Photomedicine in Gynecology & Reproduction

Photomedicine in Gynecology and Reproduction .. ............................ Photomedi...

Author: P. Wyss | Yona Tadir | Bruce J. Tromberg | Urs Haller

149 downloads 2224 Views 11MB 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

............................ Photomedicine in Gynecology and Reproduction

..

............................

Photomedicine in Gynecology and Reproduction

Editors

Pius Wyss, Zu¨rich Yona Tadir, Irvine, Calif. Bruce J. Tromberg, Irvine, Calif. Urs Haller, Zu¨rich

179 figures, 4 in color, and 27 tables, 2000

............................ Pius Wyss, MD Department of Obstetrics and Gynecology University Hospital of Zu¨rich, Switzerland

Yona Tadir, MD Beckman Laser Institute and Medical Clinic University of California, Irvine, Calif., USA

Bruce J. Tromberg, PhD Beckman Laser Institute and Medical Clinic University of California, Irvine, Calif., USA

Urs Haller, MD Department of Obstetrics and Gynecology University Hospital of Zu¨rich, Switzerland

Library of Congress Cataloging-in-Publication Data Photomedicine in gynecology and reproduction / editors, Pius Wyss ... [et al.]. p. cm. Includes bibliographical references and index. ISBN 3–8055–6905–X (hardcover : alk. paper) 1. Generative organs, Female – Diseases – Photochemotherapy. 2. Light – Physiological effect. 3. Phototherapy. I. Wyss, Pius, Dr. [DNLM: 1. Genital Diseases, Female – therapy. 2. Photochemotherapy. 3. Phototherapy. WP 650 P575 2000] RG103. P47 2000 618.10631 – dc21 99-048643 CIP

Drug Dosage. The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher. Ó Copyright 2000 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) www.karger.com Printed in Switzerland on acid-free paper by Reinhardt Druck, Basel ISBN 3–8055–6905–X

............................

Contents

Introduction 1 General Introduction Haller, U. (Zu¨rich) 4 History of Photomedicine Wyss, P. (Zu¨rich) 12 Features of Photodynamic Techniques: Diagnostics and Therapeutics Wyss, P. (Zu¨rich)

Basics 14 Characteristics of Different Photosensitizers Kimel, S. (Haifa); Orenstein, A.; Lavie, G. (Tel Hashomer) 39 Localization and Dynamics of Photosensitizers Ru¨ck, A.; Akgu¨n, N.; Heckelsmiller, K.; Beck, G.; Kunzi-Rapp, K. (Ulm) 53 Light and Chemistry of Polyatomic Organic Molecules Suppan, P. (Fribourg) 76 Photobleaching and Photodynamic Effect of Protoporphyrin IX Wyss-Desserich, M.-T. (Zu¨rich); Sun, C.H. (Irvine, Calif.); Wyss, P. (Zu¨rich); Svaasand, L.O. (Irvine, Calif./Trondheim) 86 Photoproduct Formation during Porphyrin Photodynamic Therapy Ko¨nig, K. (Jena) 96 Optical Dosimetry for Photodynamic Therapy Svaasand, L.O. (Trondheim/Irvine, Calif.); Wyss, P.; Fehr, M.K.; Major, A.L.; Tadir, Y. (Irvine, Calif.)

116 Noninvasive Characterization of Tissue Optical Properties Using

Frequency Domain Photon Migration Tromberg, B.J. (Irvine, Calif.); Coquoz, O. (Lausanne); Fishkin, J.B.; Butler, J. (Irvine, Calif.); Svaasand, L.O. (Trondheim)

133 Fluorescence Microscopy for Pharmacokinetics Krasieva, T.B. (Irvine, Calif.) 147 Correlation of Tissue Fluorescence and Photodynamic Effect Grahn, M.F.; Ansell, J.K.; de Jode, M.L. (London) 157 Mitochondria as Targets for the Induction of Apoptosis in

Photodynamic Therapy Richter, C. (Zu¨rich)

169 Cellular Effects of Photodynamic Therapy with Clinical Relevance Patrice, T.; Rousset, N.; Eleouet, S.; Carre´, J.; Vonarx, V.; Lajat, Y. (Nantes)

Endometrium 176 Introduction Wyss, P.; Steiner, R.A. (Zu¨rich); Gannon, M.J. (Leeds); Reid, R.L. (Kingston); Tadir, Y. (Irvine, Calif.) 183 Preclinical Studies: Photodynamic Therapy in the Rat and Rabbit

Endometrium with Various Photosensitizers. Pharmacokinetic, Histological and Reproductive Studies Steiner, R.A. (Chur); Wyss, P.; Wyss-Desserich, M.-T. (Zu¨rich); Tadir, Y.; Tromberg, B.J.; Krasieva, T.B.; Berns, M.W. (Irvine, Calif.); Haller, U. (Zu¨rich)

206 Pharmacology and Toxicology of 5-Aminolevulinic Acid and

Protoporphyrin IX Used for Photodynamic Endometrial Ablation in Primates and Nonprimates Wyss, P. (Zu¨rich)

213 Photodynamic Endometrial Ablation in Nonhuman Primates Van Vugt, D.A.; Krzemien, A