and Alitalo, K. (1999). Clinical applications of angiogenic growth factors and their inhibitors. Nat. Med. 5, 1359–1364. 16. Ferrara, N., Carver-Moore, K., Chen, H., Dowd, M., Lu, L., O’Shea, K. S., Powell-Braxton, L., Hillan, K. J., and Moore, M. W. (1996). Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 380, 439–442. 17. Ferrara, N., Chen, H., Davis-Smyth, T., Gerber, H. P., Nguyen, T. N., Peers, D., Chisholm, V., Hillan, K. J., and Schwall, R. H. (1998). Vascular endothelial growth factor is essential for corpus luteum angiogenesis. Nat. Med. 4, 336–340. 18. Folkman, J. (1998). Antiangiogenic gene therapy. Proc. Natl. Acad. Sci. USA 95, 9064–9066. 19. Folkman, J. (1995). Seminars in Medicine of the Beth Israel Hospital, Boston. Clinical applications of research on angiogenesis [see comments]. N. Engl. J. Med. 333, 1757–1763. 20. Folkman, J., and D’Amore, P. A. (1996). Blood vessel formation: what is its molecular basis? [comment]. Cell 87, 1153–1155. 21. Fong, G. H., Rossant, J., Gertsenstein, M., and Breitman, M. L. (1995). Role of the Flt-1 receptor tyrosine kinase in regulating the assembly of vascular endothelium. Nature 376, 66–70. 22. Fong, G. H., Zhang, L., Bryce, D. M., and Peng, J. (1999). Increased hemangioblast commitment, not vascular disorganization, is the primary defect in Flt-1 knock-out mice. Development 126, 3015–3025. 23. Fong, T. A., Shawver, L. K., Sun, L., Tang, C., App, H., Powell, T. J., Kim, Y. H., Schreck, R., Wang, X., Risau, W., Ullrich, A., Hirth, K. P., and McMahon, G. (1999). SU5416 is a potent and selective inhibitor of the vascular endothelial growth factor receptor (Flk-1/KDR) that inhibits tyrosine kinase catalysis, tumor vascularization, and growth of multiple tumor types. Cancer Res. 59, 99–106. 24. Fukumura, D., Xavier, R., Sugiura, T., Chen, Y., Park, E. C., Lu, N., Selig, M., Nielsen, G., Taksir, T., Jain, R. K., and Seed, B. (1998). Tumor induction of VEGF promoter activity in stromal cells. Cell 94, 715–725. 25. Gale, N. W., and Yancopoulos, G. D. (1999). Growth factors acting via endothelial cell-specific receptor tyrosine kinases: VEGFs, angiopoietins, and ephrins in vascular development. Genes Dev. 13, 1055–1066. 26. Gerber, H. P., Hillan, K. J., Ryan, A. M., Kowalski, J., Keller, G. A., Rangell, L., Wright, B. D., Radtke, F., Aguet, M., and Ferrara, N. (1999). VEGF is required for growth and survival in neonatal mice. Development 126, 1149–1159. 27. Gerber, H. P., Vu, T. H., Ryan, A. M., Kowalski, J., Werb, Z., and Ferrara, N. (1999). VEGF couples hypertrophic cartilage remodeling, ossification and angiogenesis during endochondral bone formation [see comments]. Nat. Med. 5, 623–628. 28. Goldman, C. K., Kendall, R. L., Cabrera, G., Soroceanu, L., Heike, Y., Gillespie, G. Y., Siegal, G. P., Mao, X., Bett, A. J., Huckle, W. R., Thomas, K. A., and Curiel, D. T. (1998). Paracrine expression of a native soluble vascular endothelial growth factor receptor inhibits tumor growth, metastasis, and mortality rate. Proc. Natl. Acad. Sci. USA 95, 8795–8800. 29. Hall, N. R., Fish, D. E., Hunt, N., Goldin, R. D., Guillou, P. J., and Monson, J. R. (1992). Is the relationship between angiogenesis and metastasis in breast cancer real? Surg. Oncol. 1, 223–229. 30. Hanahan, D., and Folkman, J. (1996). Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell 86, 353– 364. 31. Holash, J., Maisonpierre, P. C., Compton, D., Boland, P., Alexander, C. R., Zagzag, D., Yancopoulos, G. D., and Wiegand, S. J. (1999). Vessel cooption, regression, and growth in tumors mediated by angiopoietins and VEGF. Science 284, 1994–1998. 32. Houck, K. A., Leung, D. W., Rowland, A. M., Winer, J., and Ferrara, N. (1992). Dual regulation of vascular endothelial growth factor bioavailability by genetic and proteolytic mechanisms. J. Biol. Chem. 267, 26031–26037.

433

33. Joseph, I. B., and Isaacs, J. T. (1997). Potentiation of the antiangiogenic ability of linomide by androgen ablation involves down-regulation of vascular endothelial growth factor in human androgen-responsive prostatic cancers. Cancer Res. 57, 1054–1057. 34. Kaiser, W. A., and Zeitler, E. (1989). MR imaging of the breast: fast imaging sequences with and without Gd-DTPA. Preliminary observations. Radiology 170, 681–686. 35. Kendall, R. L., Wang, G., and Thomas, K. A. (1996). Identification of a natural soluble form of the vascular endothelial growth factor receptor, Flt-1, and its heterodimerization with KDR. Biochem. Biophys. Res. Commun. 226, 324–328. 36. Keyt, B. A., Nguyen, H. V., Berleau, L. T., Duarte, C. M., Park, J., Chen, H., and Ferrara, N. (1996). Identification of vascular endothelial growth factor determinants for binding KDR and Flt-1 receptors. Generation of receptor-selective VEGF variants by site-directed mutagenesis. J. Biol. Chem. 271, 5638–5646. 37. Khuri, F. R., Nemunaitis, J., Ganly, I., Arseneau, J., Tannock, I. F., Romel, L., Gore, M., Ironside, J., MacDougall, R. H., Heise, C., Randlev, B., Gillenwater, A. M., Bruso, P., Kaye, S. B., Hong, W. K., and Kirn, D. H. (2000). A controlled trial of intratumoral ONYX-015, a selectively-replicating adenovirus, in combination with cisplatin and 5-fluorouracil in patients with recurrent head and neck cancer. Nat. Med. 6, 879–885. 38. Kim, K. J., Li, B., Winer, J., Armanini, M., Gillett, N., Phillips, H. S., and Ferrara, N. (1993). Inhibition of vascular endothelial growth factorinduced angiogenesis suppresses tumour growth in vivo. Nature 362, 841–844. 39. Kong, H. L., and Crystal, R. G. (1998). Gene therapy strategies for tumor antiangiogenesis [see comments]. J. Natl. Cancer. Inst. 90, 273– 286. 40. Kong, H. L., Hecht, D., Song, W., Kovesdi, I., Hackett, N. R., Yayon, A., and Crystal, R. G. (1998). Regional suppression of tumor growth by in vivo transfer of a cDNA encoding a secreted form of the extracellular domain of the Flt-1 vascular endothelial growth factor receptor. Hum. Gene Ther. 9, 823–833. 41. Leedy, D. A., Trune, D. R., Kronz, J. D., Weidner, N., and Cohen, J. I. (1994). Tumor angiogenesis, the p53 antigen, and cervical metastasis in squamous carcinoma of the tongue. Otolaryngol. Head Neck Surg. 111, 417–422. 42. Levy, N. S., Chung, S., Furneaux, H., and Levy, A. P. (1998). Hypoxic stabilization of vascular endothelial growth factor mRNA by the RNAbinding protein HuR. J. Biol. Chem. 273, 6417–6423. 43. Libutti, S. K., Choyke, P., Carrasquillo, J. A., Bacharach, S., and Neumann, R. D. (1999). Monitoring responses to antiangiogenic agents using noninvasive imaging tests. Cancer J. Sci. Am. 5, 252– 256. 44. Millauer, B., Longhi, M. P., Plate, K. H., Shawver, L. K., Risau, W., Ullrich, A., and Strawn, L. M. (1996). Dominant-negative inhibition of Flk-1 suppresses the growth of many tumor types in vivo. Cancer Res. 56, 1615–1620. 45. Millauer, B., Shawver, L. K., Plate, K. H., Risau, W., and Ullrich, A. (1994). Glioblastoma growth inhibited in vivo by a dominant-negative Flk-1 mutant. Nature 367, 576–579. 46. Mountain, A. (2000). Gene therapy: the first decade. Trends Biotechnol. 18, 119–128. 47. O’Reilly, M. S., Boehm, T., Shing, Y., Fukai, N., Vasios, G., Lane, W. S., Flynn, E., Birkhead, J. R., Olsen, B. R., and Folkman, J. (1997). Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell 88, 277–285. 48. O’Reilly, M. S., Holmgren, L., Shing, Y., Chen, C., Rosenthal, R. A., Moses, M., Lane, W. S., Cao, Y., Sage, E. H., and Folkman, J. (1994). Angiostatin: a novel angiogenesis inhibitor that mediates the suppression of metastases by a Lewis lung carcinoma. Cell 79, 315– 328.

434


49. Ogawa, S., Lee, T. M., Nayak, A. S., and Glynn, P. (1990). Oxygenation-sensitive contrast in magnetic resonance image of rodent brain at high magnetic fields. Magn. Reson. Med. 14, 68–78. 50. Olofsson, B., Pajusola, K., Kaipainen, A., von Euler, G., Joukov, V., Saksela, O., Orpana, A., Pettersson, R. F., Alitalo, K., and Eriksson, U. (1996). Vascular endothelial growth factor B, a novel growth factor for endothelial cells. Proc. Natl. Acad. Sci. USA 93, 2576–2581. 51. Paley, P. J., Staskus, K. A., Gebhard, K., Mohanraj, D., Twiggs, L. B., Carson, L. F., and Ramakrishnan, S. (1997). Vascular endothelial growth factor expression in early stage ovarian carcinoma. Cancer 80, 98–106. 52. Pepper, M. S., and Montesano, R. (1990). Proteolytic balance and capillary morphogenesis. Cell. Differ. Dev. 32, 319–327. 53. Prewett, M., Huber, J., Li, Y., Santiago, A., O’Connor, W., King, K., Overholser, J., Hooper, A., Pytowski, B., Witte, L., Bohlen, P., and Hicklin, D. J. (1999). Antivascular endothelial growth factor receptor (fetal liver kinase 1) monoclonal antibody inhibits tumor angiogenesis and growth of several mouse and human tumors. Cancer Res. 59, 5209– 5218. 54. Rifkin, D. B., Moscatelli, D., Bizik, J., Quarto, N., Blei, F., Dennis, P., Flaumenhaft, R., and Mignatti, P. (1990). Growth factor control of extracellular proteolysis. Cell. Differ. Dev. 32, 313–318. 55. Salven, P., Lymboussaki, A., Heikkila, P., Jaaskela-Saari, H., Enholm, B., Aase, K., von Euler, G., Eriksson, U., Alitalo, K., and Joensuu, H. (1998). Vascular endothelial growth factors VEGF-B and VEGF-C are expressed in human tumors. Am. J. Pathol. 153, 103– 108. 56. Scott, P. A., Gleadle, J. M., Bicknell, R., and Harris, A. L. (1998). Role of the hypoxia sensing system, acidity and reproductive hormones in the variability of vascular endothelial growth factor induction in human breast carcinoma cell lines. Int. J. Cancer 75, 706–712. 57. Shalaby, F., Rossant, J., Yamaguchi, T. P., Gertsenstein, M., Wu, X. F., Breitman, M. L., and Schuh, A. C. (1995). Failure of blood-island formation and vasculogenesis in Flk-1-deficient mice. Nature 376, 62– 66. 58. Shields, A. F., Grierson, J. R., Dohmen, B. M., Machulla, H. J., Stayanoff, J. C., Lawhorn-Crews, J. M., Obradovich, J. E., Muzik, O., and Mangner, T. J. (1998). Imaging proliferation in vivo with [F-18]FLT and positron emission tomography. Nat. Med. 4, 1334– 1336. 59. Sipkins, D. A., Cheresh, D. A., Kazemi, M. R., Nevin, L. M., Bednarski, M. D., and Li, K. C. (1998). Detection of tumor angiogenesis in vivo by alphav beta3 –targeted magnetic resonance imaging. Nat. Med. 4, 623–626. 60. Soker, S., Takashima, S., Miao, H. Q., Neufeld, G., and Klagsbrun, M. (1998). Neuropilin-1 is expressed by endothelial and tumor cells as an isoform-specific receptor for vascular endothelial growth factor. Cell 92, 735–745.

61. Sze, G., Shin, J., Krol, G., Johnson, C., Liu, D., and Deck, M. D. (1988). Intraparenchymal brain metastases: MR imaging versus contrastenhanced CT. Radiology 168, 187–194. 62. Takahashi, Y., Bucana, C. D., Cleary, K. R., and Ellis, L. M. (1998). p53, vessel count, and vascular endothelial growth factor expression in human colon cancer. Int. J. Cancer 79, 34–38. 63. Takayama, K., Ueno, H., Nakanishi, Y., Sakamoto, T., Inoue, K., Shimizu, K., Oohashi, H., and Hara, N. (2000). Suppression of tumor angiogenesis and growth by gene transfer of a soluble form of vascular endothelial growth factor receptor into a remote organ. Cancer Res. 60, 2169–2177. 64. Tolnay, E., Kuhnen, C., Wiethege, T., Konig, J. E., Voss, B., and Muller, K. M. (1998). Hepatocyte growth factor/scatter factor and its receptor c-Met are overexpressed and associated with an increased microvessel density in malignant pleural mesothelioma. J. Cancer Res. Clin. Oncol. 124, 291–296. 65. Tsurusaki, T., Kanda, S., Sakai, H., Kanetake, H., Saito, Y., Alitalo, K., and Koji, T. (1999). Vascular endothelial growth factor-C expression in human prostatic carcinoma and its relationship to lymph node metastasis. Br. J. Cancer 80, 309–313. 66. Unger, E. C., Winokur, T., MacDougall, P., Rosenblum, J., Clair, M., Gatenby, R., and Tilcock, C. (1989). Hepatic metastases: liposomal Gd-DTPA-enhanced MR imaging. Radiology 171, 81–85. 67. Veikkola, T., and Alitalo, K. (1999). VEGFs, receptors and angiogenesis. Semin. Cancer Biol. 9, 211–220. 68. Waltenberger, J., Claesson-Welsh, L., Siegbahn, A., Shibuya, M., and Heldin, C. H. (1994). Different signal transduction properties of KDR and Flt1, two receptors for vascular endothelial growth factor. J. Biol. Chem. 269, 26988–26995. 69. Wedge, S. R., Ogilvie, D. J., Dukes, M., Kendrew, J., Curwen, J. O., Hennequin, L. F., Thomas, A. P., Stokes, E. S., Curry, B., Richmond, G. H., and Wadsworth, P. F. (2000). ZD4190: an orally active inhibitor of vascular endothelial growth factor signaling with broad-spectrum antitumor efficacy. Cancer Res. 60, 970–975. 70. Weidner, N. (1995). Intratumor microvessel density as a prognostic factor in cancer. Am. J. Pathol. 147, 9–19. 71. Weidner, N., Semple, J. P., Welch, W. R., and Folkman, J. (1991). Tumor angiogenesis and metastasis—correlation in invasive breast carcinoma. N. Engl. J. Med. 324, 1–8. 72. Yamada, Y., Nezu, J., Shimane, M., and Hirata, Y. (1997). Molecular cloning of a novel vascular endothelial growth factor, VEGF-D. Genomics 42, 483–488. 73. Zucker, S., Mirza, H., Conner, C. E., Lorenz, A. F., Drews, M. H., Bahou, W. F., and Jesty, J. (1998). Vascular endothelial growth factor induces tissue factor and matrix metalloproteinase production in endothelial cells: conversion of prothrombin to thrombin results in progelatinase A activation and cell proliferation. Int. J. Cancer 75, 780–786.

C

H

A

P

T

E

R

28 Strategies for Combining Gene Therapy with Ionizing Radiation to Improve Antitumor Efficacy DAVID H. GORSKI

HELENA J. MAUCERI

RALPH R. WEICHSELBAUM

The Cancer Institute of New Jersey UMDNJ—Robert Wood Johnson Medical School New Brunswick, New Jersey 08901

Department of Radiation and Cellular Oncology University of Chicago Hospitals Chicago, Illinois 60637

Department of Radiation and Cellular Oncology University of Chicago Hospitals Chicago, Illinois 60637

I. Introduction 435 II. Strategies Using Gene Therapy To Increase the Efficacy of Radiation Therapy 436

cell cytotoxicity. Ionizing radiation (IR) is a conventional and effective local treatment for many different tumors. Unfortunately, many human tumors remain refractory to treatment with IR. Several gene products (p53 and p21, for example) that have been proposed for gene therapy approaches to treat cancer are also involved in determining tumor cell sensitivity or resistance to IR, making the concept of combining anticancer gene therapy with IR an attractive one. Strategies that involve employing gene therapy to improve the antitumor effect of IR fall into two general categories: (1) the use of gene therapy vectors to deliver genes whose protein products improve the antitumor effect of IR, and (2) the use of radiation to enhance the antitumor effect of replication-competent viruses such as herpes virus. The first strategy generally involves the use of gene products that, when expressed within the radiation field, result in either radiosensitization or improved antitumor effects compared with IR or gene therapy alone. The second strategy relies on the observation that some viruses are capable of replicating preferentially in tumor cells, thus killing them, and that radiation enhances this viral proliferation. Finally, the variant of the first strategy, which we term “genetic radiotherapy,” relies on the existence of promoters whose activity is inducible by IR. When introduced into a tumor to be irradiated, administration of IR results in the enhanced production of a toxic gene product. This approach thus provides both temporal and spatial targeting of the toxic gene product, as well as the possibility of additive or even synergistic effects between the gene product and IR. In this chapter, we discuss these approaches to using gene therapy to improve the efficacy of IR.

A. B. C. D.

Introduction 436 p53 Gene-Transfer-Mediated Radiosensitization 436 p21 Gene Therapy and Ionizing Radiation 437 Prodrug Converting Enzyme Suicide Gene Therapy Radiosensitizes Tumor Cells 437 E. Enhancement of the Cytotoxic Effects of Ionizing Radiation by Antiangiogenic Gene Therapy 439

III. Enhancing the Replicative Potential of Antitumor Viruses with Ionizing Radiation 440 IV. Transcriptional Targeting of Gene Therapy with Ionizing Radiation (Genetic Radiotherapy) 441 A. Introduction 441 B. Induction of the Immediate Early Gene Egr-1 Following Exposure to Ionizing Radiation 441 C. Tumor Necrosis Factor-α: A Toxin for Radiation-Inducible Gene Therapy 442 D. Ad.Egr–TNF-α: Gene Therapy Spatially and Temporally Controlled by Ionizing Radiation 442

V. Summary and Future Directions

443

References 444

I. INTRODUCTION The utility of gene therapy in the treatment of cancer results from its ability to deliver therapeutic genes to tumor cells in order to alter the malignant phenotype or to induce tumor


435


436

David H. Gorski, Helena J. Mauceri and Ralph R. Weichselbaum

II. STRATEGIES USING GENE THERAPY TO INCREASE THE EFFICACY OF RADIATION THERAPY

DNA damage + +

A. Introduction p53

Generally, viruses are the method of choice in most gene therapy applications because they are capable of delivering genes of interest to the largest number of cells most efficiently and drive production of the desired gene product. In most of these strategies, gene transfer to a target tissue or tumor is accomplished using replication-incompetent viral shuttle vectors such as retroviruses or replication-deficient adenoviruses. Genes inserted into these vectors can include immunomodulatory cytokines (e.g., interleukins such as IL-6, IL-12, IL-13) that recruit immune cells to the site of the tumor [1–5]; tumor suppressor genes that slow tumor growth, reverse the activity of mutant tumor suppressor genes, or induce tumor cell apoptosis [6–8]; or prodrug converting enzymes (e.g., herpes simplex virus thymidine kinase, cytosine deaminase) that result in the metabolism of intravenously administered nontoxic prodrugs to cytotoxic compounds within the target tissue [9–11]. More recently, genes encoding antiangiogenic peptides have been inserted into viral vectors and used to block tumor-associated angiogenesis in experimental models [12–16]. Gene therapy strategies in which such genes are expressed in tumor cells and that show promise of increasing the effectiveness of radiation therapy through either radiosensitization or synergistic antitumor effects are discussed below.

B. p53 Gene-Transfer-Mediated Radiosensitization Among the tumor suppressors being considered for gene replacement therapy in cancer, the p53 protein is especially attractive, because it plays a critical role in regulating the cellular response to DNA damage [17]. When DNA is damaged by an agent such as IR or cytotoxic chemotherapy, p53, a transcription factor capable of activating multiple downstream genes [17,18], mediates several processes critical to preventing the propagation of the DNA damage when the cell divides: (1) G1 cell-cycle arrest through the induction of the cyclindependent kinase p21, which allows the cell time to repair its DNA before entering the cell cycle [17]; (2) DNA repair and synthesis through the activation of the growth arrest and DNA damage-dependent (GADD) genes and proliferating cell nuclear antigen [19,20]; and (3) inducing apoptosis in cells whose DNA is too damaged to be successfully repaired [21] (Fig. 1). Mutations in p53 are the most common genetic alterations in tumors, with p53 mutations or deletions being present in over 50% of all human cancers. Loss of or mutations in p53 significantly contribute to tumor development, progression, and chemotherapy resistance [22,23]. Indeed,

Cell cycle arrest

Apoptosis (transcription-independent)

+ +

−

p21

Bax

Apoptosis (transcription-dependent)

GADD45, PCNA

DNA repair

MDM2

FIGURE 1 p53-mediated responses to DNA damage. p53 activates multiple processes in response to various stimuli, especially DNA damage. It is also downregulated by mouse double minute-2 (MDM2), which accelerates its degradation.

a survey of the toxicity of hundreds of anticancer drugs indicates that the vast majority of clinically useful drugs are most effective in cells that express wild-type p53 [24]. Moreover, inhibition of p53 action is associated with radioresistance [25]. Thus, the rationale for transducing tumor cells with wild-type p53 depends on the observation that apoptosis induced by chemotherapy and IR is at least partially dependent on the expression of wild-type p53 and that mutated p53 is associated with radioresistance [26,27]. In theory, at least, expressing wild-type p53 in tumor cells that either lack it or express mutant p53 represents a rational strategy for overcoming tumor cell resistance to IR. Ionizing radiation activates p53 through posttranslational modifications, including the phosphorylation of N-terminal serine residues in the transactivating domain of p53 and C-terminal modifications such as lysine acetylation that increase transcriptional activation of downstream mediators by p53. It is not clear whether p53 status correlates with radiosensitivity. For instance, it has been shown that thymocytes from mice transgenic for a p53 null mutation were radioresistant, whereas thymocytes from wild-type mice were radiosensitive [28]. The clear implication was that radiationinduced apoptosis depends on the presence of functional p53. Other studies link radiosensitivity to p53 status in normal and transformed fibroblasts [29] and cells of hematopoietic origin [30]. For instance, transfection of the human papillomavirus 16 (HPV-16) E6 gene, which binds and inactivates p53, into human diploid fibroblasts rendered these cells more resistant to irradiation, presumably due to loss of p53 [29]. In contrast, other studies failed to find a conclusive correlation between p53 function and radiosensitivity in tumor cell lines [31–34]. Several studies have addressed the question of whether p53 gene therapy can improve the efficacy of IR in experimental tumor models. Most reports have focused on enhancing the radiosensitivity of tumor cells, although the results appear to depend upon the specific cell type. p53 has been

Strategies for Combining Gene Therapy with Ionizing Radiation to Improve Antitumor Efficacy

transduced into a homozygous mutant p53 ovarian tumor cell line. Tumor xenograft data showed that 45% of mice treated with IR and a replication-deficient adenovirus encoding p53 (Ad-p53) had long-term cures, compared with the mice treated with IR or Ad-p53 alone [35]. In p53-/- human colorectal tumor cell lines, delivery of Ad-p53 did not result in a significant increase in apoptosis in response to IR, but Ad-p53-infected cells underwent significantly more apoptosis than control cells. Tumor cell apoptosis was further enhanced by IR [36,37]. In the same study, tumor xenografts injected with Ad-p53 and then treated with a single 5-Gy dose of radiation demonstrated a significant increase in apoptotic cells and tumor growth delay [36]. Similar evidence supporting the potential benefit of combining IR and p53 gene therapy comes from studies utilizing head and neck squamous cell xenografts [38] and glioma cell lines and xenografts [39]. The presence of p53 is not always sufficient or necessary for radiosensitization [32]. Some human colorectal adenoma and carcinoma cell lines lacking wild-type p53 still undergo apoptosis in response to IR [31]. In addition, introduction of the HPV-16 E6 gene, which binds and inactivates p53, into several tumor cell lines did not increase their radioresistance [33], in contrast to the results seen with human diploid fibroblasts, where E6 was observed to increase radioresistance [29]. Moreover, replacement of wild-type p53 is not always sufficient to reverse the cellular defects caused by the presence of a mutant p53 because certain mutated forms of p53 appear to act in a dominant fashion [40]. The role of p53 in modulating radiosensitivity and radioresistance is likely a cell-type-specific phenomenon, making it necessary to establish whether exogenous p53 alters radiosensitivity for each tumor type before using p53 gene therapy as a radiosensitizer.

C. p21 Gene Therapy and Ionizing Radiation The cyclin-dependent kinase inhibitor p21 is an immediate downstream target of p53 and is responsible for the p53-dependent checkpoint that results in G1 arrest after DNA damage [41–43]. As such, it has been examined as a potential gene therapy target in conjunction with IR. In contrast to p53, p21 causes much less apoptosis following introduction into many cell lines, including head and neck cancer [6], lung cancer [44], prostate cancer [45], gliomas [46], and melanomas [47]. Although p21 is not as strongly proapoptotic as p53, its overexpression has been shown to promote radiosensitivity in a glioma tumor model [48]. Similarly, p21 overexpression has also been shown to promote chemosensitivity in tumor cells [49]. One hypothesis to explain these observations is that, because p21 is important in causing cell-cycle arrest in response to DNA damage, loss of p21 may cause a deficiency in repair leading to chemosensitivity or radiosensitivity [50]. However, this has not been a universal finding, and more

437

recent studies have contradicted these results and postulated a protective role against radiation damage. For instance, in a colon cancer cell line (HCT116), it was found that a lack of p21 expression produced an increase in apoptosis in vitro but no decrease in clonogenic survival and no radiosensitization. However, in HCT116 xenografts, loss of p21 led to an increased sensitivity to killing by IR that was independent of induction of cell-cycle arrest and apoptosis. More interestingly, this effect is specific to cells growing as a tumor and is not observed in vitro, implying that the tumor microenvironment likely influences whether p21 affects the radiosensitivity of a tumor cell [51]. Consistent with this observation is a recent study reporting that p21 antisense therapy sensitizes a colon carcinoma cell line to IR [52]; therefore, it is possible that the role of p21 (radiation sensitizer or protector) may depend upon cell type or the specific genetic derangements present in each tumor type that effect downstream or upstream effectors of p21 leading to cell cycle arrest and/or apoptosis.

D. Prodrug Converting Enzyme Suicide Gene Therapy Radiosensitizes Tumor Cells 1. Introduction The treatment of cancer is different from treatment of genetic diseases because effective antitumor therapy requires the complete eradication of all tumor cells. One obstacle that gene therapy must overcome, therefore, is the requirement that the therapeutic gene be introduced into every tumor cell. One strategy to overcome this problem is to use prodrug converting enzymes, which rely on the transfer of nonmammalian genes encoding enzymes to convert nontoxic, systemically administered prodrugs to toxic antimetabolites [53,54]. These strategies aim to increase intratumoral concentration of the toxic metabolite in order to kill tumor cells. When used in conjunction with IR, the enzymes used are selected in order to generate drugs that are radiosensitizers. The two most commonly used prodrug converting enzyme/prodrug strategies that have been used in conjunction with IR include herpes simplex virus thymidine kinase (HSV-TK)/ganciclovir (GCV) and cytosine deaminase (CD)/ 5-fluorocytosine (5-FC). 2. HSV-TK Radiosensitization HSV-TK phosphorylates the nucleoside analogs (E)-5(2-bromovinyl)-2 -deoxyuridine (BvdUrd), acyclovir, and GCV to toxic antimetabolites. This reaction is the basis for the effectiveness of acyclovir or GCV in the treatment of HSV infections. Monophosphorylated forms of acyclovir and GCV are then phosphorylated to nucleotide triphosphates by cellular kinases. These aberrant nucleotide triphosphates disrupt DNA replication at the level of DNA chain elongation by interfering with DNA polymerase α. The

438


effects of phosphorylated BvdUrd are caused by its inhibition of thymidylate synthase, which results in a depletion of thymidine pools within the cell [55]. Based on this knowledge, studies have been performed to demonstrate the efficacy of transferring HSV-TK to tumor cells with subsequent systemic administration of GCV. Only cells transfected with the HSV-TK gene convert GCV to its toxic phosphorylated form, resulting in tumor cell death. Several studies have demonstrated the efficacy of such an antitumor approach [10,56–58]. Radiosensitization combining antiviral nucleoside analogs with IR interferes with potential lethal damage repair or the modification of DNA to a more radiosensitive form [59]. Studies with 5-bromodeoxyuridine (BrdU) and acyclovir combined with IR have demonstrated a radiosensitizing effect in experimental systems, but with doses of IR or acyclovir too high to be clinically applicable [60,61]. Several studies have shown that HSV-TK/prodrug treatment of tumor cells results in radiosensitization and enhanced tumor regression in experimental tumor xenografts. These effects have been demonstrated in glioma cells transduced with HSV-TK followed by acyclovir administration and IR. Cells transfected retrovirally with HSV-TK followed by administration of either BvdUrd or acyclovir and IR resulted in a radiosensitizing effect, with a sensitizing enhancement ration of 1.3–1.6 [62,63]. BvdUrd and acyclovir are hypothesized to radiosensitize cells by different mechanisms. BvdUrd radiosensitizes HSV-TK glial cells only when it is administered prior to IR administration. This is necessary because the phosphorylated form of BvdUrd inactivates thymidylate synthase, resulting in depleted intracellular thymidine pools within the cell. In contrast, acyclovir radiosensitizes cells if administered before or after IR. A potential mechanism to account for this observation is that acyclovir may radiosensitize because its metabolite is incorporated in to DNA before IR, thus making the DNA more susceptible to IR. Alternatively, its metabolite may inhibit repair of DNA damage, thus increasing the toxicity of IR to the cell. The efficacy of this strategy has been demonstrated in animal tumor models. In one study, glial cells were infected with HSV-TK in cell culture and then implanted in rats. Systemic administration of prodrug and a single 20-Gy dose of IR resulted in a threefold increase in rat survival compared with IR alone [64]. 3. Cytosine Deaminase Radiosensitization A second strategy for using a prodrug converting enzyme for radiosensitization involves cytosine deaminase (CD) [65], an enzyme found in many bacteria and fungi that catalyzes the deamination of cytosine to uracil, providing uracil for the organism in time of nutritional stress [65]. It is employed to convert the nontoxic drug 5-FC to the antitumor drug 5-fluorouracil (5-FU) [9, 66–69]. 5-FU and its

metabolites kill tumor cells by interfering with both DNA and RNA metabolism through incorporation in nucleic acids and through their inhibition of thymidylate synthase and depletion of the cellular TTP pool [70,71]. 5-FU has activity against some solid tumors and is a mainstay of adjuvant therapy of colorectal cancer. It is also used as a radiosensitizer to treat a variety of human tumors [72,73]. The mechanism of radiosensitization by CD/5-FU appears to be inhibition of DNA repair due to inhibition of thymidylate synthase by the monophosphorylated form of 5-FU, as its radiosensitizing effects can be abrogated by exogenously administered thymidine [70,71,74]. Also, triphosphorylated 5-FU is incorporated into RNA, disrupting protein translation [70,71]. Because systemically administered 5-FU has dose-limiting toxicities of mucositis, diarrhea, and myelosuppression, attempts have been made to use gene therapy with CD/5-FC to produce high intratumoral concentrations of 5-FU, thus providing the benefit of its radiosensitization effect in the tumor bed but sparing patients the systemic toxicities associated with 5-FU administration. Also, 5-FU is diffusible, which would allow CD-transduced tumor cells to convert 5-FC to 5-FU, which could then diffuse into surrounding untransduced tumor cells. Several studies have combined CD/5-FC therapy with IR to enhance tumor cell killing. For example, in cell culture, the transduction of human colorectal tumor cells with a retrovirus encoding CD followed by 5-FC and irradiation produced markedly increased tumor cell killing, although the 5-FC had to be administered at least 24 hours prior to IR [67]. Pederson et al. [75,76] treated cholangiocarcinoma cells with CD delivered by a replication-deficient adenovirus and IR and demonstrated specific radiosensitization with CD/5-FC. Next, in xenograft models of colon cancer and cholangiocarcinoma, adenovirusdelivered CD plus 5-FC resulted in improved tumor growth delay when these xenografts were treated with IR [77]. Hanna et al. [78] also demonstrated similar results treating human squamous cell carcinoma xenografts grown in athymic nude mice and treated with intratumoral injections of a replicationdefective adenovirus expressing CD. Xenografts treated with Ad.CD/5-FC and IR showed a significant tumor growth delay compared with IR or Ad.CD/5-FC alone. Because tumors consist of a heterogeneous population of cells, it is likely that certain subclones of cells will be more resistant to either HSV-TK/GCV or CD/5-FC. Rogulski et al. [79] have constructed a bifunctional fusion gene expressing both HSV-TK and CD. The CD gene was fused to HSV-TK through a polyglycine linker to allow for proper folding of both prodrug converting enzymes, producing the CDglyHSV–TK construct. Transducing gliosarcoma cells with a retrovirus expressing this fusion protein, then treating with 5-FC and BvdU followed by IR resulted in two- to threefold greater cell killing than would be expected if the two prodrugs interacted with IR in an additive fashion. These results were confirmed in vivo in human tumor xenograft


models [80,81]. Because it will be necessary to transfer antitumor genes to a large percentage of tumor cells to achieve clinically relevant antitumor effects in humans, Freytag et al. [82] placed the CDglyHSV-TK construct into a replication-conditional adenovirus, ONYX-015. Because preliminary data show that only 20–30% of tumor cells within a xenograft show infection by ONYX-015, further work must validate that this viral construct combined with both 5-FC/GCV and IR results in synergistic control of tumor xenografts due to virus replication and transduction of a large percentage of the tumor [82].

E. Enhancement of the Cytotoxic Effects of Ionizing Radiation by Antiangiogenic Gene Therapy 1. Angiogenesis and Antiangiogenesis Tumors are critically dependent upon inducing the ingrowth of blood vessels from the host to supply their oxygen and nutrient needs [83]. To this end they secrete proangiogenic factors such as basic fibroblast growth factor (bFGF) [84] and vascular endothelial growth factor (VEGF) [85]. Consequently, inhibition of angiogenesis, either by blocking these proangiogenic factors or treatment with antiangiogenic factors has emerged as a promising strategy to treat primary and metastatic tumors [86–89]. Strategies to block the activity of proangiogenic factors include the administration of neutralizing antibodies to proangiogenic cytokines, such as VEGF [86,90–93] or bFGF [94–96]; antisense against VEGF [97,98] or bFGF [99]; and the engineering and expression of soluble receptors that bind to VEGF and inactivate it [92,100]. It has also become apparent that tumors can induce the production of antiangiogenic peptides that directly inhibit vascular endothelial cell proliferation and angiogenesis. The most potent and specific of these are angiostatin, a proteolytic fragment of plasminogen containing its first four kringle domains [89,101,102], and endostatin, a proteolytic fragment of collagen XVIII [88,103]. Finally, a number of smaller molecules are under study that act with varying degrees of specificity on endothelial cells to block angiogenesis. Among these are drugs such as VEGF receptor tyrosine kinase inhibitors [104,105] and TNP-1470 [106,107]. All of these strategies target tumor endothelium and disrupt angiogenesis. 2. Antiangiogenic Therapy Potentiates the Antitumor Activity of Ionizing Radiation Antiangiogenic proteins, although effective at shrinking tumors, are not tumoricidal. Tumor regrowth frequently occurs once treatment with the angiogenesis inhibitor is terminated [87,108], although there is evidence that antiangiogenic therapy can be used to induce tumor dormancy

439

[87,108]. IR is a major cytotoxic therapeutic modality that is primarily effective in the treatment of relatively small tumors while large tumors respond only with considerable toxicity to normal tissues. One strategy to overcome these therapeutic limitations is to combine angiogenesis inhibitors with cytotoxic therapies. Such an approach has been tried and has shown promise thus far. In our laboratory, we have demonstrated that combining IR with angiostatin derived from the proteolytic digestion of human plasminogen produces a greater than additive antitumor effect (Fig. 2A) [109,110]. Moreover, this effect requires that angiostatin be present in the circulation at the time IR is administered [109] and involves sensitization of the tumor endothelial cells to the cytotoxic effects of IR [110]. Similarly, in several mouse tumor models, we have also observed that some tumors secrete increased levels of VEGF in response to IR and that blocking that response by pretreatment of the mouse with a neutralizing antibody to VEGF results in greatly increased antitumor efficacy of IR treatment (Fig. 2B) [111]. These observations suggest that combining antiangiogenic peptides with IR or other cytotoxic therapies may well represent the most promising potential use of these potent new compounds. 3. Ionizing Radiation: A Means of Targeting Antiangiogenic Gene Therapy Unfortunately, large antiangiogenic peptides, especially angiostatin and endostatin, present several practical problems to overcome for clinical use. Of these, aberrant folding of the recombinant peptides when they are synthesized in vitro represents the main difficulty encountered in making active angiostatin and endostatin. This problem has hindered the ability of pharmaceutical companies to manufacture sufficient quantities of pharmaceutical-grade material for use in humans, and it is only recently that clinical trials involving endostatin have gotten under way. In addition, because these peptides are not tumoricidal, continuous administration for long periods will be necessary if they are to be used as single agents. Consequently, there has been great interest in developing gene therapy approaches for the in situ production of antiangiogenic peptides such as angiostatin and endostatin, as well other proteins such as the soluble VEGF receptor. In several tumor models, it has been shown that delivery of the angiostatin or endostatin cDNA by various means, including viral vectors [12,13,15,16,112,113], liposome-mediated methods [13,14,112,114,115], and even injection of naked DNA into skeletal muscle [116] can result in antitumor effects and marked systemic inhibition of angiogenesis. Inhibition of tumor growth and angiogenesis has also been achieved using a variation of this strategy, in which a vector expressing one of the proteases responsible for generating angiostatin in vivo is used to inhibit tumor growth [117]. Similarly, constructs expressing the extracellular domain of the VEGF receptor [92,100] or antisense to VEGF [118] can also inhibit tumor

440


FIGURE 2 Improved antitumor effect by combining antiangiogenic therapy with ionizing radiation. The effect of combining antiangiogenic therapy with ionizing radiation was examined in different tumor models. (A.) Lewis lung carcinoma (LLC) and angiostatin: C57BL6 mice were inoculated in the hindleg with LLC cells, and tumors were allowed to grow as subcutaneous tumors to a starting volume of 1012–111 mm3 (approximately 5% of the mouse body weight) prior to the commencement of treatment. Mice were then treated with either IR (20 Gy × 2 doses on days 0 and 1) or angiostatin (25 mg/kg/d) throughout the time course of the experiment, or both. (B.) SQ20B and anti-VEGF antibody: Athymic nude mice were inoculated in the hindleg with SQ20B squamous cell carcinoma cells (derived from a radioresistant human head and neck tumor), and the cells were allowed to grow as subcutaneous xenografts to a starting volume of 372–16 mm3 prior to the commencement of treatment. Mice were then treated with either IR (10 Gy on days 0, 1, 2, and 3) or a neutralizing monoclonal anti-VEGF antibody (10 g on days 0, 1, 2, and 3), or a combination of both, with the anti-VEGF antibody administered 3 hours prior to IR. The combination of blocking VEGF activity and treating with IR produced superior tumor growth delay. Squares = untreated controls; diamonds = IR alone; triangle = angiostatin (A) or anti-VEGF antibody (B) alone; circle = combination therapy. (Graphs adapted from data in Mauceri et al.110 and Gorski et al.111 )

growth and angiogenesis. Strategies such as these form the basis for combining antiangiogenic gene therapy with other cytotoxic modalities. Given the success in demonstrating the efficacy of combining at least two different antiangiogenic strategies with IR, a logical next step would be to combine antiangiogenic gene therapy with IR. Such an approach would theoretically produce much higher intratumoral levels of antiangiogenic peptide than is possible by exogenous administration and, therefore, presumably a greater antitumor effect when combined with IR. In addition, given that some antiangiogenic gene therapy strategies can generate systemic levels of angiostatin or endostatin and inhibit angiogenesis at distant sites [12,14,114–116], such strategies may have the potential advantage of also inhibiting the growth of metastatic disease. In one study, Griscelli et al. [13] reported the use of a replication-defective adenovirus expressing the secretable angiostatin-like molecule K3 (AdK3). K3 contains the first three kringle domains of plasminogen and has antiangiogenic activity comparable to angiostatin [13,119]. C6 glioma xenografts implanted in athymic nude mice were treated with either AdK3 alone or IR alone, or a combination of the two. The combination produced a significantly higher antitumor effect that tightly correlated with a marked decrease in intra-

tumoral vascularization. Seetharam et al. [120] have reported that the addition of an adenovirus expressing IL-12, which, in addition to its effects on the immune system, is also antiangiogenic, to IR not only enhances the local antitumor effect of IR but also suppresses microscopic growth of tumors at distant sites, probably through an immune enhancement [120]. These experiments suggest that the combination of antiangiogenic gene therapy with IR shows promise as a means of increasing the efficacy of IR. Further experiments will be necessary to verify the general utility of this approach and determine the best strategies for applying it to human tumors.

III. ENHANCING THE REPLICATIVE POTENTIAL OF ANTITUMOR VIRUSES WITH IONIZING RADIATION Antitumor replication-competent viruses derive their antitumor effect from direct tumor cell lysis after completion of the viral replicative cycle. Ideally, such viruses replicate preferentially in tumor cells relative to normal tissue. One strategy to abrogate reproduction in normal but not tumor cells is to delete or mutate genes necessary for replication in


normal cells but not tumor cells. The last two decades have witnessed explosive growth in our knowledge of molecular biology. Many viruses have been characterized and completely sequenced, and the specific gene products for necessary viral functions have been identified. These include genes encoding proteins involved in cell cycle, pathogenesis, and avoidance of cellular immunity. This knowledge, coupled with the ability to genetically construct viruses to reduce their pathogenicity or target tumor cells, has led to strategies for herpes, adenoviruses, and reoviruses. Such strategies show the most promise for success for treating tumors of the central nervous system, where, in contrast to growing tumor cells, the neurons are quiescent and genes responsible for the neurovirulence of various viruses have been identified [121,122]. Herpes simplex virus-1 is a 152-kb DNA virus that encodes the γ1 134.5 gene, which has reportedly been involved in herpes neurovirulence [121,122]. γ1 134.5-deleted herpesviruses are severely attenuated in their ability to replicate in neurons and thereby cause encephalitis in murine models. Wild-type-HSV-1(F) has an LD50 of 102 PFU upon intracranial injection into mice. However, γ1 134.5-deleted virus has an LD50 of 107 PFU, thus accounting for its observed decreased neurovirulence compared with wild-type HSF-1(F) [121,123,124]. R3616 is an HSV with both copies of γ1 134.5 inactivated. The Roizman laboratory has focused on the use of R3616 with IR in the treatment of gliomas [125]. One of the problems demonstrated by attenuated herpesviruses is their inability to cause tumor xenograft regression. Instead, these herpesviruses have resulted solely in tumor growth delay. The relative lack of antitumor efficacy of attenuated HSV is largely based on the failure of attenuated viruses to replicate within the tumor. However, the combination of R3616 with IR results in significant tumor regression with greater than 60% of combined treated subcutaneous glioma xenografts regressing completely [125]. Additional studies have shown that IR results in two- to fivefold greater viral recovery from irradiated tumors xenografts than unirradiated infected xenografts. An orthotopic intracranial glioma model has confirmed the enhancement of mouse survival in gliomas treated with both R3616 and IR, as well as prolonged detection of R3616 within irradiated tumors compared with unirradiated tumor [126]. These results implicate a role for radiotherapy in enhancing attenuated viral replication within tumors. Our current hypothesis for the mechanism by which IR results in enhanced herpesvirus proliferation is that IR results in a cellular environment more conducive to HSV replication through the induction of cellular proteins that promote HSV replication. Evidence from another system supporting this general hypothesis includes the observation that transfection of cells with a construct in which the luciferase reporter gene is linked to the cytomegalovirus (CMV) promoter results in increased luciferase activity after transfected cells are irradiated [127], implying that IR is inducing proteins

441

that activate the CMV promoter. This effect is not observed in all cell types, however. For instance, Cheng et al. [128] observed no induction of CMV promoter activity after COS-7 cells transfected with plasmids in which the chloramphenicol acetyl transferase gene was linked to the CMV promoter were irradiated.

IV. TRANSCRIPTIONAL TARGETING OF GENE THERAPY WITH IONIZING RADIATION (GENETIC RADIOTHERAPY) A. Introduction The utility of gene therapy as a cancer therapy is often limited by inherent tumor resistance to the gene product, difficulty introducing the gene into a sufficient number of tumor cells to cause a therapeutic effect, or poor diffusion of gene product. One of the more daunting challenges in delivering gene therapy to tumors is spatial and temporal control of the expression and effect of exogenously delivered gene. This is important in strategies using genes whose products are toxic, where it is necessary to deliver these toxic gene products to tumor cells selectively and spare normal cells as much as possible. Such strategies require precise spatial targeting of the gene to the appropriate cells, because widespread expression or diffusion could be detrimental to the patient, such as in the case of cytotoxic proteins such as ricin and Pseudomonas endotoxin. Antitumor cytokines, such as tumor necrosis factor-α, (TFN-α), can also be toxic if they diffuse away from the tumor site and reach high systemic levels. Strategies for keeping gene expression localized to the tumor have included techniques as simple as intratumoral injection [129] and techniques as sophisticated as engineering constructs in which the gene of interest is under the control of a tissue- or tumor-specific promoter [53,130]. Temporal control is also difficult to achieve. When using plasmid or viral constructs in which expression of the therapeutic gene is driven by a strong constitutive promoter, this is usually accomplished simply by injecting the gene when expression is desired. The drawback of this approach is that the length of time the gene is expressed is highly variable, depending upon the method of gene delivery (naked plasmid, liposome-based methods, adenovirus, vaccinia virus, etc.).

B. Induction of the Immediate Early Gene Egr-1 Following Exposure to Ionizing Radiation The study of gene induction by IR is important to the understanding of how cells and organisms respond to radiation exposure. IR activates the transcription of a number of genes, implying the existence of radiation-responsive elements residing upstream of IR-induced genes. These elements could

442


be exploited in genetic constructs to activate gene expression after exposure to IR. Among the genes induced soon after cells are exposed to IR are the tissue plasminogen activator (t-PA) gene [131] and the immediate early genes such as c-jun and the early growth response-1 (Egr-1) gene [132–134]. Egr-1, also known as zif/268, NGFI-A, Krox-24, and TIS-8, encodes a nuclear phosphoprotein with a cysteine/histidine zinc finger motif that is partially homologous to the corresponding domain of the Wilm’s tumor susceptibility gene, and its expression is rapidly induced after cells are stimulated or reenter the cell cycle [135,136]. Despite the relatively large number of genes that are induced after exposure to IR, relatively few radiation-inducible promoters or enhancers have been characterized. DNA sequences that activate transcription after irradiation include AP-1 [137], the NF-κB binding sequence [132], and the CArG element [138]. The Egr-1 promoter has been examined as an inducible promoter for gene therapy because it is inducible by radiation in several types of human tumor cells [133,138]. Datta et al. [138] have studied deletion mutants of the 5 promoter region of the Egr-1 promoter to identify elements responsible for radiation inducibility. The radiation response element was identified as the CArG box [CC(A + T rich)GG], a DNA sequence motif originally identified in the serum response element and found in the promoters of several immediate early genes, as well as muscle-specific promoters [139–141]. By linking CArG boxes together and placing them upstream of a CAT reporter gene, Datta et al. demonstrated a three- to fourfold increase in the expression of the CAT reporter gene after irradiation, and promoter deletion analysis revealed that the first three 5 CArG boxes were the most important for the induction of Egr-1 promoter activation [142]. Not surprisingly, the mechanism of CArG box activation depends upon the generation of free radical intermediates by IR [142]. Overall, this work defined a radiation response element (RRE) that could be placed upstream of a cDNA encoding a therapeutic protein and used to turn on expression of this gene in irradiated cells.

C. Tumor Necrosis Factor-α: A Toxin for Radiation-Inducible Gene Therapy Tumor necrosis factor-α is a polypeptide cytokine that activates a wide variety of biological responses, predominately in the immune system [143]. TNF was first identified based on its ability to induce hemorrhagic necrosis in murine tumors and damage to tumor vasculature [144]. The 55-kDa TNF receptor initiates a signaling cascade that results in the apoptosis of some tumor cells [145,146]. Clinical trials have demonstrated that the levels of TNF-α protein achieved in animal studies could not be achieved in human subjects due to systemic toxicity, including hypotension and respiratory insufficiency. A phase I trial using TNF and concomitant

radiotherapy was done to determine the maximal tolerated dose of TNF that could be used to enhance radiation effects on tumors and to establish patterns of both in-field and systemic toxicity [147]. Radiotherapy combined with human recombinant TNF-α were administered for 5 days for each consecutive week until completion of the planned course of radiotherapy. When locally advanced primary tumors were treated to doses ≥60 Gy (given as 1.80–2.25 Gy/d), minimal in-field toxicity was observed; however, acute systemic toxicity, including rigors, fever, and nausea, was observed in nearly all patients. Response to treatment was evaluated in 20 of the 31 patients. Complete regression was observed in four patients. It was proposed that tumor localization of TNF-α using gene therapy combined with radiotherapy might eliminate the observed systemic toxicity and enhance the antitumor effects of IR through the production of high local levels of intratumoral TNF-α [148].

D. Ad.Egr–TNF-α: Gene Therapy Spatially and Temporally Controlled by Ionizing Radiation Viral-mediated transfer of cytotoxic genes whose expression is controlled by RRE allows for spatial and temporal control of gene expression using IR as the means of throwing the molecular “switch” (Fig. 3). Hallahan et al. [149] synthesized a genetic construct in which the Egr-1 promoter including the CArG elements responsible for radiation inducibility was placed upstream from a cDNA encoding TNF-α

Radiation-inducible promoter Egr-1 promoter CArG CArG CArG CArG

TNF-α Therapeutic gene

+

+ +

Oxygen free radicals

+

O− O− O− − O O−

Tumor cell apoptosis Ionizing radiation Destruction of tumor vasculature

FIGURE 3 Genetic radiotherapy. The Egr-1 promoter is placed upstream of a cDNA encoding TNF-α. Ionizing radiation activates the CArG elements in the Egr-1 promoter, driving enhanced expression of the TNF-α gene in the tumor bed. The combination of the enhanced TNF-α expression and ionizing radiation results in increased tumor cell apoptosis and vascular destruction, as well as improved antitumor activity.


(Egr–TNF) [149]. Replication-deficient adenovirus type 5 (Ad-5) virus was then employed to deliver the Egr–TNF construct to radioresistant head and neck squamous cell carcinoma cells growing as xenografts in athymic nude mice in order to study the interaction of radiation-targeted TNF gene therapy in tumor cell lines and mouse xenograft models. For the human tumor xenograft SQ20B derived from an oropharyngeal squamous cell carcinoma, the treatment protocol consisted of Ad.Egr–TNF (four injections) and IR given as 5-Gy fractions to a total dose of 50 Gy. Xenografts treated with both Ad.Egr–TNF and IR demonstrated significantly greater tumor shrinkage and growth delay than xenografts treated with either modality alone. TNF expression in irradiated tumor xenografts was elevated threefold at 7 days after initiation of treatment and by eightfold by day 21. Staba et al. [150] have demonstrated similar efficacy of combined therapy in human glioma xenografts. D54 glioma cells are resistant to the cytotoxic effects of TNF-α, and no enhancement of radiation killing was observed following treatment with TNF-α and IR in vitro. However, when nude mice bearing D54 xenografts received intratumoral injections of Ad.Egr–TNF or null adenovirus (Ad-null) with and without fractionated IR (5 Gy/d, total 30 Gy), combined treatment produced complete tumor regression in 71% of xenografts, as opposed to the 7% observed in those treated with IR alone and 0% in those treated with Ad.Egr–TNF alone. Combined treatment also resulted in a significantly longer growth delay and produced marked tumor vessel thrombosis, an effect not seen with either therapy alone, suggesting that Ad.Egr–TNF and IR target the tumor vasculature [149–152]. Results similar to those observed in SQ20B and D54 xenografts have also been noted in xenografts of the prostate tumor cell line PC3 [153]. This strategy has also been applied to another therapeutic gene, HSV-TK. Joki et al. [154] combined transcriptional regulation with converting enzyme/prodrug strategies to further regulate the interaction with IR. When the Egr-1 promoter was linked to the HSV-TK gene, not only has irradiation of transfected cells resulted in enhanced HSV-TK expression driven by the Egr-1 promoter, but elevated HSV-TK also has allowed for more complete activation of GCV. Phosphorylated GCV then acts as a radiosensitizer upon subsequent IR administration. The results of these experiments show that GCV is phosphorylated to radiosensitizing levels in transfected irradiated cells, but that the basal transcription rate of the Egr-1–HSV-TK construct without IR was insufficient to phosphorylate GCV to its toxic antimetabolite. Therefore, HSV-TK expression can be regulated both temporally and spatially by IR. Taken together, all these experiments demonstrate the feasibility of using radiation-inducible promoters linked to therapeutic genes or prodrug converting enzymes to control gene expression both temporally and spatially and to enhance tumor response to IR. They also show some of the many potential strategies for exploiting such promoters for therapy.

443

Efforts are presently under way to improve upon this system by developing synthetic promoters whose activity is more tightly regulated by radiation exposure than the Egr-1 promoter. Marples et al. [155], for instance, have reported that a synthetic promoter made of multiple CArG elements is at least as effectively induced by low doses of IR as the Egr-1 promoter. More recently, in another variation on this approach, Scott et al. [156] have developed a promoter that combines the CArG elements from the Egr-1 promoter and the cre-Lox-P site-specific recombination system of the P1 bacteriophage. In this system, a single, minimally toxic dose of radiation induces cre-mediated excision of a Lox-P flanked stop cassette in a silenced expression vector, resulting in amplified levels of CMV promoter-driven expression of HSV-TK [156]. Experiments such as these demonstrate the feasibility of making promoters whose activity is very tightly regulated by exposure of the cell to IR. Once such promoters are developed, genetic radiotherapy with more tumoricidal or toxic genes will become possible.

V. SUMMARY AND FUTURE DIRECTIONS Combining radiation and gene therapy has multiple advantages. Both gene therapy and radiation therapy are used in the treatment of local disease and kill tumor cells by independent mechanisms, thus minimizing the likelihood of the tumor developing treatment-resistant clones during treatment. Moreover, in some cases, the gene therapy can impact systemic disease as well as local disease, as is the case when antiangiogenic gene therapy is combined with radiation. In theory, the locally administered antiangiogenic peptide will have a greater than additive local antitumor effect [109,110] and suppress distant metastases while the peptide is being expressed [12,14,114–116]. Moreover, spatial and temporal control can be achieved through conforming radiotherapy to the virally inoculated tumor bed expressing the therapeutic gene. Viruses delivering radiosensitizing agents or antiangiogenic peptides may allow for higher intratumoral concentrations of these drugs than is possible by systemic administration, thus theoretically enhancing the interaction between these drugs in the tumor itself and minimizing systemic toxicity due to drug. With radiosensitization, enhanced local tumor control may be achieved in radioresistant tumors , and radiosensitive tumors may be controlled with lower doses of radiation, thereby minimizing radiation-induced damage to surrounding normal tissue as much as possible. As an adjunct to this, it is possible to imagine the use of additional gene therapy using antioxidant proteins such as manganese superoxide dismutase to protect surrounding normal tissue further [157,158], thus increasing the therapeutic ratio of radiation even further. Finally, viral replication enhancement by IR can be confined to the tumor by conformal radiotherapy, allowing

444


for high titers of virus in the tumor. In light of the rapid advances in the development of these approaches to combining gene therapy with radiation in experimental models, it will be of great interest to begin to move these approaches into clinical use by developing clinical trials to test their efficacy.

References 1. Bramson, J. L., Hitt, M., Addison, C. L., Muller, W. J., Gauldie, J., and Graham, F. L. (1996). Direct intratumoral injection of an adenovirus expressing interleukin-12 induces regression and long-lasting immunity that is associated with highly localized expression of interleukin-12. Hum. Gene Ther. 7, 1995–2002. 2. Chen, L., Chen, D., Block, E., O’Donnell, M., Kufe, D. W., and Clinton, S. K. (1997). Eradication of murine bladder carcinoma by intratumor injection of a bicistronic adenoviral vector carrying cDNAs for the IL-12 heterodimer and its inhibition by the IL-12 p40 subunit homodimer. J. Immunol. 159, 351–359. 3. Drozdzik, M., Qian, C., Xie, X., Peng, D., Bilbao, R., Mazzolini, G., and Prieto, J. (2000). Combined gene therapy with suicide gene and interleukin-12 is more efficient than therapy with one gene alone in a murine model of hepatocellular carcinoma. J. Hepatol. 32, 279– 286. 4. Kasaoka, Y., Nakamoto, T., Wang, J., Usui, T., and Hamada, H. (2000). Gene therapy for murine renal cell carcinoma using genetically engineered tumor cells to secrete interleukin-12. Hiroshima J. Med. Sci. 49, 29–35. 5. Lotze, M. T., Shurin, M., Esche, C., Tahara, H., Storkus, W., Kirkwood, J. M., Whiteside, T. L., Elder, E. M., Okada, H., and Robbins, P. (2000). Interleukin-2: developing additional cytokine gene therapies using fibroblasts or dendritic cells to enhance tumor immunity. Cancer J. Sci. Am. 6, S61–66. 6. Clayman, G. L., Liu, T. J., Overholt, S. M., Mobley, S. R., Wang, M., Janot, F., and Goepfert, H. (1996). Gene therapy for head and neck cancer. Comparing the tumor suppressor gene p53 and a cell cycle regulator WAF1/CIP1 (p21). Arch. Otolaryngol. Head Neck Surg. 122, 489–493. 7. Roth, J. A., Swisher, S. G., and Meyn, R. E. (1999). p53 tumor suppressor gene therapy for cancer. Oncology (Huntington) 13, 148–154. 8. Takeda, S., Nakao, A., Miyoshi, K., and Takagi, H. (1998). Gene therapy for pancreatic cancer. Semin. Surg. Oncol. 15, 57–61. 9. Mullen, C. A., Kilstrup, M., and Blaese, R. M. (1992). Transfer of the bacterial gene for cytosine deaminase to mammalian cells confers lethal sensitivity to 5-fluorocytosine: a negative selection system. Proc. Natl. Acad. Sci. USA 89, 33–37. 10. Oldfield, E. H., Ram, Z., Culver, K. W., Blaese, R. M., DeVroom, H. L., and Anderson, W. F. (1993). Gene therapy for the treatment of brain tumors using intra-tumoral transduction with the thymidine kinase gene and intravenous ganciclovir. Hum. Gene Ther. 4, 39–69. 11. Singhal, S., and Kaiser, L. R. (1998). Cancer chemotherapy using suicide genes. Surg. Oncol. Clin. N. Am. 7, 505–536. 12. Feldman, A. L., Restifo, N. P., Alexander, H. R., Bartlett, D. L., Hwu, P., Seth, P., and Libutti, S. K. (2000). Antiangiogenic gene therapy of cancer utilizing a recombinant adenovirus to elevate systemic endostatin levels in mice. Cancer Res. 60, 1503–1506. 13. Griscelli, F., Li, H., Cheong, C., Opolon, P., Bennaceur-Griscelli, A., Vassal, G., Soria, J., Soria, C., Lu, H., Perricaudet, M., and Yeh, P. (2000). Combined effects of radiotherapy and angiostatin gene therapy in glioma tumor model. Proc. Natl. Acad. Sci. USA 97, 6698– 6703. 14. Liu, Y., Thor, A., Shtivelman, E., Cao, Y., Tu, G., Heath, T. D., and Debs, R. J. (1999). Systemic gene delivery expands the repertoire of effective antiangiogenic agents. J. Biol. Chem. 274, 13338–13344.

15. Nguyen, J. T., Wu, P., Clouse, M. E., Hlatky, L., and Terwilliger, E. F. (1998). Adeno-associated virus-mediated delivery of antiangiogenic factors as an antitumor strategy. Cancer Res. 58, 5673–5677. 16. Nguyen, J. T. (2000). Adeno-associated virus and other potential vectors for angiostatin and endostatin gene therapy. Adv. Exp. Med. Biol. 465, 457–466. 17. Agarwal, M. L., Taylor, W. R., Chernov, M. V., Chernova, O. B., and Stark, G. R. (1998). The p53 network. J. Biol. Chem. 273, 1–4. 18. Wang, Y., Schwedes, J. F., Parks, D., Mann, K., and Tegtmeyer, P. (1995). Interaction of p53 with its consensus DNA-binding site. Mol. Cell. Biol. 15, 2157–2165. 19. Fornace, A. J., Jackman, J., Hollander, M. C., Hoffman-Liebermann, B., and Liebermann, D. A. (1992). Genotoxic-stress-response genes and growth arrest genes: gadd, MyD, and other genes induced by treatments eliciting growth arrest. Ann. N.Y. Acad. Sci. 663, 139–153. 20. Kastan, M. B., Zhan, Q., El-Deiry, W. S., Carrier, F., Jacks, T., Walsh, W. V., Plunkett, B. S., Vogelstein, B., and Fornace, A. J. (1992). A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. Cell 71, 587–597. 21. Neubauer, A., Thiede, C., Huhn, D., and Wittig, B. (1996). p53 and induction of apoptosis as a target for anticancer therapy. Leukemia 10, S2–S4. 22. Greenblatt, M. S., Bennett, W. P., Hollstein, M., and Harris, C. C. (1994). Mutations in the p53 tumor suppressor gene: clues to cancer etiology and molecular pathogenesis. Cancer Res. 54, 4855–4878. 23. Velculescu, V. E., and El-Deiry, W. S. (1996). Biological and clinical importance of the p53 tumor suppressor gene. Clin. Chem. 42, 858– 868. 24. Weinstein, J. N., Myers, T. G., O’Connor, P. M., Friend, S. H., Fornace, Jr., A. J., Kohn, K. W., Fojo, T., Bates, S. E., Rubinstein, L. V., Anderson, N. L., Buolamwini, J. K., van Osdol, W. W., Monks, A. P., Scudiero, D. A., Sausville, E. A., Zaharevitz, D. W., Bunow, B., Viswanadhan, V. N., Johnson, G. S., Wittes, R. E., and Paull, K. D. (1997). An information-intensive approach to the molecular pharmacology of cancer. Science 275, 343–349. 25. Komarov, P. G., Komarova, E. A., Kondratov, R. V., ChristovTselkov, K., Coon, J. S., Chernov, M. V., and Gudkov, A. V. (1999). A chemical inhibitor of p53 that protects mice from the side effects of cancer therapy. Science 285, 1733–1737. 26. Fan, S., el-Deiry, W. S., Bae, I., Freeman, J., Jondle, D., Bhatia, K., Fornace, Jr., A. J., Magrath, I., Kohn, K. W., and O’Connor, P. M. (1994). p53 gene mutations are associated with decreased sensitivity of human lymphoma cells to DNA damaging agents. Cancer Res. 54, 5824–5830. 27. Lowe, S. W., Ruley, H. E., Jacks, T., and Housman, D. E. (1993). p53dependent apoptosis modulates the cytotoxicity of anticancer agents. Cell 74, 957–967. 28. Lowe, S. W., Schmitt, E. M., Smith, S. W., Osborne, B. A., and Jacks, T. (1993). p53 is required for radiation-induced apoptosis in mouse thymocytes. Nature 362, 847–849. 29. Tsang, N. M., Nagasawa, H., Li, C., and Little, J. B. (1995). Abrogation of p53 function by transfection of HPV16 E6 gene enhances the resistance of human diploid fibroblasts to ionizing radiation. Oncogene 10, 2403–2408. 30. Su, L. N., and Little, J. B. (1992). Transformation and radiosensitivity of human diploid skin fibroblasts transfected with SV40 T-antigen mutants defective in RB and p53 binding domains. Int. J. Radiat. Biol. 62, 461–468. 31. Bracey, T. S., Miller, J. C., Preece, A., and Paraskeva, C. (1995). Gamma-radiation-induced apoptosis in human colorectal adenoma and carcinoma cell lines can occur in the absence of wild type p53. Oncogene 10, 2391–2396. 32. Bristow, R. G., Benchimol, S., and Hill, R. P. (1996). The p53 gene as a modifier of intrinsic radiosensitivity: implications for radiotherapy. Radiother. Oncol. 40, 197–223.

Strategies for Combining Gene Therapy with Ionizing Radiation to Improve Antitumor Efficacy 33. Huang, H., Li, C. Y., and Little, J. B. (1996). Abrogation of p53 function by transfection of HPV16 E6 gene does not enhance resistance of human tumour cells to ionizing radiation. Int. J. Radiat. Biol. 70, 151–160. 34. Kohli, M., and Jorgensen, T. J. (1999). The influence of SV40 immortalization of human fibroblasts on p53-dependent radiation responses. Biochem. Biophys. Res. Commun. 257, 168–176. 35. Gallardo, D., Drazan, K. E., and McBride, W. H. (1996). Adenovirusbased transfer of wild-type p53 gene increases ovarian tumor radiosensitivity. Cancer Res. 56, 4891–4893. 36. Spitz, F. R., Nguyen, D., Skibber, J. M., Meyn, R. E., Cristiano, R. J., and Roth, J. A. (1996). Adenoviral-mediated wild-type p53 gene expression sensitizes colorectal cancer cells to ionizing radiation. Clin. Cancer Res. 2, 1665–1671. 37. Spitz, F. R., Nguyen, D., Skibber, J. M., Cusack, J., Roth, J. A., and Cristiano, R. J. (1996). In vivo adenovirus-mediated p53 tumor suppressor gene therapy for colorectal cancer. Anticancer Res. 16, 3415–3422. 38. Chang, E. H., Jang, Y. J., Hao, Z., Murphy, G., Rait, A., Fee, Jr., W. E., Sussman, H. H., Ryan, P., Chiang, Y., and Pirollo, K. F. (1997). Restoration of the G1 checkpoint and the apoptotic pathway mediated by wild-type p53 sensitizes squamous cell carcinoma of the head and neck to radiotherapy. Arch. Otolaryngol. Head Neck Surg. 123, 507– 512. 39. Badie, B., Kramar, M. H., Lau, R., Boothman, D. A., Economou, J. S., and Black, K. L. (1998). Adenovirus-mediated p53 gene delivery potentiates the radiation-induced growth inhibition of experimental brain tumors. J. Neurooncol. 37, 217–222. 40. Vinyals, A., Peinado, M. A., Gonzalez-Garrigues, M., Monzo, M., Bonfil, R. D., and Fabra, A. (1999). Failure of wild-type p53 gene therapy in human cancer cells expressing a mutant p53 protein. Gene Ther. 6, 22–33. 41. el-Deiry, W. S., Harper, J. W., O’Connor, P. M., Velculescu, V. E., Canman, C. E., Jackman, J., Pietenpol, J. A., Burrell, M., Hill, D. E., Wang, Y., and et al. (1994). WAF1/CIP1 is induced in p53-mediated G1 arrest and apoptosis. Cancer Res. 54, 1169–1174. 42. Waldman, T., Kinzler, K. W., and Vogelstein, B. (1995). p21 is necessary for the p53-mediated G1 arrest in human cancer cells. Cancer Res. 55, 5187–5190. 43. Xiong, Y., Hannon, G. J., Zhang, H., Casso, D., Kobayashi, R., and Beach, D. (1993). p21 is a universal inhibitor of cyclin kinases. Nature 366, 701–704. 44. Katayose, D., Wersto, R., Cowan, K. H., and Seth, P. (1995). Effects of a recombinant adenovirus expressing WAF1/Cip1 on cell growth, cell cycle, and apoptosis. Cell Growth Differ. 6, 1207–1212. 45. Gotoh, A., Kao, C., Ko, S. C., Hamada, K., Liu, T. J., and Chung, L. W. (1997). Cytotoxic effects of recombinant adenovirus p53 and cell cycle regulator genes (p21 WAF1/CIP1 and p16CDKN4) in human prostate cancers. J. Urol. 158, 636–641. 46. Gomez-Manzano, C., Fueyo, J., Kyritsis, A. P., McDonnell, T. J., Steck, P. A., Levin, V. A., and Yung, W. K. (1997). Characterization of p53 and p21 functional interactions in glioma cells en route to apoptosis. J. Natl. Cancer Inst. 89, 1036–1044. 47. Meng, R. D., Shih, H., Prabhu, N. S., George, D. L., and el-Deiry, W. S. (1998). Bypass of abnormal MDM2 inhibition of p53-dependent growth suppression. Clin. Cancer Res. 4, 251–259. 48. Hsiao, M., Tse, V., Carmel, J., Costanzi, E., Strauss, B., Haas, M., and Silverberg, G. D. (1997). Functional expression of human p21(WAF1/CIP1) gene in rat glioma cells suppresses tumor growth in vivo and induces radiosensitivity. Biochem. Biophys. Res. Commun. 233, 329–335. 49. Li, W. W., Fan, J., Hochhauser, D., and Bertino, J. R. (1997). Overexpression of p21waf1 leads to increased inhibition of E2F-1 phosphorylation and sensitivity to anticancer drugs in retinoblastoma- negative human sarcoma cells. Cancer Res. 57, 2193–219.

445

50. Sheikh, M. S., Chen, Y. Q., Smith, M. L., and Fornace, Jr., A. J. (1997). Role of p21Waf1/Cip1/Sdi1 in cell death and DNA repair as studied using a tetracycline-inducible system in p53-deficient cells. Oncogene 14, 1875–1882. 51. Wouters, B. G., Giaccia, A. J., Denko, N. C., and Brown, J. M. (1997). Loss of p21Waf1/Cip1 sensitizes tumors to radiation by an apoptosisindependent mechanism. Cancer Res. 57, 4703–4706. 52. Tian, H., Wittmack, E. K., and Jorgensen, T. J. (2000). p21WAF1/CIP1 antisense therapy radiosensitizes human colon cancer by converting growth arrest to apoptosis. Cancer Res. 60, 679–684. 53. Dachs, G. U., Dougherty, G. J., Stratford, I. J., and Chaplin, D. J. (1997). Targeting gene therapy to cancer: a review. Oncol. Res. 9, 313–325. 54. Martin, L. A., and Lemoine, N. R. (1996). Direct cell killing by suicide genes. Cancer Metastasis Rev. 15, 301–316. 55. Balzarini, J., Bohman, C., and De Clercq, E. (1993). Differential mechanism of cytostatic effect of (E)-5-(2-bromovinyl)-2 - deoxyuridine, 9-(1,3-dihydroxy-2-propoxymethyl)guanine, and other antiherpetic drugs on tumor cells transfected by the thymidine kinase gene of herpes simplex virus type 1 or type 2. J. Biol. Chem. 268, 6332–6337. 56. Culver, K. W., Ram, Z., Wallbridge, S., Ishii, H., Oldfield, E. H., and Blaese, R. M. (1992). In vivo gene transfer with retroviral vectorproducer cells for treatment of experimental brain tumors. Science 256, 1550–1552. 57. Ezzeddine, Z. D., Martuza, R. L., Platika, D., Short, M. P., Malick, A., Choi, B., and Breakefield, X. O. (1991). Selective killing of glioma cells in culture and in vivo by retrovirus transfer of the herpes simplex virus thymidine kinase gene. New Biol. 3, 608–614. 58. Moolten, F. L., and Wells, J. M. (1990). Curability of tumors bearing herpes thymidine kinase genes transferred by retroviral vectors. J. Natl. Cancer Inst. 82, 297–300. 59. Kinsella, T. J., Mitchell, J. B., Russo, A., Morstyn, G., and Glatstein, E. (1984). The use of halogenated thymidine analogs as clinical radiosensitizers: rationale, current status, and future prospects: non-hypoxic cell sensitizers. Int. J. Radiat. Oncol. Biol. Phys. 10, 1399–1406. 60. Bagshaw, M. A., Doggett, R. L., Smith, K. C., Kaplan, H. S., and Nelsen, T. S. (1967). Intra-arterial 5-bromodeoxyuridine and X-ray therapy. Am. J. Roentgenol. Radium Ther. Nucl. Med. 99, 886–894. 61. Sougawa, M., Akagi, K., Murata, T., Kawasaki, S., Sawada, S., Yoshii, G., and Tanaka, Y. (1986). Enhancement of radiation effects by acyclovir. Int. J. Radiat. Oncol. Biol. Phys. 12, 1537–1540. 62. Kim, J. H., Kim, S. H., Brown, S. L., and Freytag, S. O. (1994). Selective enhancement by an antiviral agent of the radiation-induced cell killing of human glioma cells transduced with HSV-tk gene. Cancer Res. 54, 6053–6056. 63. Kim, J. H., Kim, S. H., Kolozsvary, A., Brown, S. L., Kim, O. B., and Freytag, S. O. (1995). Selective enhancement of radiation response of herpes simplex virus thymidine kinase transduced 9L gliosarcoma cells in vitro and in vivo by antiviral agents. Int. J. Radiat. Oncol. Biol. Phys. 33, 861–868. 64. Kim, S. H., Kim, J. H., Kolozsvary, A., Brown, S. L., and Freytag, S. O. (1997). Preferential radiosensitization of 9L glioma cells transduced with HSV- tk gene by acyclovir. J. Neurooncol. 33, 189–194. 65. Andersen, L., Kilstrup, M., and Neuhard, J. (1989). Pyrimidine, purine and nitrogen control of cytosine deaminase synthesis in Escherichia coli K 12. Involvement of the glnLG and purR genes in the regulation of codA expression. Arch. Microbiol. 152, 115–118. 66. Blaese, R. M., Ishii-Morita, H., Mullen, C., Ramsey, J., Ram, Z., Oldfield, E., and Culver, K. (1994). In situ delivery of suicide genes for cancer treatment. Eur. J. Cancer 8, 1190–1193. 67. Khil, M. S., Kim, J. H., Mullen, C. A., Kim, S. H., and Freytag, S. O. (1996). Radiosensitization by 5-fluorocytosine of human colorectal carcinoma cells in culture transduced with cytosine deaminase gene. Clin. Cancer Res. 2, 53–57. 68. Mullen, C. A., and Blaese, R. M. (1994). Gene therapy of cancer. Cancer Chemother. Biol. Response Modif. 15, 176–189.

446


69. Mullen, C. A., Coale, M. M., Lowe, R., and Blaese, R. M. (1994). Tumors expressing the cytosine deaminase suicide gene can be eliminated in vivo with 5-fluorocytosine and induce protective immunity to wild type tumor. Cancer Res. 54, 1503–1506. 70. Parker, W. B., and Cheng, Y. C. (1990). Metabolism and mechanism of action of 5-fluorouracil. Pharmacol. Ther. 48, 381–395. 71. Weckbecker, G. (1991). Biochemical pharmacology and analysis of fluoropyrimidines alone and in combination with modulators. Pharmacol. Ther. 50, 367–424. 72. Moertel, C. G., Gunderson, L. L., Mailliard, J. A., McKenna, P. J., Martenson, Jr., J. A., Burch, P. A., and Cha, S. S. (1994). Early evaluation of combined fluorouracil and leucovorin as a radiation enhancer for locally unresectable, residual, or recurrent gastrointestinal carcinoma. The North Central Cancer Treatment Group. J. Clin. Oncol. 12, 21–27. 73. O’Connell, M. J., Martenson, J. A., Wieand, H. S., Krook, J. E., Macdonald, J. S., Haller, D. G., Mayer, R. J., Gunderson, L. L., and Rich, T. A. (1994). Improving adjuvant therapy for rectal cancer by combining protracted- infusion fluorouracil with radiation therapy after curative surgery. N. Engl. J. Med. 331, 502–507. 74. Bruso, C. E., Shewach, D. S., and Lawrence, T. S. (1990). Fluorodeoxyuridine-induced radiosensitization and inhibition of DNA double strand break repair in human colon cancer cells. Int. J. Radiat. Oncol. Biol. Phys. 19, 1411–1417. 75. Pederson, L. C., Buchsbaum, D. J., Vickers, S. M., Kancharla, S. R., Mayo, M. S., Curiel, D. T., and Stackhouse, M. A. (1997). Molecular chemotherapy combined with radiation therapy enhances killing of cholangiocarcinoma cells in vitro and in vivo. Cancer Res. 57, 4325– 4332. 76. Pederson, L. C., Vickers, S. M., Buchsbaum, D. J., Kancharla, S. R., Mayo, M. S., Curiel, D. T., and Stackhouse, M. A. (1998). Combined cytosine deaminase expression, 5-fluorocytosine exposure, and radiotherapy increases cytotoxicity to cholangiocarcinoma cells. J. Gastrointest. Surg. 2, 283–291. 77. Stackhouse, M. A., Pederson, L. C., Grizzle, W. E., Curiel, D. T., Gebert, J., Haack, K., Vickers, S. M., Mayo, M. S., and Buchsbaum, D. J. (2000). Fractionated radiation therapy in combination with adenoviral delivery of the cytosine deaminase gene and 5–fluorocytosine enhances cytotoxic and antitumor effects in human colorectal and cholangiocarcinoma models. Gene Ther. 7, 1019–1026. 78. Hanna, N. N., Mauceri, H. J., Wayne, J. D., Hallahan, D. E., Kufe, D. W., and Weichselbaum, R. R. (1997). Virally directed cytosine deaminase/5–fluorocytosine gene therapy enhances radiation response in human cancer xenografts. Cancer Res. 57, 4205–4209. 79. Rogulski, K. R., Kim, J. H., Kim, S. H., and Freytag, S. O. (1997). Glioma cells transduced with an Escherichia coli CD/HSV-1 TK fusion gene exhibit enhanced metabolic suicide and radiosensitivity. Hum. Gene Ther. 8, 73–85. 80. Rogulski, K. R., Wing, M. S., Paielli, D. L., Gilbert, J. D., Kim, J. H., and Freytag, S. O. (2000). Double suicide gene therapy augments the antitumor activity of a replication-competent lytic adenovirus through enhanced cytotoxicity and radiosensitization. Hum. Gene Ther. 11, 67–76. 81. Rogulski, K. R., Zhang, K., Kolozsvary, A., Kim, J. H., and Freytag, S. O. (1997). Pronounced antitumor effects and tumor radiosensitization of double suicide gene therapy. Clin. Cancer Res. 3, 2081–2088. 82. Freytag, S. O., Rogulski, K. R., Paielli, D. L., Gilbert, J. D., and Kim, J. H. (1998). A novel three-pronged approach to kill cancer cells selectively: concomitant viral, double suicide gene, and radiotherapy. Hum. Gene Ther. 9, 1323–1333. 83. Folkman, J. (1995). Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat. Med. 1, 27–31. 84. Klein, S., Roghani, M., and Rifkin, D. B. (1997). Fibroblast growth factors as angiogenesis factors: new insights into their mechanism of action. Exs 79, 159–192.

85. Thomas, K. A. (1996). Vascular endothelial growth factor, a potent and selective angiogenic agent. J. Biol. Chem. 271, 603–606. 86. Asano, M., Yukita, A., Matsumoto, T., Kondo, S., and Suzuki, H. (1995). Inhibition of tumor growth and metastasis by an immunoneutralizing monoclonal antibody to human vascular endothelial growth factor/vascular permeability factor-121. Cancer Res. 55, 5296–5301. 87. Boehm, T., Folkman, J., Browder, T., and O’Reilly, M. S. (1997). Antiangiogenic therapy of experimental cancer does not induce acquired drug resistance. Nature 390, 404–407. 88. O’Reilly, M. S., Boehm, T., Shing, Y., Fukai, N., Vasios, G., Lane, W. S., Flynn, E., Birkhead, J. R., Olsen, B. R., and Folkman, J. (1997). Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell 88, 277–285. 89. O’Reilly, M. S., Holmgren, L., Shing, Y., Chen, C., Rosenthal, R. A., Moses, M., Lane, W. S., Cao, Y., Sage, E. H., and Folkman, J. (1994). Angiostatin: a novel angiogenesis inhibitor that mediates the suppression of metastases by a Lewis lung carcinoma. Cell 79, 315–328. 90. Asano, M., Yukita, A., Matsumoto, T., Matsuo, K., Kondo, S., and Suzuki, H. (1995). Isolation and characterization of neutralizing monoclonal antibodies to human vascular endothelial growth factor/vascular permeability factor121 (VEGF/VPF121). Hybridoma 14, 475–480. 91. Borgstrom, P., Hillan, K. J., Sriramarao, P., and Ferrara, N. (1996). Complete inhibition of angiogenesis and growth of microtumors by anti-vascular endothelial growth factor neutralizing antibody: novel concepts of angiostatic therapy from intravital videomicroscopy. Cancer Res. 56, 4032–4039. 92. Lin, P., Sankar, S., Shan, S., Dewhirst, M. W., Polverini, P. J., Quinn, T. Q., and Peters, K. G. (1998). Inhibition of tumor growth by targeting tumor endothelium using a soluble vascular endothelial growth factor receptor. Cell Growth Differ. 9, 49–58. 93. Yuan, F., Chen, Y., Dellian, M., Safabakhsh, N., Ferrara, N., and Jain, R. K. (1996). Time-dependent vascular regression and permeability changes in established human tumor xenografts induced by an anti-vascular endothelial growth factor/vascular permeability factor antibody. Proc. Natl. Acad. Sci. USA 93, 14765–14770. 94. Coppola, G., Atlas-White, M., Katsahambas, S., Bertolini, J., Hearn, M. T., and Underwood, J. R. (1997). Effect of intraperitoneally, intravenously and intralesionally administered monoclonal anti-beta-FGF antibodies on rat chondrosarcoma tumor vascularization and growth. Anticancer Res. 17, 2033–2039. 95. Hori, A., Sasada, R., Matsutani, E., Naito, K., Sakura, Y., Fujita, T., and Kozai, Y. (1991). Suppression of solid tumor growth by immunoneutralizing monoclonal antibody against human basic fibroblast growth factor. Cancer Res. 51, 6180–6184. 96. Takahashi, J. A., Fukumoto, M., Kozai, Y., Ito, N., Oda, Y., Kikuchi, H., and Hatanaka, M. (1991). Inhibition of cell growth and tumorigenesis of human glioblastoma cells by a neutralizing antibody against human basic fibroblast growth factor. FEBS Lett. 288, 65–71. 97. Belletti, B., Ferraro, P., Arra, C., Baldassarre, G., Bruni, P., Staibano, S., De Rosa, G., Salvatore, G., Fusco, A., Persico, M. G., and Viglietto, G. (1999). Modulation of in vivo growth of thyroid tumor-derived cell lines by sense and antisense vascular endothelial growth factor gene. Oncogene 18, 4860–4869. 98. Im, S. A., Gomez-Manzano, C., Fueyo, J., Liu, T. J., Ke, L. D., Kim, J. S., Lee, H. Y., Steck, P. A., Kyritsis, A. P., and Yung, W. K. (1999). Antiangiogenesis treatment for gliomas: transfer of antisense-vascular endothelial growth factor inhibits tumor growth in vivo. Cancer Res. 59, 895–900. 99. Wang, Y., and Becker, D. (1997). Antisense targeting of basic fibroblast growth factor and fibroblast growth factor receptor-1 in human melanomas blocks intratumoral angiogenesis and tumor growth. Nat. Med. 3, 887–893. 100. Goldman, C. K., Kendall, R. L., Cabrera, G., Soroceanu, L., Heike, Y., Gillespie, G. Y., Siegal, G. P., Mao, X., Bett, A. J., Huckle, W. R., Thomas, K. A., and Curiel, D. T. (1998). Paracrine expression of a


101.

102.

103.

104.

105.

106.

107. 108.

109.

110.

111.

112.

113.

114.

115.

native soluble vascular endothelial growth factor receptor inhibits tumor growth, metastasis, and mortality rate. Proc. Natl. Acad. Sci. USA 95, 8795–8800. Gately, S., Twardowski, P., Stack, M. S., Cundiff, D. L., Grella, D., Castellino, F. J., Enghild, J., Kwaan, H. C., Lee, F., Kramer, R. A., Volpert, O., Bouck, N., and Soff, G. A. (1997). The mechanism of cancer-mediated conversion of plasminogen to the angiogenesis inhibitor angiostatin. Proc. Natl. Acad. Sci. USA 94, 10868–10872. Stathakis, P., Fitzgerald, M., Matthias, L. J., Chesterman, C. N., and Hogg, P. J. (1997). Generation of angiostatin by reduction and proteolysis of plasmin: catalysis by a plasmin reductase secreted by cultured cells. J. Biol. Chem. 272, 20641–20645. Hohenester, E., Sasaki, T., Olsen, B. R., and Timpl, R. (1998). Crystal structure of the angiogenesis inhibitor endostatin at 1.5 A resolution. EMBO J. 17, 1656–1664. Laird, A. D., Vajkoczy, P., Shawver, L. K., Thurnher, A., Liang, C., Mohammadi, M., Schlessinger, J., Ullrich, A., Hubbard, S. R., Blake, R. A., Fong, T. A., Strawn, L. M., Sun, L., Tang, C., Hawtin, R., Tang, F., Shenoy, N., Hirth, K. P., McMahon, G., and Cherrington, J. M. (2000). SU6668 is a potent antiangiogenic and antitumor agent that induces regression of established tumors. Cancer Res. 60, 4152–4160. Mendel, D. B., Laird, A. D., Smolich, B. D., Blake, R. A., Liang, C., Hannah, A. L., Shaheen, R. M., Ellis, L. M., Weitman, S., Shawver, L. K., and Cherrington, J. M. (2000). Development of SU5416, a selective small molecule inhibitor of VEGF receptor tyrosine kinase activity, as an anti-angiogenesis agent. Anticancer Drug Des. 15, 29–41. Castronovo, V., and Belotti, D. (1996). TNP-470 (AGM-1470): mechanisms of action and early clinical development. Eur. J. Cancer 32A, 2520–2527. Gervaz, P., and Fontolliet, C. (1998). Therapeutic potential of the antiangiogenesis drug TNP-470. Int. J. Exp. Pathol. 79, 359–362. O’Reilly, M. S., Holmgren, L., Chen, C., and Folkman, J. (1996). Angiostatin induces and sustains dormancy of human primary tumors in mice. Nat. Med. 2, 689–692. Gorski, D. H., Mauceri, H. J., Salloum, R. M., Gately, S., Hellman, S., Beckett, M. A., Sukhatme, V. P., Soff, G. A., Kufe, D. W., and Weichselbaum, R. R. (1998). Potentiation of the antitumor effect of ionizing radiation by brief concomitant exposures to angiostatin. Cancer Res. 58, 5686–5689. Mauceri, H. J., Hanna, N. N., Beckett, M. A., Gorski, D. H., Staba, M. J., Stellato, K. A., Bigelow, K., Heimann, R., Gately, S., Dhanabal, M., Soff, G. A., Sukhatme, V. P., Kufe, D. W., and Weichselbaum, R. R. (1998). Combined effects of angiostatin and ionizing radiation in antitumour therapy. Nature 394, 287–291. Gorski, D. H., Beckett, M. A., Jaskowiak, N. T., Calvin, D. P., Mauceri, H. J., Salloum, R. M., Seetharam, S., Koons, A., Hari, D. M., Kufe, D. W., and Weichselbaum, R. R. (1999). Blockade of the vascular endothelial growth factor stress response increases the antitumor effects of ionizing radiation. Cancer Res. 59, 3374–3378. Griscelli, F., Li, H., Bennaceur-Griscelli, A., Soria, J., Opolon, P., Soria, C., Perricaudet, M., Yeh, P., and Lu, H. (1998). Angiostatin gene transfer: inhibition of tumor growth in vivo by blockage of endothelial cell proliferation associated with a mitosis arrest. Proc. Natl. Acad. Sci. USA 95, 6367–6372. Tanaka, T., Cao, Y., Folkman, J., and Fine, H. A. (1998). Viral vectortargeted antiangiogenic gene therapy utilizing an angiostatin complementary DNA. Cancer Res. 58, 3362–3369. Chen, Q. R., Kumar, D., Stass, S. A., and Mixson, A. J. (1999). Liposomes complexed to plasmids encoding angiostatin and endostatin inhibit breast cancer in nude mice. Cancer Res. 59, 3308–3312. Sacco, M. G., Caniatti, M., Cato, E. M., Frattini, A., Chiesa, G., Ceruti, R., Adorni, F., Zecca, L., Scanziani, E., and Vezzoni, P. (2000). Liposome-delivered angiostatin strongly inhibits tumor growth and metastatization in a transgenic model of spontaneous breast cancer. Cancer Res. 60, 2660–2665.

447

116. Blezinger, P., Wang, J., Gondo, M., Quezada, A., Mehrens, D., French, M., Singhal, A., Sullivan, S., Rolland, A., Ralston, R., and Min, W. (1999). Systemic inhibition of tumor growth and tumor metastases by intramuscular administration of the endostatin gene. Nat. Biotechnol. 17, 343–348. 117. Matsuda, K. M., Madoiwa, S., Hasumi, Y., Kanazawa, T., Saga, Y., Kume, A., Mano, H., Ozawa, K., and Matsuda, M. (2000). A novel strategy for the tumor angiogenesis-targeted gene therapy: generation of angiostatin from endogenous plasminogen by protease gene transfer. Cancer Gene Ther. 7, 589–596. 118. Saleh, M., Stacker, S. A., and Wilks, A. F. (1996). Inhibition of growth of C6 glioma cells in vivo by expression of antisense vascular endothelial growth factor sequence. Cancer Res. 56, 393–401. 119. Cao, Y., Ji, R. W., Davidson, D., Schaller, J., Marti, D., Sohndel, S., McCance, S. G., O’Reilly, M. S., Llinas, M., and Folkman, J. (1996). Kringle domains of human angiostatin. Characterization of the antiproliferative activity on endothelial cells. J. Biol. Chem. 271, 29461– 29467. 120. Seetharam, S., Staba, M. J., Schumm, L. P., Schreiber, K., Schreiber, H., Kufe, D. W., and Weichselbaum, R. R. (1999). Enhanced eradication of local and distant tumors by genetically produced interleukin-12 and radiation. Int. J. Oncol. 15, 769–773. 121. Chou, J., Kern, E. R., Whitley, R. J., and Roizman, B. (1990). Mapping of herpes simplex virus-1 neurovirulence to gamma 134.5, a gene nonessential for growth in culture. Science 250, 1262–1266. 122. Martuza, R. L., Malick, A., Markert, J. M., Ruffner, K. L., and Coen, D. M. (1991). Experimental therapy of human glioma by means of a genetically engineered virus mutant. Science 252, 854–856. 123. Advani, S. J., Chung, S. M., Yan, S. Y., Gillespie, G. Y., Markert, J. M., Whitley, R. J., Roizman, B., and Weichselbaum, R. R. (1999). Replication-competent, nonneuroinvasive genetically engineered herpes virus is highly effective in the treatment of therapyresistant experimental human tumors. Cancer Res. 59, 2055–2058. 124. Markert, J. M., Gillespie, G. Y., Weichselbaum, R. R., Roizman, B., and Whitley, R. J. (2000). Genetically engineered HSV in the treatment of glioma: a review. Rev. Med. Virol. 10, 17–30. 125. Advani, S. J., Sibley, G. S., Song, P. Y., Hallahan, D. E., Kataoka, Y., Roizman, B., and Weichselbaum, R. R. (1998). Enhancement of replication of genetically engineered herpes simplex viruses by ionizing radiation: a new paradigm for destruction of therapeutically intractable tumors. Gene Ther. 5, 160–165. 126. Bradley, J. D., Kataoka, Y., Advani, S., Chung, S. M., Arani, R. B., Gillespie, G. Y., Whitley, R. J., Markert, J. M., Roizman, B., and Weichselbaum, R. R. (1999). Ionizing radiation improves survival in mice bearing intracranial high-grade gliomas injected with genetically modified herpes simplex virus. Clin. Cancer Res. 5, 1517–1522. 127. Tang, D. C., Jennelle, R. S., Shi, Z., Garver, Jr., R. I., Carbone, D. P., Loya, F., Chang, C. H., and Curiel, D. T. (1997). Overexpression of adenovirus-encoded transgenes from the cytomegalovirus immediate early promoter in irradiated tumor cells. Hum. Gene Ther. 8, 2117– 2124. 128. Cheng, X., and Iliakis, G. (1995). Effect of ionizing radiation on the expression of chloramphenicol acetyltransferase gene under the control of commonly used constitutive or inducible promoters. Int. J. Radiat. Biol. 67, 261–267. 129. Mauceri, H. J., Seung, L. P., Grdina, W. L., Swedberg, K. A., and Weichselbaum, R. R. (1997). Increased injection number enhances adenoviral genetic radiotherapy. Radiat. Oncol. Invest. 5, 220–226. 130. Robertson, 3rd, M. W., Wang, M., Siegal, G. P., Rosenfeld, M., Ashford, 2nd, R. S., Alvarez, R. D., Garver, R. I., and Curiel, D. T. (1998). Use of a tissue-specific promoter for targeted expression of the herpes simplex virus thymidine kinase gene in cervical carcinoma cells. Cancer Gene Ther. 5, 331–336. 131. Boothman, D. A., Lee, I. W., and Sahijdak, W. M. (1994). Isolation of an X-ray-responsive element in the promoter region of tissue-type

448

132.

133.

134.

135.

136. 137.

138.

139.

140.

141. 142.

143. 144.

145.


plasminogen activator: potential uses of X-ray-responsive elements for gene therapy. Radiat. Res. 138, S68–S71. Brach, M. A., Hass, R., Sherman, M. L., Gunji, H., Weichselbaum, R., and Kufe, D. (1991). Ionizing radiation induces expression and binding activity of the nuclear factor kappa B. J. Clin. Invest. 88, 691– 695. Hallahan, D. E., Sukhatme, V. P., Sherman, M. L., Virudachalam, S., Kufe, D., and Weichselbaum, R. R. (1991). Protein kinase C mediates x-ray inducibility of nuclear signal transducers EGR1 and JUN. Proc. Natl. Acad. Sci. USA 88, 2156–2160. Sherman, M. L., Datta, R., Hallahan, D. E., Weichselbaum, R. R., and Kufe, D. W. (1990). Ionizing radiation regulates expression of the c-jun protooncogene. Proc. Natl. Acad. Sci. USA 87, 5663–5666. Chavrier, P., Zerial, M., Lemaire, P., Almendral, J., Bravo, R., and Charnay, P. (1988). A gene encoding a protein with zinc fingers is activated during G0/G1 transition in cultured cells. EMBO J. 7, 29–35. Sukhatme, V. P. (1990). Early transcriptional events in cell growth: the Egr family. J. Am. Soc. Nephrol. 1, 859–866. Hallahan, D. E., Gius, D., Kuchibhotla, J., Sukhatme, V., Kufe, D. W., and Weichselbaum, R. R. (1993). Radiation signaling mediated by Jun activation following dissociation from a cell type-specific repressor. J. Biol. Chem. 268, 4903–4907. Datta, R., Rubin, E., Sukhatme, V., Qureshi, S., Hallahan, D., Weichselbaum, R. R., and Kufe, D. W. (1992). Ionizing radiation activates transcription of the EGR1 gene via CArG elements. Proc. Natl. Acad. Sci. USA 89, 10149–10153. Muscat, G. E., Gustafson, T. A., and Kedes, L. (1988). A common factor regulates skeletal and cardiac alpha-actin gene transcription in muscle. Mol. Cell. Biol. 8, 4120–4133. Liu, Z. J., Moav, B., Faras, A. J., Guise, K. S., Kapuscinski, A. R., and Hackett, P. (1991). Importance of the CArG box in regulation of beta-actin-encoding genes. Gene 108, 211–217. Treisman, R. (1992). The serum response element. TIBS 17, 423– 427. Datta, R., Taneja, N., Sukhatme, V. P., Qureshi, S. A., Weichselbaum, R., and Kufe, D. W. (1993). Reactive oxygen intermediates target CC(A/T)6GG sequences to mediate activation of the early growth response 1 transcription factor gene by ionizing radiation. Proc. Natl. Acad. Sci. USA 90, 2419–2422. Weichselbaum, R. R. (1995). Growth factors alter the therapeutic ratio in radiotherapy. Cancer J. Sci. Am. 1, 28. Carswell, E. A., Old, L. J., Kassel, R. L., Green, S., Fiore, N., and Williamson, B. (1975). An endotoxin-induced serum factor that causes necrosis of tumors. Proc. Natl. Acad. Sci. USA 72, 3666–3670. Tartaglia, L. A., Ayres, T. M., Wong, G. H., and Goeddel, D. V. (1993). A novel domain within the 55 kD TNF receptor signals cell death. Cell 74, 845–853.

146. Tartaglia, L. A., Rothe, M., Hu, Y. F., and Goeddel, D. V. (1993). Tumor necrosis factor’s cytotoxic activity is signaled by the p55 TNF receptor. Cell 73, 213–216. 147. Hallahan, D. E., Vokes, E. E., Rubin, S. J., O’Brien, S., Samuels, B., Vijaykumar, S., Kufe, D. W., Phillips, R., and Weichselbaum, R. R. (1995). Phase I dose-escalation study of tumor necrosis factor-alpha and concomitant radiation therapy. Cancer J. Sci. Am. 1, 204. 148. Weichselbaum, R. R., Hallahan, D. E., Sukhatme, V. P., and Kufe, D. W. (1992). Gene therapy targeted by ionizing radiation. Int. J. Radiat. Oncol. Biol. Phys. 24, 565–567. 149. Hallahan, D. E., Mauceri, H. J., Seung, L. P., Dunphy, E. J., Wayne, J. D., Hanna, N. N., Toledano, A., Hellman, S., Kufe, D. W., and Weichselbaum, R. R. (1995). Spatial and temporal control of gene therapy using ionizing radiation. Nat. Med. 1, 786–791. 150. Staba, M. J., Mauceri, H. J., Kufe, D. W., Hallahan, D. E., and Weichselbaum, R. R. (1998). Adenoviral TNF-alpha gene therapy and radiation damage tumor vasculature in a human malignant glioma xenograft. Gene Ther. 5, 293–300. 151. Mauceri, H. J., Hanna, N. N., Staba, M. J., Beckett, M. A., Kufe, D. W., and Weichselbaum, R. R. (1999). Radiation-inducible gene therapy. C. R. Acad. Sci. III 322, 225–228. 152. Mauceri, H. J., Hanna, N. N., Wayne, J. D., Hallahan, D. E., Hellman, S., and Weichselbaum, R. R. (1996). Tumor necrosis factor alpha (TNF-alpha) gene therapy targeted by ionizing radiation selectively damages tumor vasculature. Cancer Res. 56, 4311–4314. 153. Chung, T. D., Mauceri, H. J., Hallahan, D. E., Yu, J. J., Chung, S., Grdina, W. L., Yajnik, S., Kufe, D. W., and Weichselbaum, R. R. (1998). Tumor necrosis factor-alpha-based gene therapy enhances radiation cytotoxicity in human prostate cancer. Cancer Gene Ther. 5, 344–349. 154. Joki, T., Nakamura, M., and Ohno, T. (1995). Activation of the radiosensitive EGR-1 promoter induces expression of the herpes simplex virus thymidine kinase gene and sensitivity of human glioma cells to ganciclovir. Hum. Gene Ther. 6, 1507–1513. 155. Marples, B., Scott, S. D., Hendry, J. H., Embleton, M. J., Lashford, L. S., and Margison, G. P. (2000). Development of synthetic promoters for radiation-mediated gene therapy. Gene Ther. 7, 511–517. 156. Scott, S. D., Marples, B., Hendry, J. H., Lashford, L. S., Embleton, M. J., Hunter, R. D., Howell, A., and Margison, G. P. (2000). A radiationcontrolled molecular switch for use in gene therapy of cancer. Gene Ther. 7, 1121–1125. 157. Gorecki, M., Beck, Y., Hartman, J. R., Fischer, M., Weiss, L., Tochner, Z., Slavin, S., and Nimrod, A. (1991). Recombinant human superoxide dismutases: production and potential therapeutical uses. Free Radic. Res. Commun. 12–13, 401–410. 158. Zwacka, R. M., Dudus, L., Epperly, M. W., Greenberger, J. S., and Engelhardt, J. F. (1998). Redox gene therapy protects human IB-3 lung epithelial cells against ionizing radiation-induced apoptosis. Hum. Gene Ther. 9, 1381–1386.

C

H

A

P

T

E

R

29 Virotherapy with Replication-Selective Oncolytic Adenoviruses: A Novel Therapeutic Platform for Cancer DAVID KIRN Imperial Cancer Research Fund Program for Viral and Genetic Therapy of Cancer Imperial College School of Medicine Hammersmith Hospital London, W11 OHS, United Kingdom

I. Introduction 449 II. Attributes of Replication-Selective Adenoviruses for Cancer Treatment 451 III. Biology of Human Adenovirus 451 IV. Mechanisms of Adenovirus-Mediated Cell Killing 451 V. Approaches to Optimizing Tumor-Selective Adenovirus Replication 452 VI. Background: dl1520 (ONYX-015) 452 VII. Clinical Trial Results with Wild-Type Adenovirus: Flawed Study Design 453 VIII. A Novel Staged Approach to Clinical Research with Replication-Selective Viruses: dl1520 (ONYX-015) 454

IX. Results from Clinical Trials with dl1520 (ONYX-015) 455 A. B. C. D.

Toxicity 455 Viral Replication 456 Immune Response 456 Efficacy with dl1520 (ONYX-015) as a Single Agent 458 E. Efficacy in Combination with Chemotherapy: Potential Synergy Discovered 458

X. Results from Clinical Trials with dl1520 (ONYX-015): Summary 459 XI. Future Directions 460 A. Why Has dl1520 ONYX-015 Failed as a Single Agent for Refractory Solid Tumors? 460 B. Improving the Efficacy of Replication-Selective Agents 461

A. Intratumoral Indications 455 B. Intracavitary Indications 455 C. Vascular Delivery: Intraarterial and Intravenous Administration 455

XII. Summary 462 Acknowledgments 462 References

I. INTRODUCTION

chemotherapies and radiotherapy target a variety of different structures within cancer cells, almost all of them kill cancer cells through the induction of apoptosis. Apoptosis-resistant clones almost universally develop following standard therapy for metastatic solid epithelial cancers (e.g., nonsmall-cell lung, colon, breast, prostate, pancreatic), even if numerous high-dose chemotherapeutic agents are used in combination. The overall survival rates for most metastatic solid tumors have changed relatively little despite decades of work with

Most currently available therapies for metastatic solid tumors fail as a result of inadequate antitumoral potency and/or an overly narrow therapeutic index between cancerous and normal cells. Countless changes in dose, frequency and/or combinations of standard cytotoxic chemotherapies or radiotherapy have had at best a modest impact on patient outcome in the metastatic setting. Although standard


462

449


450

David Kirn

this approach (adjuvant therapy, in contrast, applied by definition when tumor burden is low, has resulted in clinically significant improvements in mortality). Novel therapeutic approaches must therefore have greater potency and greater selectivity than currently available treatments, and they should have novel mechanisms of action that will not lead to cross-resistance with existing approaches (i.e., do not rely exclusively on apoptosis induction in cancer cells). Tumor-targeted oncolytic viruses (virotherapy with replication-selective viruses) appear to have these characteristics. Viruses have evolved over millions of years to infect target cells, multiply, cause cell death and release of viral particles, and finally spread in human tissues. Their ability to replicate in tumor tissue allows for amplification of the input dose (e.g., 1000- to 10,000-fold increases) at the tumor site, while their lack of replication in normal tissues results in efficient clearance and reduced toxicity (Fig. 1 [see also color insert]). This selective replication within tumor tissue can theoretically increase the therapeutic index of these agents dramatically over standard replication-incompetent approaches. Also, viruses lead to infected cell death through a number of unique and distinct mechanisms. In addition, to direct lysis at the conclusion of the replicative cycle, viruses can kill cells through expression of toxic proteins, induction of both inflammatory cytokines and T-cell-mediated immunity, and enhancement of cellular sensitivity to their effects. Therefore, because activation of classical apoptosis pathways in the cancer cell is not the exclusive mode of killing, crossresistance with standard chemotherapeutics or radiotherapy is much less likely to occur. Revolutionary advances in molecular biology and genetics have led to a fundamental understanding of both (1) the replication and pathogenicity of viruses and (2) carcinogenesis. These advances have allowed novel agents to be engineered to enhance their safety and/or their antitumoral potency. Over the past decade, genetically engineered viruses in development have included adenoviruses, herpesviruses, and vaccinia. Inherently tumor-selective viruses such as reovirus, autonomous parvoviruses, Newcastle disease virus, measles virus strains, and vesicular stomatitis virus have each been characterized. Each of these agents has shown tumor selectivity in vitro and/or in vivo, and efficacy has been demonstrated in murine tumor models, with many of these agents following intratumoral, intraperitoneal, and/or intravenous routes of administration. Although preclinical data reported with these agents has been encouraging, many critical questions have awaited results from clinical trials. Viral agents such as adenovirus have complex biologies, potentially including species-specific interactions with host-cell machinery and/or immune response effectors [1,2]. Antitumoral efficacy and safety studies with these viruses have been performed in rodent or primate

FIGURE 1 Schematic representation of tumor-selective viral replication and cell killing (panel A) and tumor-selective tissue necrosis (panel B). (See color insert.)

models, and all published animal tumor model data with replication-selective adenoviruses have come from immunodeficient mouse–human tumor xenograft models [3–5]; therefore, data from cancer patients have been eagerly awaited. Now, after over 5 years of clinical development with dl1520, roughly 15 clinical trials have been completed and recently analyzed involving approximately 250 patients. This article will review the discovery and development of replication-selective oncolytic adenoviruses, with an emphasis on recently acquired data from phase I and II clinical trials. The goal will be to summarize: (1) the genetic targets and mechanisms of selectivity for these agents; (2) clinical trial data and what they have taught us to date about the promise but also the potential hurdles to be overcome with this approach; and (3) future approaches to overcome these hurdles.

Virotherapy with Replication-Selective Oncolytic Adenoviruses: A Novel Therapeutic Platform for Cancer

II. ATTRIBUTES OF REPLICATION-SELECTIVE ADENOVIRUSES FOR CANCER TREATMENT A number of efficacy, safety, and manufacturing issues must be assessed when considering a virus species for development as an oncolytic therapy. First, by definition the virus must replicate in and destroy human tumor cells. An understanding of the genes modulating infection, replication, or pathogenesis is necessary if rational engineering of the virus is to be possible. Because most solid human tumors have relatively low growth fractions, the virus should ideally infect noncycling cells. In addition, receptors for viral entry must be expressed on the target tumor(s) in patients [6]. From a safety standpoint, the parental wild-type virus should ideally cause only mild, well-characterized human disease(s). Nonintegrating viruses have potential safety advantages, as well. A genetically stable virus is desirable from both safety and manufacturing standpoints. Finally, the virus must be amenable to high-titer production and purification under good manufacturing practices (GMP) guidelines for clinical studies. Human adenoviruses have these characteristics and are therefore excellent candidates for therapeutic development.

III. BIOLOGY OF HUMAN ADENOVIRUS Adenovirus biology is reviewed in detail elsewhere [7]. Roughly 50 different serotypes of human adenovirus have been discovered; the two most commonly studied are types 2 and 5 (group C). All adenoviruses have linear, doublestranded DNA genomes of approximately 38 kb. The capsid is nonenveloped and is comprised of the structural proteins hexon, penton (binds αV β3,5 integrins for virus internalization), and fiber (binds coxsackie and adenovirus receptor, CAR) (Fig. 2 [see also color insert]). The adenovirus life-

cycle includes the following steps: (1) virus entry into the cell following CAR and integrin binding, (2) release from the endosome and entry into the nucleus, (3) expression of early region gene products, (4) cell entry into S-phase, (5) prevention of p53-dependent and –independent apoptosis, (6) shut-off of host cellular protein synthesis, (7) viral DNA replication, (8) viral structural protein synthesis, (9) virion assembly in the nucleus, (10) cell death, and (11) virus release. The E3 region encodes a number of gene products responsible for immune response evasion [8,9]. The gp-19-kDa protein inhibits major histocompatibility complex (MHC) class I expression on the cell surface (i.e., avoidance of cytotoxic T-lymphocytemediated killing) [10], and the E3 10.4/14.5-kDa (RID complex) and 14.7-kDa proteins inhibit apoptosis mediated by FasL or tumor necrosis factor (TNF) [9,11].

IV. MECHANISMS OF ADENOVIRUS-MEDIATED CELL KILLING Adenovirus replication within a target tumor cell can lead to cell destruction by several mechanisms (Table 1). Viral proteins expressed late in the course of infection are directly cytotoxic, including the E3-11.6 adenovirus death protein [12] and E4orf4. Deletion of these gene products results in a significant delay in cell death. In addition, E1A expression early during the adenovirus lifecycle induces cell sensitivity to TNF-mediated killing [13]. This effect is inhibited by the E3 proteins 10.4/14.5 and 14.7; deletion of these E3 proteins leads to an increase in TNF expression in vivo and enhanced cell sensitivity to TNF [2]. Finally, viral replication in and lysis of tumor cells has been shown to promote the induction

TABLE 1 Potential Mechanisms of Antitumoral Efficacy with Replication-Selective Adenoviruses Mechanism Direct cytotoxicity due to viral proteins Augmentation of antitumoral immunity CTL infiltration, killing Tumor cell death, antigen release Immunostimulatory cytokine induction Antitumoral cytokine induction (e.g., TNF) Enhanced sensitivity to cytokines (e.g., TNF)

Examples of adenoviral genes modulating effect E3-11.6-kDa, E4ORF4 E3-gp-19-kDaa E3-11.6-kDa E3-10.4/14.5, 14.7 kDaa E3-10.4/14.5, 14.7 kDaa E1A

Sensitization to chemotherapy

Unknown (? E1A, others)

Expression of exogenous therapeutic genes

NA

a Viral

FIGURE 2 Human adenovirus coat structure. (See color insert.)

451

protein may inhibit antitumoral mechanism. Note: CTL, cytotoxic T-lymphocyte; TNF, tumor necrosis factor; NA, not applicable.

452

David Kirn

of cell-mediated immunity to uninfected tumor cells in model systems with other viruses [14,15]; whether this will occur in patients and with adenovirus remains to be determined.

V. APPROACHES TO OPTIMIZING TUMOR-SELECTIVE ADENOVIRUS REPLICATION Two broad approaches are currently being used to engineer tumor-selective adenovirus replication. One is to limit the expression of the E1A gene product to tumor tissues through the use of tumor- and/or tissue-specific promoters. E1A functions to stimulate S-phase entry and to transactivate both viral and cellular genes that are critical for a productive viral infection [16]. A second broad approach to optimizing tumor selectivity is to delete gene functions that are critical for efficient viral replication in normal cells but not in tumor cells (described later). Tissue- or tumor-specific promoters can replace endogenous viral sequences in order to restrict viral replication to a particular target tissue. For example, the prostate-specific antigen (PSA) promoter/enhancer element has been inserted upstream of the E1A gene; the result is that viral replication correlates with the level of PSA expression in a given cell [3]. This virus, CN706 (Calydon Pharmaceuticals, CA), is currently in a phase I clinical trial of intratumoral injection for patients with locally recurrent prostate carcinoma. A second prostate-specific enhancer sequence has been inserted upstream of the E1B region in the CN706 virus; the use of these two prostate-specific enhancer elements to drive separate early gene regions has led to improved selectivity over the first generation virus [17]. A similar approach has been pursued by other groups using tissue-specific promoters to drive E1A expression selectively in specific carcinomas (e.g., alpha-fetoprotein, carcinoembryonic antigen, MUC-1) [18] (D. Kufe, in press). A second general approach is to complement loss-offunction mutations in cancers with loss-of-function mutations within the adenovirus genome. Many of the same critical regulatory proteins that are inactivated by viral gene products during adenovirus replication are also inactivated during carcinogenesis [19,20–22]. Because of this convergence, the deletion of viral genes that inactivate these cellular regulatory proteins can be complemented by genetic inactivation of these proteins within cancer cells [23,24]. The deletion approach was first described with herpesvirus. Martuza et al. [25] deleted the thymidine kinase gene (dlsptk) and subsequently the ribonucleotide reductase gene (G207) [26] to engineer replication selectivity. Two adenovirus deletion mutants have been described. The first, dl1520 (ONYX-015) was hypothesized to replicate selectively in p53-deficient tumor cells (see later discussion). A second class of deletion mutants has now been described in E1A. Mutants in the E1A con-

served region 2 are defective in pRB binding. These viruses are being evaluated for use against tumors with pRB pathway abnormalities [24,27]. With dl922/947, for example, S-phase induction and viral replication are reduced in quiescent normal cells, whereas replication and cytopathic effects are not reduced in tumor cells; interestingly, dl922/947 demonstrates significantly greater potency than dl1520 both in vitro and in vivo [24]. In a nude mouse–human tumor xenograft model, intravenously administered dl922/947 had significantly superior efficacy to even wild-type adenovirus [28]. Unlike the complete deletion of E1B-55-kDa in dl1520, these mutations in E1A are targeted to a single conserved region and may therefore leave intact other important functions of the gene product.

VI. BACKGROUND: dl1520 (ONYX-015) One approach to engineering replication selectivity is to delete viral genes that are necessary for efficient replication in normal cells but are expendable in tumor cells. This pioneering approach was first described with herpesvirus. Martuza et al. [25] deleted the thymidine kinase gene (dlsptk) and subsequently the ribonucleotide reductase gene (G207) [26] to engineer replication-selectivity. dl1520 (ONYX-015) was the first adenovirus described to mirror this approach. McCormick hypothesized that an adenovirus with deletion of a gene encoding a p53-inhibitory protein, E1B-55-kDa, would be selective for tumors that already had inhibited or lost p53 function [34]. p53 function is lost in the majority of human cancers through mechanisms including gene mutation, overexpression of p53-binding inhibitors (e.g., MDM2, human papillomavirus E6) and loss of the p53inhibitory pathway modulated by p14ARF [29–31]. However, the precise role of p53 in the inhibition of adenoviral replication has not been defined to date. In addition, other adenoviral proteins also have direct or indirect effects on p53 function (e.g., E4orf6, E1B-19-kDa, E1A) [32]. Finally, E1B-55-kDa itself has important viral functions that are unrelated to p53 inhibition (e.g., viral mRNA transport, host cell protein synthesis shut-off) (Fig. 3) [33]. Not surprisingly, therefore, the exact role of p53 in the replication-selectivity of dl1520 has been difficult to confirm despite extensive in vitro experimentation by many groups. E1B-55-kDa gene deletion was associated with decreased replication and cytopathogenicity in p53(+) tumor cells versus matched p53(−) tumor cells, relative to wild-type adenovirus, in RKO and H1299 cells [34,35]. However, conflicting data on the role of p53 in modulating dl1520 replication and/or cytopathic effects (cpe) have come from different cell systems; no p53 effect was demonstrated in matched U2OS cells, for example [36]. It is clear that many other cellular factors independent of p53 play critical roles in determining the sensitivity of cells to dl1520 [35,37–39] and that the role of


453

FIGURE 3 Diagram of both p53 pathway interactions with adenoviral gene products and functions of E1B-55-kDa: complexity of cancer cell and adenoviral biology. Note that adenoviral proteins target multiple components of this pathway at sites upstream of p53, downstream of p53, and at the level of p53 itself. Examples of p53-regulated cell functions are shown, as are the known functions of E1B55-kDa. In addition to the loss of p53 binding when E1B-55-kDa was deleted in dl1520 (ONYX-015), these functions are also lost.

p53 in regulating the replication of dl1520 can vary depending on the cell line. In addition, the significant attenuation of this virus relative to wild-type adenovirus in most tumor cells studied to date, presumably due to the loss of other critical viral functions, is a potential drawback to this therapeutic candidate. Clinical trials were ultimately necessary to determine the selectivity and clinical utility of dl1520.

VII. CLINICAL TRIAL RESULTS WITH WILD-TYPE ADENOVIRUS: FLAWED STUDY DESIGN Over the last century a diverse array of viruses were injected into cancer patients by various routes, including adenovirus, Bunyamwara, coxsackie, dengue, feline panleukemia,

Ilheus, mumps, Newcastle disease, vaccinia, and West Nile [40–43]. These studies illustrated both the promise and the hurdles to overcome with oncolytic viral therapy. Unfortunately, these previous clinical studies were not performed to current clinical research standards; therefore, none gives interpretable and definitive results. At best, these studies are useful in generating hypotheses that can be tested in future trials. Although suffering from many of the trial design flaws listed below, a trial with wild-type adenovirus is one of the most useful for hypothesis generation, as well as for illustrating how clinical trial design flaws severely curtail the utility of the study results. The knowledge that adenoviruses could eradicate a variety of tumor cells in vitro led to a clinical trial in the 1950s with wild-type adenovirus. Ten different serotypes were used to treat 30 cervical cancer

454

David Kirn

patients [43]. Forty total treatments were administered by either direct intratumoral injection (n = 23), injection into the artery perfusing the tumor (n = 10), treatment by both routes (n = 6), or intravenous administration (n = 1). Characterization of the material injected into patients was minimal. The volume of viral supernatant injected is reported, but actual viral titers/doses are not; injection volumes (and by extension doses) varied greatly. When possible, the patients were treated with a serotype to which they had no neutralizing antibodies present. Corticosteroids were administered as nonspecific immunosuppressive agents in roughly half of the cases. Therefore, no two patients were treated in identical fashion. Nevertheless, the results are intriguing. No significant local or systemic toxicity was reported. This relative safety is notable, given the lack of preexisting immunity to the serotype used and concomitant corticosteroid use in many patients. Some patients reported a relatively mild viral syndrome lasting 2–7 days (severity not defined); this viral syndrome resolved spontaneously. Infectious adenovirus was recovered from the tumor in two thirds of the patients for up to 17 days postinoculation. Two thirds of the patients had a “marked to moderate local tumor response” with necrosis and ulceration of the tumor (definition of “response” not reported). None of the seven control patients treated with either virus-free tissue culture fluid or heat-inactivated virus had a local tumor response (statistical significance not reported). Therefore, clinically evident tumor necrosis was only reported with viable virus. Neutralizing antibodies increased within 7 days after administration. Although the clinical benefit to these patients is unclear, and all patients eventually had tumor progression and died, this study did demonstrate that wild-type aden-

Intra tumoral

Intra peritoneal

Intra arterial/hepatic artery

Intra venous

(+) safety/ (+) antitumoral activity

oviruses can be safely administered to patients and that these viruses can replicate and cause necrosis in solid tumors despite a humoral immune response. The maximally tolerated dose, dose-limiting toxicity, objective response rate, and time to tumor progression, however, remain unknown for any of these serotypes by any route of administration.

VIII. A NOVEL STAGED APPROACH TO CLINICAL RESEARCH WITH REPLICATION-SELECTIVE VIRUSES: dl1520 (ONYX-015) For the first time since viruses were first conceived as agents to treat cancer over a century ago, we now have definitive data from numerous phase I and II clinical trials with a well-characterized and –quantitated virus. dl1520 (ONYX-015, Onyx Pharmaceuticals, Richmond, CA) is a novel agent with a novel mechanism of action. This virus was to become the first virus to be used in humans that had been genetically engineered for replication selectivity. We predicted that both toxicity and efficacy would be dependent on multiple factors, including: (1) the inherent ability of a given tumor to replicate and shed the virus; (2) the location of the tumor to be treated (e.g., intracranial vs. peripheral), and (3) the route of administration of the virus. In addition, we felt it would be critical to obtain biological data on viral replication, antiviral immune responses, and their relationship to antitumoral efficacy in the earliest phases of clinical development. We therefore designed and implemented a staged clinical research and development approach (Fig. 4). The goal of

(+) chemotherapy combination data

(+) safety

(+) safety

(+) safety

FIGURE 4 Staged clinical research and development approach for a replication-selective agent in cancer patients. Following demonstration of safety and biological activity by the intratumoral route, trials were sequentially initiated to study intracavitary instillation (initially intraperitoneal), intraarterial infusion (initially hepatic artery), and eventually intravenous administration. In addition, only patients with advanced and incurable cancers were initially enrolled on trials. Only after safety had been demonstrated in terminal cancer patients were trials initiated for patients with premalignant conditions. Finally, clinical trials of combinations with chemotherapy were initiated only after the safety of dl1520 as a single agent had been documented by the relevant route of administration.


this approach was to sequentially increase systemic exposure to the virus only after safety with more localized delivery had been demonstrated. Following demonstration of safety and biological activity by the intratumoral route, trials were sequentially initiated to study intracavitary instillation (initially intraperitoneal), intraarterial infusion (initially hepatic artery), and eventually intravenous administration. In addition, only patients with advanced and incurable cancers were initially enrolled on trials. Only after safety had been demonstrated in terminal cancer patients were trials initiated for patients with premalignant conditions. Finally, clinical trials of combinations with chemotherapy were initiated only after the safety of dl1520 as a single agent had been documented by the relevant route of administration.

A. Intratumoral Indications Cancer patients can benefit from the effective local therapy of an established tumor mass if the target tumor mass causes morbidity or death before other masses do. For example, patients with recurrent glioblastoma multiforme or head and neck cancer frequently die from local tumor progression without evidence of distant metastases. In contrast, eradication of a localized skin lesion in a patient with widespread pulmonary or CNS metastases is unlikely to be of benefit. Patients with recurrent head and neck carcinomas were enrolled in the initial clinical trials because most suffer severe morbidity, and even mortality, from the local/regional progression of treatment-refractory tumors; therefore, intratumoral administration had the potential to cause substantial palliation and even survival prolongation. This population was also chosen because of the accessibility of superficial tumors for direct injection and biopsy in the outpatient clinic setting. Finally, patients with tumors in superficial neck and oral locations would presumably better tolerate peritumoral inflammation and swelling than patients with intraparenchymal tumors (e.g., intracranial, intrapulmonary, or intrahepatic). Once the safety of intratumoral injection was demonstrated in the superficial neck and oral regions, trials of intratumoral injection in solid organs (pancreas, liver) were carried out.

B. Intracavitary Indications Tumor types that spread and/or cause complications primarily within specific body cavities are potentially amenable to intracavitary administration of therapeutic agents. Examples include mesothelioma (pleural cavity), ovarian carcinoma (peritoneal cavity), and recurrent superficial bladder carcinoma (bladder). In addition, several premalignant conditions are also amenable to superficial intracavitary administration, including Barrett’s esophagus and oral dysplasias (e.g., oral leukoplakia). Intraperitoneal administration to patients with advanced, refractory ovarian carcinoma was followed by intraesophageal instillation in patients with

455

Barrett’s esophagus. The virus was sequestered within the affected region of the esophagus following instillation through a Wilson–Cook catheter by occlusive proximal and distal balloons. Finally, oral dysplasias were targeted through administration as a mouthwash.

C. Vascular Delivery: Intraarterial and Intravenous Administration Although patients with the indications listed above can potentially benenfit from local–regional therapy, systemic antitumoral efficacy can have a much greater impact on overall cancer-related mortality. Preclinical studies proved that intravenous adenovirus could infect, replicate within, and inhibit the growth of established metastatic tumor [4,23]. However, nude mouse–human tumor xenograft models were of unknown relevance to human cancer patients from both safety and efficacy standpoints. Targeted intraarterial infusions were studied in the first intravascular trial. Colorectal carcinoma metastases to the liver cause morbidity and death in a large proportion of these patients, and these metastases receive ≥90% of their bloodflow from the hepatic artery. Hepatic artery infusions had therefore been used previously to target colorectal liver metastases with a variety of agents [44]. Once safety by this route of administration had been demonstrated, intravenous trials were initiated in patients with lung metastases.

IX. RESULTS FROM CLINICAL TRIALS WITH dl1520 (ONYX-015) A. Toxicity No maximally tolerated dose or dose-limiting toxicities were identified at doses up to 2 × 1012 particles administered by intratumoral injeciton. This safety is true not only for tumors injected in superficial neck and oral sites, but also for intrahepatic and intrapancreatic tumor masses, as well. No clinically significant hepatitis or pancreatitis was demonstrated. Flu-like symptoms were the most common associated toxicities. No clear association between flu-like symptoms and viral dose was demonstrable. Phase I/II and phase II trials reported a similar lack of clinically significant toxicities. This safety is remarkable given the daily or even twice-daily dosing that was repeated every 1–3 weeks in the head and neck region or pancreas. Local complications of intratumoral injections in the pancreas appeared to be related to the endoscopic ultrasound procedure rather than to the agent itself; these included bacteremia, cyst formation, and a tear of the doudenal wall (45). Each of these complications was avoided once procedural changes were made and prophylactic antibiotic treatment was mandated. Intraperitoneal, intraarterial, and intravenous administration were also remarkably well tolerated, in general. Intraperitoneal administration was feasible at doses up to 1013

456

David Kirn

particles divided over 5 days [46]. The most common toxicites included fever, abdominal pain, nausea/vomiting and bowel motility changes (diarrhea, constipation). The severity of the symptoms appeared to correlate with tumor burden. Patients with heavy tumor burdens reached a maximally tolerated dose at 1012 particles (dose-limiting toxicities were abdominal pain and diarrhea), whereas patients with a low tumor burden tolerated 1013 without significant toxicity. No dose-limiting toxicities were reported following repeated intravascular injection at doses up to 2 × 1012 particles (hepatic artery) [47] or 2 × 1013 particles (intravenous) [48,61]. Fever, chills, and asthenia following intravascular injection were more common and more severe than after intratumoral injections (grade 2–3 fever and chills vs. grade 1). Dose-related transaminitis was reported infrequently. The transaminitis was typically transient (10-fold increase over the lower-limit-of-detection after 48–72 hours). False-negative results are possible, however, as the lower limit of detection is 104 genomes/mL, and viral shedding into the bloodstream

is required for detection. It was encouraging that the frequency of replication detected in head and neck cancer trials by biopsy staining was nearly identical to that determined by indirectly plasma sample polymerase chain reaction (PCR) testing. At this time, it appears that these two approaches each have merits and that they are complementary. Viral replication has been documented at early time points after intratumoral injection in head and neck cancer patients by both tests (Fig. 5). Roughly 70% of patients with either biopsy analysis or plasma testing by PCR had evidence of replication on days 1–3 after the last treatment. In contrast, day 14–17 samples were uniformly negative. This time-course for replication mirrors closely the clinical evidence for biological activity (e.g., local inflammation and necrosis). Intratumoral injection of liver metastases (primarily colorectal) led to similar PCR results at the highest doses of a phase I trial; high-quality biopsy samples could not be collected given the location of these tumors. Patients with injected pancreatic tumors, in contrast, showed no evidence of viral replication by plasma PCR or fine-needle aspiration. Similarly, intraperitoneal dl1520 could not be shown to reproducibly infect ovarian carcinoma cells within the peritoneum. None of the plasma PCR samples was positive, and none of the 12 peritoneal fluid samples was positive. Therefore, different tumor types can vary dramatically in their permissiveness for viral infection and replication. Proof of concept for tumor infection following intraarterial or intravenous administration with human adenovirus has been achieved. Following initial clearance of input genomes, approximately half of the roughly 25 patients receiving hepatic artery infusions of 2 × 1012 particles were positive by PCR 3–5 days following treatment [60]. Three of four patients with metastatic carcinoma to the lung treated intravenously with ≥2 ×1012 particles were positive for genomes by PCR on day 3 (±1). A single lung metastasis biopsy was positive for viral replication [61]. Therefore, it is feasible to infect distant tumor nodules following intravenous or intraarterial administration.

C. Immune Response Neutralizing antibody titers to the coat (Ad-5) of dl1520 were positive but relatively low in roughly 50–60% of all clinical trial patients at baseline. Antibody titers increased uniformly following administration of dl1520 by any of the routes tested, in some cases to levels >1:80,000. Antibody increases occurred regardless of evidence for replication or shedding into the bloodstream. Flu-like symptoms (fevers, rigors) were significantly more frequent and severe with intravascular administration than with intratumoral injections. The acute inflammatory cytokine response to hepatic arterial infusion was evaluated using reverse-transcription PCR (RT-PCR) for specific cytokine mRNAs from buffy coat

TABLE 2 Viral Replication Data from Phase I and II Trials of dl1520 (ONYX-015): Intratumoral, Intraperitoneal, Intraarterial or Intravenous Injection Tumor biopsy Route of administration Intratumoral

Tumor type Head and neck

Phase

Dose/cycle (particles)

I

2 × 108 –2 × 1012

Regimen (cycle frequency)

Days posttreatment

No. positive

No. evaluable

Percent positive

Days posttreatment

Single dose (q 4 wk)

5 1–3

4 5

16 7

25 71

10

7–10 14–17

2 0

4 10

50 0

11

19

0


5–10: Overall High dose

7 5

19 6

37 83


15 5

0 0

19 22

0 0

22

0

0 0

16 16

0 0

6

15

40

0 3

4 5

0 60

1012

Daily × 5 (q 3 wk)

Intratumoral

Gastrointestinal, primarily colorectal

I

2 × 108 –2 × 1012

Intratumoral

Pancreatic

I

2 × 108 –2 × 1012

Intratumoral

Pancreatic

I/II

2 × 1011 1011 –1013

Intraarterial (hepatic artery)

Gastrointestinal, primarily colorectal

I

2 × 109 –2 × 1011

II

6 × 1011 –2 × 1012

I

2 × 108 –2 × 1012

Intravenous

Metastatic carcinoma in lung

Not done 2

Percent positive

19

II

I

No. evaluable

0

Head and neck

Ovarian

No. positive

17

Intratumoral

Intraperitoneal

Blood quantitative (PCR)

Single dose (day 1, 5, 8, 15)

15 8

Daily × 5 (q 3 wk)

1–4

Single dose (q 1 wk; 2 wk on, 2 off)

4–7

Single dose (q 1 wk; 3 wk on, 1 off)

4: Overall High dose

0

12

5 15

0 Not done

3

1 1

9 3

11 33

8: Low dose High dose

458

David Kirn

FIGURE 5 Replication of adenoviral agent (dl1520) in the nucleus of a squamous carcinoma cell from the head and neck region of a patient 3 days after intratumoral virus injection. The dark-staining individual particles (arrow) and clusters within the nucleus (box) are adenoviral particles.

leukocyte samples. The levels of the following were determined prior to treatment, 3 hours post- and 18 hours posttreatment: IL-1, IL-6, IL-10, interferon-gamma (IFN-γ ), and TNF. Significant increases were demonstrated within 3 hours for IL-1, IL-6, and TNF and to a lesser extent IFN-γ ; all cytokines were back down to pretreatment levels by 18 hours. In contrast, IL-10 did not increase until 18 hours. Future analyses will attempt to correlate clinical outcomes with cytokine levels.

D. Efficacy with dl1520 (ONYX-015) as a Single Agent The single-agent efficacy of ONYX-015 is outlined in Table 3. Two phase II trials enrolled a total of 40 patients with recurrent head and neck cancer. Tumors were treated very aggressively with 6–8 daily needle passes for 5 consecutive days (30–40 needle passes per 5- day cycle; n = 30) and 10 to 15 per day on a second trial (50–75 needle passes per cycle; n = 10). The median tumor volume on these studies was approximately 25 cm3 ; an average cubic centimeter, of tumor therefore, received an estimated 4–5 needle passes per cycle. Despite the intensity of this treatment, the unconfirmed response rate was only 14%. Therefore, even in a tumor that can be extensively and repeatedly injected, the majority of injected tumors did not respond. Interestingly, there was no correlation between evidence of antitumoral activity and neutralizing antibody levels at baseline or post-treatment. Phase I and I/II data are available for other tumor types. No objective responses were demonstrated in patients with tumor types that could not be so aggressively injected (due to their deep locations). Although some evidence of minor shrinkage or necrosis was obtained, no objective responses were documented with intratumoral injection of either

pancreatic cancer (phase I and II trials; n = 43 patients) or gastrointestinal carcinomas (phase I trial, primarily colorectal; n = 19 patients). Similarly, no responses were seen following intraperitoreal (i.p.) administration in 16 ovarian cancer patients (phase I) or intravenous (i.v.) administration to 10 patients with metastatic carcinomas (phase I). Although some of these patients were treated during phase I portions of these trials, during which segment tumor response is not a primary endpoint, the lack of responses is notable. In summary, single-agent responses across all studies were rare; therefore, combinations with chemotherapy were explored.

E. Efficacy in Combination with Chemotherapy: Potential Synergy Discovered Evidence for a potentially synergistic interaction between adenoviral therapy and chemotherapy have been obtained on multiple trials (Table 4). Encouraging clinical data have been obtained in patients with recurrent head and neck cancer treated with intratumoral dl1520 in combination with intravenous cisplatin and 5-fluorouracil [49]. Of the 37 patients treated, 19 responded (54%, intent to treat; 63%, evaluable); this compares favorably with response rates to chemotherapy alone in previous trials (30–40%, generally). The time to tumor progression was also superior to previously reported studies; however, comparisons to historical controls are unreliable. We therefore used patients as their own controls whenever possible (n = 11 patients). Patients with more than one tumor mass had a single tumor injected with dl1520, while the other mass was left uninjected. Because both masses were exposed to chemotherapy, the effect of the addition of viral therapy to chemotherapy could be assessed. The dl1520injected tumors were significantly more likely to respond


459

TABLE 3 Antitumoral Efficacy Data of dl1520 (ONYX-015) as a Single Agent: Intratumoral, Intraperitoneal, Intra-Arterial or Intravenous Injection Route of administration

≥50% tumor regressiona number of responders/total (%)

Tumor type

Phase


Regimen/ cycle frequency

Intratumoral

Head and neck

I

2 × 108 –2 × 1012

Single dose/ q 4 week

3/22 (14)

Intratumoral

Head and neck

II

1012

Daily × 5/ q 3 week

Intent-to-treat: 4/30 (13) Confirmed,intent-to-treatb : 2/30 (7) Unconfirmed, evaluable: 4/19 (20)

Intratumoral

Gastrointestinal Liver metastasescolorectal, gastric, pancreatic

I

2 × 109 –2 × 1012


0/19 (0)

Intratumoral

Pancreatic (CT-guided)

I

2 × 108 –2 × 1012


0/22 (0)

Intratumoral

Pancreatic (endoscopic US)

I

2 × 1010 (n = 3)

Single dose/ days 1, 5, 8, 15

0/3 (0)

II I

2 × 1011 (n = 18) 1011 –1013

I

2 × 108 –2 × 1012

Intraperitoneal Intravenous

Ovarian Carcinoma metastatic to lung

Daily × 5/ q 3 wk. Single dose/ q 1 wk. (3 wk. on, 1 off)

0/18 (0)c 0/16 (0) 0/ 9 (0)

a

Non-necrotic cross-sectional area used for response assessment (i.e., necrotic area subtracted from total cross-sectional area). All responses refer to shrinkage of the injected tumor mass only (i.e., distant, noninjected tumors not included. All responses were in tumors with a p53 gene mutation. b Evaluable patients defined as those receiving >1 cycle of therapy and measurable tumor at baseline and at least one occasion > 6 weeks after treatment initiation (i.e., patients without follow-up tumor measurements after 1 + cycles of treatment were excluded). Intent-to-treat analysis includes all patients receiving at least one dose of ONYX-015. The confirmed responses reflect those that were confirmed to be durable for ≥4 weeks on an intent-to-treat basis. c Responses of single agent ONYX-015 determined after 4 cycles (on day 35) on the pancreatic EUS phase I/II trial. Subsequent cycles given with chemotherapy.

( p = 0.017) and less likely to progress ( p = 0.06) than were noninjected tumors. Noninjected control tumors that progressed on chemotherapy alone were subsequently treated with ONYX-015 in some cases; two of the four injected tumors underwent complete regressions. These data illustrate the potential of viral and chemotherapy combinations. The clinical utility of dl1520 in this indication will be definitively determined in an on-going phase III randomized trial. A phase I/II trial of dl1520 administered by hepatic artery infusion in combination with intravenous 5-fluorouracil and leukovorin was carried out (n = 33 total) [47,60]. Following phase I dose escalation, 15 patients with colorectal carcinoma who had previously failed the same chemotherapy were treated with combination therapy after failing to respond to dl1520 alone; one patient underwent a partial response and roughly 10 had stable disease (2–7+ months). Chemotherapyrefractory tumors can therefore respond following the same chemotherapy in combination with hepatic artery infusions of adenovirus; the magnitude and frequency of this effect re-

main to be determined. In contrast, data from a phase I/II trial studying the combination of dl1520 and gemcitabine chemotherapy were disappointing (n = 21); the combination resulted in only two responses, and these patients had not received prior gemcitabine [45]. Therefore, potential synergy was demonstrated with dl1520 and chemotherapy in two tumor types that supported viral replication (head and neck, colorectal) but not in a tumor type that was resistant to viral replication (pancreatic).

X. RESULTS FROM CLINICAL TRIALS WITH dl1520 (ONYX-015): SUMMARY dl1520 has been extremely well-tolerated at the highest practical doses that could be administered (2 × 1012 to 2 × 1013 ) by intratumoral, intraperitoneal, intraarterial, and intravenous routes. The lack of clinically significant toxicity in the liver or other organs was remarkable. Flu-like symptoms (fever, rigors, asthenia) were the most common toxicities

460

David Kirn

TABLE 4 Evidence for Potential Synergya Between dl1520 (ONYX-015) and Chemotherapy from Clinical Trials Route of administration

Tumor type

Phase


Regimen (cycle frequency)

Evidence for potential synergya

Intratumoral

Head and neck

II

1012

ONYX-015 daily × 5 + cisplatin day 1 i.v.b; 5-FU days 1–5 c.i. (q3 wk)

ONYX-015-injected tumors significantly more likely to respond than matched, noninjected control tumors ( p = 0.017; McNemar’s test) ONYX-015-injected tumors less likely to progress than matched, noninjected control tumors ( p = 0.06; log rank test) 2 of 4 tumors progressing on chemotherapy responded to same chemotherapy plus ONYX-015 Uncontrolled: response rate 63% vs. historical 30–40% with chemotherapy and 14% with ONYX-015 alone

Intratumoral

Pancreatic (endoscopic US)

I II

2 × 1010 (n = 3) 2 × 1011 (n = 18)

ONYX-015 single dose + gemcitabine i.v.b. (q 1 wk)

None; 2 of 21 patients responded to combination

Intraperitoneal

Ovarian

I

1011 –1013

ONYX-015 daily × 5 (q 3 wk)

One patient had tumor responses (>50% reduction in CA-125) on platinum-based chemotherapy following ONYX-015, despite previous tumor progression on platinum-based chemotherapy alone and on ONYX-015 alone

Intraarterial (hepatic artery)

Gastrointestinal, liver metastases, primarily colorectal

I II

2 × 109 –6 × 1011 2 × 1012

ONYX-015 single dose + 5-FU/leucovorin i.v.b. (q 4 wk)

One partial regression, approx. 10 stable disease (2–7+ months) to combination ONYX-015 plus 5-FU/leucovorin in patients with tumor progression on both single-agent ONYX-015 and on 5-FU/leucovorin alone

Intravenous

Metastatic carcinoma

I

2 × 1010 –2 × 1013

Single dose (q 1 wk; 3 wk on, 1 off) then with weekly carboplatin/paclitaxel

N.A.

a Although synergy cannot be definitively proven in phase II clinical trials, these clinical trial results are consistent with synergy and/or a positive interaction between ONYX-015 and chemotherapy with cisplatin and 5-FU. Note: i.v.b., intravenous bolus; c.i., continuous infusion; 5-FU, 5-fluorouracil; N.A., not available.

and were increased in patients receiving intravascular treatment. Acute inflammatory cytokines (including IL-1, IL-6, TNF, and interferon-γ ) increased within hours following intraarterial infusion. Neutralizing antibodies increased in all patients, regardless of dose, route, or tumor type. Viral replication was documented in head and neck and colorectal tumors following intratumoral or intraarterial administration. Neutralizing antibodies did not block antitumoral activity in head and neck cancer trials of intratumoral injection; however, viral replication/shedding into the blood was inhibited by neutralizing antibodies. Single-agent antitumoral activity was minimal (∼ =15%) in head and neck cancers that could be repeatedly and aggressively injected. No objective responses were documented with single-agent therapy in phase I or I/II trials in patients with pancreatic, colorectal, or ovarian carcinomas; these were not definitive efficacy studies. A favorable

and potentially synergistic interaction with chemotherapy was discovered in some tumor types and by different routes of administration.

XI. FUTURE DIRECTIONS A. Why Has dl1520 ONYX-015 Failed as a Single Agent for Refractory Solid Tumors? Future improvements with this approach will be possible if the reasons for dl1520 failure as a single agent and success in combination with chemotherapy are uncovered. Factors specific to this adenoviral mutant, as well as factors that may be generalizable to other viruses, should be considered. Regarding this particular adenoviral mutant, it is important


to remember that this virus is significantly attenuated relative to wild-type adenovirus in most tumor cell lines in vitro and in vivo, including even p53 mutant tumors [35,36,38,50]. This is not an unexpected phenotype, as this virus has lost critical E1B-55-kDa functions that are unrelated to p53, including viral mRNA transport. This attenuated potency is not apparent with other adenovirus mutants such as dl922/947 [28]. In addition, a second deletion in the E3 gene region (10.4/14.5 complex) may make this virus more sensitive to the antiviral effects of TNF; an immunocompetent animal model will need to be identified in order to resolve this issue. Factors likely to be an issue with any virus include barriers to intratumoral spread, antiviral immune responses, and inadequate viral receptor expression (e.g., CAR, integrins). Viral coat modifications may be beneficial if inadequate CAR expression plays a role in the resistance of particular tumor types [51].

B. Improving the Efficacy of Replication-Selective Agents Given the high degree of safety, but to date disappointing single-agent efficacy, of dl1520 (ONYX-015) against

461

advanced solid tumors, second-generation viruses will clearly be engineered for greater potency [71]. Mutations in the adenoviral genome can enhance selectivity and/or potency. For example, a promising adenoviral E1A CR-2 mutant (dl922/947) has been described that demonstrates not only tumor selectivity (based on the G1-S checkpoint status of the cell) but also significantly greater antitumoral efficacy in vivo compared to dl1520 (all models tested) and even wild-type adenovirus (in a breast cancer metastasis model) [24]. Another E1A mutant adenovirus has demonstrated replication and cytopathic effects based on the pRB status of the target cell [27]. Deletion of the E1B-19-kDa gene (antiapoptotic bcl-2 homolog) is known to result in a “large plaque” phenotype due to enhanced speed of cell killing [52]. This observation has now been extended to multiple tumor cell lines and primary tumor cell cultures [53,54]. A similar phenotype resulted from overexpression of the E3-11.6 adenovirus death protein [55]. It remains to be seen whether these in vitro observations are followed by evidence for improved efficacy in vivo over wildtype adenovirus. Potency can also be improved by arming viruses with therapeutic genes (e.g., prodrug-activating enzymes and cytokines) [56–59]. Viral coat modifications may be beneficial

TABLE 5 Replication-Selective Microbiological Agents in Clinical Trials for Cancer Patients Parental strain

Cell phenotype allowing selective replication

Agent

Genetic alterations

Engineered Adenovirus (2/5 chimera)

ONYX-015

Cells lacking p53 function (e.g., deletion, mutation, HPV infection)

E1B55kDa gene deletion

Herpes simplex virus -1

G207

Proliferating cells

r Ribonucleotide reductase disruption (lac-Z insertion into ICP6 gene) r Neuropathogenesis gene mutation (γ 34.5 gene)

Adenovirus (serotype 5)

CN706

Prostate cells (malignant, normal)

Adenovirus (2/5 chimera)

Ad5-CD/tk-rep

Cells lacking p53 function (e.g., deletion, mutation, HPV infection)

Vaccinia virus

Wildtype ± GM-CSF

Unknown

Salmonella typhimurium

Vion /VNP20009

Extracellular proliferation in tumor milleu (mechanism unknown: ? nutrient, hypoxia, immune clearance differences)

E1A expression driven by PSE element r E1B-55kD gene deletion r Insertion of HSV-TK/CD fusion r None for selectivity r Immunostimulatory gene (GM-CSF) insertion r Deletion of msbB (lipidA metabolism) r Deletion of purI (purine synthesis)

Nonengineered Newcastle Disease virus

73-T

Unknown

Unknown (serial passage on tumor cells)

Autonomous parvoviruses

H-1

Transformed cells r↑ proliferation r↓ differentiation r ras, p53 mutation

None

Reovirus

Reolysina

Ras-pathway activation (e.g., ras mutation, EGFR signaling) and loss of interferon responsiveness

None

a not

yet in clinical trials; to enter clinical trials in 2000. Note: HPV, human papillomavirus; PSE, prostate-specific enhancer; LPS, lipopolysaccharide; EGFR, epidermal growth factor receptor.

462

David Kirn

if inadequate CAR expression plays a role in the resistance of particular tumor types [51]. Improved systemic delivery may require novel formulations or coat modifications, as well as suppression of the humoral immune response. Determination of the viral genes (e.g., E3 region) and immune response parameters mediating efficacy and toxicity will lead to immunomodulatory strategies. Finally, identification of the mechanisms leading to the potential synergy between replicating adenoviral therapy and chemotherapy may allow augmentation of this interaction. This understanding may then allow us to bolster this interaction.

XII. SUMMARY Adenovirus has a number of attractive features as a replication-selective agent for cancer treatment. Clinical studies have demonstrated that replication-competent adenovirus treatment can be well tolerated and that tumor necrosis can result. The feasibility of adenovirus delivery to tumors through the bloodstream has also been demonstrated [4,60,61]. The inherent ability of replication-competent adenoviruses to sensitize tumor cells to chemotherapy was a novel discovery that has led to chemosensitization strategies. These data will support the further development of adenoviral agents, including second-generation constructs containing exogenous therapeutic genes to enhance both local and systemic antitumoral activity [56,62,63]. In addition to adenovirus, other viral species are being developed including herpesvirus, vaccinia, reovirus, and measles virus (Table 5). [15,25,40,64–68]. Because intratumoral spread also appears to be a substantial hurdle for viral agents, inherently motile agents such as bacteria may hold great promise for this field (Table 5) [69,70]. Given the limited ability of in vitro cell-based assays and murine tumor model systems to accurately predict the efficacy and therapeutic index of replication-selective adenoviruses in patients, we believe that the timely translation of encouraging adenoviral agents into well-designed clinical trials with relevant biological endpoints is critical [71]. Only then can the true therapeutic potential of these agents be realized. The clinical development of the first-generation adenovirus ONYX-015 (dl1520) has taught us a great deal about the hurdles to be overcome with the replication-selective adenovirus approach. It has also demonstrated, however, the potential of this novel therapeutic platform to improve and prolong the lives of cancer patients.

Acknowledgments The following individuals have been instrumental in making this manuscript possible: Frank McCormick, John Nemunaitis, Stan Kaye, Tony Reid, Fadlo Khuri, James Abruzzesse, Eva Galanis, Joseph Rubin, Antonio Grillo-Lopez, Carla

Heise, Larry Romel, Chris Maack, Sherry Toney, Nick LeMoine, Britta Randlev, Patrick Trown, Fran Kahane, and Margaret Uprichard.

References 1. Wold, W. S., Hermiston, T. W., and Tollefson, A. E. (1994). Adenovirus proteins that subvert host defenses. Trends Microbiol. 2, 437–443. 2. Sparer, T. E., Tripp, R. A., Dillehay, D. L., Hermiston, T. W., Wold, W. S., and Gooding, L. R. (1996). The role of human adenovirus early region 3 proteins (gp19K, 10.4K, 14.5K, and 14.7K) in a murine pneumonia model. J. Virol. 70, 2431–2439. 3. Rodriguez, R., Schuur, E. R., Lim, H. Y., Henderson, G. A., Simons, J. W., and Henderson, D. R. (1997). Prostate attenuated replication competent adenovirus (ARCA) CN706: a selective cytotoxic for prostate-specific antigen-positive prostate cancer cells. Cancer Res. 57, 2559–2563. 4. Heise, C., Williams, A., Xue, S., Propst, M., and Kirn, D. (1999). Intravenous administration of ONYX-015, a selectively-replicating adenovirus, induces antitumoral efficacy. Cancer Res. 59, 2623– 2628. 5. Heise, C., Williams, A., Olesch, J., and Kirn, D. (1999). Efficacy of a replication-competent adenovirus (ONYX-015) following intratumoral injection: intratumoral spread and distribution effects. Cancer Gene Ther. 6, 499–504. 6. Wickham, T. J., Segal, D. M., Roelvink, P. W., Carrion, M. E., Lizonova, A., Lee, G. M., and Kovesdi, I. (1996). Targeted adenovirus gene transfer to endothelial and smooth muscle cells by using bispecific antibodies. J. Virol. 70, 6831–6838. 7. Shenk, T. (1996). Adenoviridae: the viruses and their replication, in Fields Virology. (K. Fields, ed.), pp. 2135–2137. Lippincott-Raven, Philadelphia. 8. Wold, W. S., Tollefson, A. E., and Hermiston, T. W. (1995). E3 transcription unit of adenovirus. Curr. Top. Microbiol. Immunol. 199, 237–274. 9. Dimitrov, T., Krajcsi, P., Hermiston, T. W., Tollefson, A. E., Hannink, M., and Wold, W. S. ( 1997). Adenovirus E3-10.4K/14.5K protein complex inhibits tumor necrosis factor-induced translocation of cytosolic phospholipase A2 to membranes. J. Virol. 71, 2830–2837. 10. Hermiston, T. W., Tripp, R. A., Sparer, T., Gooding, L. R., and Wold, W. S. (1993). Deletion mutation analysis of the adenovirus type 2 E3-gp19K protein: identification of sequences within the endoplasmic reticulum lumenal domain that are required for class I antigen binding and protection from adenovirus-specific cytotoxic T lymphocytes. J. Virol. 67, 5289–5298. 11. Shisler, J., Duerksen, H. P., Hermiston, T. M., Wold, W. S., and Gooding, L. R. (1996). Induction of susceptibility to tumor necrosis factor by E1A is dependent on binding to either p300 or p105-Rb and induction of DNA synthesis. J. Virol. 70, 68–77. 12. Tollefson, A. E., Ryerse, J. S., Scaria, A., Hermiston, T. W., and Wold, W. S. (1996). The E3-11.6-kDa adenovirus death protein (ADP) is required for efficient cell death: characterization of cells infected with ADP mutants. Virology 220, 152–162. 13. Gooding, L. R. (1994). Regulation of TNF-mediated cell death and inflammation by human adenoviruses. Infect. Agents Dis. 3, 106–115. 14. Toda, M., Rabkin, S., Kojima, H., and Martuza, R. (1999). Herpes simplex virus as an in situ cancer vaccine for the induction of specific antitumo immunity. Hum. Gene Ther. 10, 385–393. 15. Martuza, R. (2000). Conditionally replicating herpes viruses for cancer therapy. J. Clin. Invest. 105, 841–846. 16. Whyte, P., Ruley, H., and Harlow, E. (1988). Two regions of the adenovirus early region 1A proteins are required for transformation. J. Virol. 62, 257–265.

Virotherapy with Replication-Selective Oncolytic Adenoviruses: A Novel Therapeutic Platform for Cancer 17. Yu, D., Sakamoto, G., and Henderson, D. R. (1999). Identification of the transcriptional regulatory sequences of human kallikrein 2 and their use in the construction of calydon virus 764, an attenuated replication competent adenovirus for prostate cancer therapy. Cancer Res. 59, 1498–1504. 18. Hallenback, P. L., Chang, Y. N., Hay, C., Golightly, D., et al. (1999). A novel tumor-specific replication-restricted adenoviral vector for gene therapy of hepatocellular carcinoma. Hum. Gene Ther. 10, 1721–1733. 19. Barker, D. D., and Berk, A. J. (1987). Adenovirus proteins from both E1B reading frames are required for transformation of rodent cells by viral infection and DNA transfection. Virology 156, 107–121. 20. Nielsch, U., Fognani, C., and Babiss, L. E. (1991). Adenovirus E1A– p105(Rb) protein interactions play a direct role in the initiation but not the maintenance of the rodent cell transformed phenotype. Oncogene 6, 1031–1036. 21. Sherr, C. J. (1996). Cancer cell cycles. Science 274, 1672–1677. 22. Olson, D. C., and Levine, A. J. (1994). The properties of p53 proteins selected for the loss of suppression of transformation. Cell Growth Differ. 5, 61–71. 23. Heise, C., Sampson-Johannes, A., Williams, A., McCormick, F., Von Hoff, D. D., and Kirn, D. H. (1997). ONYX-015, an E1B geneattenuated adenovirus, causes tumor-specific cytolysis and antitumoral efficacy that can be augmented by standard chemotherapeutic agents [see comments]. Nat. Med. 3, 639–645. 24. Kirn, D., Heise, C., Williams, M., Propst, M., and Hermiston, T. (1998). Adenovirus E1A CR2 mutants as selectively-replicating agents for cancer, in Cancer Gene Therapy (E. C. Lattime and S. L. Gerson, eds.). San Diego. 25. Martuza, R. L., Malick, A., Markert, J. M., Ruffner, K. L., and Coen, D. M. (1991). Experimental therapy of human glioma by means of a genetically engineered virus mutant. Science 252, 854–856. 26. Mineta, T., Rabkin, S. D., Yazaki, T., Hunter, W. D., and Martuza, R. L. (1995). Attenuated multi-mutated herpes simplex virus-1 for the treatment of malignant gliomas. Nat. Med. 1, 938–943. 27. Fueyo, J., Gomez-Manzano, C., Alemany, R., Lee, P., McDonnell, T., Mitlianga, P., Shi, Y., Levin, V., Yung, W., and Kyritsis, A. (2000). A mutant oncolytic adenovirus targeting the Rb pathway produces antiglioma effect in vivo. Oncogene 19, 2–12. 28. Heise, C., Hermiston, T., and Kirn, D. (2000). An adenovirus E1A mutant that has potent and selective antitumoral efficacy following intravenous administration in murine tumor models. Proc. Am. Assoc. Cancer Res. 41, 350. 29. Scheffner, M., Munger, K., Byrne, J. C., and Howley, P. M. (1991). The state of the p53 and retinoblastoma genes in human cervical carcinoma cell lines. Proc. Natl. Acad. Sci. USA 88, 5523–5527. 30. Zhang, Y., Xiong, Y., and Yarbrough, W. G. (1998). ARF promotes MDM2 degradation and stabilizes p53: ARF-INK4a locus deletion impairs both the Rb and p53 tumor suppression pathways. Cell 92, 725–734. 31. Hollstein, M., Sidransky, D., Vogelstein, B., and Harris, C. C. (1991). p53 mutations in human cancers. Science 253, 49–53. 32. Dobner, T., Horikoshi, N., Rubenwolf, S., and Shenk, T. (1996). Blockage by adenovirus E4orf6 of transcriptional activation by the p53 tumor suppressor. Science 272, 1470–1473. 33. Yew, P. R., Liu, X., and Berk, A. J. (1994). Adenovirus E1B oncoprotein tethers a transcriptional repression domain to p53. Genes Dev. 8, 190–202. 34. Bischoff, J. R., Kirn, D. H., Williams, A., Heise, C., Horn, S., Muna, M., Ng, L., Nye, J. A., Sampson-Johannes, A., Fattaey, A., and McCormick, F. (1996). An adenovirus mutant that replicates selectively in p53deficient human tumor cells [see comments]. Science 274, 373–376. 35. Harada, J., and Berk, A. (1999). p53-independent and -dependent requirements for E1B-55kD in adenovirus type 5 replication. J. Virol. 73, 5333–5344.

463

36. Rothmann, T., Hengstermann, A., Whitaker, N. J., Scheffner, M., and zur Hausen, H. (1998). Replication of ONYX-015, a potential anticancer adenovirus, is independent of p53 status in tumor cells. J. Virol. 72, 9470–9478. 37. Heise, C., Sampson, J. A., Williams, A., McCormick, F., Von, H. D., and Kirn, D. H. (1997). ONYX-015, an E1B gene-attenuated adenovirus, causes tumor-specific cytolysis and antitumoral efficacy that can be augmented by standard chemotherapeutic agents [see comments]. Nat. Med. 3, 639–645. 38. Goodrum, F. D., and Ornelles, D. A. (1997). The early region 1B 55-kilodalton oncoprotein of adenovirus relieves growth restrictions imposed on viral replication by the cell cycle. J. Virol. 71, 548– 561. 39. Goodrum, F. D., and Ornelles, D. A. (1998). p53 status does not determine outcome of E1B 55-kilodalton mutant adenovirus lytic infection. J. Virol. 72, 9479–9490. 40. Kirn, D. (2000). Replication-selective micro-organisms: fighting cancer with targeted germ warfare. J. Clin. Invest. 105, 836–838. 41. Southam, C. M., and Moore, A. E. (1952). Clinical studies of viruses as antineoplastic agents, with particular reference to Egypt 101 virus. Cancer 5, 1025–1034. 42. Asada, T. (1974). Treatment of human cancer with mumps virus. Cancer 34, 1907–1928. 43. Smith, R., Huebner, R. J., Rowe, W. P., Schatten, W. E., and Thomas, L. B. (1956). Studies on the use of viruses in the treatment of carcinoma of the cervix. Cancer 9, 1211–1218. 44. Kemeny, N., Huang, Y., Cohen, A., Shi, W., Conti, J., Brennan, M., Bertino, J., Turnbull, A., Sullivan, D., Stockman, J., Blumgart, L., and Fong, Y. (1999). Hepatic arterial infusion of chemotherapy following resection of hepatic metastases from colorectal cancer. N. Engl. J. Med. 341, 2039–2048. 45. Hecht, R., Abbruzzese, J., Bedford, R., Randlev, B., Romel, L., Lahodi, S., and Kirn, D. (2000). Endoscopic ultrasound-guided intratumoral injection of pancreatic carcinomas with a replication-selective adenovirus: a phase I/II clinical trial. Proc. Am. Soc. Clin. Oncol. 19, 1039 (abstract). 46. Vasey, P., Shulman, L., Gore, M., Kirn, D., and Kaye, S. (2000). A phase I trial of an E1B-55kD gene-deleted adenovirus administered by intraperitoneal injection into patients with advanced, refractory ovarian carcinoma. Proc. Am. Soc. Clin. Oncol. 47. Reid, T., Galanis, E., Abbruzzese, J., Randlev, B., Romel, L., Rubin, J., and Kirn, D. (2000). Hepatic arterial infusion of a replication-selective adenovirus, ONYX-015: a phase I/II clinical trial. Proc. Am. Soc. Clin. Oncol. 19, 953 (abstract). 48. Kirn, D. (2001). Clinical trial results with the replication-selective adenovirus dl1520 (Onyx-015): What have we learned? Gene Therapy 8, 89–98. 49. Khuri, F., Nemunaitis, J., Ganly, I., Gore, M., MacDougal, M., Tannock, I., Kaye, S., Hong, W., and Kirn, D. (2000). A controlled trial of ONYX-015, an E1B gene-deleted adenovirus, in combination with chemotherapy in patients with recurrent head and neck cancer. Nat. Med. 6, 879–885. 50. Kirn, D., Hermiston, T., and McCormick, F. (1998). ONYX-015: clinical data are encouraging [letter; comment]. Nat. Med. 4, 1341– 1342. 51. Roelvink, P., Mi, G., Einfeld, D., Kovesdi, I., and Wickham, T. (1999). Identification of a conserved reseptor-binding site on the fiber proteins of CAR-recognizing adenoviridae. Science 286, 1568–1571. 52. Chinnadurai, G. (1983). Adenovirus 2 Ip+ locus codes for a 19 kd tumor antigen that plays an essential role in cell transformation. Cell 33, 759–766. 53. Sauthoff, H., Heitner, S., Rom, W., and Hay, J. (2000). Deletion of the adenoviral E1B-19kD gene enhances tumor cell killing of a replicating adenoviral vector. Hum. Gene Ther. 11, 379–388.

464

David Kirn

54. Medina, D. J., Sheay, W., Goodell, L., Kidd, P., White, E., Rabson, A. B., and Strair, R. K. (1999). Adenovirus-mediated cytotoxicity of chronic lymphocytic leukemia cells. Blood 94, 3499–3508. 55. Doronin, K., Toth, K., Kuppuswamy, M., Ward, P., Tollefson, A., and Wold, W. (2000). Tumor-specific, replication-competent adenovirus vectors overexpressing the adenovirus death protein. J. Virol. 74, 6147–6155. 56. Hermiston, T. (2000). Gene delivery from replication-selective viruses: arming guided missiles in the war against cancer. J. Clin. Invest. 105, 1169–1172. 57. Hawkins, L., Nye, J., Castro, D., Johnson, L., Kirn, D., and Hermiston, T. (1999). Replicating adenoviral gene therapy. Proc. Am. Assoc. Cancer Res. 40, 476. 58. Freytag, S. O., Rogulski, K. R., Paielli, D. L., Gilbert, J. D., and Kim, J. H. (1998). A novel three-pronged approach to kill cancer cells selectively: concomitant viral, double suicide gene, and radiotherapy [see comments]. Hum. Gene Ther. 9, 1323–1333. 59. Wildner, O., Blaese, R. M., and Morris, J. M. (1999). Therapy of colon cancer with oncolytic adenovirus is enhanced by the addition of herpes simplex virus-thymidine kinase. Cancer Res. 59, 410–413. 60. Reid, A., Galanis, E., Abbruzzese, J., Romel, L., Rubin, J., and Kirn, D. (1999). A phase I/II trial of ONYX-015 administered by hepatic artery infusion to patients with colorectal carcinoma, EORTC-NCI-AACR Meeting on Molecular Therapeutics of Cancer. 19, 953 (abstract). 61. Nemunaitis, J., Cunnungham, C., Buchanan, A., Blackburn, A., Edelman, G., Maples, P., Netto, G., Tong, A., Olson, S., and Kirn, D. (2001). Intravenous infusion of a replication-selective adenovirus (Onyx-015) in cancer patients: safety, feasibility, and biological activity. Gene Therapy 8(10), 746–759.

62. Heise, C., and Kirn, D. (2000). Replication-selective adenviruses as oncolytic agents. J. Clin. Invest. 105, 847–851. 63. Agha-Mohammadi, S., and Lotze, M. (2000). Immunomodulation of cancer: potential use of replication-selective agents. J. Clin. Invest. 105, 1173–1176. 64. Norman, K., and Lee, P. (2000). Reovirus as a novel oncolytic agent. J. Clin. Invest. 105, 1035–1038. 65. Mastrangelo, M., Eisenlohr, L., Gomella, L., and Lattime, E. (2000). Poxvirus vectors: orphaned and underappreciated. J. Clin. Invest. 105, 1031–1034. 66. Coffey, M., Strong, J., Forsyth, P., and Lee, P. (1998). Reovirus therapy of tumors with activated ras pathway. Science 282, 1332– 1334. 67. Kirn, D. (2000). A tale of two trials: selectively replicating herpesviruses for brain tumors. Gene Ther. 7, 815–816. 68. Lattime, E. C., Lee, S. S., Eisenlohr, L. C., and Mastrangelo, M. J. (1996). In situ cytokine gene transfection using vaccinia virus vectors. Semin. Oncol. 23, 88–100. 69. Low, K., Ittensohn, M., Le, T., Platt, J., Sodi, S., Amoss, M., Ash, O., Carmichael, E., Chakraborty, A., Fischer, J., Lin, S., Luo, X., Miller, S., Zheng, L., King, I., Pawelek, J., and Bermudes, D. (1999). Lipid A mutant Salmonella with suppressed virulence and TNFalpha induction retain tumor-targeting in vivo. Nat. Biotechnol. 17, 37–41. 70. Sznol, M., Lin, S., Bermudes, D., Zheng, L., and King, I. (2000). Use of preferentially replicating bacteria for the treatment of cancer. J. Clin. Invest. 105, 1027–1030. 71. Kirn, D., Martuza, R., Zwiebel, J. (2001). Replication-selective virotherapy for cancer: biological principles, risk management and future directions. Nature Med. 7(7), 781–787.

C

H

A

P

T

E

R

30 E1A Cancer Gene Therapy DUEN-HWA YAN

RUPING SHAO

MIEN-CHIE HUNG

Departments of Molecular and Cellular Oncology and Surgical Oncology M. D. Anderson Cancer Center The University of Texas Houston, Texas 77030

Department of Molecular and Cellular Oncology M. D. Anderson Cancer Center The University of Texas Houston, Texas 77030

Departments of Molecular and Cellular Oncology and Surgical Oncology M. D. Anderson Cancer Center The University of Texas Houston, Texas 77030

I. Introduction 465 II. HER2 Overexpression and E1A-Mediated Antitumor Activity 465 III. Mechanisms of E1A-Mediated Anti-tumor Activity 467 A. B. C. D. E.

form established cell lines [4], adenovirus type-5 or type-2 E1A cannot transform established cell lines [5] but could cooperate with other viral and cellular oncogenes to transform primary culture cells [6]. Therefore, adenovirus type-5 and type-2 E1A were considered as immortalization oncogenes. In this review, “E1A” is refers to the nontransforming E1A, and most of the experimental results described here were based on the use of type-5 E1A. In the last decade, E1A was found to be associated with multiple antitumor activities [7–1] (Fig. 1). Multicenter E1A clinical trials on ovarian, breast, and head and neck cancers are currently underway. Cancer model studies have confirmed the E1A-mediated antitumor activity. Study on the molecular mechanisms underlying the E1A-mediated antitumor activity has been an enlightening endeavor that has yielded many insightful observations (Fig. 2). In this review, we attempt to summarize these results with an emphasis on the observations obtained during the past decade.

HER2 Downregulation 467 Inhibition of Metastasis 468 Axl Downregulation 468 NF-κB Inactivation 469 Bystander Effect 470

IV. E1A Gene Therapy: Preclinical Models

470

A. Ovarian Cancer Model 470 B. Breast Cancer Model 471 C. Safety Studies 471

V. E1A Gene Therapy: Clinical Trials

472

A. Phase I Breast and Ovarian Cancer with Intracavity Administration 472 B. Phase II Head and Neck Cancer with Intratumor Administration 472

VI. Conclusion

473

References 473

II. HER2 OVEREXPRESSION AND E1A-MEDIATED ANTITUMOR ACTIVITY I. INTRODUCTION

HER2 (also known as neu or c-erbB-2) overexpression in breast, ovarian, and head and neck tumors is known to be the indicator for poor prognosis and poor survival of these cancer patients [12–17]. Although HER2 overexpression alone is not sufficient to confer a chemoresistance in normal mammary epithelial cells [18], our results and others indicate that HER2 overexpression found in human tumors usually correlates

The E1A gene products of human adenovirus type 5— 12S (243 amino acids) and 13S (289 amino acids)— are known to activate viral gene transcription and regulate the host gene expression as viruses propagate inside the cell [1–3]. In contrast to adenovirus type-12 E1A, a potent oncogene that can trans-


465


466

Duen-Hwa Yan, Ruping Shao and Mien-Chie Hung

E1A

(1) Sensitization to apoptosis induced by TNF-α, γ-irradiation, serum deprivation, Taxol, and other chemo-drugs

(1) Tumor growth

(2) Bystander effect

(3) Angiogenesis

(2) Metastasis

FIGURE 1 E1A-mediated antitumor activities. E1A enhances sensitization to apoptosis and produces bystander effect. E1A also inhibits tumor growth and suppresses metastasis and angiogenesis.

HER-2 overexpression or Axl/Gas6

E1A PI3K TNF-α or γ-irradiation Akt

IKK

Apoptosis

BAD

NF-κB/IκB

Degradation

Apoptosis

IκB-p NF-κB

Antiapoptotic gene expression

Survival FIGURE 2 E1A-mediated signaling pathways involved in antitumor activities. E1A targets HER2 overexpression, Ax1/Gas6, and IKK in Akt and NF-κB survival pathways to sensitize cells to apoptosisinduced by stress signals.

E1A Cancer Gene Therapy

with chemoresistance [19–23] and insensitivity to radiation treatment [24,25]. These results argue that, in addition to HER2 overexpression, other genetic changes during tumorigenesis are required to render a tumor cell chemoresistant. One should also keep in mind that the level of HER2 expression in cancer cells may also play a role in this process; that is, a certain level of HER2 overexpression is required to exhibit a chemoresistance phenotype [21]. It is therefore conceivable that E1A may convert a chemoresistant phenotype of HER2-overexpressing cells to a chemosensitive phenotype by downregulating HER2 expression. Although HER2 mRNA stability may contribute in part to HER2 overexpression in certain cancer cells [26], gene amplification [12–14] and transcriptional upregulation [27–31] are the primary causes for HER2 overexpression in cancer cells. One of the best ways to achieve an antitumor activity in HER2-overexpressing tumor is to turn off HER2 gene transcription that leads to HER2 downregulation. Our initial effort to identify nuclear factors that could do just that led us to the adenovirus E1A [32,33]. We showed that E1A could readily downregulate the transcription of the HER2 gene in vitro. Importantly, we demonstrated in vivo that E1A expression could reduce tumorigenecity and suppress metastatic potential of murine fibroblast cells transformed by murine HER2 gene [34–36]. We subsequently demonstrated the tumor suppressor activity of E1A in HER2-overexpressing human cancer cells derived from breast [37], ovary [38,39], and lung [40]. Thus, an antitumor activity appears to be one of the multifunctional features of E1A. In particular, E1A seems to exert its antitumor activity in HER2-overexpressing cancer cells by reducing their metastatic potentials [35,36,41] and by increasing their sensitivity to the chemotherapy [42–44] and radiation [45] treatments. The above observations form a scientific basis for the development of an E1A-based gene therapy strategy against HER2-overexpressing tumors, and, as we shall see, the E1A gene therapy could be applied to the low HER2-expressing cancer cells, as well.

III. MECHANISMS OF E1A-MEDIATED ANTI-TUMOR ACTIVITY A. HER2 Downregulation 1. HER2 Gene Transcriptional Repression E1A expression represses the steady-state HER2 mRNA and protein expression [33] by downregulating HER2 promoter activity [32]. To investigate the mechanism of the E1A-mediated HER2 transcriptional repression, we showed that a forced expression of an E1A-binding protein, p300 [46], could override the E1A-mediated HER2 transcriptional repression. Conversely, E1A expression inhibits p300mediated transcriptional activation via a p300 consensus binding sequence on the HER2 promoter [47]. These results

467

strongly suggest that p300 is a coactivator for HER2 promoter activity, and that E1A targets p300, resulting in HER2 transcriptional repression. This idea was further supported by the finding that the p300-binding domain of E1A (4–25 or 40–80 amino acids) is required for HER2 transcriptional repression [48]. However, it may not be the only mechanism that accounts for the HER2 transcriptional repression by E1A. Our earlier data indicated that a HER2 promoter element containing a consensus sequence responsive to E1A-mediated transcriptional repression (i.e., TGGAATG [49–52]) and an E-box sequence (i.e., CAGTTG [53]) could alleviate the E1A-mediated HER2 transcriptional repression when provided exogenously [32]. This result suggests that E1A may repress HER2 promoter activity by targeting the trans-acting factors interacting with these promoter elements. Further experiments are required to pinpoint the protein/DNA interaction on the HER2 promoter that is the target for E1A-mediated transcriptional repression. However, it appears that multiple cis-acting elements and trans-acting factors on the HER2 promoter are involved in the E1A-mediated transcriptional repression. 2. Sensitization to TNF-α-Induced Apoptosis Functionally speaking, downregulation of HER2 overexpression has another expected consequence: sensitization to tumor necrosis factor-alpha (TNF-α)-induced apoptosis [54]. Although the molecular mechanism responsible for this sensitization was not known at that time, recent study on the pathway of HER2 overexpression showed that Akt (a serine/threonine kinase) activation is one of the main events leading to TNF-α resistance (Fig. 2) [55]. In short, HER2 overexpression activates phosphatidylinositol 3-kinase (PI3K) [56], which, in turn, activates Akt by phosphorylation [57]. Under certain condition, Akt activation turns on multiple downstream signal pathways [58], including the nuclear factor kappa B (NF-κB) pathway leading to cell survival in response to stress signals [59–63]. In other cases, however, NF-κB is not activated in response to Akt activation [64,65]. The difference between these two observations could be due to different cell types used in their respective systems. In our experimental system, however, it makes sense that HER2 overexpression could protect the cells from TNF-α-induced apoptosis by activating the NF-κB pathway. Intriguingly, introducing a dominant-negative Akt (dnAkt) that abolishes Akt function in HER2-overexpressing cells could restore their sensitivity to TNF-α killing. This result clearly establishes a positive relationship between HER2 overexpression and Akt activation. Indeed, this observation was confirmed by the immunohistochemical analysis on tumor tissues that showed a positive correlation between the level of active Akt and the level of HER2 expression [55]. Given that, one would predict a NF-κB activation in HER2-overexpressing and Aktactivating tumor cells even in the absence of TNF-α. It is

468


indeed the case, as the activity of a NF-κB-activated promoter is higher in HER2-overexpressing cells than in low-HER2expressing cells. Furthermore, study on the Akt/NF-κB pathway showed an increased level in HER2-overexpressing cells of the phosporylated form of a NF-κB inhibitor, IκB-α (the phosporylation of which leads to subsequent ubiquitination and degradation of IκB-α protein [66]). Thus, IκB-α inactivation correlates well with a higher NF-κB activity in HER2overexpressing cells than that in low-HER2-expressing cells. The fact that the activity of IκB kinases (IKKs), which phosphorylate IκB-α and result in IκB-α degradation [67,68], is higher in HER2-overexpressing cells than in dnAkttransfected cells further confirms the notion that activation of the HER2/Akt/NF-κB pathway is responsible for cell survival in response to TNF-α. Taken together, HER2 downregulation by E1A would predict a sensitization of HER2-overexpressing cells to apoptosis induced by stress signals such as TNF-α that triggers the NFκB pathway. As will be discussed later, it is indeed the case. Moreover, we shall see that, in addition to HER2, the molecules involved in the NF-κB pathway turn out to be targets for E1A action as well. In that light, E1A-mediated sensitization to certain stress-signals-induced apoptosis may be a general effect independent of HER2 expression level (Fig. 2). 3. Sensitization to Taxol-Induced Apoptosis Is HER2 overexpression a hallmark for chemoresistance in cancer cells? Little agreement regarding the answer to this question can be found among investigators in the HER2 cancer biology field. The controversy surrounding this issue in both clinical and laboratory settings has been discussed recently [21,69]. The possibility of a threshold HER2 expression required for chemoresistance in cancer cells (e.g., Taxol resistance) appears to be a plausible explanation for the discrepancies between the seemingly conflicting results [70,71]. To further complicate the matter, in the case of “normal” human mammary epithelial cell lines (e.g., MCF-10A [72,73] and human mammary epithelial cells (HMEC) [18]), enforced HER2 expression causes Taxol resistance in MCF-10A cells [22], but not in the HMECs [18]. Obviously, the different genetic background associated with these cell lines will no doubt contribute to the apparent opposite results. The caveat, however, is that it is possible for a different HER2 expression level between these cell lines to account for the differences in their sensitivity to Taxol. Given the precaution mentioned above, one would expect that E1A-mediated HER2 downregulation may sensitize HER2-overexpressing cancer cells to Taxol-induced apoptosis, and that is indeed the case [42–44]. Either in E1A-stable transfectants [42,44] or with adenovirus infection [43], E1A expression in HER2-overexpressing cancer cells makes them more susceptible to Taxol killing. E1Aexpressing SKOV3.ip1 ovarian cancer cells (ip1-E1A) are

sensitive to Taxol killing as well as cisplatin, doxorubicin, TNF-α, or serum-withdrawl-induced killing [44]. In contrast, the E1A-expressing, low-HER2-expressing ovarian [42] or breast [43] cancer cells do not exhibit such chemosensitization. Although these data may argue for the importance of HER2 overexpression in chemoresistance, it cannot exclude the possibility that E1A could sensitize cells to apoptosis independent of the HER2 expression level. In fact, previous reports support that possibility [74,75]. At any rate, these results are in agreement with the E1A action in repressing HER2 transcription and inactivating the NF-κB survival pathway in response to stress signals.

B. Inhibition of Metastasis The ability of E1A to inhibit metastasis in vitro and in vivo has been demonstrated in different laboratories [35,37,39, 76–80]. The E1A-regulated molecules the involved in reducing the metastatic potential of cancer cells include upregulated E-cadherins [81,82], a nucleoside diphosphate kinase (NM23) [78], and tissue inhibitors of metalloproteinase (TIMPs) [83], as well as downregulated matrix metalloproteinases (MMPs) (e.g., MMP-1 [80], MMP-3 [84,85], MMP9 [80,86–89]) urokinase-type plasminogen activator (uPA) [80], adhesion molecules (e.g., CD44s) [90], and HER2 [35,41]. Because HER2 overexpression enhances metastatic potential of the cancer cell [35,91], it is conceivable that E1A may inhibit metastasis by downregulating HER2 expression. Indeed, E1A expression in HER2-overexpressing cancer cells rendered the cells less metastatic [35,41]. In addition, enforced HER2 expression, which results in MMP upregulation, partially restored the metastatic potential of the otherwise less metastatic cancer cells in vitro and in vivo [92], suggesting that HER2 overexpresion is involved in the metastatic phenotype of cancer cells. As in E1A-mediated sensitization, it is possible that E1A could suppress metastasis of cancer cells regardless of HER2 expression level. In support of this idea, reexpressing HER2 in E1A-expressing cancer cells could restore their tumorigenecity but failed to restore the suppressed metastatic potential or the repressed MMP expression [41]. Again, this result demonstrates the multifunctional features of E1A in suppressing metastasis that include the downregulation of HER2 and MMP, and the fact that E1A may suppress metastasis through other mechanisms independent of HER2 downregulation.

C. Axl Downregulation Using a differential display technique based on polymerase chain reaction (PCR) to analyze the receptor tyrosine kinase (RTK) expression profiles in the E1A-expressing cancer cells, Axl, a member of the UFO membrane receptor family [93–95], was found consistently down-regulated in


E1A-expressing cells [96]. Similar to HER2 downregulation, Axl transcription was repressed by E1A. Using an E1Aexpressing cell line in which Axl expression was enforced (E1A–Axl), the functional significance of Axl downregulation in the E1A-mediated antitumor activity could be probed. E1A–Axl cells have a growth rate similar to that of the E1A-expressing cells (albeit much reduced as compared with that of the parental cells) in the absence of the Axl-specific ligand, Gas6 [97–99], whereas in the presence of Gas6 E1A–Axl cells resumed their growth rate comparable to that of the parental cells. These results suggest that Axl downregulation may be one of the critical events in the process of the E1A-mediated growth retardation. Moreover, in contrast to the E1A-expressing cells, for which Gas6 has no effect on the E1A-mediated sensitization to serum-deprivationinduced apoptosis, E1A–Axl cells kept that sensitization in the absence of Gas6 but lost it when Gas6 is present. This observation confirms the notion that the Axl/Gas6 signal pathway is one of the cellular survival pathways in response to stress such as serum starvation, and the inhibition of this pathway by E1A sensitizes cells to apoptosis induced by these stress signals (Fig. 2). Further elucidation of the molecular mechanisms underlying the E1A-mediated inhibition of Axl/Gas6 pathway led to the finding that a critical Axl/Gas6 downstream molecule, Akt [100], was inactivated as a result of E1A expression (Lee and Hung, unpubl. results). In E1A–Axl cells, however, Akt is reactivated in the presence of Gas6, indicating a direct relationship between Axl/Gas6 signal transduction and Akt activation leading to cell survival. Functional confirmation of this relationship came from the observation that either blocking Akt by a dominant-negative Akt or blocking PI3K, an upstream molecule of Akt, by a specific inhibitor could render E1A–Axl cells sensitive to serum-deprivation-induced apoptosis in the presence of Gas6. Upon examining the Akt downstream molecules, including BAD (a proapoptotic molecule) [101,102], IKK [63,103], and FKHRL [104], the data showed an increased level of phosphorylated BAD (which leads to BAD inactivation) in E1A–Axl cells in the presence of Gas6. However, no detectable biochemical changes in NF-κB activation (resulting from IKK activation by Akt) or Fas ligand downregulation (resulting from FKHRL activation by Akt) (Lee and Hung, unpubl. results) were observed in these cells. These results strongly suggest a scenario that E1A represses Axl transcription leading to BAD activation by inhibiting Akt activity through the Gas6/Axl/PI3K/Akt/BAD pathway. As a corollary, cells become sensitive to apoptosis induced by serum deprivation (Fig. 2).

D. NF-κB Inactivation It has been known that E1A could sensitize cells to apoptosis induced by TNF-α [105,106] or ionizing radiation [75,107]. After investigating the E1A action in the TNF-α

469

and γ -irradiation pathways, the finding suggests that NF-κB, a common molecule shared by these pathways [60,108], may be a critical target of E1A to mediate apoptosis induced by TNF-α or γ -irradiation [45,109] (Fig. 2). Nuclear factor-κB can be activated by inflammatory cytokines such as TNF-α and a host of other stimuli including ionizing radiation [66,110], however, the role of the active NF-κB is to prevent apoptosis induced by these stimuli [59–62]. Indeed, in most cells, NF-κB activation prevents apoptosis through activating antiapoptotic genes, though under certain conditions and in certain cell types, active NF-κB could also induce apoptosis [111,112]. Importantly, aberrant NF-κB activation is involved in a variety of human diseases [113], including cancer [114,115]. These findings provide a possible explanation for the reason why a majority of cancer cells are resistant to TNF-α, radiation, or chemotherapy treatment. Thus, NF-κB has become an attractive target for various cancer gene therapy designs. ip1-E1A cells are highly susceptible to γ -irradiationinduced apoptosis [45] as compared with the parental cells that do not express E1A. When a NF-κB/DNA binding assay was performed, it became apparent that ip1-E1A cells lose the ability to generate active NF-κB in response to γ -irradiation, as indicated by the lack of p50/p65 heterodimer (the active form of NF-κB) binding to DNA. This result suggests that E1A expression may block the activation of NF-κB upon γ -irradiation in ip1-E1A cells. That enforced expression of NF-κB in ip1-E1A partially rescued the γ -irradiationinduced apoptosis confirms this observation [45]. Thus, E1Amediated NF-κB inactivation may be responsible for the sensitization to γ -irradiation-induced apoptosis. But, how does E1A inactivate NF-κB? The answer to this question came from the study of the effect of TNF-α on E1A-expressing cells [109]. Similar to γ irradiation, TNF-α could preferentially induce ip1-E1A cells to apoptosis as compared with cells that do not express E1A. Again, the active form of NF-κB is missing in ip1-E1A cells in response to TNF-α treatment, suggesting that NF-κB inactivation may also play a role in the E1A-mediated sensitization to TNF-α-induced apoptosis. One possible mechanism for NF-κB inactivation by E1A is that E1A may downregulate NF-κB protein expression. Although the NF-κB protein level was not downregulated by E1A in response to TNF-α, the phosporylated form of IκB-α was concurrently reduced in ip1-E1A cells. This phenomenon takes place without any changes on IκB-α protein level either before or after TNF-α treatment in ip1-E1A cells, suggesting that E1A may inactivate NF-κB by keeping IκB-α underphosporylated. Because IKK phosphorylates IκB-α, it is possible that E1A may inhibit IKK activity so that IκB-α could not be properly phosporylated. This possibility was supported by the observation that the endogenous IKK (both α and β forms) activity was inhibited in TNF-α-treated ip1-E1A cells as determined by its ability to phosphorylate IκB-α, whereas IKK

470


was active in the TNF-α-treated, non-E1A-expressing cells. Interestingly, E1A could also inhibit the exogenous IKK activity when IKK was transiently transfected into ip1-E1A cells. Taken together, these results suggest a scenario in which E1A may inactivate NF-κB by inhibiting IKK activity, leading to the stablization of IκB-α protein and retention of the IκB-α/NF-κB complex in the cytoplasm. Thus, in ip1-E1A cells, NF-κB is prevented from entering the nucleus and from activating the antiapoptotic genes. In this fashion, E1A mediates the TNF-α (and very likely γ -irradiation)-induced apoptosis (Fig. 2). Despite vigorous effort to elucidate how E1A inactivates IKK activity, no direct interaction between E1A and IKK could be found, suggesting that E1A may inactivate IKK in an indirect manner. Determining the molecular mechanism underlying IKK inactivation by E1A presents an exciting area of study into the function of E1A as a cancer therapeutic gene.

E. Bystander Effect Bystander effect is one of the important features of a useful therapeutic gene for cancer gene therapy. The classical example is the bystander effect generated by the herpes simplex virus thymidine kinase (TK) gene. In the presence of the prodrug ganciclovir, TK expression kills not only the TK-transfected cells but also the nearby untransfected cells [116]. The gap junction that mediates the intercellular communication appears to be responsible for the TK-induced bystander killing [117]. The evidence that E1A may possess a bystander effect came from a tumorigenicity assay that coimplanted ip1-E1A cells with the control cell lines that do not express E1A [118]. The result was surprising in that ip1-E1A cells (which possess a reduced tumorigenecity) could somehow suppress the tumor growth of two highly tumorigenic cancer cell lines: HER2-overexpressing SKOV3 and lowHER2-expressing MDA-MB-435. The histological sections of tumors generated by coimplantation showed a significant reduction of microvessels determined by the staining of a blood vessel marker (i.e., Factor VIII) as compared with that in SKOV3 tumors. This result indicates a reduced angiogenesis taken place in the coimplanted tumors. Using TdT (terminal deoxynucleotidy transferase)-mediated dUTP nick end labeling (TUNEL) assay to detect apoptotic cells in tumor sections, the coimplanted tumors have a level of apoptosis comparable to that found in the ip1-E1A tumors. In contrast, SKOV3 tumors have a minimum level of apoptosis. These data suggest that reduced angiogenesis and enhanced apoptosis may cause the reduced tumorigenecity observed in the coimplanted tumors. One possible mechanism for the E1A-mediated bystander effect is that ip1-E1A cells may secret certain factor(s) that suppresses tumor growth by its ability to generate antiproliferative and proapoptotic effects on the neighboring cells. Indeed, when SKOV3 cells were cultured in the medium

that has been used to culture ip1-E1A cells, the growth of SKOV3 cells was significantly inhibited. A similar result was also observed when MDA-MB-435 cells were cultured in ip1-E1A medium. Furthermore, as determined by TUNEL assay, more apoptotic SKOV3 cells were observed when cultured in ip1-E1A medium than in SKOV3 medium. These results strongly suggest the existence of a secreted factor generated from ip1-E1A cells, and this secreted factor could suppress proliferation and induce apoptosis in the neighboring, non-E1A-expressing cells. The identification and characterization of this E1A-induced factor would certainly facilitate our understanding about the E1A-mediated bystander effect.

IV. E1A GENE THERAPY: PRECLINICAL MODELS We have so far described the encouraging in vitro results regarding the phenomenon of E1A-mediated HER2 downregulation and chemosensitization. We also proposed several mechanisms by which E1A may act to achieve these functions. To test the efficacy of an E1A-based gene therapy in mice bearing human HER2-overexpressing tumors, three orthotopic cancer xenograft models—ovarian, breast, and lung—were established and two E1A delivery systems were used: a cationic liposome, DCChol:DOPE {3β[N-(N -dimethylaminoethane)-carbamoyl] cholesterol:dioleoylphatidylethanolamine (3:2)} (DC-Chol) [119] and adenovirus E1A (Ad.E1A) [33]. A toxicity study was subsequently conducted in immunocompetent mice to ensure the safety of the procedure and to determine the minimum side effects associated with the E1A gene therapy treatment.

A. Ovarian Cancer Model The orthotopic ovarian cancer model was established by injecting human HER2-overexpressing ovarian cancer cells (SKOV3) intraperitonealy (i.p.) into female nu/nu mice. The implanted ovarian tumors obtained from mesentery and the inside of the peritoneal cavity showed HER2-positive staining [39]. The tumor-bearing mice received i.p. injection of E1A expression vector complexed with either DC-Chol (E1A/DCChol) [39] or Ad.E1A [38]. 1. E1A/DC-Chol Treatment Necropsy analysis showed that some of E1A/DC-Choltreated mice, though dying of tumor-related symptoms, had no detectable tumor invasion and metastasis as commonly seen in mice in the control groups (i.e., no treatment, mutant E1A/DC-Chol, E1A alone, or DC-Chol alone) [39]. Upon examination of tumor tissues excised from the E1A/DC-Choltreated mice, it became clear that E1A expression correlated

471


well with downregulation of HER2 protein but there was no decrease in HER2 protein level in the tumors obtained from the control groups. Remarkably, 70% of the E1A/DC-Choltreated mice survived more than a year, while the controls all died within 200 days [39]. The surviving mice appeared normal and healthy, as there were no detectable tumors inside the mice or any obvious side effects associated with the treatment. These results showed that (1) by i.p. injection, E1A/DC-Chol complex is a useful vehicle to transduce E1A into ovarian cancer cells in vivo; and (2) E1A/DC-Chol treatment could repress HER2 expression, suppress tumor growth, reduce metastasis, increase survival, and have no obvious side effects. This observation became one of the first indications of the efficacy and feasibility of using E1A/DC-Chol-based gene therapy to effectively treat ovarian cancer in a xenograft model. 2. Ad.E1A Treatment The efficacy of using Ad.E1A in the above orthotopic ovarian cancer model appeared similar to that obtained from the E1A/DC-Chol treatment [38,39]. In addition to SKOV3, the Ad.E1A study also included a low-HER2-expressing human ovarian cancer cell line, 2774. Intriguingly, while Ad.E1A could effectively increase survival in the SKOV3 tumor model, it failed to do so in the 2774 tumor model [38,120]. This result was not due to a difference of viral infection efficiency between SKOV3 and 2774 cell lines, as both could be infected equally as determined by adenovirus carrying β-galactosidase gene (Ad.LacZ) [38]. This observation raises a possibility that E1A may mediate a preferential antitumor effect on HER2-overexpressing ovarian cancer cells but not on ovarian cancer cells with low HER2 expression. Alternatively, a more rigorous treatment may be needed for the low-HER2-expressing cancer cells such as 2774 to achieve a efficacy similar to that seen in treating the HER2overexpressing cancer cells. SKOV3 tumors excised from Ad.E1A-infected mice showed a positive staining for E1A proteins and a concurrent reduction of HER2 protein expression on the same tumor samples. This result, therefore, confirms in vivo a causal relationship between E1A expression and HER2 downregulation. Using Ad.LacZ to monitor the E1A expression spectrum in the SKOV3 tumor model, it is encouraging to know that a high LacZ expression was found in malignant ascites and tumors as compared with that in other tissues and organs, suggesting that Ad.E1A may preferentially target these tumor sites [38]. 3. E1A/DC-Chol and Taxol Combined Treatment As mentioned before, E1A could sensitize HER2overexpressing ovarian cancer cells to Taxol-induced apoptosis [42,44], and this phenomenon seems to be HER2 overexpression specific, as no such sensitization was observed in Taxol-treated, E1A-expressing 2774 cells [42]. To test a

possible enhancement of E1A-mediated antitumor activity in conjunction with Taxol treatment, E1A/DC-Chol + Taxol was i.p. injected to treat mice bearing SKOV3 tumors. The E1A/DC-Chol + Taxol treatment yielded the best survival result among all treatment groups including the E1A/DC-Chol alone treatment [42]. This observation is congruous to the in vitro data [42,44] and suggests that the E1A/DC-Chol + Taxol treatment may enhance the E1A-mediated antitumor activity in mice bearing HER2-overexpressing ovarian tumors.

B. Breast Cancer Model Ad.E1A infection preferentially inhibited the growth of HER2-overexpressing breast cancer cells (e.g., MDAMB-361 and SKBR3), whereas there was little or no E1A-mediated growth inhibitory effect on the low-HER2expressing cancer cells (e.g., MDA-MB-435 and MDA-MB231) [37]. Based on this observation, both Ad.E1A and E1A/DC-Chol were used to assess the potential efficacy in an orthotopic, HER2-overexpressing breast cancer model. MDA-MB-361 cells were transplanted into the mammary fat pads of female nu/nu mice. The mammary tumors become palpable usually about 45 days after implantation. Ad.E1A or E1A/DC-Chol was intratumor injected. Six months of E1A treatment by either Ad.E1A or E1A/DC-Chol prolonged survival (the mean survival was greater than 2 years as opposed to less than 15 months in the control groups) and inhibited tumor growth. The Ad.E1A treatment appeared slightly better than E1A/DC-Chol treatment. Remarkably, no metastasis was found in intraperitoneal organs such as liver, intestine, spleen, and kidney [37]. These results are consistent with the ability of E1A to inhibit metastasis and are reminiscent of the E1A-mediated antitumor effect on HER2-overexpressing ovarian tumors, for which no detectable metastasis was found in E1A-treated mice [39]. The mammary tumor suppression correlated well with the expression of E1A and the downregulation of HER2 protein as determined by western blot and immunohistochemical analysis on the tumor samples [37]. The above data suggest the feasibility of an E1A-based gene therapy (either by Ad.E1A or by E1A/DC-Chol) against HER2-overexpressing breast cancer in vivo.

C. Safety Studies To ensure the safety of E1A/DC-Chol administration by intraperitoneal injection in clinical trials, it is imperative that a safe and tolerable dosage of E1A/DC-Chol is well defined. A series of studies that evaluate any adverse effects associated with a range of E1A and DC-Chol combinations were conducted in immunocompetent ICR female mice [121]. In an acute toxicity study, a range of E1A/DC-Chol doses that were 0.5–10 times the starting dose (10 nmol of DC-Chol complexed with 1 μg E1A DNA, 10:1) proposed in the phase I clinical trial did not cause apparent acute or residual toxic effects on mice. Hepatic and renal functions appeared normal

472


and other major organs showed normal pathology. These results suggest that the E1A/DC-Chol dosage proposed in the clinical trial may be safe, as it is much lower than that used in the animal toxicity study. A repeated E1A/DC-Chol treatment by i.p. injection did not show significant lesions on major organs in treated mice even 6 weeks after the injections. Again, this observation supports the idea that the proposed E1A/DC-Chol dosage in the human clinical trial may be safe. Although the DC-Chol/E1A DNA ratio of 10:1 was proposed for the clinical trial, a ratio of 13:1 has been used in vitro and in vivo studies and successfully demonstrated the gene delivery efficiency and treatment efficacy [39,119]. Can a comparable treatment efficacy be achieved by a minimum E1A/DC-Chol dosage so that a potential toxicity associated with a high concentration of DC-Chol could be avoided? To find that minimum effective dosage, a dose study using i.p.injected DC-Chol/LacZ DNA to monitor the β-galactosidase (β-gal) expression in SKOV3 tumor model was conducted [120]. There was no significant difference in β-gal expression between the DC-Chol/LacZ ratios of 0.5:1 and 26:1. Thus, it is possible that a much lower concentration of DCChol than was proposed in the clinical trial could be used without compromising the efficiency of gene delivery. Another concern of using E1A-based gene therapy in cancer treatment is the ability of E1A to immortalize and transform the otherwise normal cells under certain circumstances [4,122]. Different from retroviral vector that usually integrates into the chromosomes of the recipient cells, the E1A/DC-Chol or Ad.E1A delivery system ordinarily expresses E1A transiently. Therefore, the issue of E1A-mediated immortalization may not be a serious problem using the above-mentioned delivery systems. However, if it is necessary to circumvent this potential complication, we generated a modified E1A, named mini-E1A, that lacks the CR2 region responsible for immortalization and Rb protein binding. Mini-E1A remains competent in tumor suppression, as shown by its ability to repress HER2 transcription and suppress HER2-mediated transformation phenotype and tumorigenecity [48]. To test the efficacy of mini-E1A/DC-Chol treatment in SKOV3 model, mini-E1A/DC-Chol was i.p. injected into mice that bore SKOV3 tumors. Similar to wildtype E1A, mini-E1A/DC-Chol treatment prolonged survival at the two DC-Chol/DNA ratios tested: 1:1 and 13:1 [120]. Thus, it is may be possible to substitute wild-type E1A with mini-E1A in gene therapy treatment. By doing so, we may avoid the potentially undesirable complications associated with wild-type E1A. Although E1A/DC-Chol treatment generally allows the E1A gene to be expressed transiently, it is possible that a small percentage of the transfected E1A gene may be integrated into the host chromosome. This potential problem could be especially serious if E1A remains in the cells of the reproductive organs. If that was the case, it is likely that E1A may be transmitted to the next generation. For this reason, the organs of E1A/DC-Chol-treated mice (tumor free and surviv-

ing for 1.5 years) were analyzed by PCR technique to detect the presence of E1A DNA [120]. Only two organs consistently contained E1A DNA: lungs and kidneys. Other organs such as liver, heart, spleen, brain, uterus, or ovaries had no detectable E1A DNA. This result suggests that, under the E1A/DC-Chol treatment conditions, lungs and kidneys are the most susceptible organs for chromosomal integration of E1A DNA, and, more importantly, E1A DNA is undetectable in the uterus and ovaries.

V. E1A GENE THERAPY: CLINICAL TRIALS A. Phase I Breast and Ovarian Cancer with Intracavity Administration To evaluate the feasibility of using E1A gene therapy for patients with HER2-overexpressing cancer, a phase I clinical trial was conducted in a group of patients with advanced breast and ovarian cancers. An E1A/DC-Chol cationic liposome complex was injected weekly into the thoracic or peritoneal cavity of 18 patients. The most common toxicity were fever, nausea, vomiting, and/or discomfort at the injection sites. E1A gene expression was readily detectable in tumor cells that showed a concurrent HER2 downregulation. Analyzing the intracavitary fluid from patients over the course of E1A/DC-Chol treatment, the total number of tumor clumps was significantly decreased after the treatment. In addition, the expression of Ki-67, a nuclear antigen expressed on all human proliferating cells [123], was also decreased, suggesting a decreased proliferation of tumor cells as a result of the E1A/DC-Chol treatment. When apoptosis was analyzed by TUNEL assay, a significant increase in the percentage of apoptosis in tumor cells was seen in all patients analyzed. Thus, the E1A/DC-Chol gene therapy suppresses tumor cell growth by both reducing cell replication and enhancing apoptosis. It is interesting to note that E1A/DC-Chol treatment caused a more drastic effect on the tumor cell clump reduction than on the HER2 downregulation. It is likely, therefore, that other molecular mechanisms also contribute to the E1A-mediated antitumor activity. In the treated patient fluids, TNF-α expression was significantly enhanced, which may account for the increased apoptosis seen in patient tumors. Thus, the phase I data clearly indicate the feasibility of using E1A/DC-Chol gene therapy to treat patients with HER2overexpressing tumors, and the study successfully proves the working concept developed from our preclinical studies [11].

B. Phase II Head and Neck Cancer with Intratumor Administration A multicenter phase II study of E1A gene therapy on head and neck cancers has recently been completed [124]. E1A/DC-Chol complex was used as a single agent and administered by intratumor injection. Among 20 treated patients,

473


5% (1 out of 20) showed a complete response and 45% (9 out of 20) showed an objective response or reaching a state of stable disease. The most common side effect was pain at the injection site, but there were no serious adverse events relating to E1A/DC-Chol administration. Based on the encouraging results of the phase II trials, a possible combined E1A/DCChol therapy with ionizing radiation and/or chemotherapy should be feasible in the near future.

VI. CONCLUSION The E1A-mediated antitumor activity has manifested itself in the ability of E1A to inhibit tumor growth, suppress metastasis and angiogenesis, and sensitize tumor cells to apoptosis induced by therapeutic treatments such as TNF-α, γ -irradiation, and Taxol. The utility of an E1A-based gene therapy was validated by the apparent success of such application in ovarian, breast, and head and neck cancer models without obvious toxic side effect. The study on the molecular mechanisms underlying the E1A-mediated antitumor activity revealed that E1A represses HER2 and Axl transcription and targets the NF-κB pathway that connects HER2 overexpression and the Axl/Gas6-mediated signal pathway and the TNF-α (or γ -irradiation) signal pathway. In addition, E1A inactivates IKK, leading to NF-κB inactivation and subsequent shutting down of the survival program, which sensitizes cells to apoptosis. Based on the success of the phase I breast and ovarian cancer clinical trials using E1A/DC-Chol via intracavity administration, a phase II trial in ovarian cancer patients is currently underway. Also, because advanced breast cancer patients usually have distant metastasis in other organs such as bone and brain, to treat metastatic breast cancer it is imperative that E1A be delivered systemically. Thus, a phase I trial of E1A/DC-Chol by intravenous administration for metastatic breast cancer, patients has been proposed. Moreover, a tumor-specific E1A gene therapy strategy will be valuable to enhance targeting specificity and reduce potential side effects. To that end, several approaches have been developed. For examples, taking advantage of high HER2 promoter activity in many HER2-overexpressing tumors, a HER2 promoter-driven E1A could specifically express E1A in HER2-overexpressing tumor cells [125]. Or, using HER2 antisense iron-responsible element, one could preferentially direct E1A expression in cancer cells that overexpress HER2 mRNA [126]. Another approach is to use HER2-specific binding filamentous bacteriophage to deliver E1A gene into HER2-overexpressing cancer cells [127]. It has been a long journey from the first demonstration of the ability of E1A to repress HER2 transcription up to today when E1A gene therapy is in phase II clinical trials. We now know more about the antitumor activity of E1A than when we first witnessed such activity in HER2overexpressing cancer cells about a decade ago, when the

predominant view about E1A was that it was an “oncogene.” In light of the E1A-mediated sensitization effects, the challenge in future E1A gene therapy is the development of effective combined E1A/chemo- or radiation-therapy strategies supported by solid in vitro and in vivo studies. With a better understanding of the mechanism by which E1A suppresses tumors and a better design of the delivery vehicle, it is hoped that an E1A-based gene therapy could become an effective treatment for cancer patients.

References 1. Nevins, J. R. (1995). Adenovirus E1A: transcription regulation and alteration of cell growth control. Curr. Top. Microbiol. Immunol. 199, 25–32. 2. Jones, N. (1995). Transcriptional modulation by the adenovirus E1A gene. Curr. Top. Microbiol. Immunol. 199, 59–80. 3. Brockmann, D., and Esche, H. (1995). Regulation of viral and cellular gene expression by E1A proteins encoded by the oncogenic adenovirus type 12. Curr. Top. Microbiol. Immunol. 199, 81–112. 4. Moran, E., and Mathews, M. B. (1987). Multiple functional domains in the adenovirus E1A gene. Cell 48, 177–178. 5. Houweling, A., van den Elsen, P. J., and van der Eb, A. J. (1980). Partial transformation of primary rat cells by the leftmost 4.5% fragment of adenovirus 5 DNA. Virology 105, 537–550. 6. Ruley, H. E. (1990). Transforming collaborations between ras and nuclear oncogenes. Cancer Cells 2, 258–268. 7. Chinnadurai, G. (1992). Adenovirus E1A as a tumor-suppressor gene. Oncogene 7, 1255–1258. 8. Frisch, S. M. (1996). Reversal of malignancy by the adenovirus E1A gene. Mutat. Res. 350, 261–266. 9. Mymryk, J. S. (1996). Tumour suppressive properties of the adenovirus 5 E1A oncogene. Oncogene 13, 1581–1589. 10. Yu, D., and Hung, M.-C. (1998). The erbB2 gene as a cancer therapeutic target and the tumor- and metastasis-suppressing function of E1A. Cancer Metastasis Rev. 17, 195–202. 11. Hung. M.-C., Wang, S.-C., and Hortobagyi, G. (1999). Targeting HER2/neu-overexpressing cancer cells with transcriptional repressor genes delivered by cationic liposome, in Nonviral Vectors for Gene Therapy, Huang, L., Hung, M.-C., and Wagner, E., eds., pp. 357–375. Academic Press, San Diego. 12. Slamon, D. J., Clark, G. M., Wong, S. G., Levin, W. J., Ullrich, A., and McGuire, W. L. (1987). Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235, 177–182. 13. Slamon, D. J., and Clark, G. M. (1988). Amplification of c-erbB-2 and aggressive human breast tumors? Science 240, 1795-1798. 14. Slamon, D. J., Godolphin, W., Jones, L. A., Holt, J. A., Wong, S. G., Keith, D. E., Levin, W. J., Stuart, S. G., Udove, J., Ullrich, A., and Press, M. F. (1989). Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 244, 707–712. 15. Xia, W., Lau, Y. K., Zhang, H. Z., Liu, A. R., Li, L., Kiyokawa, N., Clayman, G. L., Katz, R. L., and Hung, M.-C. (1997). Strong correlation between c-ErbB-2 overexpression and overall survival of patients with oral squamous cell carcinoma. Clin. Cancer Res. 3, 3–9. 16. Xia, W., Lau, Y. K., Zhang, H. Z., Xiao, F. Y., Johnston, D. A., Liu, A. R., Li, L., Katz, R. L., and Hung, M.-C. (1999). Combination of EGFR, HER-2/neu, and HER-3 is a stronger predictor for the outcome of oral squamous cell carcinoma than any individual family members. Clin. Cancer Res. 5, 4164–4174. 17. Berchuck, A., Kamel, A., Whitaker, R., Kerns, B., Olt, G., Kinney, R., Soper, J. T., Dodge, R., Clarke-Pearson, D. L., and Marks, P. (1990). Overexpression of HER-2/neu is associated with poor

474

18.

19.

20.

21. 22.

23.

24.

25.

26.

27. 28.

29.

30.

31.

32.

33.


survival in advanced epithelial ovarian cancer. Cancer Res. 50, 4087–4091. Orr, M. S., O’Connor, P. M., and Kohn, K. W. (2000). Effects of c-erbB2 overexpression on the drug sensitivities of normal human mammary epithelial cells. J. Natl. Cancer Inst. 92, 987–994. Tsai, C. M., Chang, K. T., Perng, R. P., Mitsudomi, T., Chen, M. H., Kadoyama, C., and Gazdar, A. F. (1993). Correlation of intrinsic chemoresistance of non-small-cell lung cancer cell lines with HER2/neu gene expression but not with ras gene mutations. J. Natl. Cancer Inst. 85, 897–901. Gusterson, B. A., Gelber, R. D., Goldhirsch, A., Price, K. N., SäveSöderbörgh, S. J., Anbazhagan, R., Styles, J., Rudenstam, C. M., Golouh, R., Reed, R., Marinez-Tello, F., Tiltman, F., Torhorst, J., Grigolato, P., Bettelheim, R., Neville, A. M., Bürki, K., Castiglione, M., Collins, J., Lindtner, J., and Senn, H. J., (1992). Prognostic importance of c-erbB-2 expression in breast cancer. International (Ludwig) Breast Cancer Study Group. J. Clin. Oncol. 10, 1049–1056. Yu, D., and Hung, M.-C., (2000). Role of erbB2 in breast cancer chemosensitivity. Bioessays 22, 673–680. Ciardiello, F., Caputo, R., Pomatico, G., De Laurentiis, M., De Placido, S., Bianco, A. R., and Tortora, G. (2000). Resistance to taxanes is induced by c-erbB-2 overexpression in human MCF-10A mammary epithelial cells and is blocked by combined treatment with an antisense oligonucleotide targeting type I protein kinase. A. Int. J. Cancer 85, 710–715. Baselga, J., Tripathy, D., Mendelsohn, J., Baughman, S., Benz, C. C., Dantis, L., Sklarin, N. T., Seidman, A. D., Hudis, C. A., Moore, J., Rosen, P. P., Twaddell, T., Henderson, I. C., and Norton, L. (1999). Phase II study of weekly intravenous trastuzumab (Herceptin) in patients with HER2/neu-overexpressing metastatic breast cancer. Semin. Oncol. 26, 78–83. Pirollo, K. F., Hao, Z., Rait, A., Ho, C. W., and Chang, E. H., (1997). Evidence supporting a signal transduction pathway leading to the radiation-resistant phenotype in human tumor cells. Biochem. Biophys. Res. Commun. 230, 196–201. Burke, H. B., Hoang, A., Iglehart, J. D., and Marks, J. R. (1998). Predicting response to adjuvant and radiation therapy in patients with early stage breast carcinoma. Cancer 82, 874–877. Doherty, J. K., Bond, C. T., Hua, W., Adelman, J. P., and Clinton, G. M., (1999). An alternative HER-2/neu transcript of 8 kb has an extended 3 UTR and displays increased stability in SKOV-3 ovarian carcinoma cells. Gynecol. Oncol. 74, 408–415. Miller, S. J., and Hung, M.-C. (1995). Regulation of HER-2/neu gene expression. Oncol. Rep. 2, 497–503. Kraus, M. H., Popescu, N. C., Amsbaugh, S. C., and King, C. R. (1987). Overexpression of the EGF receptor-related proto-oncogene erbB-2 in human mammary tumor cell lines by different molecular mechanisms. EMBO J. 6, 605–610. Bosher, J. M., Williams, T., and Hurst, H. C. (1995). The developmentally regulated transcription factor AP-2 is involved in c-erbB-2 overexpression in human mammary carcinoma. Proc. Natl. Acad. Sci. USA 92, 744–747. Bosher, J. M., Totty, N. F., Hsuan, J. J., Williams, T., and Hurst, H. C. (1996). A family of Ap-2 proteins regulates c-erbB-2 experssion in mammary carcinoma. Oncogene 13, 1701–1707. Hollywood, D. P., and Hurst, H. C. (1993). A novel transcription factor, OB2-1, is required for overexpression of the proto-oncogene c-erbB-2 in mammary tumour lines. EMBO J. 12, 2369–2375. Yu, D., Suen, T. C., Yan, D.-H., Chang, L. S., and Hung, M.-C. (1990). Transcriptional repression of the neu protooncogene by the adenovirus 5 E1A gene products. Proc. Natl. Acad. Sci. USA 87, 4499–4503. Yan, D.-H., Chang, L. S., and Hung, M.-C. (1991). Repressed expression of the HER-2/c-erbB-2 proto-oncogene by the adenovirus E1A gene products. Oncogene 6, 343–345.

34. Yu, D., Scorsone, K., and Hung, M.-C. (1991). Adenovirus type 5 E1A gene products act as transformation suppressors of the neu oncogene. Mol. Cell. Biol. 11, 1745–1750. 35. Yu, D., Hamada, J., Zhang, H., Nicolson, G. L., and Hung, M.-C. (1992). Mechanisms of c-erbB2/neu oncogene-induced metastasis and repression of metastatic properties by adenovirus 5 E1A gene products. Oncogene 7, 2263—2270. 36. Yu, D., Wolf, J. K., Scanlon, M., Price, J. E., and Hung, M.-C. (1993). Enhanced c-erbB-2/neu expression in human ovarian cancer cells correlates with more severe malignancy that can be suppressed by E1A. Cancer Res. 53, 891–898. 37. Chang, J. Y., Xia, W., Shao, R., Sorgi, F., Hortobagyi, G. N., Huang, L., and Hung, M.-C. (1997). The tumor suppression activity of E1A in HER-2/neu-overexpressing breast cancer. Oncogene 14, 561–568. 38. Zhang, Y., Yu, D., Xia, W., and Hung, M.-C. (1995). HER-2/neutargeting cancer therapy via adenovirus-mediated E1A delivery in an animal model. Oncogene 10, 1947–1954. 39. Yu, D., Matin, A., Xia, W., Sorgi, F., Huang, L., and Hung, M.-C. (1995). Liposome-mediated in vivo E1A gene transfer suppressed dissemination of ovarian cancer cells that overexpress HER-2/neu. Oncogene 11, 1383−1388. 40. Chang, J. Y., Xia, W. Y., Shao, R. P., and Hung, M.-C. (1996). Inhibition of intratracheal lung cancer development by systemic delivery of E1A. Oncogene 13, 1405–1412. 41. Yu, D., Shi, D., Scanlon, M., and Hung, M.-C. (1993). Reexpression of neu-encoded oncoprotein counteracts the tumor-suppressing but not the metastasis-suppressing function of E1A. Cancer Res. 53, 5784– 5790. 42. Ueno, N. T., Bartholomeusz, C. L., Hermann, J. L., Estrov, Z., Shao, R., Andreeff, M., Price, J., Paul, R. W., Anklesaria, P., Yu, D., and Hung, M.-C. (2000). E1A-mediated paclitaxel sensitization in HER-2/neuoverexpressing ovarian cancer through apoptosis involving caspase-3 pathway. Clin. Cancer Res. 6, 250–259. 43. Ueno, N. T., Yu, D., and Hung, M.-C. (1997). Chemosensitization of HER-2/neu-overexpressing human breast cancer cells to paclitaxel (Taxol) by adenovirus type 5 E1A. Oncogene 15, 953–960. 44. Brader, K. R., Wolf, J. K., Hung, M. C., Yu, D., Crispens, M. A., van Golen, K. L., and Price, J. E. (1997). Adenovirus E1A expression enhances the sensitivity of an ovarian cancer cell line to multiple cytotoxic agents through an apoptotic mechanism. Clin. Cancer Res. 3, 2017–2024. 45. Shao, R., Karunagaran, D., Zhou, B. P., Li, K., Lo, S. S., Deng, J., Chiao, P., and Hung, M.-C. (1997). Inhibition of nuclear factor-kappaB activity is involved in E1A-mediated sensitization of radiation-induced apoptosis. J. Biol. Chem. 272, 32739–32742. 46. Rikitake, Y., and Moran, E. (1992). DNA-binding properties of the E1A-associated 300-kilodalton protein. Mol. Cell. Biol. 12, 2826– 2836. 47. Chen, H., and Hung, M.-C. (1997). Involvement of co-activator p300 in the transcriptional regulation of the HER-2/neu gene. J. Biol. Chem. 272, 6101–6104. 48. Chen, H., Yu, D., Chinnadurai, G., Karunagaran, D., and Hung, M.-C. (1997). Mapping of adenovirus 5 E1A domains responsible for suppression of neu-mediated transformation via transcriptional repression of neu. Oncogene 14, 1965–1971. 49. Borrelli, E., Hen, R., and Chambon, P. (1984). Adenovirus-2 E1A products repress enhancer-induced stimulation of transcription. Nature 312, 608–612. 50. Velcich, A., Kern, F. G., Basilico, C., and Ziff, E. B., (1986). Adenovirus E1A proteins repress expression from polyomavirus early and late promoters. Mol. Cell. Biol. 6, 4019–4025. 51. Hen, R., Borrelli, E., and Chambon, P. (1985). Repression of the immunoglobulin heavy chain enhancer by the adenovirus-2 E1A products. Science 230, 1391–1394.

E1A Cancer Gene Therapy 52. Stein, R. W., and Ziff, E. B. (1987). Repression of insulin gene expression by adenovirus type 5 E1A proteins. Mol. Cell. Biol. 7, 1164– 1170. 53. Funk, W. D., Ouellette, M., and Wright, W. E. (1991). Molecular biology of myogenic regulatory factors. Mol. Biol. Med. 8, 185–195. 54. Hudziak, R. M., Lewis, G. D., Winget, M., Fendly, B. M., Shepard, H. M., and Ullrich, A. (1989). p185HER2 monoclonal antibody has antiproliferative effects in vitro and sensitizes human breast tumor cells to tumor necrosis factor. Mol. Cell. Biol. 9, 1165–1172. 55. Zhou, B. P., Hu, M. C., Miller, S. A., Yu, Z., Xia, W., Lin, S. Y., and Hung, M.-C. (2000). HER-2/neu blocks tumor necrosis factor-induced apoptosis via the Akt/NF-kappaB pathway. J. Biol. Chem. 275, 8027– 8031. 56. Hu, P., Margolis, B., Skolinik, E. Y., Lammers, R., Ullrich, A., and Schlessinger, J. (1992). Interaction of phosphatidylinositol 3-kinaseassociated p85 with epidermal growth factor and platelet-derived growth factor receptors. Mol. Cell. Biol. 12, 981–990. 57. Franke, T. F., Kaplan, D. R., and Cantley, L. C. (1997). PI3K: downstream AKTion blocks apoptosis. Cell 88, 435–437. 58. Downward, J. (1998). Mechanisms and consequences of activation of protein kinase B/Akt. Curr. Opin. Cell. Biol. 10, 262–267. 59. Beg, A. A., and Baltimore, D. (1996). An essential role for NF-kappaB in preventing TNF-alpha-induced cell death. Science 274, 782–784. 60. Wang, C. Y., Mayo, M. W., and Baldwin, Jr., A. S. (1996). TNF- and cancer therapy-induced apoptosis: protentiation by inhibition of NFkappaB. Science 274, 784–787. 61. Van Antwerp, D. J., Martin, S. J., Kafri, T., Green, D. R., and Verma, I. M. (1996). Suppression of TNF-alpha-induced apoptosis by NFkappaB. Science 274, 787–789. 62. Liu, Z. G., Hsu, H., Goeddel, D. V., and Karin, M. (1996). Dissection of TNF receptor 1 effector functions: JNK activation is not linked to apoptosis while NF-kappaB activation prevents cell death. Cell 87, 565–576. 63. Ozes, O. N., Mayo, L. D., Gustin, J. A., Pfeffer, S. R., Pfeffer, L. M., and Donner, D. B. (1999). NF-kappaB activation by tumour necrosis factor requires the Akt serine-threonine kinase. Nature 401, 82–85. 64. Delhase, M., Li, N., and Karin, M. (2000). Kinase regulation in inflammatory response. Nature 406, 367–368. 65. Madge, J. A., and Pober, J. S. (2000). A phosphatidylinositol 3kinase/Akt pathway, activated by tumor necrosis factor or interleukin-1, inhibits apoptosis but does not activate NFkappaB in human endothelial cells. J. Biol. Chem. 275, 15458–15465. 66. Baldwin, A. S. J. (1996). The NF-κB and IκB proteins: New discoveries and insights. Annu. Rev. Immunol. 14, 649–681. 67. Mercurio, F., Zhu, H., Murray, B. W., Shevchenko, A., Bennett, B. L., Li, J., Young, D. B., Barbosa, M., Mann, M., Manning, A., and Rao, A. (1997). IKK-1 and IKK-2: cytokine-activated IkappaB kinases essential for NF-kappaB activation. Science 278, 860–866. 68. Traenckner, E. B., Pahl, H. L., Henkel, T., Schmidt, K. N., Wilk, S., and Baeuerle, P. A. (1995). Phosphorylation of human I kappa B-alpha on serines 32 and 36 controls I kappa B-alpha proteolysis and NF-kappa B activation in response to diverse stimuli. EMBO J. 14, 2876–2883. 69. Hung, M.-C., and Yu, D. (2000). Therapeutic resistance of erb B-2overexpressing cancers and strategies to overcome this resistance, in DNA Alterations in Cancer (M. Ehrlich, ed.), pp. 457–470. Eaton publishing, Natick, MA. 70. Pegram, M. D., Finn, R. S., Arzoo, K., Beryt, M., Pietras, R. J., and Slamon, D. J. (1997). The effect of HER-2/neu overexpression on chemotherapeutic drug sensitivity in human breast and ovarian cancer cells. Oncogene 15, 537–547. 71. Yu, D., Liu, B. L., Tan, M., Li, J. Z., Wang, S. S., and Hung, M.-C. (1996). Overexpression of c-erbB-2/neu in breast cancer cells confers increased resistance to taxol via MDR-1-independent mechanisms. Oncogene 13, 1359–1365.

475

72. Ciardiello, F., McGeady, M. L., Kim, N., Basolo, F., Hynes, N., Langton, B. C., Yokozaki, H., Saeki, T., Elliott, J. W., Masui, H. et al. (1990). Transforming growth factor-alpha expression is enhanced in human mammary epithelial cells transformed by an activated c-Ha-ras protooncogene but not by the c-neu protooncogene, and overexpression of the transforming growth factor-alpha complementary DNA leads to transformation. Cell Growth Differ. 1, 407– 420. 73. Ciardiello, F., Gottardis, M., Basolo, F., Pepe, S., Normanno, N., Dickson, R. B., Bianco, A. R., and Salomon, D. S. (1992). Additive effects of c-erbB-2, c-Ha-ras, and transforming growth factor-alpha genes on in vitro transformation of human mammary epithelial cells. Mol. Carcinog. 6, 43–52. 74. de Stanchina, E., McCurrach, M. E., Zindy, F., Shieh, S. Y., Ferbeyre, G., Samuelson, A. V., Prives, C., Roussel, M. F., Sherr, C. J., and Lowe, S. W. (1998). E1A signaling to p53 involves the p19 (ARF) tumor suppressor. Genes Dev. 12, 2434–2442. 75. Frisch, S. M., and Dolter. K. E. (1995). Adenovirus E1A-mediated tumor suppression by a c-erbB-2/neu-independent mechanism. Cancer Res. 55, 5551–5555. 76. Pozzatti, R., McCormick, M., Thompson, M. A., and Khoury, G. (1988). The E1A gene of adenovirus type 2 reduces the metastatic potential of ras-transformed rat embryo cells. Mol. Cell. Biol. 8, 2984– 2988. 77. Pozzatti, R., Muschel, R., Williams, J., Padmanabhan, R., Howard, B., Liotta, L., and Khoury, G. (1986). Primary rat embryo cells transformed by one or two oncogenes show different metastatic potentials. Science 232, 223–227. 78. Steeg, P. S., Bevilacqua, G., Pozzatti, R., Liotta, L. A., and Sobel, M. E. (1988). Altered expression of NM23, a gene associated with low tumor metastatic potential, during adenovirus 2 E1A inhibition of experimental metastasis. Cancer Res. 48, 6550–6554. 79. Steeg, P. S., Bevilacqua, G., Kopper, L., Thorgeirsson, U. P., Talmadge, J. E., Liotta, L. A., and Sobel, M. E. (1988). Evidence for a novel gene associated with low tumor metastatic potential. J. Natl. Cancer Inst. 80, 200–204. 80. Frisch, S. M., Reich, R., Collier, I. E., Genrich, L. T., Martin, G., and Goldberg, G. I. (1990). Adenovirus E1A represses protease gene expression and inhibits metastasis of human tumor cells. Oncogene 5, 75–83. 81. Frisch, S. M. (1994). E1A induces the expression of epithelial characteristics. J. Cell. Biol. 127, 1085–1096. 82. Hennig, G., Behrens, J., Truss, M., Frisch, S., Reichmann, E., and Birchmeier, W. (1995). Progression of carcinoma cells is associated with alterations in chromatin structure and factor binding at the Ecadherin promoter in vivo. Oncogene 11, 475–484. 83. Santoro, M., Battaglia, C., Zhang, L., Carlomagno, F., Martelli, M. L., Salvatore, D., and Fusco, A. (1994). Cloning of the rat tissue inhibitor of metalloproteinases type 2 (TIMP-2) gene: analysis of its expression in normal and transformed thyroid cells. Exp. Cell. Res. 213, 398– 403. 84. Offringa, R., Smits, A. M., Houweling, A., Bos, J. L., and van der Eb, A. J. (1988). Similar effects of adenovirus E1A and glucocorticoid hormones on the expression of the metalloprotease stromelysin. Nucleic Acids Res. 16, 10973–10984. 85. Linder, S., Popowicz, P., Svensson, C., Marshall, H., Bondesson, M., and Akusjarvi, G. (1992). Enhanced invasive properties of rat embryo fibroblasts tansformed by adenovirus E1A mutants with deletions in the carboxy-terminal exon. Oncogene 7, 439–443. 86. Garbisa, S., Pozzatti, R., Muschel, R. J., Saffiotti, U., Ballin, M., Goldfarb, R. H., Khoury, G., and Liotta, L. A. (1987). Secretion of type IV collagenolytic protease and metastatic phenotype: induction by transfection with c-Ha-ras but not c-Ha-ras plus Ad2-E1A. Cancer Res. 47, 1523–1528.

476


87. Offringa, R., Gebel, S., van Dam, H., Timmers, M., Smits, A., Zwart, R., Stein, B., Bos, J. L., van der Eb, A., and Herrlich, P. (1990). A novel function of the transforming domain of E1A: repression of AP-1 activity. Cell 62, 527–538. 88. Bernhard, E. J., Muschel, R. J., and Hughes, E. N. (1990). Mr 92,000 gelatinase release correlates with the metastatic phenotype in transformed rat embryo cells. Cancer Res. 50, 3872–3877. 89. Bernhard, E. J., Hagner, B., Wong, C., Lubenski, I., and Muschel, R. J. (1995). The effect of E1A transfection on MMP-9 expression and metastatic potential. int. J. Cancer 60, 718–724. 90. Hofmann, M., Rudy, W., Gunthert, U., Zimmer, S. G., Zawadzki, V., Zoller, M., Lichtner, R. B., Herrlich, P., and Ponta, H. (1993). A link between ras and metastatic behavior of tumor cells: ras induces CD44 promoter activity and leads to low-level expression of metastasis-specific variants of CD44 in CREF cells. Cancer Res. 53, 1516–1521. 91. Yu, D., Wang, S. S., Dulski, K. M., Tsai, C. M., Nicolson, G. L., and Hung, M.-C. (1994). C-erbB-2/neu overexpression enhances metastatic potential of human lung cancer cells by induction of metastasis-associated properties. Cancer Res. 54, 3260–3266. 92. Tan, M., Yao, J., and Yu, D. (1997). Overexpression of the c-erbB-2 gene enhanced intrinsic metastasis potential in human breast cancer cells without increasing their transformation abilities. Cancer Res. 57, 1199–1205. 93. O’Bryan, J. P., Frye, R. A., Cogswell, P. C., Neubauer, A., Kitch B., Prokop, C., Espinosa, R. D., Le Beau, M. M., Earp, H. S., and Liu, E. T. (1991). Ax1, a transforming gene isolated from primary human myeloid leukemia cells, encodes a novel receptor tyrosine kinase. Mol. Cell. Biol. 11, 5016–5031. 94. Fantl, W. J., Johnson, D. E., and Williams, L. T. (1993). Signalling by receptor tyrosine kinases. Annu. Rev. Biochem. 62, 453–481. 95. van der Geer, P., Hunter, T., and Lindberg, R. A. (1994). Receptor protein-tyrosine kinases and their signal transduction pathways. Annu. Rev. Cell. Biol. 10, 251–337. 96. Lee, W.-P., Liao, Y., Robinson, D., Kung, H. J., Liu, E. T., and Hung, M.-C. (1999). Ax1–gas6 interaction counteracts E1A-mediated cell growth suppression and proapoptotic activity. Mol. Cell. Biol. 19, 8075–8082. 97. Manfioletti, G., Brancolini, C., Avanzi, G., and Schneider, C. (1993). The protein encoded by a growth arrest-specific gene (gas6) is a new member of the vitamin K-dependent proteins related to protein S, a negative coregulator in the blood coagulation cascade. Mol. Cell. Biol. 13, 4976–4985. 98. Stitt, T. N., Conn, G., Gore, M., Lai, C., Bruno, J., Radziejewski, C., Mattsson, K., Fisher, J., Gies, D. R., Jones, P. F. et al. (1995). The anticoagulation factor protein S and its relative, Gas6, are ligands for the Tyro 3/Axl family of receptor tyrosine kinases. Cell 80, 661– 670. 99. Varnum, B. C., Young, C., Elliott, G., Garcia, A., Bartley, T. D., Fridell, Y. W., Hunt, R. W., Trail, G., Clogston, C., Toso, R. J. et al. (1995). Axl receptor tyrosine kinase stimulated by the vitamin K-dependent protein encoded by growth-arrest-specific gene 6. Nature 373, 623–626. 100. Goruppi, S., Ruaro, E., Varnum, B., and Schneider, C. (1997). Requirement of phosphatidylinositol 3-kinase-dependent pathway and Src for Gas6-Axl mitogenic and survival activities in NIH 3T3 fibroblasts. Mol. Cell. Biol. 17, 4442–4453. 101. Datta, S. R., Dudek, H., Tao, X., Masters, S., Fu, H., Gotoh, Y., and Greenberg, M. E. (1997). Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91, 231–241. 102. del Peso, L., Gonzalez-Garcia, M., Page, C., Herrera, R., and Nunez, G. (1997). Interleukin-3-induced phosphorylation of BAD through the protein kinase Akt. Science 278, 687–689. 103. Romashkova, J. A., and Makarov, S. S. (1999). NF-kappaB is a target of AKT in anti-apoptotic PDGF signalling. Nature 401, 86–90.

104. Brunet, A., Bonni, A., Zigmond, M. J., Lin, M. Z., Juo, P., Hu, L. S., Anderson, M. J., Arden, K. C., Blenis, J., and Greenberg, M. E. (1999). Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor. Cell 96, 857–868. 105. Chen, M. J., Holskin, B., Strickler, J., Gorniak, J., Clark, M. A., Johnson, P. J., Mitcho, M., and Shalloway, D. (1987). Induction by E1A oncogene expression of cellular susceptibility to lysis by TNF. Nature 330, 581–583. 106. Duerksen-Hughes, P., Wold, W. S., and Gooding, L. R. (1989). Adenovirus E1A renders infected cells sensitive to cytolysis by tumor necrosis factor. J. Immunol. 143, 4193–4200. 107. Lowe, S. W., and Ruley, H. E. (1993). Stabilization of the p53 tumor suppressor is induced by adenovirus 5 E1A and accompanies apoptosis. Genes Dev. 7, 535–545. 108. Foo, S. Y., Nolan, G. P. (1999). NF-κB to the rescure: RELs , apoptosis and cellular transformation. TIG 15, 229–235. 109. Shao, R., Hu, M. C., Zhou, B. P., Lin, S. Y., Chiao, P. J., von Lindern, R. H., Spohn, B., and Hung M.-C. (1999). E1A sensitizes cells to tumor necrosis factor-induced apoptosis through inhibition of IkappaB kinases and nuclear factor kappaB activities. J. Biol. Chem. 274, 21495–21498. 110. Siebenlist, U., Franzoso, G., and Brown, K. (1994). Structure, regulation and function of NF-kappa B. Annu. Rev. Cell. Biol. 10, 405–455. 111. Baichwal, V. R., and Baeuerle, P. A. (1997). Activate NF-kappa B or die? Curr. Biol. 7, R94–R96. 112. Sonenshein, G. E. (1997). Rel/NF-kappa B transcription factors and the control of apoptosis. Semin. Cancer Biol. 8, 113–119. 113. Karin, M., and Ben-Neriah, Y. (2000). Phosphorylation meets ubiquitination: the control of NF-[kappa]B activity. Annu. Rev. Immunol. 18, 621–663. 114. Gilmore, T. D., Koedood, M., Piffat, K. A., and White, D. W. (1996). Rel/NF-kappaB/IkappaB proteins and cancer. Oncogene 13, 1367– 1378. 115. Luque, I., and Gelinas, C. (1997). Rel/NF-kappa B and I kappa B factors in oncogenesis. Semin. Cancer Biol. 8, 103–111. 116. Culver, K. W., Ram, Z., Wallbridge, S., Ishii, H., Oldfield, E. H., and Blaese, R. M. (1992). In vivo gene tansfer with retroviral vectorproducer cells for treatment of experimental brain tumors. Science 256, 1550–1552. 117. Mesnil, M., Piccoli, C., Tiraby, G., Willecke, K., and Yamasaki, H. (1996). Bystander killing of cancer cells by herpes simplex virus thymidine kinase gene is mediated by connexins. Proc. Natl. Acad. Sci. USA 93, 1831–1835. 118. Shao, R., Xia, W., and Hung, M. C. (2000). Inhibition of angiogenesis and induction of apoptosis are involved in E1A-mediated bystander effect and tumor suppression. Cancer Res. 60, 3123–3126. 119. Gao, X., and Huang, L. (1991). A novel cationic liposome reagent for efficient transfection of mammalian cells. Biochem. Biophys. Res. Commun. 179, 280–285. 120. Xing, X., Zhang, S., Chang, J. Y., Tucker, S. D., Chen, H., Huang, L., and Hung, M.-C. (1998). Safety study and characterization of E1Aliposome complex gene delivery in an ovarian cancer model. Gene Ther. 5, 1538–1544. 121. Xing, X., Liu, V., Xia, W., Stephens, L. C., Huang, L., Lopez-Berestein, G., and Hung, M.-C. (1997). Safety studies of the intraperitoneal injection of E1A–liposome complex in mice. Gene Ther. 4, 238–243. 122. Ruley, H. E. (1983). Adenovirus early region 1A enables viral and cellular transforming genes to transform primary cells in culture. Nature 304, 602–606. 123. Gerdes, J., Schwab, U., Lemke, H., and Stein, H. (1983). Production of a mouse monoclonal antibody reactive with a human nuclear antigen associated with cell proliferation. Int. J. Cancer 31, 13–20. 124. Reynolds, T. C., Alberts, D., Gershenson, D., Gleich, L., Glisson, B., Hanna, E., Huang, L., Hung, M.-C., Kenady, D., Ueno, N., Villaret, D.,

E1A Cancer Gene Therapy and Yoo, G. (2000). Activity of E1A in human clinical trials, in ASCO 19, 461a. 125. Pandha, H. S., Martin, L. A., Rigg, A., Hurst, H. C., Stamp, G. W., Sikora, K., and Lemoine, N. R. (1999). Genetic prodrug activation therapy for breast cancer: a phase I clinical trial of erbB-2-directed suicide gene expression. J. Clin. Oncol. 17, 2180–2189.

477

126. Yan, D.-H. (1998). Targeting human breast cancer cells that overexpress HER-2/neu mRNA by an antisense iron responsive element. Biochem. Biophys. Res. Commun. 246, 353–358. 127. Poul, M. A., and Marks, J. D. (1999). Targeted gene delivery to mammalian cells by filamentous bacteriophage. J. Mol. Biol. 288, 203– 211.


P A R T VI

PRODRUG ACTIVATION STRATEGIES FOR GENE THERAPY OF CANCER


C

H

A

P

T

E

R

31 Preemptive and Therapeutic Uses of Suicide Genes for Cancer and Leukemia

I. II. III. IV.

FREDERICK L. MOOLTEN

PAULA J. MROZ

Edith Nourse Rogers Memorial Veterans Hospital Bedford, Massachusetts Boston University School of Medicine Boston, Massachusetts 02118

Edith Nourse Rogers Memorial Veterans Hospital Bedford, Massachusetts 02188

Suicide genes constitute an alternative approach. Rather than manipulating, positively or negatively, existing cellular functions, they introduce new functions that sensitize cells to drugs at concentrations that would otherwise be innocuous. Most suicide genes encode enzymes that catalyze the conversion of prodrugs into cytotoxic antimetabolites. The best known among these genes, the herpes thymidine kinase (HSV-TK) gene, sensitizes cells to the guanosine analog ganciclovir (GCV) as a consequence of HSV-TK-catalyzed phosphorylation of GCV to intermediates that lethally inhibit DNA synthesis. Since the initial reports introducing the suicide gene concept [1,2], many animal studies have demonstrated that systemically administered GCV can eradicate transplanted tumors bearing transduced HSV-TK genes (reviewed in Moolten [3] and Tiberghien [4])(Fig. 1). A serendipitous property of the HSV-TK/GCV combination is the bystander effect, a phenomenon that manifests itself as an ability of GCV to kill not only HSV-TK transduced cells but also untransduced cells in their proximity. The mechanism probably involves the transfer of activated GCV metabolites [1,5–8], at least in vitro, although stimulation of host immune/inflammatory reactions and damage to tumor blood vessels may also play a role in vivo. To the extent that immune phenomena are involved, systemic antitumor effects may sometimes be observable. Numerous other suicide gene/prodrug systems have since been described [9–23]. Some of the better characterized combinations are listed in Table 1. Of interest, p450-2B1 and nitroreductase genes generate products that are not antimetabolites but alkylating agents and therefore potentially more effective than antimetabolites in quiescent cells. The products of the Fas/FKBP and caspase/FKBP fusion genes

Introduction 481 Therapeutic Uses of Suicide Genes 482 Preemptive Uses of Suicide Genes in Cancer 483 Creation of Stable Suicide Functions by Combining Suicide Gene Transduction with Endogenous Gene Loss 485 A. Loss of Purine or Thymidine Salvage Pathways Creates Chemosensitivity 485 B. Stability of Suicide Functions in HPRT-Negative/gpt-Positive Cells 486 C. Stability of Suicide Functions in TK-Negative/HSV-TK-Positive Cells 487

V. Preemptive Uses of Suicide Genes To Control Graft-Versus-Host Disease in Leukemia 487 VI. Future Prospects for Preemptive Use of Suicide Genes 488 References 489

I. INTRODUCTION The emergence of cancer gene therapy as a new discipline bears testimony to a need unmet by conventional therapies: selectivity. Cytokine gene therapy, suppressor genes, and antisense/ribozymes each aim at targeting cancer cells selectively. Implicit in these approaches is the presumption that there will be something about neoplastic cells that distinguishes them sufficiently from vital normal cells to permit therapeutic modalities to suppress or kill them without subjecting their normal counterparts to intolerable host toxicity. The presumption is probably true for some cancers but false for others, perhaps for a majority.


481


482

Frederick L. Moolten and Paula J. Mroz

II. THERAPEUTIC USES OF SUICIDE GENES

FIGURE 1 Differential effects of GCV on HSV-TK-positive and -negative tumors in the same mouse. HSV-TK-positive sarcoma cells were injected subcutaneously into the right flank and HSV-TK-negative cells into the left flank. (A) AT 13 days, small tumors were observed at each site. (B) At day 16, the tumors were growing progressively; an 8-day course of GCV administered intraperitoneally was begun. (C) By day 23, the genepositive tumor had shrunk while the negative tumor had enlarged. (D) At day 37, the positive tumor had regressed completely, while the negative tumor continued to grow. (From Moolten, F. L., Cancer Res., 46, 5276–5281, 1986. With permission.)

induce apoptosis. Both components of each pair are human proteins; similarly, p450-2B1, although of rat origin, has human p450 counterparts. Fas/FKBP, caspase/FKBP, and p450 genes are therefore less likely to provoke host immune reactions against transduced cells than other suicide genes that generate proteins of bacterial or viral origin.

In theory, suicide genes can be used both therapeutically in cancer patients and preemptively in individuals not yet afflicted with a cancer, as described below. Clinical trials to date, however, have been limited to patients with established malignancies. These trials have principally involved tumors at limited sites, including brain tumors [24,25], ovarian cancer that has extended to the peritoneal cavity [26 –28], and mesotheliomas [29]. Most of the trials utilize the HSVTK /GCV combination, with the majority of these employing a modified virus as a vehicle (“vector”) for introducing the HSV-TK gene into tumor cells after intratumoral injection or other instillation techniques that restrict the gene to the known location of the tumor. HSV-TK transduction is then followed by systemic GCV administration. Suicide gene trials are included in a comprehensive listing of gene therapy trials compiled as of late 1999 [30]. Most ongoing clinical trials have utilized one of two different vector systems to transduce the tumor cells [31]. In each case, the vector consists of a virus capable of infecting human cells that has been genetically engineered to eliminate genes responsible for viral replication and cellular pathology, substituting in their place the suicide gene to be used for therapy. The first system entails the use of vectors derived from murine retroviruses. Retroviral vectors mediate transduction that is relatively stable, at least in the short term, as a consequence of the integration of vector sequences into the DNA of the host genome, but to date it has been difficult to produce cellfree suspensions containing these vectors at titers sufficient to yield more than minimal transduction levels in vivo. Because of this limitation, most protocols have not attempted to introduce the vectors themselves into tumors, but rather producer cells, which are murine fibroblasts that generate and release the HSV-TK vectors at their in vivo injection site to yield a continuous supply until the cells are rejected by the host or killed by the administration of GCV. The second system entails the use of vectors derived from human adenoviruses. Because adenoviral vectors do not integrate into genomic DNA, they mediate only transient transduction but possess the advantage of high titers that obviate the need for producer cells. Neither the adenoviral nor retroviral vector system, however, is currently capable of transducing suicide genes into more than a minority of tumor cells in vivo. A major component of the rationale underlying current trials is the expectation that bystander effects might permit GCV to eradicate untransduced cells by virtue of their proximity to transduced cells. Most of the clinical trials are in early stages. To date, reported results include signs of tumor regression in some individuals [32], but few patients have experienced significant clinical benefit. A clear limitation is the difficulty of delivering suicide genes to all areas of a large tumor, even if the tumor has not metastasized [33]. This problem is not

483

Preemptive and Therapeutic Uses of Suicide Genes for Cancer and Leukemia

TABLE 1 Suicide Gene/Prodrug Combinations Gene

Prodrug

Active product

Ref.

HSV-TK

GCV

GCV mono- and diphosphates

Cytosine deaminase

5-Fluorocytosine

5-Fluorouracil

[3] [9]

Gpt

6-Thioxanthine

6-Thioxanthine ribonucleotide

[10]

p450-2B1

Cyclophosphamide

Phosphoramide mustard

Purine nucleoside phosphorylase

6-Methylpurine deoxyribonucleoside

6-Methylpurine

[13]

Deoxycytidine kinase

Ara-C

Ara-C monophosphate

[14]

Nitroreductase

CB1954

5-Azaridin-1-yl-4-hydroxylamino2-nitrobenzamide

[15]

Fas-FKBP

AP1903a

Multimerized fas

Caspase-FKBP

AP1903a

Multimerized caspase

[18,19]

Sodium/iodide symporter

Radioiodide

Concentrated intracellular radioiodide

[20,21]

Carboxypeptidase

Peptide-linked alkylating agent or methotrexate

Free alkylating agent or methotrexate

[22,23]

[11,12]

[16,17]

a Strictly

speaking, AP1903 is not a prodrug, as it is not activated by the product of the suicide gene but rather activates that product by cross-linking it to form the multimers needed for the Fas or caspase proteins to trigger apoptotic pathways. FKBP is an abbreviation for FK506 binding protein.

fully solved by bystander effects, as these effects tend to be powerful only at short ranges. Another obvious limitation of a localized injection approach stems from the fact that the lethality of most cancer results from metastatic rather than localized disease. Metastatic disease will require systemic approaches that expose normal as well as neoplastic cells to the therapeutic modality. Attempts to address this problem include the linkage of suicide genes to promoters that might be highly active in tumor cells, with little or no activity in vital normal tissues. These include a tyrosinase promoter for melanomas [34,35], an alphafetoprotein promoter for hepatomas [36,37], Ebstein–Barr virus (EBV)-encoded transcriptional regulatory elements for EBV-related lymphomas and other EBV-associated malignancies [38,39], an osteocalcin promoter for osteosarcomas [40], and an ErbB2 promoter for breast carcinomas (based on evidence that a subset of breast cancers may exhibit ErbB2 promoter hyperactivity [41]). Promising initial evidence for therapeutic specificity has been reported in murine systems involving transplanted tumors [34,35,37,42], including a reduction in lung metastases of a melanoma after intravenous administration of a retroviral HSV-TK vector followed by GCV therapy [35]. It remains to be determined how much specificity might be achievable with these promoters in human cancers that arise endogenously and whether or not these genes can be delivered in bulk to metastatic deposits in sufficient quantity and uniformity to ensure tumor eradication by prodrug therapy. In addition, these cancers are exceptional; most cancers have yet to exhibit evidence of promoter activities unshared by vital normal stem cells.

III. PREEMPTIVE USES OF SUICIDE GENES IN CANCER Our recent work in murine systems has focused on exploring the feasibility of a different application of suicide genes: their preemptive use before a cancer develops, with particular emphasis on individuals at excessive risk for cancer. The goal of preemption is to achieve selectivity without requiring neoplastic cells to possess the one property whose frequent absence has confounded other approaches to cancer therapy, genetic or otherwise — a targetable difference from normal cells. To obviate the need for targetability, preemption is designed to exploit the clonal (i.e., single cell) origin of human cancers [43– 45] by introducing suicide genes not into an established cancer but into a tissue from which cancers may arise. Because it is clonal, any cancer that subsequently arises within that tissue from a transduced cell should uniformly carry the suicide gene in all its cells as a clonal property, including metastases. Within a transduced clone of cancer cells, suicide gene expression might be lost in an occasional cell through mutations that delete or inactivate the gene, but in theory such cells might be susceptible to bystander killing by their proximity to gene-positive cells. The several studies that report the curability of tumors that arise from transplanted clonal populations of HSV-TK-positive tumor cells [1,2,6,46], even when the tumors are known to harbor gene-negative mutants [1,46,47], are consistent with this expectation. In a nonvital tissue such as breast or prostate epithelium, preemption can aim at transducing a chosen suicide gene into

484


as many cells as possible to maximize the probability that a subsequent cancer will arise from a transduced cell. Cells outside the transduced tissue would remain unsensitized, and measures to promote selectivity within the transduced tissue itself would be unnecessary, as loss of nonneoplastic breast or prostate epithelial cells during cancer therapy would not be life threatening. For preemptive sensitization of a vital tissue such as bone marrow or gastrointestinal epithelium, selectivity must be achieved differently by introducing one or more suicide genes in mosaic rather than homogeneous fashion [1,48]. Mosaicism creates selectivity by ensuring that whatever cell later spawns a cancer will share its clonal sensitivity pattern with only a fraction of the normal cells (Fig. 2). As a first step in testing the preemption paradigm, we have asked whether suicide gene transduction into cells that are not yet malignant might permit effective therapy of cancers that later arose from them [49]. TM4 is a line of preneo-

plastic murine mammary epithelial cells that can be propagated in tissue culture for subsequent in vivo transplantation [50,51]. A retroviral vector, STK [2], was used to transduce the HSV-TK gene into these cells in vitro. The cells were then injected subcutaneously into syngeneic BALB/c mice, where they formed small, nongrowing nodules from which cancers later arose in 40% of the mice. When the mice were treated with GCV, 7 out of 20 responded with complete and durable tumor regressions, and the remainder exhibited a significant retardation of tumor growth (Table 2). Control tumors (transduced and untreated, or untransduced and GCV-treated) invariably exhibited progressive growth. In comparison with controls, the HSV-TK gene by itself exerted no adverse effects on cancer incidence or growth rates, indicating that its presence was not a liability for the preemptively transduced preneoplastic cells and that its observable therapeutic effects operated through GCV.

FIGURE 2 Preemptive introduction of suicide genes in mosaic fashion to ensure that later cancers are sensitized to eradication by a prodrug while their tissue of origin is protected by the presence of cells that do not share their sensitivity [48]. In the diagram, genes A and B are suicide genes that sensitize cells to prodrugs A and B, respectively. The clonal origin of a cancer arising from one of the B-Sensitized cells renders it uniformly sensitive to the prodrug B. Normal cells with the same sensitivity are also killed, but the remaining normal cells in the mosaic survive to repopulate the tissue. It should be noted that the principle of mosaicism illustrated in the diagram with suicide genes can also be implemented with resistance genes [48]. Thus, for example, if genes A and B encode resistance to drugs A and B, administration of drug A will selectively kill the cells that lack gene A, and drug B will kill cells that lack gene B. Unlike sensitivity mosaicism, however, mosaicism created by resistance genes cannot utilize drug doses that fully exploit the difference between sensitive and resistant cells without risking excessive drug toxicity to other tissues that have not acquired the genes. A potential advantage of resistance mosaicism is stability. Thus, a cell that carries a suicide gene can rid itself of its selecive sensitivity through loss or inactivation of the gene via a one-step mutation, whereas a cell that is selectively sensitive because it lacks a gene for high-level drug resistance will in many cases require multiple independent mutations to achieve that level of resistance.


TABLE 2 GCV Therapy of Tumors Arising from Preneoplastic Mammary Epithelial Cells GCV therapy of subsequent tumorsa

Mice with durable tumor regressions/totalb

Median survival (days)

HSV-TK transducedc

0 +

0/25 7/20

46 152

Untransduced

0 +

0/8 0/11

70 72

Transduced with control vectorc

0 +

0/8 0/7

38 53

Preneoplastic cells injected

a Tumor-bearing mice received 150 mg/kg GCV twice daily for 5 days by intraperitoneal injection. b Seven GCV-treated mice in the HSV-TK group exhibited durable regressions, defined as complete tumor regressions without recurrence over a 300-day observation interval; tumor regressions were not observed in any of the other mice. c The STK vector used to transduce the HSV-TK gene was constructed by inserting this gene into vector LNL6 [2]; the latter was used as an HSV-TK negative control. Source: Adapted from Moolten et al. [49].

The results represent the first experimental validation of the principle of preemption, demonstrating that a process applied to premalignant cells could alter the response to therapy of a future cancer. They also illustrate a number of obstacles that stand between this principle and its effective human implementation. The majority of mice were not cured. In tumors that were not eradicated, HSV-TK enzyme activity was low, consistent with an in vivo downregulation of gene expression that occurred during the brief (weeks to months) interval of experimental observation. Durable regressions were limited to small tumors; larger ones responded only with growth delays. Finally, the study was feasible only because the epithelial cells at risk for cancer could be cultivated and transduced in vitro and later reintroduced into host mice, thus obviating the need to reach mammary epithelial cells in situ. These limitations define issues that must ultimately be addressed to convert the principle of preemption into a modality that can be applied to individuals at risk for breast cancer or other malignancies. Paramount among them is the need to achieve high-efficiency integration of suicide genes into the genomic DNA of tissues in vivo and the need to improve the long-term stability of suicide functions in cells harboring the integrated genes beyond what is currently achievable with retroviral transduction. Stable chemosensitivity is threatened not only by changes in gene regulation that cells experience consequent to exposure to an in vivo environment but also by mutations that permanently delete or inactivate transduced genes. An additional concern, the possibility of long-term ill effects of transduction by retroviral vectors (including oncogenesis), has been ameliorated by theoretical calculations [52] and by the absence of vector-related cancers over the course of multiple gene therapy studies in animals and humans [53,54].

485

The efficiency problem is one that has long vexed much of the gene therapy field and may yield only to the eventual development of redesigned, and perhaps synthetic, vectors. Given the constraints of available technology, we have recently focused on the second issue: long-term stability, limiting our current efforts to cells that can be manipulated ex vivo. Cells in this category include lymphocytes that might be transfused into recipients after ex vivo manipulation, embryonic or tissue-specific stem or progenitor cells of various lineages (hematopoietic, mesenchymal, neural, etc.) that might be cultured as a source of tissue replacement, and cells transplanted as a source of therapeutic genes (for example, genes for Factor IX in some hemophiliacs, growth hormone in deficient individuals, insulin in diabetics, angiogenic factors for cardiovascular disease, or antiangiogenic factors for cancer). Introduction of a suicide gene into such cells is an attractive prospect as a “fail-safe” maneuver to permit their subsequent ablation if they later exhibit malignant or other aberrant behavior [52,55]. Critical to the prospective use of genes in this fashion is the requirement that all, or almost all sensitized cells and their progeny maintain normal viability and retain their chemosensitivity over intervals that may range from months to years and encompass many cell generations. Steps to preserve the gene-bearing cells include the use of genes that encode nonimmunogenic proteins or the use of conditions that promote tolerance to potentially immunogenic proteins. Steps to preserve chemosensitivity include strategies to maximize the persistence of suicide genes by transducing multiple gene copies or to preserve their expression through the use of vectors constructed to render promoter regions insusceptible to methylation or other “silencing” mechanisms. As described later, an additional approach we have explored is designed to maintain stable chemosensitivity by using suicide genes as replacements for, rather than additions to, selected cellular functions.

IV. CREATION OF STABLE SUICIDE FUNCTIONS BY COMBINING SUICIDE GENE TRANSDUCTION WITH ENDOGENOUS GENE LOSS A. Loss of Purine or Thymidine Salvage Pathways Creates Chemosensitivity Our effort to maximize the stability of suicide functions exploited the observation that when endogenous cellular functions are lost through mutation, the frequency with which they are regained is typically much lower than the frequency with which the functions of exogenously transduced genes are lost; that is, most loss of function mutations are highly stable. Two well-characterized loss of function mutations are those involving the genes for hypoxanthine/guanine phosphoribosyltransferase (HPRT) and cellular thymidine kinase

486


(not to be confused with the HSV-TK gene, which encodes a different enzyme). The salient feature of HPRT and cellular TK is that mutational loss of either creates chemosensitivity, with HPRT deficiency sensitizing cells to inhibitors of purine synthesis and thymidine kinase deficiency creating sensitivity to inhibitors of thymidylate synthesis. A regimen that inhibits both of these biosynthetic pathways is hypoxanthine/aminopterin/thymidine (HAT) [56]. HAT is well tolerated by normal cells because, unlike HPRT-deficient cells, they can utilize the hypoxanthine and, unlike TK-deficient cells, can utilize the thymidine to circumvent the respective blocks in purine and thymidylate synthesis imposed by the antifolate drug aminopterin. HPRT-negative cells can be selected by virtue of their resistance to 6-thioguanine (6TG) [56], and TK-negative cells can be selected for their resistance to iododeoxyuridine or bromodeoxyuridine [57]. The advantage of exploiting the stability that might characterize HAT-sensitizing mutations is offset by a significant potential limitation. The normal role of the HPRT and TK enzymes is to incorporate hypoxanthine and thymidine, respectively, into salvage pathways that reclaim these compounds for nucleic acid synthesis. Although loss of these pathways is not lethal, it appears to create subtle growth disadvantages in certain cell populations, eventually resulting, for example, in the loss of detectable HPRT-negative cells in cell populations of hematopoietic origin in women who begin life with both HPRT-positive and -negative cells [58]. The growth disadvantage appears not to be universal, as some other tissues (fibroblasts, hair follicles) retain their mosaic character in these women. Nevertheless, in the disadvantaged tissues, HAT sensitivity will ultimately prove unstable, not because cells lose sensitivity but because the cells themselves will fail to persist in the absence of a substitute means of accomplishing salvage pathway functions. To determine whether stable suicide functions could be created, a two-pronged approach was utilized that combined HPRT or TK deficiency with the addition of a new gene that replaced the lost salvage pathway functions and also mediated a suicide function of its own. HPRT-negative cells were obtained by 6TG selection and then exposed to a retroviral vector [10] that transduced the Escherichia coli gpt gene, which sensitizes cells to 6-thioxanthine (6TX). Like HPRT, the enzyme encoded by the gpt suicide gene, xanthine/guanine phosphoribosyltransferase (XGPRT), is capable of catalyzing the incorporation of hypoxanthine or guanine into nucleotide synthesis salvage pathways; its suicide function derives from its additional ability to use 6TX as a substrate. The gpt gene thus serves a dual purpose. It adds to the already stable chemosensitivity of HPRT-deficient cells by introducing an additional suicide function that must be lost by mutation for cells to lose all chemosensitivity. At the same time, it preserves the salvage pathway competency that HPRT-deficient cells would otherwise lack.

TABLE 3 Growth Rates of Wild-Type Cell Lines and Subclones

Cellsa

Salvage pathway competency

K3T3 H+ G−

+

Doubling time (hours)b 15.6

H− G−

−

16.2, 18.6

K3T3 H− G+

+

15.3, 15.8, 17.1, 19.2, 20.6, 22.6, 24.0

CLS1 H+ G−

+

19.5

CLS1 H− G− CLS1 H− G+

−

21.7

+

13.7, 15.7, 16.0, 24.4, 29.1, 29.4, 41.0, 43.1

LY18 H+ G−

+

20.1

LY18 H− G− LY18 H− G+

−

21.5, 27.3

+

18.5, 20.0, 21.6,22.1, 22.6, 23.8, 25.1, 27.5

K3T3

a The cell lines tested were of fibroblastic (K3T3), epithelial (CLS1), and pre-B-lymphocytic (LY18) origin. b Doubling times represent the mean calculated from one to four replicates of duplicate cultures for each line and subclone tested. c H, hprt; G, gpt.

Based on the same rationale, the stability of the suicide function was also examined in cells that were deficient in cellular TK and had acquired the HSV-TK gene. Before transduction, each of five HPRT-deficient clones tested exhibited a growth rate that was slightly to moderately slower than that of their wild-type parents (Table 3). Transduction of the gpt gene into HPRT-deficient cells yielded a new population that substituted 6TX sensitivity for their previously acquired HAT sensitivity (Table 4). Unlike their predecessors, these gpt-transduced clones exhibited a broad range of doubling times, with some growing slowly and others growing as fast or faster than parental cells. The observed variations may reflect position effects or other attributes of the integrated gpt vector.

B. Stability of Suicide Functions in HPRT-Negative/gpt-Positive Cells When four HPRT-deficient, gpt-transduced subclones of murine K3T3 fibrosarcoma cells were exposed to 6TX, surviving colonies ranged in number from 1 to 14 per 1.2 × 103 cells, representing mutant frequencies (corrected for plating efficiency) of 1.3 × 10−3 to 1.9 × 10−2 . The loss of 6TX sensitivity was accompanied in each case by reacquisition of HAT sensitivity, consistent with loss of gpt-mediated salvage functions (Table 4). To determine the frequency with which both suicide functions were lost, expanded populations of three 6TX-resistant

487


TABLE 4 Frequency of Acquisition and Subsequent Loss of Phenotypes Associated with Chemosensitivity Sensitivity profile Transition process

Phenotype

Isolation frequency

HAT

6TX

None (wild type)

6TG

H+ G−

—

R

R

S

6TG selection

H− G−

5 × 10−6 –1.3 × 10−5

S

R

R

gpt transduction

H− G+

∼0.5

R

S

S

Natural mutation

H− G−

1.3 × 10−3 –1.9 × 10−2

S

R

R

Natural mutation

HATR

3.0 × 10−8

R

R

NT

Note. The parental K3T3, CLS1, and LY18 cells used in the study were subjected to 6TG selection to yield clones that lacked HPRT enzyme activity and were sensitive to HAT, thus manifesting a suicide function consistent with their loss of hypoxanthine salvage capacity. Their 6TX ID50 exceeded 200 μM for K3T3 and LY18, and 250 μM for CLS1. After gpt transduction, the ID50 of the tested clones ranged from 0.5 to 10 μM for K3T3, from 1 to 3 μM for CLS1 cells, and from 2 to 10 μM for LY18. Data on transitions from H+ G− to H− G− to H− G+ were obtained for all three cell lines; further transitions to H+ G− and then HATR were measured only for K3T3. In addition, an H+ G− subclone of CLS1 cells that had never been subjected to gpt transduction was tested and yielded no HAT-resistant colonies from 2 × 107 cells. H, hprt; G, gpt, S, sensitive (all cells destroyed except for resistant mutants); R, resistant (cells grow at normal or near normal rates in the presence of the drug); NT, not tested; ID50 , 6TX concentration reducing cell numbers to 50% of the numbers in untreated control cultures during a 6-day assay interval.

subclones, each grown in 17 T-75 flasks containing 106 cells/flask, were tested for the presence of HAT-resistant mutants. Of the 51 flasks, 50 yielded no HAT-resistant colonies, and the remaining flask yielded a single colony. The result corresponds to a corrected mutant frequency of 3.0 × 10−8 . The subclones originated from an HPRTnegative, gpt-transduced clone that yielded 6TX-resistant mutants at a frequency of 5.4 × 10−3 . When this figure is multiplied by the frequency of HAT-resistant mutants, the resulting value, 5.4 × 10−3 × 3.0 × 10−8 = 1.6 × 10−10 constitutes an estimate of the predicted frequency of the combined loss of both suicide functions. The same calculations applied to the clones with the greatest and poorest gpt stability yielded a frequency range of 3.9 × 10−11 to 5.7 × 10−10 . The rarity of HAT-resistant colonies among HPRT-negative K3T3 cells appeared to be matched by an HPRT-negative subclone of CLS1 cells, which yielded no surviving colonies among 20 flasks totaling 2 × 107 cells exposed to HAT.

C. Stability of Suicide Functions in TK-Negative/HSV-TK-Positive Cells NIH3T3 fibroblasts that lack cellular TK but had been transduced with the HSV-TK gene were exposed to 8.8 μM GCV to select for mutants that had lost the HSV-TK suicide function. GCV-resistant clones were obtained from replicate cultures in numbers that corresponded to mutant frequencies of 1.5 × 10−4 to 3.4 × 10−3 . Subsequent reexposure

to GCV confirmed their resistant status. Analysis of two GCV-resistant clones confirmed that they were now HAT sensitive, as expected, and revealed TK enzyme levels that were only 1.5 and 1.8% of the levels of the GCV-sensitive cells from which they were derived, a decline consistent with their loss of GCV sensitivity and reacquisition of HAT sensitivity. When the two clones were subsequently exposed to HAT, HAT-resistant mutants were obtained at a frequency of 2 × 10−7 and 1.2 × 10−6 . The combined frequencies, representing the frequency with which a GCVchemosensitive population would be expected to revert to a GCV-insensitive, HAT-insensitive, wild-type phenotype, thus ranged from 3.0 × 10−11 to 4.1 × 10−9 . This implies a stability similar to the stability of suicide functions observed in HPRT-negative, gpt-transduced K3T3 cells.

V. PREEMPTIVE USES OF SUICIDE GENES TO CONTROL GRAFT-VERSUS-HOST DISEASE IN LEUKEMIA The relevance of suicide genes to neoplastic disease extends beyond their direct presence in neoplastic cells. A promising area currently under active investigation involves the use of the HSV-TK gene to impart GCV sensitivity not to malignant cells but rather to cells used to treat the malignancy: donor T lymphocytes administered in conjunction with allogeneic bone marrow in patients with leukemia and related diseases. Allogeneic bone marrow transplantation (allo-BMT) is currently associated with long-term remissions in a

488


substantial number of patients with acute leukemia, chronic myelogenous leukemia, multiple myeloma, and myelodysplasia at a frequency that may exceed 50% in favorable circumstances; many of these remissions are thought to represent cures [59,60]. Most of the reduction in leukemic cell numbers is accomplished by the intensive chemoradiotherapy that precedes the allo-BMT, but donor T cells play a critical role in eradicating residual cells. This achievement comes at a cost—the frequent occurrence of graft-versus-host disease (GVHD) severe enough to result, directly or indirectly, in substantial treatment-related mortality. In addition, the severity of GVHD reflects in part the degree of antigenic disparity between donor and recipient and thus limits the availability of suitable donors; HLA mismatching poses the greatest threat of lethal GVHD, and, despite HLA matching, unrelated donors represent a greater hazard than HLA-matched sibling donors. Because immunosuppressive drugs have often failed to control GVHD adequately and impose hazards of their own, T-depleted marrow has been employed in an effort to avoid this complication. Unfortunately, the absence of T cells has been associated with poor leukemia control, reduced marrow engraftment, and a serious immunodeficiency that renders patients vulnerable to a variety of infections. Among the infectious sequelae are severe cytomegalovirus infections and potentially lethal EBV-induced lymphoproliferative disease [60]. One approach to preserving T-cell function involves alloBMT with T-depleted marrow followed later by infusions of donor peripheral blood leukocytes, a rich source of T-cells. In some cases, the infusions have been delayed until specifically necessitated by leukemia relapse or viral sequelae [60]. Delayed infusion of T cells appears to reduce the threat of GVHD but does not eliminate it. An additional advantage of utilizing separate marrow and T cell infusions, however, is the opportunity to manipulate the T cells. In particular, this opportunity has been exploited to transduce the HSVTK gene into donor T cells to sensitize them to GCV and thereby permit their subsequent ablation for severe GVHD [61–63]. The administration of HSV-TK-transduced T cells thus extends the benefits of T cells to all patients while later eliminating the cells only in those patients in whom they induce life-threatening pathology. This rationale is the basis for ongoing clinical trials in Milan [61,63] and more recently in this country and elsewhere [30], as well as additional protocols that have been approved or are under review, all involving patients receiving allo-BMT for leukemia or myeloma [30]. In each case, donor peripheral blood leukocytes are isolated, stimulated to proliferate, transduced with retroviral vectors bearing the HSV-TK gene plus a selectable marker, subjected to selection, and infused into patients after further growth to achieve adequate cell numbers. Results reported to date from the first clinical trial (from Milan) are preliminary but encouraging [63]. Transduced

cells retained their ability to exert antileukemic effects in most cases, including complete remissions in three out of eight patients. Two patients developed acute GVHD; in each case, administration of GCV quickly eliminated the transduced cells from the circulation and induced nearly complete resolution of the clinical and biochemical signs of GVHD. The T cells thus appear to have responded as expected. In an additional patient who developed chronic GVHD, GCV resulted in only partial amelioration. The lesser efficacy may reflect the existence in chronic GVHD of a substantial fraction of cells that are not proliferating at the time of GCV administration and are therefore insusceptible to the inhibitory effects of GCV phosphates on DNA synthesis [64].

VI. FUTURE PROSPECTS FOR PREEMPTIVE USE OF SUICIDE GENES Until in vivo transduction efficiency improves, the failsafe use of suicide genes is likely to remain a phenomenon that can only be applied to cells that are manipulated in vitro and later reintroduced into human hosts. Potential applications include their use as a precaution against either malignant behavior of the reintroduced cells or immune pathology that they might induce [1,2,4,55]. Additionally, suicide genes added to cells transplanted to supply a missing function constitute a potential mechanism to control hyperactivity of the transplanted cells, such as hyperinsulinemia resulting from excessive growth or function of cells expressing native or transduced insulin genes [65]. If suicide genes of nonhuman origin are to be used preemptively, their success will require that their presence not provoke host immune reactions that result in the elimination of the transduced cells. Such reactions have been observed in some [55,66] but not other [49,67] studies involving cells transduced with the HSV-TK gene. The development of improved methods for inducing immune tolerance, the use of genes transcribed from inducible promoters that remain inactive until an appropriate stimulus is applied, or the creation of suicide genes that are expressed at the level of nucleic acid rather than protein (e.g., as catalytic RNA) [68,69] are possible approaches to this problem. If reliable methods for controlling immune rejection are developed to the point that they permit the use of xenografted tissues in humans, the introduction of suicide genes as transgenes into animals used as a source of the xenografts constitutes a further fail-safe use of suicide genes, one designed to protect against undesired effects of the grafted cells. An intriguing application of the HSV-TK gene that is likely to be tested soon in clinical trials is its use as a marker for in vivo gene transduction. In addition to its phosphorylation of GCV, HSV-TK phosphorylates a number of other nucleoside analogs that are poor substrates for cellular


kinases, including halogenated pyrimidine analogs such as 5-iodo-2 -fluoro-2 -deoxy-1-β-D-arabinofuranosyluracil (FIAU). Tjuvajev et al. [70] have shown that when 131 Ilabeled FIAU is administered to mice bearing tumors carrying transduced HSV-TK genes, the location and extent of HSV-TK expression can be precisely delineated by in vivo imaging with a gamma camera and single-photon emission tomography (SPECT). Extending this concept, they have also demonstrated that when the HSV-TK vector also transduces a separate gene (lacZ), the imaging analysis not only correlates with HSV-TK expression but also locates and quantifies expression of the linked gene [71]. This use of HSV-TK as a marker in conjunction with FIAU or other substrates that are currently under investigation [72] harbors the potential for it to serve a dual purpose: measuring the function of whatever therapeutic gene it might be linked to in a gene therapy subject and additionally serving to protect that subject against unwanted behavior by the transduced cells. A final prospect relates to the possibility, discussed above, that efficient incorporation of one or more suicide genes into one or more tissues might eventually permit cancers that arise later to be treated effectively, based on their clonal origin from a sensitized cell. The previous discussion emphasized the prospect that clonality might ensure the presence of a suicide gene even in metastatic or disseminated cancers (i.e., the late stages of a cancer/host relationship). It is also possible, however, that early, preclinical stages might be targetable as well. Recent evidence indicates that DNA derived from cancer cells is sometimes detectable in blood or secretions by PCR analysis. Thus, mutant K-ras [73–76] genes have been detected in both plasma [73,74] and feces [75,76] of patients with colorectal [74,75] and pancreatic [73,76] carcinomas, mutant p53 genes have been demonstrated in the urine of patients with bladder cancer [77], and specific microsatellite DNA alterations have been detected in the plasma [78] and sputum [79] of lung cancer patients and in serum from patients with head and neck cancer [80]. Some of the detected alterations represented changes that were also present in premalignant lesions that accompanied the cancer or in one case were found in the absence of a cancer [76]. In theory, suicide genes harbored by the cells of cancers that arose in preemptively transduced tissues would also be detectable, and analysis of flanking genomic sequences could be used to determine whether they represented the monoclonal pattern of a neoplasm or the polyclonal pattern of nonneoplastic tissue. If the detection sensitivity of this type of DNA analysis increases to the point where incipient clonal proliferations are detectable in individuals who harbor suicide genes in various vulnerable tissues (breast, lung, bone marrow, etc.), then detection would permit early action, such as a search for the neoplasm, biopsy, and surgery or radiotherapy as indicated. If the neoplasm is found, prodrug administration could be added to surgery or radiotherapy in an adjuvant role. If the

489

neoplasm is small enough to elude attempts to locate it, administration of a prodrug could be used to ablate it before it surfaces clinically, in essence exploiting preemption as a form of cancer prevention. If in vivo transduction efficiency in nonvital tissues such as breast or prostate eventually improves to the point where a suicide gene can be transduced into almost all the epithelial cells of these tissues, prevention should also be feasible at an even earlier stage, if desired. Thus, individuals at high risk for breast or prostate cancer might, at some stage in their life, choose to receive a prodrug as a form of molecular “epitheliectomy” in preference to surgical bilateral mastectomy or prostatectomy.

References 1. Moolten, F. L. (1986). Tumor chemosensitivity conferred by inserted herpes thymidine kinase genes: paradigm for a prospective cancer control strategy. Cancer Res. 46, 5276–5281. 2. Moolten, F. L., and Wells, J. M. (1990). Curability of tumors bearing herpes thymidine kinase genes transferred by retroviral vectors. J. Natl. Cancer Inst. 82, 297–300. 3. Moolten, F. L. (1994). Drug sensitivity (“suicide”) genes for selective cancer chemotherapy. Cancer Gene Therap. 1, 279–287. 4. Tiberghien, P. (1994). Use of suicide genes in gene therapy. J. Leukocyte Biol. 56, 203–209. 5. Bi, W. L., Parysek, L. M., Warnick, R., and Stambrook, P. J. (1993). In vitro evidence that metabolic cooperation is responsible for the bystander effect observed with HSV tk retroviral gene therapy. Human Gene Therap. 4, 725–731. 6. Freeman, S. M., Abboud, C. N., Whartenby, K. A., Packman, C. H., Koeplin, D. S., Moolten, F. L., and Abraham, G. N. (1993). The “bystander effect”: tumor regression when a fraction of the tumor mass is genetically modified. Cancer Res. 53, 5274 –5283. 7. Hooper, M. L., and Subak-Sharpe, J. H. (1981). Metabolic cooperation between cells. Int. Rev. Cytol. 69, 45–104. 8. Culver, K. W., Ram, Z., Wallbridge, S., Ishii, H., Oldfield, E. H., and Blaese, R. M. (1992). In vivo gene transfer with retroviral vector producer cells for treatment of experimental brain tumors. Science 256, 1550–1552. 9. Mullen, C. A., Kilstrup, M., and Blaese, R. M. (1992). Transfer of the bacterial gene for cytosine deaminase to mammalian cells confers lethal sensitivity to 5-fluorocytosine: a negative selection system. Proc. Natl. Acad. Sci. USA 89, 33–37. 10. Mroz, P. J., and Moolten, F. L. (1993). Retrovirally transduced Escherichia coli gpt genes combine selectability with chemosensitivity capable of mediating tumor eradication. Hum. Gene Ther. 4, 589–595. 11. Wei, M. X., Tamiya, T., Chase, M., Boviatsis, E. J., Chang, T. K. H., Kowall, N. W., Hochberg, F. H., Waxman, D. J., Breakefield, X. O., and Chiocca, E. A. (1994). Experimental tumor therapy in mice using the cyclophosphamide-activating cytochrome P450 2B1 gene. Human Gene Ther. 5, 969–978. 12. Chen, L., Waxman, D. J., Chen, D., and Kufe, D. W. (1996). Sensitization of human breast cancer cells to cyclophosphamide and ifosfamide by transfer of a liver cytochrome P450 gene. Cancer Res. 56, 1331– 1340. 13. Sorscher, E. J., Peng, S., Bebok, Z., Allan, P. W., Bennett, L. L., and Parker, W. B. (1994). Tumor cell bystander killing in colonic carcinoma utilizing the Escherichia coli DeoD gene to generate toxic purines. Gene Ther. 1, 233–238.

490


14. Manome, Y., Wen, P. Y., Dong, Y., Tanaka, T., Mitchell, B. S., Kufe, D. W., and Fine, H. A. (1996). Viral vector transduction of the human deoxycytidine kinase cDNA sensitizes glioma cells to the cytotoxic effects of cytosine arabinoside in vitro and in vivo. Nat. Med. 2, 567– 573. 15. Bridgewater, J. A., Knox, R. J., Pitts, J. D., Collins, M. K., and Springer, C. J. (1997). The bystander effect of the nitroreductase/CB1954 enzyme/prodrug system is due to a cell-permeable metabolite. Human Gene Ther. 8, 709–717. 16. Spencer, D. M., Belshaw, P., Chen, L., Ho, S. N., Randazzo, F., Crabtree, G. R., and Schreiber, S. L. (1996). Functional analysis of Fas signaling in vivo using synthetic inducers of dimerization. Current Biol. 6, 839–847. 17. Freiberg, R. A., Spencer, D. M., Choate, K. A., Peng, P. D., Schreiber, S. L., Crabtree, G. R., and Khavari, P. A. (1996). Specific triggering of the Fas signal transduction pathway in normal human keratinocytes. J. Biol. Chem. 271, 31666–31669. 18. Fan, L., Freeman, K. W., Khan, T., Pham, E., and Spencer, D. M. (1999). Improved artificial death switches based on caspases and FADD. Hum. Gene Ther. 10, 2273–2285. 19. Amara, J. F., Courage, N. L., and Gilman, M. (1999). Cell surface tagging and a suicide mechanism in a single chimeric human protein. Hum. Gene Ther. 10, 2651–2655. 20. Mandell, R. B., Mandell, L. Z., and Link, Jr., C. J. (1999). Radioisotope concentrator gene therapy using the sodium /iodide symporter gene. Cancer Res. 59, 661–668. 21. Spitzweg, C., Zhang, S., Bergert, E. R., Castro, M. R., McIver, B., Heufelder, A. E., Tindall, D. J., Young, C. Y. F., and Morris, J. C. (1999). Prostate-specific antigen (PSA) promoter-driven androgen-inducible expression of sodium iodide symporter in prostate cancer cell lines. Cancer Res. 59, 2136–2141. 22. Stribbling, S. M., Friedlos, F., Martin, J., Davies, L., Spooner, R. A., Marais, R., and Springer, C. (2000). Regressions of established breast carcinoma xenografts by carboxypeptidase G2 suicide gene therapy and the prodrug CMDA are due to a bystander effect. Hum. Gene Ther. 11, 285–292. 23. Hamstra, D. A., Pagé, M., Maybaum, J., and Rehemtulla, A. (2000). Expression of endogenously activated secreted or cell surface carboxypeptidase A sensitizes tumor cells to methotrexate-α-peptide prodrugs. Cancer Res. 60, 657–665. 24. Oldfield, E. H. (1993). Gene therapy for the treatment of brain tumors using intra-tumoral transduction with the thymidine kinase gene and intravenous ganciclovir. Hum. Gene Ther. 4, 39–69. 25. Culver, K. W., and van Gilder, J. (1994). Gene therapy for the treatment of malignant brain tumors with in vivo tumor transduction with the herpes simplex thymidine kinase gene/ganciclovir system. Hum. Gene Ther. 5, 343–379. 26. Freeman, S. M., McCune, C., Angel, C., Abraham, G. N., and Abboud,C. N. (1992). Treatment of ovarian cancer using HSV-TK gene modified vaccine — regulatory issues. Hum. Gene Ther. 3, 342– 349. 27. Link, C. J., and Moorman, D. (1996). A phase I trial of in vivo gene therapy with the herpes simplex thymidine kinase/ganciclovir system for the treatment of refractory or recurrent ovarian cancer. Hum. Gene Ther. 7, 1161–1179. 28. Alvarez, R. D., and Curiel, D. T. (1997). A phase I study of recombinant adenovirus vector-mediated intraperitoneal delivery of herpes simplex virus thymidine kinase (HSV-TK) gene and intravenous ganciclovir for previously treated ovarian and extraovarian cancer patients. Hum. Gene Ther. 8, 597–613. 29. Treat, J., Kaiser, L. R., Sterman, D. H., Litzky, L., Davis, A., Wilson, J. M., and Albelda, S. M. (1996). Treatment of advanced mesothelioma with the recombinant adenovirus H5.010RSVTK: a phase I trial (BB-IND 6274). Hum. Gene Ther. 7, 2047–2057.

30. Anon. (2000). Human gene marker/therapy clinical protocols (complete updated listings). Human Gene Ther. 11, 919–979. 31. Jolly, D. (1994). Viral vector systems for gene therapy. Cancer Gene Ther. 1, 51–64. 32. Freeman, S. M., Whartenby, K. A., Freeman, J. L., Abboud, C. N., and Marrogi, A. J. (1996). In situ use of suicide genes for cancer therapy. Semin. Oncol. 23, 31–45. 33. Harsh, G. R., Deisboeck, T. S., Louis, D. N., Hilton, J., Colvin, M. Silver, J. S., Qureshi, N. H., Kracher, J. Finkelstein, D., Chiocca, E. A., and Hochberg, F. H. (2000). Thymidine kinase activation of ganciclovir in recurrent malignant gliomas: a gene-marking and neuropathological study. J. Neurosurg. 92, 804–811. 34. Vile, R. G., and Hart, I. R. (1993). Use of tissue-specific expression of the herpes simplex virus thymidine kinase gene to inhibit growth of established murine melanomas following direct intratumoral injection of DNA. Cancer Res. 53, 3860–3864. 35. Vile, R. G., Nelson, J. A., Castleden, S., Chong, H., and Hart, I. R. (1994). Systemic gene therapy of murine melanoma using tissue specific expression of the HSVtk gene involves an immune component. Cancer Res. 54, 6228–6234. 36. Huber, B. E., Richards, C. A., and Krenitsky, T. A. (1991). Retroviralmediated gene therapy for the treatment of hepatocellular carcinoma: an innovative approach for cancer therapy. Proc. Natl. Acad. Sci. USA 88, 8039–8043. 37. Macri, P., and Gordon, J. W. (1994). Delayed morbidity and mortality of albumin /SV40 T-antigen transgenic mice after insertion of an alphafetoprotein /herpes virus thymidine kinase transgene and treatment with ganciclovir. Hum. Gene Ther. 5, 175–182. 38. Judde, J.-G., Spangler, G., MacGrath, I., and Bhatia, K. (1996). Use of Epstein–Barr virus nuclear antigen-1 in targeted therapy of EBVassociated neoplasia. Hum. Gene Ther. 7, 647–653. 39. Franken, M., Estabrooks, A., Cavacini, L., Sherburne, B., Wang, F., and Scadden, D. T. (1996). Epstein–Barr virus-driven gene therapy for EBV-related lymphomas. Nat. Med. 2, 1379–1382. 40. Ko, S.-C., Cheon, J., Kao, C., Gotoh, A., Shirakawa,T., Sikes, R. A., Karsenty, G., and Chung, L. W. K. (1996). Osteocalcin promoter-based toxic gene therapy for the treatment osteosarcoma in experimental models. Cancer Res. 56, 4614–4619. 41. Harris, J. D., Gutierrez, A. A., Hurst, H. C., Sikora, K., and Lemoine, N. R. (1994). Gene therapy for cancer using tumour-specific prodrug activation. Gene Ther. 1, 170–175. 42. Kaneko, S., Hallenbeck, P., Kotani, T., Nakabayashi, H., McGarrity, G., Tamaoki, T., Anderson, W. F., and Chiang, Y. L. (1995). Adenovirusmediated gene therapy of hepatocellular carcinoma using cancerspecific gene expression. Cancer Res. 55, 5283–5287. 43. Fialkow, P. J. (1976). Clonal origin of human tumors. Biochim. Biophys. Acta 458, 283–321. 44. Fearon, E. R., Hamilton, S. R., and Vogelstein, B. (1987). Clonal analysis of human colorectal tumors. Science 238, 193–197. 45. Fujii, H., Marsh, C., Cairns, P., Sidransky, D., and Gabrielson, E. (1996). Genetic divergence in the clonal evolution of breast cancer. Cancer Res. 56, 1493–1497. 46. Moolten, F. L., Wells, J. M., Heyman, R. A., and Evans, R. M. (1990). Lymphoma regression induced by ganciclovir in mice bearing a herpes thymidine kinase transgene. Hum. Gene Ther. 1, 125–134. 47. Moolten, F., Wells, J. M., and Mroz, P. J. (1992). Multiple transduction as a means of preserving ganciclovir chemosensitivity in sarcoma cells carrying retrovirally transduced herpes thymidine kinase genes. Cancer Lett. 64, 257–263. 48. Moolten, F. L. (1990). Mosaicism induced by gene insertion as a means of improving chemotherapeutic selectivity. Crit. Rev. Immunol. 10, 203–233. 49. Moolten, F. L., Vonderhaar, B. K., and Mroz, P. J. (1996). Transduction of the herpes thymidine kinase gene into premalignant murine


50.

51.

52.

53.

54.

55.

56.

57. 58.

59. 60. 61.

62.

63.

64.

65.

mammary epithelial cells renders subsequent breast cancers responsive to ganciclovir therapy. Hum. Gene Ther. 7, 1197–1204. Jerry, D. J., Ozbun, M. A., Kittrell, F. S., Lane, D. P., Medina, D., and Butel, J. S. (1993). Mutations in p53 are frequent in the preneoplastic stage of mouse mammary tumor development. Cancer Res. 53, 3374– 3381. Kittrell, F. S., Oborn, C. J., and Medina, D. (1992). Development of mammary preneoplasias in vivo from mouse mammary epithelial cell lines in vitro. Cancer Res. 52, 1924–1932. Moolten, F. L., and Cupples, L. A. (1992). A model for predicting the risk of cancer consequent to retroviral gene therapy. Hum. Gene Ther. 3, 479–486. Cornetta, K., Morgan, R. A., and Anderson, W. F. (1991). Safety issues related to retroviral-mediated gene transfer in humans. Hum. Gene Ther. 2, 5–14. Cornetta, K., Morgan, R. A., Gillio, A., Sturm, S., Baltrucki, L., O’Reilly, R., and Anderson, W. F. (1991). No retroviremia or pathology in long–term follow-up of monkeys exposed to a murine amphotropic retrovirus. Hum. Gene Ther. 2, 215–220. Riddell, S. R., Elliott, M., Lewinsohn, D. A., Gilbert, M. J., Wilson, L., Manley, S. A., Lupton, S. D., Overell, R. W., Reynolds, T. C., Corey, L., and Greenberg, P. D. (1996). T-cell mediated rejection of gene-modified HIV-specific cytotoxic T lymphocytes in HIV-infected patients. Nat. Med. 2, 216–223. Sybalski, W., and Sybalska, E. H. (1961). A new chemotherapeutic principle for the treatment of drug resistant neoplasms. Cancer Chemotherapy Rep. 11, 87–89. Littlefield, J. W. (1966). The use of drug-resistant markers to study the hybridization of mouse fibroblasts. Exp. Cell Res. 41, 190 –196. Rossiter, B. J. F., and Caskey, C. T. (1995). Hypoxanthine-guanine phosphoribosyltransferase deficiency: Lesch–Nyhan syndrome and gout, in The Metabolic and Molecular Bases of Inherited Disease, 7th ed., (C. R. Beaudet et al., eds.), pp. 1679–1706. McGraw-Hill, New York. Beutler, E., Lichtman, M. K., Coller, B. S., and Kipps, T. J., eds. (1995). Williams Hematology, 5th ed., McGraw-Hill, New York. Shlomchik, W. D., and Emerson, S. G. (1996). The immunobiology of T cell therapies for leukemias. Acta Haematol. 96, 189–213. Bordignon, C., and Bonini, C. (1995). Transfer of the HSV-TK gene into donor peripheral blood lymphocytes for in vivo modulation of donor anti-tumor immunity after allogeneic bone marrow transplantation. Hum. Gene Ther. 6, 813–819. Tiberghien, P. (1997). Use of donor T-lymphocytes expressing herpes simplex-thymidine kinase in allogeneic bone marrow transplantation: a phase I–II study. Hum. Gene Ther. 8, 615–624. Bonini, C., Ferrari, G., Verzeletti, S., Servida, P., Zappone, E., Ruggieri, L., Ponzoni, M., Rossini, S., Mavilio, F., Traversari, C., and Bordignon, C. (1997). HSV-TK gene transfer into donor lymphocytes for control of allogeneic graft-versus-leukemia. Science 276, 1719– 1724. Bonini, C., and Bordignon, C., (1997). Potential and limitations of HSV-TK-transduced donor peripheral blood lymphocytes after alloBMT. Hematol. Cell. Ther. 39, 273–274. Yoshimoto, K., Murakami, R., Moritani, M., Ohta, M., Iwahana, H., Nakauchi, H., and Itakura, M. (1996). Loss of ganciclovir sensitivity by exclusion of thymidine kinase gene from transplanted proinsulinproducing fibroblasts as a gene therapy model for diabetes. Gene Ther. 3, 230–234.

491

66. Tapscott, S. J., Miller, A. D., Olson, J. M., Berger, M. S., Groudine, M., and Spence, A. M. (1994). Gene therapy of rat 9L gliosarcoma tumors by transduction with selectable genes does not require drug selection. Proc. Natl. Acad. Sci. USA 91, 8185–8189. 67. Pavlovic, J., Nawrath, M., Tu, R., Heinicke, T., and Moelling, K. (1996). Anti-tumor immunity is involved in the thymidine kinase-mediated killing of tumors induced by activated Ki-ras(G12V). Gene Ther. 3, 635–643. 68. Prudent, J. R., Uno, T., and Schultz, P. G. (1994). Expanding the scope of RNA catalysis. Science 264, 1924 –1927. 69. Wilson, C., and Szostak, J. W. (1995). In vitro evolution of a selfalkylating ribozyme. Nature 374, 777–782. 70. Tjuvajev, J. G., Finn, R., Watanabe, K., Joshi, R., Oku, T., Kennedy, J., Beattie, B., Koutcher, J., Larson, S., and Blasberg, R. G. (1996). Noninvasive imaging of herpes virus thymidine kinase gene transfer and expression: a potential method for monitoring clinical gene therapy. Cancer Res. 56, 4087–4095. 71. Tjuvajev, J., Safer, M., Sadelain, M., Avril, N., Oku, T., Joshi, R., Finn, R., Larson, S., and Blasberg, R. (1997). Noninvasive imaging of the HSV1-tk marker gene for monitoring the expression of other target genes in vivo. J. Neuro-Oncol. 35 (suppl. 1), S45. 72. Gambhir, S. S., Bauer, E., Black, M. E., Liang, Q., Kokoris, M. S., Barrio, J. R., Iyer, M., Namavari, M., Phelps, M. E., and Herschman, H. R. (2000). A mutant herpes simplex virus type 1 thymidine kinase reporter gene shows improved sensitivity for imaging reporter gene expression with positron emission tomography. Proc. Natl. Acad. Sci. USA 97, 2785–2790. 73. Sorenson, G. D., Pribish, D. M., Valone, F. H., Memoli, V. A., and Yao, S. L. (1993). Mutated K-ras sequences in plasma from patients with pancreatic carcinoma. Proc. Am. Assoc. Cancer Res. 34, A174. 74. Lefort, L., Anker, P., Vasioukhin, V., Lyautey, J., Lederrey, C., and Stroun, M. (1995). Point mutations of the K-ras gene present in the DNA of colorectal tumors are found in the blood plasma DNA of the patients. Proc. Am. Assoc. Cancer Res. 36, A3319. 75. Sidransky, D., Tokino, T., Hamilton, S. R., Kinzler, K. W., Levin, B., Frost, P., and Vogelstein, B. (1992). Identification of ras oncogene mutations in the stool of patients with curable colorectal tumors. Science 256, 102–105. 76. Caldas, C., Hahn, S., Hruban, R. H., Yeo, C., and Kern, S. (1994). Detection of K-ras mutations (mut) in the stool of patients (pts) with pancreatic adenocarcinoma (PCa). Proc. Am. Soc. Clin. Oncol. 13, A294. 77. Sidransky, D., Von Eschenbach, A., Tsai, Y. C., Jones, P., Summerhayes, I., Marshall, F., Meera, P., Green, P., Hamilton, S. R., Frost, P., and Vogelstein, B. (1991). Identification of p53 gene mutations in bladder cancers and urine samples. Science 252, 706– 709. 78. Chen, X. Q., Stroun, M., Magnenat, J. -L., Nicod, L. P., Kurt, A-M, Lyautey, J., Lederrey, C., and Anker, P. (1996). Microsatellite alterations in plasma DNA of small cell lung cancer patients. Nature Med. 2, 1033–1035. 79. Miozzo, M., Sozzi, G., Musso, K., Pilotti, S., Incarbone, M., and Pastorino, U. (1996). Microsatellite alterations in bronchial and sputum specimens of lung cancer patients. Cancer Res. 56, 2285– 2288. 80. Nawroz, H., Koch, W., Anker, P., Stroun, M., and Sidransky D. (1996). Microsatellite alterations in serum DNA of head and neck cancer patients. Nat. Med. 2, 1035–1037.


C

H

A

P

T

E

R

32 Treatment of Mesothelioma Using Adenoviral-Mediated Delivery of Herpes Simplex Virus Thymidine Kinase Gene in Combination with Ganciclovir

I. Introduction

DANIEL H. STERMAN

STEVEN M. ALBELDA

Thoracic Oncology Research Laboratory Pulmonary, Allergy, and Critical Care Division University of Pennsylvania Medical Center Philadelphia, Pennsylvania 19104

Thoracic Oncology Research Laboratory Pulmonary, Allergy, and Critical Care Division University of Pennsylvania Medical Center Philadelphia, Pennsylvania 19104

493

by the suicide gene are often of non-human origin, such as the Escherichia coli cytosine deaminase (CDA) gene [2] or the herpes simplex virus-1–thymidine kinase (HSV-TK) gene [3]. The latter was shown by Moolten et al. [4,5] to kill tumor cells when combined with administration of the antiviral agent, ganciclovir (GCV). The drug ganciclovir (9-[1,3-dihydroxy-2-propoxy)methyl]-guanine) is an acyclic nucleoside that is poorly metabolized by mammalian cells and is therefore generally nontoxic. However, after being converted to GCV-monophosphate by herpes virus family (herpes simplex virus-1, cytomegalovirus, vaccinia virus) thymidine kinases, it is rapidly converted to GCV-triphosphate by mammalian kinases [6]. Ganciclovir triphosphate is a potent inhibitor of viral DNA polymerase and is also a toxic analog that competes with normal mammalian nucleosides for DNA replication [6]. In addition, incorporation of GCV-monophosphate into the DNA template has also been demonstrated to induce significant cytotoxicity [7]. The antitumor effect of HSV–thymidine kinase/ganciclovir (HSV-TK/GCV) gene therapy was assayed originally in animal models where producer cells containing a retroviral construct encoding for HSV-TK were stereotactically injected into brain tumors. In these models, tumor regression was observed after GCV administration [8,9]. Subsequently, similar antineoplastic properties were described in in vivo studies involving direct intratumoral delivery of HSV-TK by an adenoviral vector [10,11].

A. Bystander Effects: Intercellular Passage of GCV Metabolites and/or Immunologic Effects 494 B. Adenoviral Delivery Systems 494

II. Clinical Use of HSV-TK in the Treatment of Localized Malignancies 494 A. Malignant Mesothelioma: Paradigm for HSV-TK/GCV Gene Therapy 494 B. Preclinical Data: Animal and Toxicity Studies 494 C. Initial Phase I Clinical Trial 496 D. Adjunctive Phase I Clinical Trials 497

III. Challenges and Future Directions

499

A. Strategies To Optimize Patient Selection 499 B. Strategies To Augment Gene Transfer 500 C. Strategies To Assess Gene Transfer Noninvasively 501 References 501

I. INTRODUCTION One prominent approach in cancer gene therapy is the introduction of toxic or suicide genes into tumor cells to facilitate their destruction. One such suicide gene approach involves the transduction of a neoplasm with a cDNA encoding for an enzyme that would render its cells sensitive to a benign drug by converting the prodrug to a toxic metabolite [1]. As described in previous chapters, these enzymes encoded


493


494

Daniel H. Sterman and Steven M. Albelda

A. Bystander Effects: Intercellular Passage of GCV Metabolites and/or Immunologic Effects Given the limited gene transfer efficiency of current vector systems, the primary reason for the success of in vivo HSV-TK experiments appeared to be the finding that HSVTK expression in every cell was not required for complete tumor regression. This so-called bystander effect was demonstrated in in vitro mixing experiments using retrovirally infected tumor cells. Subsequently, in vivo experiments involving tumors where only 10–20% of the cells expressed the HSV-TK gene demonstrated that complete tumor regression was noted in animals after ganciclovir treatment [5,9,12–14]. The nature of this bystander effect is complex and appears to involve passage of toxic GCV metabolites from transduced to nontransduced cells via gap junctions or apoptotic vesicles [15,16]; and induction of antitumor immune responses capable of killing distant, non-HSV-TK-transduced cells [17].

B. Adenoviral Delivery Systems The transfer of HSV-TK DNA to target tumor cells can be accomplished in a variety of ways including the use of viral vectors, liposomes, cellular delivery systems, and naked DNA electrocorporation [18]. Early in vitro and in vivo studies utilized retroviral vectors to facilitate HSV-TK DNA transfer into tumors, including the injection of producer cell lines that secrete retrovirus containing the suicide gene [8,19]. These retroviral-based approaches have been used successfully in animal models of brain tumor, ovarian cancer, and hepatocellular carcinoma [20]. Retroviruses have several limitations as delivery vehicles of therapeutic genes for cancer gene therapy insofar as they infect only actively dividing cells, carry risks of insertional mutagenesis, and are difficult to produce in large scale for human clinical trials. Contrastingly, adenoviruses are able to infect both dividing and nondividing cells, do not carry the theoretical risk of insertional mutagenesis (they deliver their DNA episomally), and are much easier to produce in lots large enough for use in clinical studies [21]. For these reasons, adenoviruses have become the vector of choice for delivery of the HSV-TK gene, as well as other therapeutic genes, in many cancer gene therapy experimental models. Based on these factors, our group and others have produced recombinant, replication-deficient adenoviral vectors encoding the HSV-TK gene (Fig. 1A) and have shown that this vector, in combination with GCV, could eradicate tumor cells in vitro and in in vivo models of various tumors such as malignant mesothelioma, lung cancer, brain tumors, colon carcinoma, hepatocellular carcinoma, glioma, and melanoma [10,22–25].

II. CLINICAL USE OF HSV-TK IN THE TREATMENT OF LOCALIZED MALIGNANCIES Based upon the success of in vivo studies from multiple laboratories, several centers have conducted, or are in the process of conducting, human trials of adenovirally delivered HSV-TK in combination with GCV in advanced malignancies [26–28]. Because current vector technology does not yet allow for systemic administration, the initial clinical trials have primarily focused on localized malignancies, where directed instillation of vector (in conjunction with a bystander effect) could have some potential for therapeutic efficacy. The primary targets have included brain tumors, ovarian carcinoma, melanoma, prostate carcinoma, and malignant mesothelioma.

A. Malignant Mesothelioma: Paradigm for HSV-TK/GCV Gene Therapy Our group has focused on malignant pleural mesothelioma as a primary target, as we feel it has many features that can serve as a paradigm for other localized malignancies. Several characteristics make mesothelioma an attractive target for gene therapy: (1) There is no standard, effective therapy for the disease; (2) mesothelioma is readily accessible in the pleural space for vector delivery, biopsy, and subsequent analysis of treatment effects; (3) local extension of disease, rather than distant metastases, is responsible for much of the morbidity and mortality of mesothelloma; and (4) current treatment options are very limited. Thus, unlike other neoplasms that metastasize earlier in their course, in patients with mesothelioma small increments of improvement in local control could engender significant improvements in palliation or survival. Accordingly, a number of gene therapy trials aimed at treating mesothelioma have begun or are in the planning stages. At least two of the active programs University of Pennsylvania and Louisiana State University are investigating delivery of the HSV-TK gene to mesothelioma cells in combination with systemic GCV, although the delivery systems differ: adenovirus and PA-1 ovarian carcinoma cell line, respectively [29,30].

B. Preclinical Data: Animal and Toxicity Studies Initial experiments demonstrated that replication-deficient adenoviral HSV-TK vectors efficiently transduced mesothelioma cells both in tissue culture and in animal models and facilitated HSV-TK-mediated killing of human mesothelioma cells in the presence of low concentrations of GCV [31,32]. Subsequently, the Ad.HSV-TK vector was used to treat established, intraperitoneal human mesothelioma tumors and lung cancers in SCID mice [10,23]. Following GCV therapy, macroscopic tumor was eradicated in 90% of animals, and

Treatment of Mesothelioma

FIGURE 1 (Top) Illustration of the adenoviral (Ad) vector used in the initial phase I trial (H5.010RSVTK). This so-called first-generation replication-incompetent Ad is deleted in the early genes E1 and E3 with the HSV-TK gene inserted in the E1 region. (Middle) A third-generation Ad vector containing deletions in the E1 and E4 regions with preservation of E3. E1/E4-deleted Ad vectors offer theoretical advantages-over first-generation vectors due to their diminished cytopathic effects and hepatoxicity and reduced cellular immune responses. (Bottom) A tumor-selective replicating Ad. TK vector with a tumor-selective promoter (Calretinin, MnSOD, Mesothelin) substituted for the Ad.E1 promoter. This would potentially allow for greater HSV-TK delivery to solid tumors with decreased collateral injury to normal tissues.

495

496


microscopic tumor was undetectable in 80% of animals. Tumor reduction was accompanied by a significant increase in survival. Marked decreases in tumor size have also been seen in an intrapleural rat model of syngeneic mesothelioma with more modest increases in survival [24]. The in vitro and in vivo sensitivites of human mesothelioma cells to HSVTK/GCV gene therapy have been confirmed by other independent investigators [33]. Based on the efficacy data in animals, we conducted preclinical toxicity testing for submission to the Recombinant DNA Advisory Committee (RAC) and the Food and Drug Administration (FDA). The trials were designed to mimic the proposed clinical trials. Rats were given high doses of virus intrapleurally followed by intraperitoneal administration of GCV at the same dose proposed for initial use in the clinical trial (10 mg/kg/day). Toxicity was limited to localized inflammation of the pleural and pericardial surfaces. Formal toxicology studies were also done in three non-human primates given high-dose virus (1012 PFU) and GCV [34]. No adverse clinical effects were seen, nor any hematological or biochemical abnormalities. Necropsy findings were limited to inflammatory changes in the chest wall and intrathoracic serosa.

TABLE 1 Mesothelioma Gene Therapy Trial #1 Results (As of 8/15/2000) No. of Stage/cell patients type

Vector dose (PFU)

Status

Gene Survival transfer (months) score

62/M

IA/Ea

1 × 109

Progressed

57

0

56/M

III/E

1 × 109

Deceased

8

0

69/M

III/B

1 × 109

Deceased

20

5

66/M

II/E

3.2 × 109

Deceased

11

0

71/M

IA/E

3.2 × 109

Progressed

53

0

71/M

II/B

1 × 1010

Deceased

4

3

70/M

II/E

1 × 1010

Deceased

6

0

60/M

II/E

1 × 1010

Deceased

27

1

74/M

II/B

3.2 × 1010

Deceased

2

—a

60/M

III/E

3.2 × 1010

Deceased

9

0

37/F

IV/E

1 × 1011

Deceased

16

0

37/M

IIIb

1 × 1011

Deceased

2

0

65/F

III/E

1 × 1011

Deceased

10

4

66/F

IA/E

3.2 × 1011

Progressed

46

3

60/M

IV/B

3.2 × 1011

Deceased

5

3

69/M

IB/E

3.2 × 1011

Deceased

8

6

C. Initial Phase I Clinical Trial

70/F

IB/E

3.2 × 1011

Deceased

16

4

69/F

IB/E

3.2 × 1011

Deceased

26

5

A phase I clinical trial for patients with mesothelioma began in November 1995 at the University of Pennsylvania Medical Center in conjunction with Penn’s Institute for Human Gene Therapy. In this dose-escalation protocol, patients with mesothelioma who met strict inclusion criteria (including patent pleural cavities) underwent intrapleural administration of a single dose of Ad.HSV-TK vector followed by 2 weeks of intravenous GCV [26,29]. The initial adenoviral vector used was a so-called first-generation replicationincompetent virus, deleted in the early genes E1 and E3 with the HSV-TK gene inserted in the E1 region (H5.010RSV-TK; Fig. 1A). The protocol was designed as a dose-escalation study, starting with a vector dose of 1 × 109 plaque forming units (PFU) and increasing in half-log increments to the maximal dose level of 1 × 1012 PFU. At the completion of the 14-day GCV course, patients were discharged to home to continue outpatient follow-up that included serial radiographic, biochemical, and hematological testing. Throughout the study, the patients were carefully evaluated for evidence of toxicity, viral shedding, immune responses to the virus, and radiographic evidence of tumor response. As summarized in Table 1, 26 patients (21 male, 5 female), ranging in age from 37 to 81, were enrolled in the study between November 1995 and November 1997 [26]. The overall goal of this trial was to determine the toxicity, efficacy of gene transfer, and immune responses generated in response to the intrapleural instillation of Ad.HSV-TK. Clinical toxicities of the Ad.TK/GCV gene therapy were minimal and a maximal tolerated dose (MTD) was not achieved. Intratumoral HSV-TK gene transfer was documented in 17 of 25 evaluable

74/M

III/E

(S) 3.2 × 1011

Deceased

7

4

68/M

III/S

(S)

3.2 × 1011

Deceased

0.5

6

71/M

IB/E

(S) 3.2 × 1011

Progressed

39

4

75/M

IB/E

(S) 3.2 × 1011

Deceased

31

5

81/M

II/E

(S) 3.2 × 1011

Deceased

15

6

72/M

II/E

1 × 1012

Deceased

21

4

65/M

II/E

1 × 1012

Deceased

6.5

6

67/M

IA/S

1 × 1012

Deceased

23

6

a Patient

009 was unable to have the follow-up thoracoscopic biopsy. 012 had a pseudomesotheliomatous adenocarcinoma. Patients 19–23 (S) received high-dose corticosteroids at time of vector instillation. Note: Gene transfer scoring system: 0 = No gene transfer by any method, 1 = + DNA PCR in single sample, 2 = + DNA PCR in >1 sample, 3 = + RT-PCR or + in situ hybridization, 4 = + immunohistochemical detection of HSV-TK in few cells or positive immunoblot within single biopsy, 5 = + immunohistochemistry in few cells on multiple biopsies, 6 = + immunohistochemistry in many cells on multiple biopsies. b Patient

patients in a dose-related fashion by DNA-polymerase chain reaction (PCR), reverse transcription–PCR (RT-PCR), in situ hybridization, and immunohistochemistry (IHC) utilizing a murine monoclonal antibody directed against HSV-TK. All patients treated at dose levels of 3.2 × 1011 PFU or greater demonstrated evidence of intratumoral HSV-TK expression via IHC [26]. In general, the treatment protocol was well tolerated at all dosage levels. Toxicities were non-dose-limiting and included mild liver function test abnormalities, anemia, fever, and bullous exanthem at the instillation site. No MTD level was attained. At the highest dose level of


1 × 1012 PFU, two of three patients developed transitory hypotension and hypoxemia within hours after vector instillation that resolved with supplemental oxygen and intravenous fluids. Miscellaneous toxicities included atrial tachyarrhythmias, lymphopenia, and migratory polyarthralgias, each in a single patient [26]. Strong antiadenoviral humoral and cellular immune responses were noted, including neutrophil-predominant intratumoral inflammation in the posttreatment biopsy, generation of high titers of antiadenoviral neutralizing antibodies in serum and pleural fluid, significant increases in inflammatory cytokine production (TNF-α, IL-6) in pleural fluid, generation of serum antibodies against adenoviral structural proteins, and increased peripheral blood mononuclear cell proliferative responses to adenoviral proteins [35]. In a small substudy, five patients (patients 19–23) underwent administration of intravenous corticosteroids prior to and immediately following vector delivery [36]. This pilot trial was designed to preliminarily assess the effects of immunosuppression upon the degree of intratumoral gene transfer and antiadenoviral immune responses and was based on animal experiments showing that immunosuppression with dexamethasone augmented antitumor efficacy [37]. Results indicated a decreased incidence of fever and hypoxemia in the corticosteroid-treated cohort but an increased incidence of reversible mental status changes (“steroid psychosis”), particularly with a higher dose of methylprednisilone [36]. No diminution in humoral or cellular immune responses to the adenoviral vector was demonstrated in the group receiving corticosteroids, nor were there any detectable differences in the degree of intratumoral gene transfer. As a phase I trial, the focus of this initial study was on safety issues and establishment of a MTD. Because of the heterogeneity of the patient population in terms of age, stage, histology, and vector dose, the clinical efficacy of Ad.RSVTK/GCV gene therapy in malignant pleural mesothelioma was difficult to assess. Of the 26 patients enrolled in the initial phase I trial, 22 have died, with a median survival posttreatment of approximately 11 months and no fatal complications attributable to the gene therapy protocol (see Table 1). One patient (20) in the corticosteroid group who had stage IV mesothelioma at the time of enrollment died in the intensive care unit 2 weeks after completion of the protocol from rapid progression of his mesothelioma with malignant involvement of the contralateral hemithorax. Four of the 26 patients enrolled in this initial protocol were alive and available for evaluation as of August 2000. All four had stage IA or IB disease at the time of enrollment, and all have had clinical and/or radiographic evidence of progression of disease. The median survival of the four surviving patients posttreatment is 50 months, significantly longer than the median survival of 8–14 months for mesothelioma patients in general. Of the trial participants who are deceased, all had progressive mesothelioma as their primary cause of death, typically with invasion of mediastinum, contralateral

497

hemithorax, and transdiaphragmatic extension, as well as widespread metastatic disease, a fairly common finding in advanced-stage mesothelioma. Only one of the 26 patients (patient 26) had radiographic evidence of intrathoracic tumor regression posttreatment on follow-up chest computed tomography (CT) scan. This patient eventually died from intraperitoneal disease progression. At autopsy, extensive intraabdominal tumor was observed but relatively minimal disease in the treated thoracic cavity.

D. Adjunctive Phase I Clinical Trials 1. Ad.HSV-TK Gene Therapy for Mesothelioma with Third-Generation Vector We demonstrated in our first phase I trial that intrapleural Ad.HSV-TK gene therapy was safe, could effectively deliver transgene to superficial areas of mesothelioma tumor nodule, and induced significant humoral and cellular responses to the Ad vector [26,35]. Nevertheless, we felt that in order to achieve significant clinical responses warranted for phase II studies, improved intratumoral gene transfer was necessary. We decided to achieve this goal initially by increasing the vector dose, but doing so with the first-generation vector became problematic because of high levels of homologous recombination during large-scale production for clinical-grade lots, producing unacceptable levels of replication-competent adenovirus. In addition, there were some concerns regarding the hepatotoxicity and systemic inflammatory responses of first-generation adenoviral vectors as doses were increased, consistent with our findings in the highest dose cohort from the first trial. For these reasons, in June 1998 we started a new phase I clinical trial employing an advanced-generation adenoviral vector, with the goal of maximizing vector dose with minimal toxicity [38]. This new vector contained deletions in the E1 and E4 regions with preservation of the E3 region (Fig. 1B). The presence of an intact E4 region, unlike E3, is critical to the late phase of the viral life cycle. E4 deletions engender decreased viral DNA synthesis and late gene expression as well as instability of late mRNAs [39]. Therefore, adenoviral vectors with lethal deletions in E1 and E4 purportedly offer a significant advantage over first-generation vectors with only a single lethal deletion in the E1 region and thus have diminished cytopathic effects and reduced cellular immune responses [40]. In addition, because two replication-necessary genes are deleted, simple recombination could not produce a replication-competent virus, allowing for production of large amounts of clinical-grade vector at lower cost. The primary goals of the second phase I clinical trial were to determine the toxicity, gene transfer efficiency, and immune responses associated with the intrapleural injection of high titers of the E1/E4-deleted Ad.RSV-TK combined with systemic ganciclovir. To date, five patients have been treated, starting at a dose 1 log lower than the highest dose used

498


with the E1/E3-deleted Ad vector. The first two patients were treated at a dose of 1.5 × 1013 viral particles. At this dose, we saw minimal toxicity, primarily transitory fever (grade 1) developing approximately 24 hours after vector instillation. Patients treated at this dose level did not exhibit other adverse systemic reactions to vector instillation nor did they develop elevated liver enzymes or bullous skin lesions. The next three patients were treated with a dose of 5.0 × 1013 viral particles with evidence of increased but non-dose-limiting toxicity. All three patients experienced acute febrile responses (grade 1) after vector instillation, with rapid defervescence. One patient (29) developed hypotension and hypoxemia (grade 2) within hours after vector administration which resolved with supplemental oxygen and intravenous fluids. Patient 29 also developed elevated serum transaminases to levels approximately two to three times normal (grade 2) after vector delivery, peaking during the first week of ganciclovir therapy, but returning to normal levels by completion of the protocol. The patient had no associated elevations in serum bilirubin or prothrombin time and no clinical evidence of hepatic dysfunction. The third patient treated at the higher dose level (patient 31) developed low-grade fever (grade 1) after intrapleural vector instillation, as well as a contralateral inflammatory pleural effusion associated with moderate pleuritic chest pain (grade 2). The latter was suggestive of an induced immune response directed against mesothelial antigens. Patient 31 had no signs of hepatotoxicity. Overall, there appeared to be equal or lower hepatoxicity in the patients treated with the E1/E4-deleted vector compared to patients treated with equivalent doses of the E1/E3-deleted adenovirus but a similar pattern of increased but non-dose-limiting systemic side effects at higher dose levels [38]. Gene transfer was detected in all patients at both dose levels via immunohistochemistry using a murine monoclonal antibody directed against the HSV-TK protein. As in the initial phase I trial, gene transfer appeared to be dose related, with the patients at the higher dose level having more extensive staining on their posttreatment biopsies. As in the initial phase I trial, significant humoral responses to the recombinant adenoviral vector were seen, with the development of high titers of total and neutralizing antiadenoviral antibodies within 15–20 days of vector instillation in all five patients. Deletion of the E3 region, therefore, did not seem to impact on the immunogenicity of the vector, at least in this small group of patients [38]. Of the five patients treated, two are surviving (patients 29, 30), both of them treated at the higher dose level of 5.0 × 1013 particles of Ad.HSV-TK. Each of the patients had evidence of stable disease for at least 12 months after treatment. Patient 29, a 34-year-old female with stage I epithelioid mesothelioma, demonstrated evidence of decreased tumor metabolic activity on follow-up 18-fluorodeoxyglucose (18 FDG) PET scan performed at day 80. She had an additional 18 FDG PET scan at the University of Adelaide, Australia, 10 months after completion of the protocol which demonstrated minimal

pleural FDG uptake. Concomitantly, the patient’s clinical status remained stable without other antitumor therapy, and serial chest CT scans have shown no evidence of progression of pleural thickening or nodularity. This delayed decrease in tumor metabolic activity several months after completion of the Ad.RSV-TK/GCV gene therapy protocol suggests the development of an induced antitumor immune response. She has had no antineoplastic therapy other than this gene therapy protocol. Patient 30, a 57-year-old with stage I pleural mesothelioma, had stable disease clinically and radiographically 12 months post completion of the protocol despite refraining from other antineoplastic treatment. At approximately 18 months post treatment, the patient developed increasing chest wall discomfort associated with slowly progressing ipsilateral pleural thickening consistent with progressive disease [38].

2. Ad.HSV-TK Gene Therapy for Mesothelioma with Dose Escalation of Ganciclovir One other approach to augment Ad.HSV-TK gene therapy is to increase the dose of administered ganciclovir (Fig. 2, top section). In vitro and animal experiments clearly show that after tumor transduction with HSV-TK, the cytotoxic

FIGURE 2 Strategies to augment HSV-TK efficacy.


response is directly related to the GCV dose [23,25–28,40]. The dose of GCV used in all of the human trials (5 mg/kg i.v. b.i.d.) was chosen based upon the in vitro sensitivity testing of viral isolates and in vivo pharmacological measurements [42], as well as clinical experience with AIDS-related cytomegalovirus (CMV) retinitis. Based upon this hypothesis, we initiated a phase I clinical trial in July 1999 involving intrapleural delivery of the E1/E4-deleted Ad vector followed by intravenous GCV with gradual dose-escalation of the nucleoside analog. We have so far completed the first of four prospective cohorts in this study, with the first group of three patients being treated with 3.0 × 1013 particles of Ad.RSVTK and 7.5 mg/kg ganciclovir i.v. b.i.d. (15 mg/kg/day). All three patients tolerated the treatment well. The most common toxicity was transitory fever within 24 hours post vector instillation. Other adverse events included grade 3 lymphopenia in patient 102 (discussed later), elevated gamma-glutamyl transferase (GGT) and lactic dehydrogenase (LDH) in one patient (grade 2), hyponatremia and hypokalemia in one patient (grade 2), and thrombocytosis in two patients. These toxicities were all non-dose-limiting and should not preclude advancement to the next GCV dose level of 10 mg/kg/dose (20 mg/kg/day × 14 days). The initial patient in cohort 1, patient 101, was a 73-yearold male diagnosed with stage I epithelioid mesothelioma who underwent Ad.HSV-TK instillation followed by 28 doses of GCV at 7.5 mg/kg/dose. Successful HSV-TK gene transfer to tumor was confirmed by immunohistochemical evaluation of biopsy samples. He had minimal side effects from the GCV infusion. Thoracostomy tube drainage and subsequent vector infusion engendered full lung expansion and sclerosis of the pleural space with significant improvement in the patient’s exercise capacity and performance status. Review of the patient’s day 80 postprotocol chest CT scan demonstrated increased diffuse pleural thickening consistent with a postvector infusion inflammatory reaction. No new pleural or parenchymal nodules or masses were noted, nor was there involvement of the chest wall, mediastinum, abdominal cavity, or contralateral lung to indicate disease progression. In addition, as per protocol, the patient underwent pretreatment, day 80, and Day 170 18 FDG PET scans. His pretreatment scan demonstrated intense 18 FDG uptake in the mediastinal pleural and right hilar regions as well as in the posterior parietal pleura. This uptake was reduced on the day 80 scan consistent with a decrease in tumor metabolic activity (Figure 3). Subsequent 18 FDG PET at day 170 showed a dramatic increase in tracer uptake consistent with increased tumor metabolic activity. This correlated with the patient’s increasing shortness of breath and right anterior chest wall fullness and discomfort, as well as with the findings of his repeat chest CT scan, which revealed progressive pleural thickening and nodularity with encasement of the right lung. Patients 102 and 103 both demonstrated increased 18 FDG uptake on their follow-up day 80 PET studies and also had clear evidence of progression on standard chest CT studies. Clear assessment of antitumor

499

FIGURE 3 Pre- and posttreatment 18-fluorodeoxyglucose (FDG) positron emission tomography (PET) scans from patient 101 in Phase I clinical trial of intrapleural Ad.TK followed by escalating doses of systemic GCV. The initial scan on the left demonstrates a rightward shift of the mediastinum caused by the large pleural effusion with intense FDG uptake in the mediastinal pleural and right hilar regions as well as the posterior parietal pleura. This uptake was dramatically reduced on the day 80 scan consistent with a marked decrease in tumor metabolic activity.

activity awaits determination of the MTDs of both GCV and Ad.HSV-TK, as well as conduct of phase II studies.

III. CHALLENGES AND FUTURE DIRECTIONS Evidence of showing limited toxicity and detectable gene transfer, as well as our anecdotal experience with tumor responses, suggest that the Ad.HSV-TK approach has exciting potential for the treatment of malignant mesothelioma, as well as other localized malignancies. In addition, one of the most valuable aspects of our trial has been the identification of specific challenges that must be addressed to make this system more useful. These include limitations in gene transfer efficiency and difficulties in noninvasively assessing gene transfer.

A. Strategies To Optimize Patient Selection Although we obtained gene transfer in areas below the surface of a tumor, penetration was limited. Thus, using the current strategy, therapeutic efficacy could only be expected in patients with relatively small tumor loads (small nodules or diffuse, thin tumors). There are at least two ways in which we could create this clinical situation. First, once a MTD is reached, patients with only small amounts of pleural disease (nodules less than 5 mm) can be treated. Second, and probably more practical, patients with more than minimal disease could undergo a surgical “debulking” to minimize tumor burden. Gene therapy could be administered in the operating room as an adjuvant therapy after most of the tumor has been removed.

500


B. Strategies To Augment Gene Transfer There is a subgroup of patients who are not good candidates for debulking surgery because of visceral pleural involvement. Because our data from the clinical trial suggest that gene transfer is possible even in patients with titers of anti-Ad neutralizing antibodies of up to 1:500, we postulate that repeated administration of vector and GCV (e.g., three doses over a 3-week period) will lead to augmented gene transfer. Animal data support this hypothesis. Recently completed studies in immunocompetent mice with established peritoneal tumors by our group [43] and others [44] showed marked increases in efficacy after multiple courses of intraperitoneal injections of Ad.HSV-TK, each followed by a course of GCV therapy. In our study, efficacy was increased equally well in those mice who had previously been immunized with adenovirus and had developed neutralizing antibodies [43]. Another approach to the gene transfer problem is to maximize the efficacy of any of the HSV-TK enzyme that is expressed (Fig. 2, lower section). The underlying principal of our suicide gene approach is that the herpes simplex virus– thymidine kinase-1 enzyme has a relaxed specificity (in comparison to mammalian thymidine kinase) that allows it to phosphorylate not only thymidine but also other nucleoside analogs such as ganciclovir (GCV) and acyclovir (ACV). Unfortunately, HSV-TK has a high affinity for thymidine (Km = 0.5μM), whereas the affinity for GCV (Km = 45μM) and ACV (Km = >400 μM) are much lower [45]. This relationship suggests at least two ways in which the efficacy of HSV-TK could be augmented. First, higher levels of GCV could be provided to drive the equilibrium away from thymidine (Fig. 2, top section). This approach is already being used in our ongoing clinical trial (see earlier discussion). Second, “molecular remodeling” of the HSV-TK enzyme has been performed with the goal of increasing the substrate specificity towards GCV and ACV and concomitantly to decreasing thymidine utilization (Fig. 2, lower section). As described in detail [45], a segment of the HSV-TK gene at the putative nucleoside-binding site was substituted with random nucleotide sequences. Mutant enzymes that demonstrated preferential phosphorylation of GCV or ACV were selected from more than one million Escherichia coli transformants (Fig. 2, lower section). These mutants show enhanced acyclovir and ganciclovir killing and bystander effects [46]. We are currently producing and testing adenoviral vectors containing the mutated HSV-TK and anticipate they will enhance cell killing and augment the bystander effect. A growing body of evidence supports the hypothesis that, in most models tested, treatment with HSV-TK/GCV results in immunologic reactions against tumor cells that enhance killing efficacy [14,17,47– 49]. One reason for this antitumor immune reaction may be that in many cases, HSVTK/GCV-mediated cell killing occurs through a nonapoptotic (i.e., necrotic) pathway, a type of cell death that effectively

generates appropriate “danger signals” which then trigger significant immune responses [49,50]. With this rationale in mind, a number of investigators have conducted experiments showing that when the HSV-TK gene plus a cytokine gene are transduced into malignant cells, augmented tumor killing efficacy is achieved. To provide a few examples, augmented tumor killing effects have been reported with HSVTK plus IL-2 in mouse liver metastasis from colon carcinoma [51], a mouse squamous cell carcinoma model [52], a murine melanoma model [53], and a rat intraperitoneal colon cancer model [54]. Synergistic effects have also been reported with HSV-TK and interferon-alpha in Friend erythroleukemia cells [55] and with HSV-TK and granulocyte– macrophage colony-stimulating factor (GM-CSF) in mouse liver metastasis from colon carcinoma [56]. Animal studies are underway in mouse models of mesothelioma to determine the best combination of cytokines with HSV-TK, as well as the best way to combine these therapies (i.e., direct injection of cytokine versus delivery of cytokine using gene therapy). Finally, we hypothesize that a vector capable of replication in tumor cells (even only one to two rounds of replication) would allow much greater gene transfer. In this system, tumor killing could occur via two mechanisms: direct tumor lysis due to viral replication and by HSV-TK-mediated killing after administration of GCV. We anticipate a host immune response will limit viral replication and prevent widespread dissemination. However, it is likely that the generation of a tumor-selective replicating virus would be an important safety feature. We therefore plan to develop and evaluate tumor-selective replicating adenovirus-HSV-TK vectors [57]. To do this, we will substitute the adenoviral E1 promoter with tumorselective promoters (Fig. 1C). This is an approach that has been successfully used with the prostate-specific antigen (PSA) promoter to create a virus that selectively replicates in prostate cancer cells [58]. A number of promising choices for mesothelioma include the manganese-superoxide dismutase (MnSOD) promoter. Recent work by the Kinnula group in Finland has shown that MnSOD is very highly expressed in human malignant mesothelioma tissues and cell lines in contrast to normal lung or pleural tissues [59]. Two alternative mesothelioma “selective” promoters are those for the genes calretinin or mesothelin. Calretinin is a 29-kDa calciumbinding protein that is expressed primarily in the central and peripheral nervous system. Interestingly, high levels of calretinin expression have also been noted in mesothelial and mesothelioma cells, with very low expression levels in almost every other peripheral tissue studied [60,61]. Mesothelin is a 40-Kda surface protein of unknown function that is expressed only on the tissues forming the pleural, pericardial, and peritoneal membranes [62]. Other more general tumor-selective promoters, such as promoters responsive to the transcription factor E2F [63] or the survivin gene [64], would also be candidates.


FIGURE 4 Schema of noninvasive imaging of HSV-TK gene transfer utilizing GCV analogs labeled with radioactive tags that can be visualized by PET scanning. Trapping of the radiolabeled substrate will occur in tissues expressing HSV-TK and allow visualization of functional gene transfer.

C. Strategies To Assess Gene Transfer Noninvasively To date, the only method available to assess gene transfer is to biopsy tumor tissue with subsequent analysis for transgene DNA, RNA, or protein. The ability to measure gene transfer in a quantitative, noninvasive manner would have significant benefits for our clinical trials, as well as others. We would be able to rationally compare different treatment regimens, dosing schedules, and new vectors. Accordingly, we and others [65,66] are developing an approach using GCV analogs labeled with radioactive tags that can be visualized by positron emission tomography (PET) scanning. Trapping of the radiolabeled substrate will occur in tissues expressing HSV-TK and allow visualization of functional gene transfer (Fig. 4.). This methodology has been validated in animal models and will be tested in our clinical trials as a method of noninvasively assessing transgene expression in later trials.

References 1. Tiberghien, P. (1994). Use of suicide genes in gene therapy. J. Leukoc. Biol. 56, 203–209. 2. Huber, B. E., Austin, E. A., Richards, C. A., Davis, S. T., and Good, S. S. (1994). Metabolism of 5-fluorocytosine to 5-fluorouacil in human colorectal tumor cells transduced with the cytosine deaminase gene: significant antitumor effects when only a small percentage of tumor cells express cytosine deaminase. Proc. Natl. Acad. Sci. USA 91, 8302– 8306. 3. Hoganson, D. K., Batra, R. K., Olsen, J. C., and Boucher, R. C. (1996). Comparison of the effects of three different toxin genes and their levels of expression on cell growth and bystander effect in lung adenocarcinoma. Cancer Res. 56, 1315–1323. 4. Moolten, F. L., and Wells, J. M. (1990). Curability of tumors bearing herpes thymidine kinase genes transferred by retroviral vectors. J. Natl. Canc. Inst. 82, 297–300. 5. Moolten, F. L., Wells, J. M., and Mroz, P. J. (1992). Multiple transduction as a means of preserving ganciclovir chemosensitivity in sarcoma cells carrying retrovirally transduced herpes thymidine kinase genes. Cancer Lett. 64, 257–263. 6. Matthews, T., and Boehme, R. (1988). Antiviral activity and mechanism of action of ganciclovir. Rev Infect Dis. 10, S490–S494.

501

7. Rubsam, L. Z., Davidson, B. L., and Shewach, D. S. (1998). Superior cytotoxicity with ganciclovir compared with acyclovir and 1-βD-arabinofuranosylthymine in herpes simplex virus–thymidine kinase expressing cells: a novel paradigm for cell killing. Cancer Res. 58, 3873–3882. 8. Culver, K. W., Ram, Z., Wallbridge, S., Ishii, I., Oldfield, E. H., and Blaese, R. M. (1992). In vivo gene transfer with retroviral vectorproducer cells for treatment of experimental brain tumors. Science 256, 1550–1552. 9. Ram, Z., Culver, K. W., Walbridge, B., Blaese, R. M., and Oldfield, E. H. (1993). In situ retroviral-mediated gene transfer for the treatment of brain tumors in rats. Cancer Res. 53, 83–88. 10. Smythe, W. R., Hwang, H. C., Elshami, A. A., Amin, K., Eck, S., Davidson, B., Wilson, J., Kaiser, L. R., and Albelda, S. M. (1995). Successful treatment of experimental human mesothelioma using adenovirus transfer of the herpes simplex–thymidine kinase gene. Ann. Surg. 222, 78–86. 11. Bonnekoh, B., Greenhalgh, D., Bundman, S., Eckhardt, J. N., Longley, M. A., Chen, S. H., Woo, S. L., and Roop, D. R. (1995). Inhibition of melanoma growth by adenoviral-mediated HSV thymidine kinase gene transfer in vivo. J. Invest. Dermatol. 104, 313–319. 12. Freeman, S. M., Abboud, C. N., Whartenby, K. A., Packman, C. H., Koeplin, D. S., Moolten, F. L., and Abraham, G. N. (1993). The “bystander effect”: tumor regression when a fraction of the tumor mass is genetically modified. Cancer Res. 53, 5274–5283. 13. Hasegawa, Y., Emi, N., Shimokata, K., Abe, A., Kawabe, T., Hasegawa, T., Kirioka, T., and Saito, H. (1993). Gene transfer of herpes simplex virus type I thymidine kinase gene as a drug sensitivity gene into human lung cancer lines using retroviral vectors. Am. J. Resp. Cell. Mol. Bio. 8, 655–661. 14. Caruso, M., Panis, Y., Gagandeep, S., Houssin, D., Salzmann, J. L., and Klatzmann, D. (1993). Regression of established macroscopic liver metastases after in situ transduction of a suicide gene. Proc. Natl. Acad. Sci. USA 90, 7024–7028. 15. Elshami, A. A., Saavedra, A., Zhang, H. B., Kucharczuk, J. C., Spray, D. C., Fishman, G. I., Kaiser, L. R., and Albelda, S. M. (1996). Gap junctions play a role in the bystander effect of the herpes simplex virus thymidine kinase/ganciclovir system in vitro. Gene Ther. 3, 85–92. 16. Mesnil, M., and Yamasaki, H. (2000). Bystander effect in herpes simples virus-thymidine kinase/ganciclovir cancer gene therapy: role of gapjunctional intercellular communication. Cancer Res. 60, 3989–3999. 17. Pope, I. M., Poston, G. J., and Kinsella, A. R. (1997). The role of the bystander effect in suicide gene therapy. Eur. J. Cancer 33, 1005–1016. 18. Vile, R. G., and Hart, I. R. (1993). Use of tissue-specific expression of the herpes simplex virus thymidine kinase gene to inhibit growth of established murine melanomas following direct intratumoral injection of DNA. Cancer Res. 53, 3860–3864. 19. Vile, R. G., Nelson, J. A., Castleden, S., Chong, H., and Hart, I. R. (1994). Systemic gene therapy of murine melanoma using tissue specific expression of the HSVtk gene involves an immune component. Cancer Res. 54, 6228–6234. 20. Moolten, F. L. (1994). Drug sensitivity (“suicide”) genes for selective cancer therapy. Cancer Gene Ther. 1, 279–287. 21. Kozarsky, K., and Wilson, J. M. (1993). Gene therapy: adenovirus vectors. Curr. Opin. Genet. Dev. 3, 499–503. 22. Chen, S. H., Shine, H. D., Goodman, J. C.,. Grossman, R., and Woo, S. L. C. (1994). Gene therapy for brain tumors: regression of experimental gliomas by adenovirus-mediated gene transfer in vivo. Proc. Natl. Acad. Sci. USA 91, 3054–3057. 23. Hwang, H. C., Smythe, W. R., Elshami, A. A., Kucharczuk, J. C., Amin, K., Williams, J. P., Litzky, L. A., Kaiser, L. R., and Albelda, S. M. (1995). Gene therapy using adenovirus carrying the herpes simplex thymidine kinase gene to treat in vitro models of human malignant mesothelioma and lung cancer. Am. J. Resp. Cell Molec. Biol. 13, 7–16.

502


24. Elshami, A., Kucharczuk, J., Zhang, H., Smythe, W., Huang, H., Amin, K., Litzky, L., Kaiser, L. R., and Albelda, S. M. (1996). Treatment of pleural mesothelioma in an immunocompetent rat model utilizing adenoviral transfer of the HSV–thymidine kinase gene. Hum. Gene Ther. 7, 141–148. 25. Perez-Cruet, M. J., Trask, T. W., Chen, S. H., Goodman, J. C., Woo, S. L. C., Grossman, R. G., and Shine, H. D. (1994). Adenovirusmediated gene therapy of experimental gliomas. J. Neurosci. Res. 39, 506–511. 26. Sterman, D. H., Treat, J., Litzky, L. A., Amin, K., Molnar-Kimber, K., Wilson, J., Albelda, S. M., and Kaiser, L. R. (1998). Adenovirusmediated herpes simplex virus thymidine kinase gene delivery in patients with localized malignancy: results of a phase I clinical trial in malignant mesothelioma. Hum. Gene Ther. 9, 1083–1092. 27. Alavi, J. B., and Eck, S. L. (1998). Gene therapy for malignant gliomas. Hematol. Oncol. Clin. N. Am. 12, 617–629. 28. Morris, J. C., Ramsey, W. J., Wildner, O., Muslow, H. A., AguilarCordova, E., and Blaese R. M. (2000). A phase I study of intralesional administration of an adenovirus vector expressing the HSV-1 thymidine kinase gene (AdV.RSV-TK ) in combination with escalating doses of ganciclovir in patients with cutaneous metastatic melanoma. Hum. Gene Ther. 11, 487–503. 29. Treat, J., Kaiser, L. R., Sterman, D. H., Litzky, L. A., Davis, A., Wilson, J. M., and Albelda, S. M. (1996). Treatment of advanced mesothelioma with the recombinant adenovirus H5.010RSVTK: a phase I trial (BB-IND 6274). Hum. Gene Ther. 7, 2047–2057. 30. Schwarzenberger, P., Harrison, L., Weinacker, A., Marrogi, A., Byrne, P., Ramesh, R., Theodossiou, C., Gaumer, R., Summer, W., Freeman, S. M., and Kolls, J. K. (1998). The treatment of malignant mesothelioma with a gene modified cancer cell line: a phase I study. Hum. Gene Ther. 9, 2641–2649. 31. Smythe, W. R., Hwang, H. C., Amin, K. M., Eck, S., Wilson, J., Kaiser, L. R., and Albelda, S. M. (1994). Use of recombinant adenovirus to transfer the herpes simplex virus thymidine kinase (HSVtk) gene to thoracic neoplasms: an effective in vitro drug sensitization system. Cancer Res. 54, 2055–2059. 32. Smythe, W. R., Kaiser, L. R., Amin, K. M., Pilewski, J., Eck, S., Wilson, J., and Albelda, S. M. (1994). Successful adenovirus-mediated gene transfer in an in vivo model of human malignant mesothelioma. Ann. Thor. Surg. 57, 1395–1401. 33. Esandi, M. C., van Someren, G. D., Vincent, A. J., van Bekkum, D. W., Valerio, D., Bout, A., and Noteboom, J. L. (1997). Gene therapy of experimental malignant mesothelioma using adenovirus vectors encoding the HSVtk gene. Gene Ther. 4, 280–287. 34. Kucharczuk, J. C., Raper, S., Elshami, A. A., Amin, K., Sterman, D. H., Litzky, L. A., Kaiser, L. R., and Albelda, S. M. (1996). Safety of adenoviral-mediated transfer of the herpes simplex thymidine kinase cDNA to the pleural cavity of rats and non-human primates. Hum. Gene Ther. 7, 2225–2233. 35. Molnar-Kimber, K. L., Sterman, D. H., Chang, M., Elbash, M., Elshami, A., Roberts, J. R., Treat, J., Wilson, J. M., Kaiser, L. R., and Albelda, S. M. (1998). Humoral and cellular immune responses induced by adenoviral-based gene therapy for localized malignancy: results of a phase I clinical trial for malignant mesothelioma. Hum. Gene Ther. 9, 2121–2133. 36. Sterman, D. H., Molnar-Kimber, K., Iyengar, T., Chan, M., Lanuti, M., Amin, K. M., Pierce, B. K., Kang, E., Treat, J., Recio, A., Litzky, L. A., Wilson, J. M., Kaiser, L. R., and Albelda, S. M. (2000). A pilot study of systemic corticosteroid administration in conjunction with intrapleural adenoviral vector administration in patients with malignant pleural mesothelioma. Cancer Gene Ther. 7, 1511–1518. 37. Elshami, A., Kucharczuk, J., Sterman, D., Smythe, W., Hwang, H., Amin, K., Litzky, L., Albelda, S., and Kaiser, L. (1995). The role of immune suppression in the efficacy of cancer gene therapy using aden-

38.

39. 40.

41.

42.

43.

44.

45.

46.

47.

48. 49.

50.

51.

52.

53.

ovirus transfer of the herpes simplex–thymidine kinase gene. Ann. Surg. 224, 298–310. Sterman, D. H., Recio, A., Molnar-Kimber, K., Knox, L., Hughes, J., Alavi, A., Lanuti, M., Litzky, L. A., Albelda, S. M., and Kaiser, L. R. (1999). Herpes simplex virus thymidine kinase (HSVtk) gene therapy utilizing an E1/E4-deleted adenoviral vector: preliminary results of a phase I clinical trial for pleural mesothelioma. Am. J. Resp. Crit. Care Med. 159, A237. Wang, Q., and Finer, M. H. (1996). Second-generation adenoviral vectors. Nat. Med. 2, 714–716. Gao, G. P., Yang, Y., and Wilson, J. M. (1996). Biology of adenovirus vectors with E1 and E4 deletions for liver-directed gene therapy. J. Virol. 70, 8934–8943. Kato, K., Yoshida, J., Mizuno, M., Sugita, K., and Emi, N. (1994). Retroviral transfer of herpes simplex thymidine kinase gene into glioma cells causes targeting of ganciclovir cytotoxic effect. Neurol. Med. Chir. (Tokyo) 34, 339–344. Jacobson, M. A. (1997). Treatment of cytomegalovirus retinitis in patients with the acquired immunodeficiency syndrome. N. Engl. J. Med. 337, 105–114. Lambright, E. S., Force, S. D., Lanuti, M., Wasfi, D. S., Amin, K., Albelda, S. M., and Kaiser, L. R. (2000). Efficacy of repeated adenoviral suicide gene therapy in a localized murine tumor model. Ann. Thor. Surg. 70, 1865–1871. Al-Hendy, A., Magliocco, A. M., Al-Tweigeri, T., Braileanu, G., Crellin, N., Li, H., Strong, T., Curiel, D., and Chedrese, P. J. (2000). Ovarian cancer gene therapy: repeated treatment with thymidine kinase in an adenovirus vector and ganciclovir improves survival in a novel immunocompetent murine model. Am. J. Obstet. Gynecol. 182, 553– 559. Black, M. E., Newcomb, T. G., Wilson, H. M. P., and Loeb, L. A. (1996). Creation of drug-specific herpes simplex virus type 1 thymidine kinase mutants for gene therapy. Proc. Natl. Acad. Sci. USA 93, 3525– 3529. Qiao, H. J., Black, M. E., and Caruso, M. (2000). Enhanced ganciclovir killing and bystander effect of human tumor cells transduced with retroviral vector carrying a herpes simplex thymidine kinase gene mutant. Hum. Gene Ther. 11, 1569–1576. Hall, S. J., Sanford, M. A., Atkinson, G., and Chen, S. H. (1998). Induction of potent antitumor natural killer cell activity by herpes simplex virus–thymidine kinase and ganciclovir therapy in an orthotopic mouse model of prostate cancer. Cancer Res. 58, 3221– 3225. Freeman, S. M., Ramesh, R., and Marogi, A. J. (1997). Immune system in suicide gene therapy. Lancet 349, 2–3. Vile, R. G., Castleden, S., Marshall, J., Camplejohn, R., Upton, C., and Chong, H. (1997). Generation of an anti-tumor immune response in a non-immunogenic tumour: HSVtk killing in vivo stimulates a mononuclear cell infiltrate and a Th1-like profile of intratumoural cytokine expression. Int. J. Cancer 71, 267–274. Melcher, A., Todryk, S., Hardwick, N., Ford, M., Jacobson, M., and Vile, R. (1998). Tumor immunogenicity is determined by the mechanism of cell death via induction of heat shock protein expression. Nat. Med. 4, 581–587. Chen, S. H., Li Chen, X. H., Wang, Y., Kosai, K. I., Finegold, M. J., Rich, S. S., and Woo, S. C. (1995). Combination gene therapy for liver metastasis of colon carcinoma in vivo. Proc. Natl. Acad. Sci. USA 92, 2577–2581. O’Malley, Jr., B., Cope, K. A., Chen, S. H., Li, D., Schwartz, M., and Woo, S. L. C. (1996). Combination gene therapy for oral cancer in a murine model. Cancer Res. 56, 1737–1741. Castleden, S. A., Chong, H., Garcia-Ribas, I., Melcher, A. A., Hutchinson, G., Roberts, B., Hart, I. R., and Vile, R. G. (1997). A family of bicistronic vectors to enhance both local and systemic antitumor


54.

55.

56.

57. 58.

59.

effects of HSVtk or cytokine expression in a murine melanoma model. Hum. Gene Ther. 8, 2087–2102. Coll, J., Mesnil, M., Lefebvre, M., Lancon, A., and Favrot, M. (1997). Long-term survival of immunocompetent rats with intraperitoneal colon carcinoma tumors using herpes simplex thymidine kinase/ganciclovir and IL-2 treatments. Gene Ther. 4, 1160–1166. Santodonato, L., D’Agostino, G., Santini, S., Carlei, D., Musiani, P., Modesti, A., Signorelli, P., Belardelli, F., and Ferrantini, M. (1997). Local and systemic antitumor response after combined therapy of mouse metastatic tumors with tumor cells expressing IFN-alpha and HSVtk: perspectives for the generation of cancer vaccines. Gene Ther. 4, 1246– 1255. Hayashi, S., Nobuhiko, E., Yokoyama, I., Namii, Y., Uchida, K., and Takagi, H. (1997). Inhibition of establishment of hepatic metastasis in mice by combination gene therapy using both herpes simplex virus– thymidine kinase and granulocyte macrophage-colony stimulating factor genes in murine colon cancer. Cancer Gene Ther. 4, 339–344. Alemany, R., Balague, C., and Curiel, D. T. (2000). Replicative adenoviruses for cancer therapy. Nat. Biotechnol. 18, 723–727. Rodriguez, R., Schuur, E. R., Lim, H. Y., Henderson, G. A., Simons, J. W., and Henderson, D. R. (1997). Prostate attenuated replication competent adenovirus (ARCA) CN706: a selective cytotoxic for prostate-specific antigen-positive prostate cancer cells. Cancer Res. 57, 2559–2563. Kahlos, K., Anttila, S., Asikainen, T., Kinnula, K., Raivio, K. O., Mattson, K., Linnainmaa, K., and Kinnula, V. L. (1998). Manganese superoxide dismutase in healthy human pleural mesothelium and in malignant pleural mesothelioma. Am. J. Respir. Cell. Mol. Biol. 18, 579–580.

503

60. Doglioni, C., Dei Tos, A. P., Laurino, L., Iuzzolino, P., Chiarelli, C., Celio, M. R., and Viale, G. (1996). Calretinin: a novel immunocytochemical marker for mesothelioma. Am. J. Surg. Pathol. 20, 1037–1046. 61. Gotzos, V., Vogt, P., and Celio, M. (1996). The calcium binding protein calretinin is a selective marker for malignant pleural mesotheliomas of the epithelial type. Pathol. Res. Pract. 192, 137–147. 62. Chang, K. and Pastan, I. (1996). Molecular cloning of mesothelin, a differentiation antigen present on mesothelium, mesotheliomas, and ovarian cancers. Proc. Natl. Acad. Sci. USA 93, 136–140. 63. Amin, K. M., Tsukuda, K., Odaka, M., Molnar-Kimber, K., Kaiser, L. R., and Albelda, S. M. (2001). The development and characterization of a mutant oncolytic adenovirus that replicates selectivity in ovarian and lung cancer cells over-expressing E2F-1 protein (abstr). Am. Assoc. Cancer Res. Annu. Meet., p. 3716. 64. Ambrosini G., Adid, C., and Altieri, D. C. (1997). A novel anti-apoptosis gene, surviving, expressed in cancer and lymphoma. Nat. Med. 3, 917– 921. 65. Gambhir, S. S., Barrio, J. R., Phelps, M. E., Iyer, M., Namavari, M., Satyamurthy, N., Wu, L., Green, L. A., Bauer, E., MacLaren, D. C., Nguyen, K., Berk, A. J., Cherry. S. R., and Herschman, H. R. (1999). Imaging adenoviral-directed repoerter gene expression in living animals with positron emission tomography. Proc. Natl. Acad. Sci. USA 96, 2333–2338. 66. Gambhir, S. S., Bauer, E., Black, M. E., Liang, Q., Kokoris, M. S., Barrio, J. R., Iyer, M., Namavari, M., Phelps, M. E., and Herschman H. R. (2000). A mutant herpes simplex virus type 1 thymidine kinase reporter gene shows improved sensitivity for imaging reporter gene expression with positron emission tomography. Proc. Natl. Acad. Sci. USA 97, 2785–2790.


C

H

A

P

T

E

R

33 The Use of Suicide Gene Therapy for the Treatment of Malignancies of the Brain KEVIN D. JUDY

STEPHEN L. ECK

HUP–Department of Neurosurgery The University of Pennsylvania Medical Center Philadelphia, Pennsylvania 19004

HUP–Department of Neurosurgery The University of Pennsylvania Medical Center Philadelphia, Pennsylvania 19004

I. II. III. IV.

Introduction 505 Retrovirus Vector for HSV-TK 506 Adenovirus Vector for HSV-TK 509 Herpes Simplex Virus Vectors Expressing Endogenous HSV-TK 510 V. Promising Preclinical Studies 510

Radiation therapy directed to the tumor bed, either alone or following maximal surgical resection, has been the most effective treatment to delay local regrowth [4,8]. Following aggressive treatment with surgery, radiation therapy, and chemotherapy, these high-grade gliomas invariably relapse. Growth of gliomas is restricted to the CNS, as these tumors do not metastasize to visceral organs or bone. The absence of widespread metastases or significant organ failure makes these patients excellent subjects for local experimental therapies. Patients with a good performance status can be treated at the time of recurrence with a second craniotomy combined with additional local therapy. Because the delivery of the genetic vector is the major limitation of current gene therapy technology, brain tumors are more attractive targets for this type of treatment compared to more common cancers that metastasize widely. Local therapies to the tumor bed have shown success in tumor control and reduce the systemic adverse effects from the agents. Implantable polymer wafers containing carmustine chemotherapy will deliver extremely high doses of chemotherapy directly to the tumor with minimal systemic exposure to the drug [9]. Disruption of the blood–brain barrier using osmotic diuretics or bradykinin analogs can enhance the penetration of chemotherapy agents into the tumor [7,10]. Local delivery of interferon and interleukin-2 (IL-2) through a tumor-embedded catheter (e.g., Ommaya reservoir) has been utilized to overcome the limitation of systemic administration of these short-lived and systemically toxic cytokines. Optimal local therapies must have the ability to kill both dividing and nondividing tumor cells, as the majority of tumor cells are not actively dividing at the time of treatment. The agent must be able to penetrate deeply into the tumor and surrounding tissues to destroy the tumor cells, which

References 511

I. INTRODUCTION High-grade gliomas (anaplastic astrocytoma and glioblastoma multiforme) are the most common and unfortunately the most lethal tumors of the brain that occur in adults. Survival for patients with anaplastic astrocytoma (WHO grade III) is usually less than 3 years and for patients with glioblastoma multiforme (WHO grade IV) is less than 1 year. Surgical resection of these tumors will reduce the mass effect of the tumor, thus improving quality of life and time to clinical progression of the tumor. It is impossible to completely resect high-grade gliomas due to the diffuse nature of the glioma cells dispersed throughout the surrounding “normal”-appearing brain [1]. These neoplastic cells contribute to tumor regrowth in the same location as the original tumor [2] as well as to migration of malignant cells to distant parts of the central nervous system (CNS). Chemotherapy has shown only modest success improving the survival of patients with these tumors [3,4]. The limited benefits of systemic chemotherapy have been attributed in part to inherent resistance of the tumor cells to the chemotherapy due to expression of alkylguanine-DNA alkyltransferase [5,6] and inability of the drugs to cross the blood–brain barrier [7].


505


506

Kevin D. Judy and Stephen L. Eck

FIGURE 1 Conventional chemotherapy for brain tumors (left panel) requires that a systemically distributed drug reach the tumor in concentrations sufficient to exert a tumoricidal effect. This approach is typically limited by systemic toxicity that prohibits dose escalation to levels sufficient for tumor eradication. Gene-directed enzyme prodrug therapy (right panel) permits the systemic administration of relatively nontoxic drugs (e.g., ganciclovir) which are only converted to their active form in cells that have been transduced to express the enzyme (e.g., HSV-TK) needed to activate them. Moreover, the activated drug can be locally redistributed within the tumor to nontransduced cells, achieving a “bystander effect.” This limits systemic exposure to the active form of the drug, which accumulates selectively within the tumor. Prodrugs can be selected for their ability to cross the blood–brain barrier, even though their activated forms may lack this ability. (See color insert.)

infiltrate the surrounding “normal” brain along white matter tracts. Gene-directed enzyme prodrug therapy (GDEPT) for brain tumors uses gene transfer as a drug delivery system. The genetic vector (adenovirus, retrovirus, and liposome vectors have been used clinically) delivers the suicide gene (typically herpes simplex virus–thymidine kinase, HSV-TK) into the tumor cell, where it phosphorylates ganciclovir (or other suitable substrate), creating a toxic metabolite that leads to abortive DNA synthesis and cell death [11]. Success of this therapy is dependent upon efficient penetration of the tumor by the vector and having rates of cell division that render the cells sensitive to DNA synthesis inhibitors. The requirement for efficient tumor cell transduction is to some degree lessened by the “bystander effect” (Fig. 1 [See also color insert]), which permits activated drug to pass from transduced to nontransduced cells [12]. Recombinant retroviruses and adenoviruses have been the most extensively studied vectors in clinical trials of GDEPT for brain tumors [13]. Recombinant herpes viruses carrying the endogenous HSVTK gene have been more recently introduced into clinical studies [14].

II. RETROVIRUS VECTOR FOR HSV-TK Replication-defective recombinant retroviruses were the first vectors to be used in human clinical trials of GDEPT. They had the perceived advantage of selectively transducing only dividing cells which would restrict therapeutic gene ex-

pression (e.g., HSV-TK) to tumor cells and spare the more slowly dividing normal brain parenchyma. The immediate limitation of retroviral vectors was the inability to highly concentrate the vector and thereby limit the volume of infusate to a reasonable size for instillation into the brain. Moreover, the vectors survive in vivo for only short periods of time relative to the rate of cell divisions (a requirement for retroviral entry into the nucleus). To circumvent these limitation, vectorproducing cells (VPCs) derived from murine fibroblasts were constructed to release the retroviral vector containing the HSV-TK gene [15–17]. In concept, multiple focal deposits of these VPCs would produce the retroviral vector over a sustained period of time (several days) before being cleared by the host immune systems (rejection of a xenograft). Multiple sites of the tumor (up to 50) have been injected with the VPCs to distribute the vector throughout the tumor [18]. Subsequent systemic administration of ganciclovir (GCV) leads to conversion by HSV-TK to GCV–monophosphate within the tumor cells and VPCs that are then rapidly phosphorylated by human cellular kinases to GCV–triphosphate [11]. This metabolite inhibits DNA replication, leading ultimately to the death of dividing tumor cells. Mammalian thymidine kinase is much less efficient in performing the initial phosphorylation which accounts for the low toxicity of systemically administered GCV (e.g., mild, reversible myelosuppression after prolonged use) [11]. A large number of in vitro and in vivo studies have demonstrated that a significant proportion of nontransduced cells in the vicinity of cells undergoing gene transfer are also killed by a bystander effect (metabolic cooperation). As noted, this

The Use of Suicide Gene Therapy for the Treatment of Malignancies of the Brain

effect arises in part from the intercellular transport of GCV– phosphates by way of gap junctions between adjacent tumor cells [19]. Other pathways have been proposed to contribute to the ‘bystander effect,’ including the release of apoptotic vesicles containing GCV metabolites and an induced immune response to nontransduced cells by released tumor antigens [16,20]. Whether or not such immune mechanisms contribute to eradication of brain tumors in patients remains to be determined. However, it would seem unlikely that an immune bystander effect plays a significant role given the immunosuppressive effects of the human brain tumors themselves and the concomitant use of high doses of immunosuppressive glucocorticoids that are routinely employed to control cerebral edema. Prior in vitro and in vivo studies have shown that only 10–50% of tumor cells need to be transduced (in laboratory models) to achieve complete tumor destruction. In addition, there is some evidence from animal and human studies to suggest that the HSV-TK gene delivery results in damage to blood vessels within the tumor microenvironment, potentially contributing to the therapeutic effect [21–23]. Ram et al. postulated that transduction of endothelial cells may have contributed to tumor response, because in some patients microhemorrhages were seen in the tumors by magnetic resonance imaging (MRI) scan during the first week of ganciclovir treatment [23]. Vascular injury has also been documented in histologic specimens obtained in non-human primate studies [21] and from post-gene-therapy resection specimens. The first application of the retroviral HSV-TK–GCV system in human brain tumors was performed by Ram et al. [18] (see Table 1). Fifteen patients, 12 with gliomas and three with metastases (two melanoma and one breast carcinoma), harboring 19 lesions were treated with 1 × 108 to 1 × 109 VPCs per treatment by stereotactic injection. Intravenous GCV was begun 7 days after the VPC injection to allow for vector release and transduction of the surrounding tumor cells. The TABLE 1 Outcome of Human Brain Tumor Trials Using Viral Vectors Expressing HSV-TK Vector

Patients alive >1 year (number/total)

Investigators

Ref.

RV–HSV-TK

3/15

Ram et al.

RV–HSV-TK

1/15

Izquierdo et al.

[25,26]

[18]

RV–HSV-TK

3/12

Klatzmann et al.

[27]

RV–HSV-TK

13/48

Shand et al.

[28]

RV–HSV-TK

62/124

Rainov et al.

[32,59]

Adeno–HSV-TK

3/13

Trask et al.

Adeno–HSV-TK

5/13

Eck et al.

Mutant HSV-1

2/21

Martuza et al.

[35] [34,60] [40]

Summary of patients surviving longer than 1 year from the time of gene therapy in clinical trials using viral vectors expressing HSV-TK.

507

GCV was given for a total of 14 days. Two patients had their tumors, resected 7 days following VPC injection, after which the tumor bed was reinjected with VPCs, and GCV was continued for 7 days following the tumor resection. Four patients showing antitumor activity received a second treatment. The principal adverse events noted were seizures and hemorrhage from the multiple injections. Not surprisingly, injection of the murine VPCs into patients stimulated production of antiVPC antibodies in the majority of patients. This immune response did not appear to contribute to toxicity or efficacy. The inadvertant introduction of replication-competent retrovirus (RCR) has been a subject of some concern following the induction of lymphoma in monkeys after RCR administration [24]. However, no RCR or vector DNA was identified in the patients’ peripheral blood samples. Two patients underwent biopsy of the tumor 7 days following treatment. In situ hybridization revealed expression of HSV-TK in the VPCs and in tumor cells immediately surrounding the VPCs. However, the most striking observation was the limited spread of the vector from the VPC sites [18]. Objective responses (i.e., 50% or more reduction in tumor volume) were seen in five lesions in four patients. Two of these patients with gliomas have remained alive with tumor control for several years. These five responding tumors had volumes of 1.4 ± 0.5 mL, suggesting that smaller tumors in which a high density of VPCs could be administered were most responsive to the treatment. This is consistent with the limited distribution of the HSV-TK gene expression in most patients and indicates that techniques to improve delivery and distribution of the therapeutic gene must be developed if clinical utility is to be achieved with this approach. A similar retroviral–HSV-TK construct was developed by Izquierdo et al. utilizing the previously described vector producing cell approach [25]. Five patients with recurrent glioblastoma multiforme were treated with multiple injections of the VPCs into the tumor followed by GCV. The treatment was tolerated well by all five patients. One patient had a significant reduction in the size of the treated tumor but not an adjacent tumor that was not treated. A second patient had a small reduction in size of the treated tumor. The remaining three patients did not have an observable response to the GDEPT. The same investigators proceeded with a second trial for recurrent glioblastoma multiforme (GBM) in which the recurrent tumor was resected, the tumor bed injected with VPCs, and an Ommaya reservoir placed into the tumor bed [26]. Three patients were entered into the trial, but one suffered an intracranial hemorrhage and never received the GDEPT. A second dose of VPCs was given to the two remaining patients through the Ommaya reservoir 1 week following the first injection, followed by 2 weeks of GCV beginning 1 week following the second injection. This allowed the VPCs deposits greater time to transduce the surrounding tumor cells. The patients were then retreated with VPCs through the Ommaya reservoir for the next 3–6 months

508


at times when follow-up MRI showed evidence of enhancement suggestive of recurrent tumor. The two patients survived 11 and 17 months with persistent enhancing tumor, suggesting incomplete eradication of all tumor cells. The apparent increase in survival suggests a potential benefit from this therapy; however, these findings have yet to be reproduced in a larger patient population. Klatzmann et al. [27] completed a phase I/II study of retrovirus expressing HSV-TK using VPCs in 12 patients with recurrent glioblastoma multiforme who underwent resection of the tumor with infiltration of the tumor bed with the VPC [27]. GCV was instituted 7 days postoperatively and continued for 14 days as in previous studies. In general, the treatment was tolerated well, with adverse events consisting of bacterial infections, hemorrhage following the surgery, and one patient experiencing progressive aseptic meningitis and death 2 months following treatment. The progressive deterioration in the latter patient was found at autopsy to be due to massive gliomatous invasion of the ventricular cavities. The adverse events were not directly attributable to the use of VPCs. A weak HSV-TK signal was detected in the blood of one patient 1 hour following injection of the VPCs using a polymerase chain reaction (PCR) technique. The remaining PCR investigations in that patient as well as the other patients did not detect transgene dissemination. Four of the 12 patients showed a lack of tumor progression by MRI after 4 months of follow-up. Three of the patients survived more than 1 year, with one patient being alive with no evidence of disease 2.8 years from treatment. MRI scan evidence of progressive tumor growth did not correlate with histological evidence of minimal tumor growth in two patients. The European/Canadian study group studied 48 patients with recurrent glioblastoma multiforme who were treated with HSV-TK VPCs following tumor resection [28]. The VPCs were directly injected into the tumor cavity, and the patients were given GCV for 14 days. The median survival was 8.6 months, with four patients having no evidence of disease at 12 months. Retrovirus could not be cultured from blood specimens. Retroviral DNA was detected in peripheral white blood cells in some patients, suggesting that retrovirus may have transduced reactive lymphocytes in the tumor bed in the brain [29]. An autopsy study of 32 patients that had received the VPC in clinical trials has been performed [30]. Twenty-four brain tumor biopsies from this group of patients were examined for RCV and vector DNA sequences by PCR assays. RCV was not detected in any of the samples, including normal tissue samples at autopsy; however, vector DNA was found in the scalp, kidney, liver, and lung. Vector DNA was found in brain tumor specimens (55%), adjacent brain tissue (22%), and contralateral brain (6.7%), with increased detection in those patients receiving multiple, rather than single, injections of vector. There was no evidence that detection of retroviral sequences by the sensitive PCR assay was in any way related to clinical outcome.

A study of serial MRI scans in patients receiving two or more cycles of VPCs with GCV was performed by Deliganis et al. [31]. They evaluated seven patients receiving between two and four infusions of VPCs via an Ommaya reservoir that was placed into the resected tumor cavity. These patients were followed with MRI scans every 40 days following the initial treatment. The changes in the areas of enhancing tissues were variable. One patient had a transient increase in tissue enhancement followed by a sustained decrease in enhancement. Two patients developed an early increase in enhancing tissue followed by a transient plateau in one and a transient decrease in the other. Another patient had a stable volume of enhancement for 132 days before developing a progressive increase in enhancing tissue. The remaining three patients showed serial increases in enhancing tissue and edema indicative of progressive disease. Patients having an initial increase in enhancing tissue were thought to be experiencing an acute inflammatory reaction to the VPC/retroviral treatment. These and other observations illustrate the potential shortcomings of conventional MRI which, using areas of enhancement, frequently cannot readily distinguish tumor from effects of the gene transfer itself. A large multicenter trial sponsored by Novartis Pharma and Genetic Therapy, Inc., has now been completed [32]. In this study, patients with newly diagnosed glioblastoma were randomized to receive either standard tumor resection and radiation therapy, or the same plus the addition of VPC injection at surgery, followed by 2 weeks of ganciclovir. The median times to tumor progression and survival were not different in the two treatment arms. One fatal complication occurred: an infection related to ganciclovir-induced neutropenia. No RCR was found in peripheral blood or autopsy specimens. Taken together, these studies using the retrovirus producer cells and the HSV-TK ganciclovir system demonstrate that the treatment can be given safely. There has been no evidence of systemic toxicity of the virus. The virus has been detected by PCR in a very small percentage of normal tissues distant from the brain. However, the limited efficacy in these highly selected patients suggests that it is probably not very effective in patients with large glial tumors, due to limited distribution of the retrovirus within the tumor and, at best, only a small contribution of the bystander effect. This treatment is more likely to be effective in patients who have had a tumor debulking prior to the vector injection. Although the procedure itself is well tolerated and uses standard neurosurgical techniques, the use of VPCs remains cumbersome from both a manufacturing and pharmacy point of view and affords little if any benefit in the currently employed applications. As noted above, a potential way to augment to effects of gene transfer is by eliciting an immune response to the tumor. One approach has been to coexpress immune modulatory agents along with the HSV-TK gene. A retroviral vector producing both HSV-TK and human IL-2 has been developed in an effort to combine the cytotoxic effects of the HSV-TK with

The Use of Suicide Gene Therapy for the Treatment of Malignancies of the Brain

a cellular immune response to tumor antigens [33]. Four patients with recurrent glioblastoma multiforme were treated by stereotactic implantation of HSV-TK retroviral VPCs. Cerebrospinal fluid levels of IL-2 were sequentially followed in one of the patients as evidence of IL-2 production. Transduction of circulating peripheral blood mononuclear cells was observed in another patient. Two patients had posttreatment biopsies of the tumors. The endothelial cells were intensely stained by the HSV-TK, probe indicating that the vector is expressed in the neovascular component of the tumor. There was no evidence of antitumor immunity in this small study which serves only to illustrate the potential feasibility of the approach. Where examined, retrovirus from the VPCs has been detected by PCR in only trace amounts in normal tissues and only in proximity to the VPCs within the tumor in these clinical studies. There is no evidence of systemic toxicity from the retroviral vector or the VPCs. Despite the ample evidence of clinical safety, the efficacy has been limited. The retroviral vector has restricted distribution in gliomas and so has shown antitumor activity predominantly in tumors ≤1.5 mL in volume. This size limitation would restrict clinical applications to tumors that have been reduced in size by surgical debulking, radiation therapy, or chemotherapy or any combination of these.

III. ADENOVIRUS VECTOR FOR HSV-TK The retrovirus HSV-TK GDEPT system established a proof of principle for the antitumor activity and safety of HSV-TK ganciclovir in patients with high-grade gliomas. Subsequent work with recombinant adenovirus vectors sought to overcome some of the inherent limitations of the retroviral vectors [13]. Adenoviruses have the potential advantage of being prepared in high titer, obviating the need for injection of producer cells. They do not integrate into the host genome and thereby lack the risk of insertional mutagenesis (a concern with retroviruses, especially those containing RCR). Adenoviruses transduce both dividing and nondividing cells and, therefore, can achieve a high level of HSV-TK expression shortly after injection, a perceived advantage in gliomas where the majority of tumor cells are not actively dividing. Preclinical studies have demonstrated low neurotoxicity despite the anticipated immune response to adenoviral vectors in the central nervous system [21]. This might be expected to be worse in humans who have preexisting immunity to the serotypes of adenovirus used as gene delivery vectors; however, this has not been seen in clinical trials [34,35]. The E1/E3-deleted, replication-defective adenoviral vectors [36] expressing HSV-TK (Ad. HSV-TK) have been evaluated in several clinical trials for the treatment of malignant gliomas. In a study by Trask et al. [35], 13 patients with

509

recurrent glioma were treated with a single stereotactic injection of Ad. HSV-TK. This phase I study evaluated escalating doses from 108 to 1011 PFU followed by 14 days of GCV. Patients were followed by clinical examination and serial MRI scans. One patient receiving 1010 PFU of vector deteriorated rapidly, apparently due to aggressive tumor growth and without an apparent effect of vector administration. Two patients developed significant toxicity at the 1011 PFU dose. These toxicities included an injection-site hematoma and brain edema in one patient and obtundation, hyponatremia, and hydrocephalus in the other patient that required placement of a ventriculoperitoneal shunt. The authors concluded that the latter patient probably suffered from inadvertent injection of the adenovirus into the lateral ventricle. Three patients from this study remained alive at more than 3 years following the treatment. We have used a similar Ad. HSV-TK in a trial of 13 patients with high-grade gliomas [34,37]. One patient with an unresectable thalamic tumor received a single stereotactic injection of the vector followed by 14 days of GCV infusion. The remaining 12 patients were treated with a stereotactic injection of vector into the tumor followed by 6 days of daily GCV, craniotomy to resect the tumor, and reinjection with vector followed by 14 days of GCV. Two patients with glioblastoma multiforme were treated at the time of first diagnosis with Ad. HSV-TK followed by standard radiation therapy, whereas the remaining 11 patients had recurrent highgrade gliomas. The dose of vector ranged from 108 to 1011 PFU for each treatment to give a total dose of 2 × 108 to 2 × 1011 PFU for patients undergoing craniotomy and resection of the tumor. Dose-limiting toxicity occurred in two patients at the 1011 PFU dose level. Three patients experienced transient increased intracranial pressure (diagnosed by direct measurement or clinical presentation) manifested as severe headache, and one of them developed an altered mental status. One patient experienced altered mental status, agitation, headache, and hypertension after vector administration during the second surgery. In all cases, the patients recovered within 24 hours with routine medical management. Other toxicities included mild reversible elevation of transaminases and transient fever. The median time to tumor progression in these patients was 3 months, and the median survival was 10 months. Five patients lived 12 months or longer, and one patient remained alive without tumor for 3 years following Ad.HSV-TK treatment before suffering a local recurrence. Despite the preexisting immunity to adenovirus, serious adverse events did not correlate with immune response to virus as assessed by changes in adenovirus antibody titers or T-cell responses [34]. Because gene distribution is critical to the success of GDEPT in brain tumors, Puumalainen et al. [38] examined the transfer of the lacZ gene (which produces β-galactosidase as a marker protein) in patients about to undergo surgery for recurrent glioma [38]. A catheter was implanted into

510


the tumor and vector injected for 3 days using either a retrovirus (no VPCs were used) or adenovirus expressing β-galactosidase. The tumor and catheter were resected several days later. They found that gene transfer efficiency varied from 10 ng/ml. Int. J. Radiat. Oncol. Biol. Phys. 35, 862–868. 4. Smit, W. G. J. M., Helle, P. A., Van Putten, W. L. J. et al. (1990). Late radiation damage in prostate cancer patients treated by high dose external radiotherapy in relation to rectal dose. Int. J. Radiat. Oncol. Biol. Phys. 18, 23–29. 5. Lawton, C. A., Wonb, M., Pilepich, M. V. et al. (1991). Long-term treatment sequelae following external beam irradiation for adenocarcinoma of the prostate: analysis of RTOG studies 7506 and 7706. Int. J. Radiat. Oncol. Biol. Phys. 21, 935–939. 6. Hanlon, A. L., Schultheiss, T. E., Hunt, M. A., Movsas, B., Peter, R. S, Hands, G. E. (1997). Chronic rectal bleeding after high-dose conformal treatment of prostate cancer warrants modification of existing morbidity scales. Int. J. Radiat. Oncol. Biol. Phys. 38(1), 59–63. 7. Zelefsky, M. J., Cowen, D., Fuks, Z. et al. (1999). Long term tolerance of high dose three dimensional conformal radiotherapy for patients with localized prostate carcinoma. Cancer 85, 2460–2468. 8. Lee, W. R., Hanks, G. E., Hanlon, A. L. et al. (1996). Lateral rectal shielding reduces late rectal morbidity following high dose three-dimensional conformal radiation therapy for clinically localized prostate cancer: further evidence for a significant dose effect. Int. J. Radiat. Oncol. Biol. Phys. 35, 251–257. 9. Chen, M. E., Johnston, D. A., Tang, K., Babaian, R. J., and Troncoso, P. (2000). Detailed mapping of prostate carcinoma foci: biopsy strategy implication. Cancer 89, 1800 –1809. 10. Kawashita, Y., Ohtsuru, A., Kaneda, Y. et al. (1999). Regression of hepatocellular carcinoma in vitro and in vivo by radiosensitizing suicide gene therapy under the inducible and spatial control of radiation. Hum. Gene Ther. 10, 1509–1519. 11. Kim, J. H., Kim, S. H., Brown, S. L. et al. (1994). Selective enhancement by an antiviral agent of the radiation-induced cell killing of human glioma cells transduced with HSV-tk gene. Cancer Res. 54, 6053– 6056.

523

12. Nishihara, E., Nagayama, Y., Mawatari, F. et al. (1997). Retrovirusmediated herpes simplex virus thymidine kinase gene transduction renders human thyroid carcinoma cell lines sensitive to ganciclovir and radiation in vitro and in vivo. Endocrinology 138(11), 4577– 4583. 13. Atkinson, G., and Hall, S. (1999). Prodrug activation gene therapy and external beam irradiation in the treatment of prostate cancer Urology 54, 1098–1104. 14. Stevens, C. W., Zeng, M., and Cerniglia, G. J. (1996). Ionizing radiation greatly improves gene transfer efficiency in mammalian cells. Hum. Gene Ther. 7, 1727–1734. 15. Zeng, M., Cerniglia, G. J., Eck, S. L., and Stevens, C. W. (1997). Highefficiency stable gene transfer of adenovirus into mammalian cells using ionizing radiation. Hum. Gene Ther. 8, 1025–1032. 16. Jain, P. T., and Gerwirtz, D. A. (1999). Sustained enhancement of liposome-mediated gene delivery and gene expression in human breast tumor cells by ionizing radiation. Int. J. Radiat. Oncol. Biol. Phys. 75(2), 217–223. 17. Simon, W. J., and Marshall, F. F. (1998). The future of gene therapy in the treatment of urologic malignancies. Urol. Clin. North Am. 25, 23–38. 18. Moolten, F. L. (1986). Tumor chemosensitivity conferred by inserted herpes thymidine kinase genes: paradigm for a prospective cancer control strategy. Cancer Res. 46, 5276–5281. 19. Moolten, F. L., and Wells, J. M. (1990). Curability of tumors bearing herpes thymidine kinase transferred by retroviral vectors. J. Natl. Cancer. Inst. 82, 297–300. 20. Vieweg, J., Rosenthal, F. M., Bannerji, R., Heston, W. D. W., Fair, W. R., Gansbacher, B., and Gilboa, E. (1994). Immunotherapy of prostate cancer in the Dunning rat model: use of cytokine gene modified tumor vaccines. Cancer Res. 54, 1760 –1765. 21. Zhang, W. W., Fang, X., Mazur, W., French, B. A., George, R. N., and Roth, J. A. (1994). High-efficiency gene transfer and high level expression of wild-type p53 in human lung cancer cells mediated by recombinant adenovirus. Cancer Gene Ther. 1, 5–13. 22. Eastham, J. A., Chen, S.-H., Sehgal, I., Yang, G., Timme, T. L., Hall, S. H., Woo, S. L. C., and Thompson, T. C. (1996). Prostate cancer gene therapy: herpes simplex virus thymidine kinase gene transduction followed by ganciclovir in mouse and human prostate cancer models. Hum. Gene Ther. 7, 515–523. 23. Prince, M. H. (1998). Gene transfer: a review of methods and applications. Pathology 30, 335–347. 24. Van der Eb, M. M., Cramer, S. J., Vergouwe, Y., Schagen, F. H., Van Krieken, J. H., Van de Eb, A. J., Rinkes, I. H., Van de Velde, C. J., and Hoeben, R. C. (1998). Severe hepatic dysfunction after adenovirusmediated transfer of the herpes simplex virus thymidine kinase gene and ganciclovir administration. Gene Ther. 5, 451–458. 25. Hall, S. J., Mutchnik, S. E., Chen, S.-H., Woo, S. L. C., and Thompson, T. C. (1997). Adenovirus-mediated herpes simplex virus thymidine kinase gene and ganciclovir therapy leads to systemic activity against spontaneous and induced metastasis in an orthotopic mouse model of prostate cancer. Int. J. Cancer 70(b), 183–187. 26. Hall, S. J., Mutchnik, S. E., Yang, G., Timme, T. L., Nasu, Y., Bangma, C. H., Woo, S. L. C., Shaker, M., and Thompson, T. C. (1999). Cooperative therapeutic effects of androgen ablation and adenovirus-mediated herpes simplex virus–thymidine kinase gene and ganciclovir therapy in experimental prostate cancer. Cancer Gene Ther. 6, 54–63. 27. Timme, T. L., Hall, S. J., Barrios, R., Woo, S. L. C., Aguilar-Cordova, E., and Thompson, T. C. (1998). Local inflammatory response and vector spread after direct intraprostatic injection of a recombinant adenovirus containing the herpes simplex virus thymidine kinase gene and ganciclovir therapy in mice. Cancer Gene Ther. 5, 74–82. 28. Herman, J. R., Adler, H. L., Aguilar-Cordova, E., Rojas-Martinez, A., Woo, S., Timme, T. L., Wheeler, T. M., Thompson, T. C., and Scardino, P. T. (1999). Hum. Gene Ther. 10, 1239–1249.

524

Bin S. Teh et al.

29. Chhikara, M., Zhu, X., Teh, B. S., Vlachaki, M. T., Chiu, J. K., Woo, S. Y., Berner, B. M., Thompson, T. C., Butler, E. B., and AguilarCordova, E. (2000). Radio-gene therapy enhanced reduction of induced metastases of prostate cancer. First International Conference on Translational Research and Pre-clinical Strategies in Radio-Oncology 2000. Int. J. Radiat. Oncol. Biol. Phys. 46, 786–787. 30. Chhikara, M., Huang, H., Vlachaki, M. T., Zhu, X., Teh, B., Chiu, K. J., Woo, S., Berner, B., Smith, E. O., Oberg, K. C., Aguilar, L. K., Thompson, T. C. Butler, E. B., and Aguilar-Cordova, E. (2001). Enhanced therapeutic effect of HSV-tk+GCV gene therapy and ionizing radiation for prostate cencer. Mol. Ther. 3(4), 536–542. 31. Chen, S. H., Shine, H. D., Goodman, J. C., Grossman, R. G., and Woo, S. L. C. (1994). Gene therapy for brain tumors: regression of experimental gliomas by adeno-virus-mediated gene transfer in vivo. Proc. Natl. Acad. Sci. USA 91, 3054–3057. 32. Hasenburg, A., Tong, X. W., Rojas-Martinez, A., Myberg-Hoffman, C., Kieback, C. C., Kaplan, A., Kaufman, R. H., Ramzy, I., AguilarCordova, E., and Kieback, D. G. (2000). Thymidine kinase gene therapy with concomitant topotecan chemotherapy for recurrent ovarian cancer. Cancer Gene Ther. 7(6), 839–844. 33. Teh, B. S., Mai, W. Y., Uhl, B. M., Augspurger, M. E., Grant, W. H., Lu, H. H., Woo, S. Y., Carpenter, L. S., Chiu, J. K., and Butler, E. B. (2001). Intensity-modulated radiation therapy (IMRT) for prostate cancer with the use of a rectal balloon for prostate immobilization: acute

34.

35.

36.

37.

38.

39.

toxicity and dose-volume analysis. Int. J. Radiat. Oncol. Biol. Phys. 49(3), 705–712. Teh, B. S., Woo, S. Y., and Butler, E. B. (1999). Intensity modulated radiation therapy (IMRT): a new frontier in radiation oncology. Oncologist 4, 433– 442. Cox, J. D., Stetz, J., and Pajak, T. F. (1995). Toxicity criteria of the Radiation Therapy Oncology Group (RTOG) and the European Organization for Research and Treatment of Cancer (EORTC). Int. J. Radiat. Oncol. Biol. Phys. 31, 1341–1346. Rodriguez, L. V., and Terris, M. K. (1998). Risks and complications of transrectal ultrasound guided prostate needle biopsy: a prospective study and review of the literature. J. Urol. 160, 2115–2120. Zagars, G. K., Pollack, A., Kavadi, V. S., and Eschenbach, A. (1995). Prostate specific antigen and radiation therapy for clinically localized prostate cancer: an update and review of the M. D. Anderson experience. Int. J. Radiat. Oncol. Biol. Phys. 32, 293–306. Zietman, A., Coen, J. J., Shipley, W. U., Willett, C. G., and Efird, J. T. (1994). Radical radiation therapy in the management of prostatic adenocarcinoma: the initial prostate specific antigen value as a predictor of treatment outcome. J. Urol. 141, 640–645. Ayala, G., Wheeler, T. M., Shalev, M., Thompson, T. C., Miles, B., Aguilar-Cordova, E., Chakraborty, S., and Kadmon, D. (2000). Cytopathic effect of insitu gene therapy in prostate cancer. Hum. Pathol. 31, 866–870.

Index

A AAV, see Adeno-associated viruses Accessory proteins, lentiviruses, 111 Acid-eluted peptides, genetic immunization, 187 Ad.E1A, ovarian cancer model, 471 Ad.Egr-TNF-α, genetic radiotherapy, 442–443 Adeno-associated viruses advantages, 59 antitumor immunity–tumor vaccines, 67–68 biology, 54–56 characterization, 58–59 DNA, 60 gene removal, 59–60 gene therapy, 301 gene transfer to hematopoietic cells, 62–64 immune responses, 60 packaging, 56–57 purification, 58 safety issues, 70–71 titration, 57–59 Adenoviruses Ad-p53 combinations in gene therapy, 307 gene therapy, 301 HSV-TK, 494, 509–510 human, biology, 451 mediated cell killing, mechanisms, 451–452 p21, see p21-expressing adenovirus p53, see p53-expressing adenovirus replication, 451–452 soluble VEGF, tumor angiogenesis control, 425 tumor burden inhibition, 425–428 wild-type, clinical trials, 453–454 Adoptive immunotherapy antitumor-reactive T cells sensitized lymph nodes, 243–246 tumor-infiltrating lymphocytes, 242–243 DC genetic modulation, 250–251 T cell manipulation chimeric TCRs, 249–250 genetic transduction, 247 immunoregulatory molecule delivery, 248 overview, 246–247 TCR gene transfer, 248–249 TIL-marking studies, 247–248


Ad-p21, see p21-expressing adenovirus Ad-p53, see p53-expressing adenovirus AdVMART1, genetic immunization, 188 AFP, see alpha-Fetoprotein AGT, see Alkyltransferase Alkyltransferase, methyltransferase-mediated drug resistance, 344 alpha-Fetoprotein, expression, 182 Alternative splicing, retroviral cis elements, 21 Androgens, HSV-TK suicide gene therapy, 519 Angiogenesis antiangiogenic gene therapy, 439 ribozymes in cancer models, 100–101 tumor biology role, 405–406 tumor growth, 422 VEGF, 424–425 VEGF receptor adenovirus control, 425 Angiopoietin-1, antiangiogenic gene therapy, 414 Angiostatin, antiangiogenic gene therapy, 407–408 Animals drug-resistant DHFR antitumor studies, 387 tumor immunotherapy, 172 Antiangiogenic gene therapy angiogenesis, 439 angiostatin, 407–408 delivery of factors, 422–423 ELR(−) CXC chemokines, 411 EMAP-II, 412 endostatin, 408–409 endothelial cell-specific gene delivery, 414–415 experimental vs. clinical settings, 423 future directions, 415 interferons, 409–410 interleukins, 410 ionizing radiation antitumor activity, 439 targeting, 439–440 p16, 412 p53, 412 proangiogenic cytokines angiopoietin-1, 414 plasminogen activators, 414

525

VEGF, 412–413 response assessment microvessel density, 429–430 MRI imaging, 430–431 PET imaging, 431 ultrasound imaging, 431–432 thrombospondins, 411–412 TIMPs, 411 transgene product safety, 428 translation context, 429 vector safety, 428 VEGF, 423–424 VEGF and angiogenesis, 424–425 VEGF receptors, 423–424 Antiangiogenic proteolytic fragments angiostatin, 407–408 endostatin, 408–409 Antibodies antitumor immunity, 137 conjugated, 83–84 fragment characteristics, 81–82 monoclonal, see Monoclonal antibodies unconjugated, 82–83 Antibody-mediated immune responses, cellular response comparison, 128–129 Antifolates drug-resistant DHFR hematopoietic cells, 388 toxicity protection in vitro, drug-resistant DHFR, 385–387 Antigen-presenting cells major histocompatibility complex, 146 peptide-pulsed professional, immunization, 137 skin penetration in PMGT, 234 Antigens cancer rejection, 179–180 carcinoembryonic cancer gene therapy, 34 expression, 181–182 HLA, dendritic cell loading, 199–200 HLA-A, MHC restriction, 180–181 HLA-A2, mutant ras peptides binding, 150–151 pathways, 146–147 prostate-specific, expression, 182


526 Antigens (continued) RNA-loaded DC applications, 202 self, tumor cell expression, 133–134 SV40 large T antigen, p53 inactivation, 283 tissue-specific differentiation, 134 tumor, see Tumor antigens viral, tumor expression, 135 Antineoplastic agents, dose-limiting toxicity, 342 Antisense oligonucleotides Bcl-2 advantages and disadvantages, 324 application in vivo, 321–322 biotechnology limitations, 317–318 breast carcinoma, 320 leukemia, 318–319 liver carcinoma cells, 320–321 lung carcinoma cells, 319–320 lymphomas, 319 myelomas, 319 prostate carcinoma, 320 ribozymes, 317 vectors, 316 Bcl-xL, 322–324 transcripts, transdominant molecules, 65–66 Antitumor activity animals with drug-resistant DHFR, 387 canine tumor vaccine, 233 CTL activation role, 132 cytokine gene transfer bladder cancer patients, 215 melanoma patients, 214–215 overview, 210 in situ overview, 210–211 tumor transfection, 213 vaccinia recombinant usage, 213–214 vaccinia virus vectors, 211–213 E1A-mediated, see E1A-mediated antitumor activity ionizing radiation, antiangiogenic gene therapy, 439 murine model, PMGT cytokine gene therapy, 233 DNA vaccination, 231–233 reactive T cells, generation by gene-modified tumors sensitized lymph nodes, 243–246 tumor-infiltrating lymphocytes, 242–243 T cell genetic manipulation chimeric TCRs, 249–250 genetic transduction, 247 immunoregulatory molecule delivery, 248 overview, 246–247 TCR gene transfer, 248–249 TIL-marking studies, 247–248 viral replicative potential, 440–441 Antitumor immunity–tumor vaccines costimulatory molecules, 68 cytokine genes, 67–68 oncotropic vectors, 70 overview, 66–67 safety issues, 70–71 suicide genes, 69 tumor antigen-specific vaccines, 68–69

Index tumor vascular supply, 69–70 vector targeting, 70 APCs, see Antigen-presenting cells Apoptosis Ad-p53, 304 Bcl protein role, 315–316 cancer model ribozymes, 101 E1A-mediated antitumor activity, 467–468 genes inducing bax, 290 Fas, 290–291 overview, 289–290 TRAIL receptors, 291 Autonomous parvoviruses, 60–61 Axl, E1A-mediated antitumor activity, 468–469

B bax, gene therapy, 290 B-cell lines, PMGT applications, 230–231 Bcl, apoptosis role, 315–316 Bcl-2, antisense vectors advantages and disadvantages, 324 application in vivo, 321–322 biotechnology limitations, 317–318 breast carcinoma, 320 leukemia, 318–319 liver carcinoma cells, 320–321 lung carcinoma cells, 319–320 lymphomas, 319 myelomas, 319 prostate carcinoma, 320 ribozymes, 317 vectors, 316 Bcl-xL, antisense oligonucleotides, 322–324 Bladder cancer, patients, intravesical vaccinia, 215 Blood flow, tumors, antiangiogenic gene therapy MRI imaging, 430–431 PET imaging, 431 ultrasound imaging, 431 B lymphoctyes, dendritic cell subset, 171–172 BMTs, see Bone marrow transplants Bone marrow drug-resistance gene therapy target, 342–344 purification, p53, 306–307 Bone marrow transplants, HSC applications, 264–266 Brain, malignancies, suicide gene therapy HSV-TK adenovirus vector, 509–510 HSV vector expression, 510 retrovirus vector, 506–509 preclinical studies, 510–511 BRCA1, see Breast cancer susceptibility gene 1 Breast cancer E1A gene therapy clinical trials, 472 model, 471 genomic approach drug resistance, DNA microarray studies, 396–398 method development, 393–396 patient management, 398–399

Breast cancer susceptibility gene 1, 275, 288–289 Breast carcinoma, Bcl-2 antisense oligonucleotides, 320 Bystander effect Ad-p53, 304–305 E1A-mediated antitumor activity, 470 ganciclovir, 494

C Calcium phosphate, effect on DNA gene transfer, 35 Cancer bladder, patients, intravesical vaccinia, 215 breast, see Breast cancer head and neck, E1A gene therapy clinical trials, 472–473 MDR1 cellular overexpression, 305 models, ribozyme application apoptosis, 101 chromosomal translocations, 98–99 malignant cell proliferation, 99–100 multidrug resistance, 100 telomerase, 101 tumor angiogenesis, 100–101 tumor metastasis, 100–101 viral infections, 101–102 ovarian, E1A gene therapy, 470–472 preemption, suicide gene uses, 483–485 Cancer gene therapy adeno-associated virus vectors, 301 adenovirus vectors, 301 Ad-p53, 304–307 antibodies, 84–91 apoptosis-inducing genes, 289–291 BRCA1, 275, 288–289 chimeric viral vectors, 301–302 dendritic cells, 173–174 E2F-1, 287–288 fragile histidine triad, 289 HSV-TK, mesothelioma, 496–501 HSV-TK–GCV, malignant mesothelioma, 494 infectious viral-based vectors, 31–32 ionizing radiation, 441–443 lentiviral vectors, 301 liposomes vector, 300 ONYX-015 adenoviruses, 275–276 p14ARF , 286 p16, 284–285 p21, 280–281, 283–284 p21WAF1/CIP1 , 280 p21 mutant, 281 p21 vs. p53, 281–283 p27, 286–287 p27Kip1 , 286 p53, 273–275, 302–304, 306–308 parvovirus vectors, 61–66 parvovirus vectors, antitumor immunity–tumor vaccines, 66–71 plasmid gene transfer, 33–37 PTEN, 288 radiation therapy combination, 513–515 radiation therapy efficacy, 436–440

527

Index retinoblastoma, 285–286 retroviral vector, 3–4, 300–301 vector administration route, 302 Von Hippel-Landau loss, 289 Cancer immunotherapy animal models, 172 anti-ras immune system interactions, 158–159 considerations, 136–137 PMGT applications, 225 applications in vivo, 226–228 canine tumor vaccine, 233 clinical trials, 234–235 human cell in vitro modification, 228–231 murine model antitumor efficacy, 231–233 technical aspects, 226 vector considerations, 228 strategies, 137–138 Cancer rejection antigens, identification, 179–180 Cancer therapy HSC applications chimeric receptor genes, 267–268 drug resistance genes, 266–267 gene marking, 264–266 overview, 262–264 lentiviral vector applications dendritic cells, 118–119 HSCs, 117–118 T lymphocytes, 118 tumor cells, 119 replication-selective adenoviruses, 451 Canine tumor vaccine, antitumor activity, 233 CAR, see Coxsackie–adenovirus receptor Carcinoembryonic antigen cancer gene therapy, 34 expression, 181–182 Carcinoma Bcl-2 antisense oligonucleotides, 319–321 ras oncogene immunogenicity in vivo, 152–156 CD, see Cytidine deaminase cDNA, see Complementary DNA CEA, see Carcinoembryonic antigen Cell cycle arrest, Ad-p53, 304 lentiviral vs. oncoretroviral vectors, 112 Cell fusion, tumor–DCs, genetic immunization, 188–189 Cell killing, adenovirus-mediated, mechanisms, 451–452 Cellular immune response pathogen elimination, 146 ras oncogene peptide-induced, CD4+ and CD8+ , 149 tumor cells, humoral response comparison, 128–129 CFU–GM, see Granulocyte-macrophage colony-forming units Chemokines, ELR(−) CXC, antiangiogenic gene therapy, 411 Chemoprotection, hematopoietic cells, 64–65 Chemosensitivity, purine and thymidine salvage pathways, 485–486

Chemotherapy antiangiogenic gene therapy, 429 ONYX-015 efficacy, 458–459 p53, 305–306 Chimeric receptors, genes, HSCs, 267–268 Chimeric T-cell receptors, adoptive immunotherapy, 249–250 Chimeric viral vectors, gene therapy, 301–302 Chromosomal translocations, ribozymes in cancer models, 98–99 Chronic myelogenous leukemia gene therapy anti-bcr–abl drug resistance gene, 332–334 anti-bcr–abl targets, 332 antisense–drug resistance in vivo efficacy, 334 gene-disruption methods, 332 molecular mechanisms, 331 overview, 331–332 chTCRs, see Chimeric T-cell receptors cis-active elements, retroviral vectors cDNA, 22 coexpression strategies, 20–22 early hematopoietic cells, 16–18 overview, 6–7, 16 regulatable promoters, 19–20 RNA elements, 20 silencing, 19 T lymphocytes, 19 tumor cells, 19 CML, see Chronic myelogenous leukemia Complementary DNA, cis elements, 22 Complementation, tumor suppressor genes, 66 Conjugated antibodies, 83–84 Costimulatory molecules, antitumor immunity–tumor vaccines, 68 Coxsackie–adenovirus receptor, gene therapy, 301 CTLs, see Cytotoxic T-lymphocytes Cytidine deaminase, drug resistance therapies, 348 Cytokine gene therapy intravesial vaccinia in bladder cancer patients, 215 melanoma patients, 214–215 overview, 210 PMGT application, 233 in situ overview, 210–211 tumor transfection by vaccinia recombinants, 213–214 tumor vaccines, 67–68 vaccinia recombinant usage, 213–214 vaccinia virus vectors, 211–213 Cytoplasmic retention, p53, 283 Cytosine deaminase, radiosensitization, 438–439 Cytotoxicity, ionizing radiation, 439–440 Cytotoxic T-lymphocytes CD4+ activation, 129–132 antitumor immunity, 132 cellular immune responses, 149 exogenously synthesized antigen recognition, 129 CD8+ activation, 129–132 antitumor immunity, 132

cellular immune responses, 149 endogenously synthesized antigen recognition, 129 human CD8+ CTL for ras 4–12(Val12) epitope, 151 neo-antigenic determinant, 156 peptide variants, 158 tumor cell lysis, 151–152 HSCs, 267 p53, 183 self antigen expression, 133–134

D DCE-MRI, see Dynamic contrast-enhanced magnetic resonance imaging DCs, see Dendritic cells Dendritic cells B lymphocyte regulation, 171–172 discovery and function, 167–168 gene therapy, 173–174 genetic immunization DC transduction and transfection, 187–188 overview, 185–186 preclinical development, 190 genetic modulation, 250–251 human subset, 170–171 immature cells, 168–169 innate immunity effectors, 172 lentiviral vectors, 118–119 loading with genetic material, 199–200 mature cells, 169 mouse subset, 169–170 regulatory T cell heterogeneity, 171 RNA loading cell applications, 201–202 process, 201 RNA amplification, 201 RNA vs. DNA loading, 200–201 tumor antigen gene delivery, 186 tumor fusion, genetic immunization, 188–189 tumor immunology, 172–173 type 1–type 2 T cell heterogeneity, 171 DHFR, see Dihydrofolate reductase DHFR–CD fusion gene, see Dihydrofolate reductase–cytidine deaminase fusion gene DHFR–TS fusion gene, see Dihydrofolate reductase–thymidylate synthase fusion gene Dihydrofolate reductase, drug-resistant animal antitumor studies, 387 antifolate-mediated in vivo selection, 388 antifolate toxicity protection in vitro, 385–387 characteristics, 384–385 myeloprotection, 367, 371–372 Dihydrofolate reductase–cytidine deaminase fusion gene double-mutant variety, 372–373 non-Hodgkin’s lymphoma treatment, 376–377 synthetic fusion gene, 369–370 Dihydrofolate reductase–thymidylate synthase fusion gene, 368–369

528

Index

Disease control in antiangiogenic gene therapy, 429 graft-versus-host, leukemia, suicide gene preemption, 487–488 residual therapy, drug-resistance gene transfer, 375–376 dl 1520, see ONYX-015 adenoviruses DNA adeno-associated virus, 60 cDNA, cis elements, 22 dendritic cell loading, RNA loading comparison, 200–201 direct injection, plasmid-based vector gene transfer, 33–34 gene transfer, calcium phosphate effect, 35 provirus, reverse transcription, 8 tumor antigen–peptide, immunization, 137 DNA–GAL4 complexes, antibodies in nonviral gene delivery, 87 DNA microarrays, drug resistance studies, 396–398 DNA–poly-L-lysine complexes, antibodies in nonviral gene delivery, 84–87 DNA vaccination, PMGT, murine model antitumor efficacy, 231–233 Dose intensification, drug-resistance gene transfer, 374–375 Drug resistance CML therapy, 332–334 DNA microarray studies, 396–398 genes, HSCs, 266–267 Drug-resistance fusion genes natural types, 368–369 synthetic types, 369–370 tailored genes, 370 Drug-resistance gene therapy bone marrow as target, 342–344 clinical trials, 350–351 cytidine deaminase, 348 dual-drug resistance approach, 349–350 glutathione-S-transferase, 348–349 MDR1, 344 myelosuppression, 342 risks, 344 Drug-resistance gene transfer DHFR mutants, myeloprotection, 371–372 hematopoietic cells, 13 myeloprotection, 373–378 oncology applications, 4–5 Dual-drug resistance, gene transfer, 349–350 Dynamic contrast-enhanced magnetic resonance imaging, gene therapy assessment, 430–431

E E1A cancer gene therapy breast cancer, 471–472 head cancer, 472–473 neck cancer, 472–473 ovarian cancer, 470–472 safety studies, 471–472 E1A–DC-Chol, ovarian cancer model, 470–471 E1A-mediated antitumor activity

Axl downregulation, 468–469 bystander effect, 470 HER2 downregulation, 467–468 HER2 overexpression, 465–467 metastasis inhibition, 468 NF-κB, 469–470 E2F-1, gene therapy, 287–288 EBV, see Epstein–Barr virus Egr-1, genetic radiotherapy, 441–442 EMAP-II, see Endothelial monocyte-activating polypeptide II Endostatin, antiangiogenic gene therapy, 408–409 Endothelial cell-specific gene delivery, 414–415 Endothelial monocyte-activating polypeptide II, 412 Endothelium, tumor, dominant-negative VEGF receptors, 425 env, translation, 9 Envelope, retroviral trans elements ligand-directed targeting, 13–15 overview, 11–12 pseudotyped vectors, 12–13 Env proteins lentiviruses, 111 ligand-directed targeting, 14 Epidermal growth factor receptors, 14 Episomes, function, 40–43 Epitopes, antigen processing, 146–147 Epstein–Barr virus, episomes, 40–41

F Fas, gene therapy, 290–291 FHIT, see Fragile histidine triad FMEV, see Friend–MCF–MESV hybrid vector FMRI, see Functional magnetic resonance imaging Fragile histidine triad, gene therapy, 289 Friend–MCF–MESV hybrid vector, cis elements, 18 Functional magnetic resonance imaging, antiangiogenic gene therapy, 430 Fusion genes DHFR–CD double-mutant variety, 372–373 non-Hodgkin’s lymphoma treatment, 376–377 synthetic fusion gene, 369–370 DHFR–TS, 368–369 drug-resistance, see Drug-resistance fusion genes Fusion proteins, retroviral cis elements, 22

G gag–pol, translation, 9 Gag proteins, lentiviruses, 110 GALV, see Gibbon ape leukemia virus Ganciclovir, mesothelioma bystander effects, 494 HSVtk–GCV gene therapy, 494, 498–499 GCV, see Ganciclovir GDEPT, see Gene-directed enzyme prodrug therapy

Gene delivery endothelial cell-specific delivery, 414–415 liposome-mediated delivery, 35–36 non-retroviral viral, antibody role, 91 nonviral, see Nonviral gene delivery retroviral, antibody role, DNA–poly-L-lysine complexes, 88–91 tumor antigens to DCs, 186 Gene-directed enzyme prodrug therapy HSV-TK adenovirus vector, 509 HSV-TK retrovirus vector, 506–507 Gene marking, oncology applications, 4 Gene-modified tumors sensitized lymph nodes, 243–246 tumor-infiltrating lymphocytes, 242–243 vaccine, sensitized lymph node cells, 243–246 Genes adeno-associated virus, 59–60 BRCA1, 275, 288–289 HSCs, 266–268 immunomodulatory, antiangiogenic properties, 409–411 lentiviruses, 110–111 MDR1, selection, myeloprotection, 366–367 oncogenes, see Oncogenes retroviruses, 6 Genetic immunization acid-eluted peptides, 187 clinical trials, 190 DC transduction and transfection, 187–188 DC usage, 185–186 peptides, 187 plasmid type, 183–185 preclinical development, 190 problems, 189–190 tumor antigen gene delivery to DCs, 186 tumor-dendritic cell fusion, 188–189 tumor lysates, 186–187 Genetic material, dendritic cell loading, 199–200 Genetic transduction, T cells, 247 Gene transfer clinical approaches, 370–371 drug-resistance, see Drug-resistance gene transfer dual-drug resistance approach, 349–350 hematopoietic cells, 61–64 HSCs, 258–262, 377–378 MDR1 into humans, 361 mesothelioma treatment, 500–501 MGMT, 345–346 particle-mediated, see Particle-mediated gene transfer plasmids, 33–37 VEGF inhibition, tumor endothelium transduction, 425 viral vs. nonviral methods, 200 Genomics, breast cancer therapy DNA microarray studies, 396–398 method development, 393–396 patient management, 398–399 Gibbon ape leukemia virus, 6 Glutathione-S-transferase, drug resistance therapies, 348–349 GM-CSF

529

Index gene-modified tumor vaccines, 244–246 plasmid immunization, 184 Graft-versus-host disease, leukemia, suicide gene preemption, 487–488 Granulocyte-macrophage colony-forming units DHFR–CD fusion gene, 372–373 genetic modification, 356–357 GST, see Glutathione-S-transferase

H Hairpin ribozymes, structure and function, 96 Hammerhead ribozymes, structure and function, 96 Head and neck cancer, E1A gene therapy clinical trials, 472–473 Heat shock proteins, immunization, 137–138 Hematopoietic cells chemoprotection, 64–65 cis elements, 16–18 DHFR, antifolate-mediated in vivo selection, 388 drug resistance gene transfer, 13 genetic modification, 356–357 gene transfer, 61–64 MDR1-transduced, early phase I studies, 359–360 P-glycoprotein, MDR1 gene therapy, 357–359 Hematopoietic stem cells cancer therapy applications, 262–268 definition, 257–258 gene transfer, 258–262, 377–378 lentiviral vectors, 117–118 Hepatitis B virus, p53 effect, 283 HER2, E1A-mediated antitumor activity, 465–468 HER2/neu, expression, 182 Herpes simplex virus, endogenous HSV-TK expression, 510 Herpes simplex virus-1–thymidine kinase adenovirus vector, 509–510 delivery system, 494 HSV-TK–GCV gene therapy, malignant mesothelioma, 494 HSV vector expression, 510 mesothelioma adjunctive phase I clinical trial, 497–499 gene transfer strategies, 500–501 patient selection optimization, 499 phase I clinical trial, 496–497 radiosensitization, 437–438 retrovirus vector, 506–509 suicide gene therapy, 515 suicide gene therapy–radiotherapy Ad.HSV-TK, 519 androgen deprivation, 519 clinical considerations, 521–522 constitutional symptoms, 520 gene vector, 518–519 genitourinary and lower gastrointestinal toxicity, 521 hematologic toxicity, 520 hepatic and renal toxicity, 520–521 novel uses, 522

patient characteristics, 520 patient evaluation, 519 patient selection, 518 preclinical studies, 515–517 radiotherapy, 519 treatment arms, 518 treatment cessation criteria, 519 treatment cessation and delays, 520 toxicity studies, 494–496 HFV, see Human foamy virus HIV-1-based vectors, packaging systems, 114 HLA, see Human leukocyte antigen HLA-As, see Human leukocyte antigen A2 HPRT-negative–gpt-positive cells, suicide function stability, 486–487 HPV16 E6, p53 effect, 281–282 HSCs, see Hematopoietic stem cells HSPs, see Heat shock proteins HSV, see Herpes simplex virus HSV-TK, see Herpes simplex virus-1–thymidine kinase Human cell modification in vitro, PMGT application, 228–231 dendritic cells, 170–171 dendritic cell tumor immunology, 172–173 MDR1 gene therapy, 361–362 MDR1 gene transfer, 361 Human foamy virus, pseudotyped retroviral vectors, 12–13 Human leukocyte antigen, dendritic cell loading, 199–200 Human leukocyte antigen A2 MHC restriction, 180–181 mutant ras peptides binding, 150–151 Humoral immune responses, cellular response comparison, 128–129

I IFN, see Interferons IL-7, see Interleukin-7 Immune response adeno-associated virus, 60 antibody-mediated, cellular response comparison, 128–129 cell-mediated, generation, 208–210 cellular, see Cellular immune response host, malignant cells, 133 humoral, cellular response comparison, 128–129 ONYX-015 clinical trials, 456–458 T-lymphocyte-mediated, humoral response comparison, 128–129 tumor cells, response comparisons, 128–129 Immune system, anti-ras cancer immunotherapy, 158–159 CD8+ CTL epitopes, 156 mutant ras CD8+ CTL epitope peptide variants, 158 tumor escape mechanisms, 156–158 Immunity, dendritic cell effects, 172 Immunization genetic, see Genetic immunization

heat shock proteins, 137–138 peptide-pulsed professional antigen presenting cells, 137 tumor antigen–peptide, 137 tumor cell-based vaccines, 137 tumor peptides, 137 Immunogenicity, ras oncogene peptides in vivo, 152–156 Immunoliposome–DNA complexes, antibodies in nonviral gene delivery, 87 Immunologic monitoring, RNA-loaded DC applications, 202 Immunomodulatory genes, antiangiogenic properties ELR(−) CXC chemokines, 411 interferons, 409–410 interleukins, 410 Immunoregulatory molecules, T cells for delivery, 248 Immunotherapy, see Adoptive immunotherapy; Cancer immunotherapy Inducible promoters, plasmids, 38–40 Infection, recombinant virions, 114–115 Injection, direct, DNA, plasmid-based vector gene transfer, 33–34 Integrase, targeting, 15 Interferons, antiangiogenic gene therapy, 409–410 Interleukin-7, gene-modified tumors, 243 Interleukins, antiangiogenic gene therapy, 410 Internal ribosomal entry site myeloprotection, 367 retroviral cis elements, 21–22 Intralesional vaccinia virus vectors, melanoma patients, 214–215 Ionizing radiation antiangiogenic gene therapy, 439–440 antitumor virus replicative potential effects, 440–441 gene therapy, 441–443 p21 gene therapy, 437 IRES, see Internal ribosomal entry site

L LCMV, see Lymphocytic choriomeningitis virus LCR, see Locus control region Lentiviral vectors classification, 110 dendritic cells, 118–119 design, 113 gene therapy, 301 HSCs, 117–118 life cycle and genes, 110–111 oncoretroviral vector comparison, 112–113 packaging systems, 114 recombinant virion infectious spectrum, 114–115 T lymphocytes, 118 transfer vectors, 115–117 tumor cells, 119 Leukemia Bcl-2 antisense oligonucleotides, 318–319 CML, see Chronic myelogenous leukemia

530

Index

Leukemia (continued) graft-versus-host disease, suicide gene preemption, 487–488 Life cycle lentiviruses, 110–111 retroviruses, 7–9 Ligand-directed targeting, retroviral envelope, 13–15 Ligand–DNA conjuates, cancer gene therapy, 36–37 Liposome-mediated gene delivery, cancer gene therapy, 35–36 Liposomes, gene therapy vector, 300 Liver carcinoma cells, Bcl-2 antisense oligonucleotides, 320–321 Locus control region, lentiviral vectors, 117 Long terminal repeats lentiviral vectors, 115 retroviral architecture, 11 retroviral cis elements, 6–7, 20–21 retroviral life cycle, 7–8 LTRs, see Long terminal repeats Lung carcinoma cells, Bcl-2 antisense oligonucleotides, 319–320 Lymph node cells, sensitized, 243–246 Lymphocytic choriomeningitis virus, 13 Lymphomas Bcl-2 antisense oligonucleotides, 319 induction by MLV, 10 non-Hodgkin’s, DHFR–CD fusion gene treatment, 376–377

M MAGE, analysis, 179–180 Magnetic resonance imaging, antiangiogenic gene therapy, 430–431 Major histocompatibility complex antigen-presenting cells, 146 CD4+ and CD8+ CTL activation, 129–132 CD4+ and CD8+ CTL antitumor immunity, 132 CD4+ CTLs, exogenously synthesized antigen recognition, 129 CD8+ CTLs, endogenously synthesized antigen recognition, 129 restriction, tumor antigens, 180–181 Matrix metalloproteinase, non-retroviral viral gene delivery, 91 MDM2, p53 effect, 282–283 MDR1, see Multidrug resistance-1 Melanoma cells, PMGT applications, 228–230 Melanoma peptides, types, 181–183 Mesothelioma HSV-TK–GCV gene therapy, 494 HSV-TK gene adjunctive phase I clinical trial, 497–499 gene transfer strategies, 500–501 patient selection optimization, 499 phase I clinical trial, 496–497 MESV, see Murine embryonic stem cell virus Metastasis E1A-mediated antitumor activity, 468 inhibition, p53, 306

ribozymes in cancer models, 100–101 Methotrexate drug-resistant DHFR, 384–385 myeloprotection, 367 Methylguanine-DNA-methyltransferase-mediated drug resistance gene transfer, 345 mechanism, 345 mutants, 345–346 myeloprotection, 346–347 overview, 344–345 selection in vivo, 347–348 tumor sensitization, 346–347 MGMT, see Methylguanine-DNAmethyltransferase-mediated drug resistance MHC, see Major histocompatibility complex Microvessel, density, antiangiogenic gene therapy response, 429–430 MLVs, see Murine leukemia viruses MMAC1, see PTEN MMP, see Matrix metalloproteinase Models antitumor studies, 387 dendritic cells, 169–170 E1A gene therapy, 470–472 PMGT, 231–233 preclinical, see Preclinical models tumor immunotherapy, 172 Moloney murine leukemia virus, cis elements, 16–17 MoMLV, see Moloney murine leukemia virus Monoclonal antibodies non-retroviral viral gene delivery, 91 nonviral gene delivery, 84–87 retroviral gene delivery, 88–91 targeting abilities, 81–82 Mouse dendritic cells, 169–170 PMGT, 231–233 MPEV, see MPSV–MESV hybrid vector MPSV–MESV hybrid vector, cis elements, 17–18 MTX, see Methotrexate Multidrug-resistance 1 drug-resistance gene therapy target, 344 gene overexpression in cancer cells, 305 gene selection, myeloprotection, 366–367 gene transfer into humans, 361 hematopoietic cell chemoprotection, 64–65 transduced hematopoietic cells, 359–360 Multidrug-resistance gene therapy hematopoietic progenitor cell targeting, 356–357 humans, 361–362 P-glycoprotein, 356–359 transduction inefficiency, 360 Murine embryonic stem cell virus, cis elements, 17–18 Murine leukemia viruses classification, 6 lymphoma induction, 10 nuclear transport and integration, 15 pseudotyped retroviral vectors, 12–13

Mutated self proteins, tumor cells, 134–135 Mutations DHFR–CD fusion gene, 372–373 dihydrofolate reductase, myeloprotection, 371–372 MGMT, 345–346 ras CD4+ and CD8+ T-cell epitopes CD8+ CTL-mediated tumor cell lysis, 151–152 HLA-A2-binding mutant ras peptides, 150–151 human CD8+ CTL for ras 4–12(Val12) epitope, 151 overview, 149–150 ras oncogene peptide immunogenicity in vivo, 152–156 reverse transcriptase, vector comparisons, 112 Myelomas, Bcl-2 antisense oligonucleotides, 319 Myeloprotection applications, 365–366 drug-resistance gene selection, 366–367 drug-resistance gene transfer, 371–378 limitations, 366 MGMT, 346–347 Myelosuppression, drug-resistance gene therapy, 342

N

Neo-antigenic determinants, CD8+ CTL epitopes, 156 Neoplasm, development, ras oncogenes, 147–148 NF-κB, E1A-mediated antitumor activity, 469–470 Non-Hodgkin’s lymphoma, DHFR–CD fusion gene treatment, 376–377 Noninfectious plasmid-based vectors, usage considerations, 33 Non-retroviral viral gene delivery, antibody role, 91 Nonviral gene delivery, antibody role DNA–GAL4 complexes, 87 DNA–poly-L-lysine complexes, 84–87 immunoliposome–DNA complexes, 87 Normal cells self antigen expression, 133–134 tissue-specific differentiation antigen expression, 134 Nuclear transport, retroviral trans elements, 15

O Oligonucleotides, see Antisense oligonucleotides Oncogenes K-ras, CD8+ CTL-mediated lysis, 151–152 ras human CD4+ T-cell responses, 149 human CD8+ T-cell responses, 149 immunogenicity in vivo, 152–156 neoplastic development, 147–148

531

Index Oncogenic proteins, tumor cell expression, 135 Oncoretroviral vectors, lentiviral vector comparison cell-cycle and transduction, 112 packaging sizes, 112 reverse transcriptase mutation rates, 112 silencing, 112–113 stability, 112 Oncotropic vectors, antitumor immunity–tumor vaccines, 70 ONYX-015 adenoviruses background, 452–453 clinical research, 454–455 clinical trial results chemotherapy combination, 458–459 immune response, 456–458 overview, 459–460 single agent efficacy, 458 toxicity, 455–456 viral replication, 456 efficacy improvements, 461–462 gene therapy clinical trials, 276 overview, 275 process, 275–276 refractory solid tumors, 460–461 Open reading frames, adeno-associated virus, 54–55 ORFs, see Open reading frames Ovarian cancer E1A gene therapy clinical trials, 472 E1A gene therapy model, 470–471

P p14 ARF , gene therapy, 286 p16, gene therapy, 412 p16 INK4 , gene therapy, 284–285 p21WAF1/CIP1, gene therapy, 280 p21-expressing adenovirus, gene therapy ionizing radiation, 437 overexpression and senescence, 283–284 overview, 280–281 p21 mutant, 281 p53 comparison, 281–283 p27, gene therapy, 286–287 p27Kip1 , gene therapy, 286 p53-expressing adenovirus, gene therapy adenovirus combinations, 307 antiangiogenic therapy, 412 apoptosis, 304 bone marrow purification, 306–307 bystander effect, 304–305 cell-cycle arrest, 304 chemotherapy, 305–306 expression, 182–183 family members, 308 gene-transfer-mediated radiosensitization, 436–437 MDR1-overexpressing cancer cells, 305 metastases inhibition, 306 overview, 302–304 p21 comparison, 281–283 radiotherapy, 306

synthetic molecules, 307 tumor suppression, 273–275 Packaging systems adeno-associated virus, 56–57 HIV-1-based vectors, 114 lentiviral vs. oncoretroviral vectors, 112 retroviral vectors, 9–10 Particle-mediated gene transfer application in vivo, 226–228 applications, 225 cancer gene therapy, 34–35 cancer immunotherapy canine tumor vaccine, 233 PMGT ex vivo, 234–235 PMGT in vivo, 235 skin penetration, 234 human cell in vitro modification, 228–231 murine model antitumor efficacy, 231–233 technical aspects, 226 vector considerations, 228 Parvoviridae adeno-associated virus advantages, 59 biology, 54–56 characterization, 58–59 DNA, 60 gene removal, 59–60 immune responses, 60 packaging, 56–57 purification, 58 titration, 57–59 autonomous parvoviruses, 60–61 Parvoviruses, gene therapy antitumor immunity–tumor vaccines costimulatory molecules, 68 cytokine genes, 67–68 oncotropic vectors, 70 overview, 66–67 safety issues, 70–71 suicide genes, 69 tumor antigen-specific vaccines, 68–69 tumor vascular supply, 69–70 vector targeting, 70 autonomous parvoviruses, 60–61 hematopoietic cells, 61–65 transdominant molecule delivery, 65–66 tumor purging, 64 tumor suppressor gene complementation, 66 Pathogens, cellular immune response, 146 PBSCTs, see Peripheral blood stem cell transplants PCNA, deficient p21 mutant, 281 Peptide-pulsed professional antigen-presenting cells, 137 Peptides delivery to dendritic cells, 186 genetic immunization, 187 melanoma types, 181–183 mutant CD8+ epitope variants, 158 ras oncogene, 149, 152–156 tumor, immunization, 137 Peripheral blood stem cell transplants, 264–266 PET, see Positron emission tomography P-glycoprotein, MDR1 gene therapy, 356–359

Plasmids genetic immunization, 183–185 gene transfer DNA, 35 DNA direct injection, 33–34 ligand–DNA conjuates, 36–37 liposome-mediated gene delivery, 35–36 particle-mediated gene delivery, 34–35 inducible promoters, 38–40 noninfectious, usage considerations, 33 overview, 37–38 PMGT, 228 replicating type, 40–43 tissue-specific promoters, 38 Plasminogen activators, antiangiogenic gene therapy, 414 PMGT, see Particle-mediated gene transfer Pol enzymes, lentiviruses, 111 Polypurine tract, retroviral life cycle, 7–8 Positron emission tomography, antiangiogenic gene therapy, 431 PP tract, see Polypurine tract Preclinical models, plasmid gene transfer DNA, 35 DNA direct injection, 33–34 ligand–DNA conjuates, 36–37 liposome-mediated gene delivery, 35–36 particle-mediated gene delivery, 34–35 Proangiogenic cytokines, antiangiogenic gene therapy angiopoietin-1, 414 plasminogen activators, 414 VEGF, 412–413 Prodrug converting enzyme, suicide gene therapy, 437–439 Pro enzymes, lentiviruses, 111 Promoters, cis elements, 19–21 Prostate carcinoma, Bcl-2 antisense oligonucleotides, 320 Prostate-specific antigens, 182 Protein cleavage, retroviral cis elements, 22 Proteins accessory, lentiviruses, 111 Bcl, apoptosis role, 315–316 delivery to dendritic cells, 186 Env, 14, 111 fusion, retroviral cis elements, 22 Gag, lentiviruses, 110 HSPs, immunization, 137–138 mutated self proteins, tumor cells, 134–135 oncogenic, tumor cell expression, 135 Ras, expression, 182 TIMPs, antiangiogenic gene therapy, 411 viral, packaging cell expression, 10 Provirus, DNA, reverse transcription, 8 PSA, see Prostate-specific antigens Pseudotyped retroviral vectors, 12–13 PTEN, gene therapy, 288 Purine salvage pathways, suicide gene transduction, 485–486

R Radiation, see Ionizing radiation Radiosensitization

532 Radiosensitization (continued) cytosine deaminase, 438–439 HSV-TK, 437–438 p53 gene-transfer-mediated, 436–437 Radiotherapy antiangiogenic gene therapy, 429 cancer gene therapy combination, 513–515 gene therapy effects cytotoxic effect enhancement, 439–440 p21 therapy and ionizing radiation, 437 p53 gene-transfer-mediated radiosensitization, 436–437 prodrug converting enzyme, 437–439 HSV-TK suicide gene therapy combination Ad.HSV-TK, 519 androgen deprivation, 519 clinical considerations, 521–522 constitutional symptoms, 520 gene vector, 518–519 genitourinary and lower gastrointestinal toxicity, 521 hematologic toxicity, 520 hepatic and renal toxicity, 520–521 novel uses, 522 patient characteristics, 520 patient evaluation, 519 patient selection, 518 preclinical studies, 515–517 radiotherapy, 519 treatment arms, 518 treatment cessation criteria, 519 treatment cessation and delays, 520 p53, 306 ras 4–12(Val12) epitope, human CD8+ CTL line, 151 ras p21, neoplastic development, 147–148 Ras protein, 182 Rb, see Retinoblastoma Recombinant virions, infectious spectrum, 114–115 Regulatable promoters, cis elements, 19–20 Replicating plasmid vectors, function, 40–43 Replication adenoviruses, tumor-selective replication, 451–452 antitumor viruses, ionizing radiation effects, 440–441 ONYX-015 clinical trials, 456 Residual disease therapy, drug-resistance gene transfer, 375–376 Retinoblastoma, gene therapy, 285–286 Retroviral gene delivery, DNA–poly-L-lysine complexes, 88–91 Retroviral vectors basic architecture, 10–11 cancer gene therapy, ideal characteristics, 3–4 cis elements cDNA, 22 coexpression strategies, 20–22 early hematopoietic cells, 16–18 overview, 6–7, 16 regulatable promoters, 19–20 RNA elements, 20 silencing, 19

Index T lymphocyes, 19 tumor cells, 19 classification, 6 genes and products, 6 gene therapy vector, 300–301 HSV-TK, 506–509 life cycle, 7–9 oncological applications, 4–5 packaging cells, 9–10 trans elements, envelope ligand-directed targeting, 13–15 overview, 11–12 pseudotyped vectors, 12–13 trans elements, nuclear transport and integration, 15 Reverse transcription lentiviral vs. oncoretroviral vectors, 112 proviral DNA, 8 Ribozymes antisense Bcl-2, 317 cancer models apoptosis, 101 chromosomal translocations, 98–99 malignant cell proliferation, 99–100 multidrug resistance, 100 telomerase, 101 tumor angiogenesis, 100–101 tumor metastasis, 100–101 viral infections, 101–102 structure and function, 96–98 transdominant molecules, 66 RNA dendritic cell loading, 200–202 retroviral cis elements, 20 tumor antigen–peptide, immunization, 137

S Safety, antitumor immunity–tumor vaccines, 70–71 Self antigens, tumor cell expression, 133–134 Self proteins, mutated, tumor cells, 134–135 Sequence homology, adeno-associated virus, 57–58 Signaling pathways, tumor-specific T cell responses, 132–133 Silencing cis elements, 19 lentiviral vs. oncoretroviral vectors, 112–113 Skin, penetration in PMGT, 234 SNV, see Spleen necrosis virus Spleen necrosis virus, retroviral gene delivery, 90 Splicing, alternative, retroviral cis elements, 21 Stem cells, see Hematopoietic stem cells SU, ligand-directed targeting, 14 Suicide gene therapy antitumor immunity–tumor vaccines, 69 gene stability HPRT-negative–gpt-positive cells, 486–487 TK-negative–HSV-TK-positive cells, 487 HSV-TK adenovirus vector, 509–510 clinical advances, 515 HSV vector expression, 510

retrovirus vector, 506–509 HSV-TK–radiotherapy combination Ad.HSV-TK, 519 androgen deprivation, 519 clinical considerations, 521–522 constitutional symptoms, 520 gene vector, 518–519 genitourinary and lower gastrointestinal toxicity, 521 hematologic toxicity, 520 hepatic and renal toxicity, 520–521 novel uses, 522 patient characteristics, 520 patient evaluation, 519 patient selection, 518 preclinical studies, 515–517 radiotherapy, 519 treatment arms, 518 treatment cessation criteria, 519 treatment cessation and delays, 520 overview, 482–483 preclinical studies, 510–511 preemptive uses cancer therapy, 483–485 future uses, 488–489 leukemia graft-versus-host disease, 487–488 prodrug converting enzyme, 437–439 purine and thymidine salvage pathways, 485–486

T Taxol induced apoptosis, E1A-mediated antitumor activity, 468 ovarian cancer model, 471 T-cell epitopes, CD4+ and CD8+ , mutant identification CD8+ CTL-mediated tumor cell lysis, 151–152 HLA-A2-binding mutant ras peptides, 150–151 human CD8+ CTL for ras 4–12(Val12) epitope, 151 overview, 149–150 ras oncogene peptide immunogenicity in vivo, 152–156 T-cell receptors antitumor immunity–tumor vaccines, 67 chimeric TCRs, 249–250 gene transfer, 248–249 HSCs, 267–268 T cells genetic manipulation, antitumor reactivity chimeric TCRs, 249–250 genetic transduction, 247 immunoregulatory molecule delivery, 248 overview, 246–247 TCR gene transfer, 248–249 TIL-marking studies, 247–248 regulatory, dendritic cell heterogeneity, 171 tumor-specific, intracellular signaling pathway defects, 132–133

533

Index type 1–type 2, dendritic cell heterogeneity, 171 TCR, see T-cell receptors TDLNs, see Tumor-drained lymph nodes Telomerase, ribozymes in cancer models, 101 TEP1, see PTEN Thrombospondins, antiangiogenic gene therapy, 411–412 Thymidine, salvage pathways, suicide gene transduction, 485–486 TILs, see Tumor-infiltrating lymphocytes TIMPs, see Tissue inhibitors of matrix metalloproteinases Tissue culture, DHFR–CD fusion gene, 372–373 Tissue inhibitors of matrix metalloproteinases, 411 Tissue-specific differentiation antigens, 134 Tissue-specific promoters, plasmids, 38 Titration, adeno-associated virus, 57–59 TK-negative–HSV-TK-positive cells, suicide function stability, 487 T-lymphocyte-mediated immune response, 128–129 T lymphocytes cis elements, 19 cytotoxic, see Cytotoxic T-lymphocytes lentiviral vectors, 118 subsets, 147 Toxicity antifolate, drug-resistant DHFR, 385–387 dose-limiting, drug-resistance gene therapy, 342 HSV-TK studies, 494–496 HSV-TK suicide gene therapy genitourinary and lower gastrointestinal toxicity, 521 hematologic toxicity, 520 hepatic and renal toxicity, 520–521 ONYX-015 clinical trials, 455–456 TRAIL, p53 gene therapy, 305 TRAIL decoy receptor, TRUNDD, overexpression, 283 TRAIL receptors, gene therapy, 291 Transdominant molecules, anti-sense transcripts, 65–66 Transduction dendritic cells, genetic immunization, 187–188 lentiviral vs. oncoretroviral vectors, 112 MDR1 inefficiency, 360 T cells, 247 tumor cells, soluble VEGF receptors, 425 tumor endothelium, dominant-negative VEGF receptors, 425 trans elements, retroviral vectors ligand-directed targeting, 13–15 nuclear transport and integration, 15 overview, 11–12 pseudotyped vectors, 12–13 Transfection dendritic cells, genetic immunization, 187–188 tumors, vaccinia virus recombinants, 213 Transfer vectors, 115–117

Transgene products, safety in antiangiogenic gene therapy, 428 Translocations, chromosomal, ribozymes in cancer models, 98–99 Transplants, peripheral blood stem cells, HSC applications, 264–266 TRUNDD, overexpression, 283 TSPs, see Thrombospondins Tumor antigens CTL activation, 129–132 endogenously synthesized, CD8+ CTL recognition, 129 exogenously synthesized, CD4+ CTL recognition, 129 gene delivery to dendritic cells, 186 identification, 179–180 immunization, 137 MHC restriction, 180–181 specific vaccines, antitumor immunity–tumor vaccines, 68–69 Tumor burden, inhibition intramuscular adenoviral administration, 425–426 intravenous adenoviral administration, 426–428 Tumor cells CD8+ CTL-mediated lysis, 151–152 cis elements, 19 immune response comparisons, 128–129 lentiviral vectors, 119 mutated self proteins, 134–135 oncogenic protein expression, 135 prodrug converting enzyme suicide gene therapy, 437–439 self antigen expression, 133–134 tissue-specific differentiation antigen expression, 134 transduction, soluble VEGF receptors, 425 tumor-specific T cell responses, 132–133 vaccines, 137 Tumor-drained lymph nodes, 243–246 Tumor escape, 156–158 Tumor immunotherapy, animal models, 172 Tumor-infiltrating lymphocytes antitumor immunity–tumor vaccines, 67 cancer rejection antigen identification, 179–180 gene-modified tumors, 242–243 marking studies, 247–248 Tumor lysates, genetic immunization, 186–187 Tumor necrosis factor α apoptosis, E1A-mediated antitumor activity, 467–468 gene-modified tumors, 243 genetic radiotherapy, 442 Tumor peptides, 137 Tumor purging, 64 Tumors angiogenesis, VEGF receptor adenovirus control, 425 angiogenesis role, 405–406 blood flow and vascularity, imaging, 430–432

dendritic cell fusion, genetic immunization, 188–189 endothelium, transduction, dominant-negative VEGF receptors, 425 growth, angiogenesis, 422 immunology, dendritic cells, 172–173 individuals bearing, tumor tolerance, 132 Rb-resistant, gene therapy, 286 refractory solid, ONYX-015 efficacy, 460–461 ribozymes in cancer models, 100–101 sensitization, MGMT, 346–347 transfection, vaccinia virus recombinants, 213 vascular supply, interruption, 69–70 viral antigen expression, 135 Tumor-specific T lymphocytes, adoptive immunotherapy, 137 Tumor suppressors complementation, 66 p53, see p53

U Ultrasound imaging, antiangiogenic gene therapy assessment, 431–432 Unconjugated antibodies, 82–83 Untranslated vector regions, retroviral cis elements, 22

V Vaccination, DNA, see DNA vaccination Vaccine-primed lymph nodes, 243–246 Vaccines antitumor immunity–tumor, see Antitumor immunity–tumor vaccines canine tumor, antitumor activity, 233 gene-modified tumor, sensitized lymph node cells, 243–246 tumor cell-based, immunization, 137 Vaccinia virus vectors cytokine gene transfer, 211–214 intralesional, see Intralesional vaccinia virus vectors intravesical, bladder cancer patients, 215 Vascular endothelial growth factor angiogenesis, 424–425 antiangiogenic gene therapy, 412–413, 423–424 gene transfer inhibition, tumor endothelium transduction, 425 Vascular endothelial growth factor receptors antiangiogenic gene therapy, 423–424 dominant-negative, tumor endothelium transduction, 425 soluble, tumor effects, 425 Vascularity, tumors antiangiogenic gene therapy, imaging, 430–431 interruption, 69–70 Vector-producing cells, HSV-TK retrovirus vector, 506–509

534 Vector targeting, antitumor immunity–tumor vaccines, 70 VEGF, see Vascular endothelial growth factor VEGF receptors, see Vascular endothelial growth factor receptors Vesicular stomatitis virus, pseudotyped retroviral vectors, 12

Index VHL, see Von Hippel-Landau loss Viral antigens, tumor expression, 135 Viral-based vectors, infectious, human gene therapy, 31–32 Viral infections, ribozymes in cancer models, 101–102 Viral proteins, packaging cell expression, 10

Viruses, replication, ONYX-015 clinical trials, 456 Von Hippel-Landau loss, gene therapy, 289 VPCs, see Vector-producing cells VPLN, see Vaccine-primed lymph nodes VSV-G, see Vesicular stomatitis virus

Gene Therapy of Cancer (2nd Edition)

Cancer Gene Therapy

Gene Therapy for Cancer

Autologous And Cancer Stem Cell Gene Therapy (Progress in Gene Therapy) (Progress in Gene Therapy)

Gene Therapy

Gene Therapy of Cancer: Methods and Protocols, 2nd Edition (Methods in Molecular Biology, Vol. 542)

Ovarian Cancer, 2nd Edition

Gene Therapy for Cancer (Cancer Drug Discovery and Development)

Gene Therapy Protocols 2nd Edition (Methods in Molecular Medicine)

Viral Therapy of Cancer

Ribozymes in Gene Therapy of Cancer (Biotechnology Intelligence Unit)

Geriatric Physical Therapy 2nd Edition

Gene Therapy Technologies

Gene Therapy Protocols

Muscle Gene Therapy

The Biology of Cancer, 2nd Edition

Viral gene therapy

Gene Therapy Protocols

Gene Targeting: A Practical Approach (2nd edition)

Everyone'S Guide To Cancer Therapy 4th Edition

Gene Cloning and Manipulation, 2nd Edition

Gene Targeting: A Practical Approach (2nd edition)

Gene Therapy in the Treatment of Cancer: Progress and Prospects (Cancer: Clinical Science in Practice)

Gene Therapy Protocols: Design and Characterization of Gene Transfer Vectors

Camptothecins in Cancer Therapy

Nutrient-Gene Interactions in Cancer

Targeted Cancer Therapy

Advances in Cancer Therapy

Gene Therapy of Cancer (2nd Edition)

Cancer Gene Therapy

Gene Therapy for Cancer

Autologous And Cancer Stem Cell Gene Therapy (Progress in Gene Therapy) (Progress in Gene Therapy)

Gene Therapy

Gene Therapy of Cancer: Methods and Protocols, 2nd Edition (Methods in Molecular Biology, Vol. 542)

Ovarian Cancer, 2nd Edition

Gene Therapy for Cancer (Cancer Drug Discovery and Development)

Gene Therapy Protocols 2nd Edition (Methods in Molecular Medicine)

Viral Therapy of Cancer

Ribozymes in Gene Therapy of Cancer (Biotechnology Intelligence Unit)

Geriatric Physical Therapy 2nd Edition

Gene Therapy Technologies

Gene Therapy Protocols

Muscle Gene Therapy

The Biology of Cancer, 2nd Edition

Viral gene therapy

Gene Therapy Protocols

Gene Targeting: A Practical Approach (2nd edition)

Everyone'S Guide To Cancer Therapy 4th Edition

Gene Cloning and Manipulation, 2nd Edition

Gene Targeting: A Practical Approach (2nd edition)

Gene Therapy in the Treatment of Cancer: Progress and Prospects (Cancer: Clinical Science in Practice)

Gene Therapy Protocols: Design and Characterization of Gene Transfer Vectors

Camptothecins in Cancer Therapy

Nutrient-Gene Interactions in Cancer

Targeted Cancer Therapy

Advances in Cancer Therapy

Recommend Documents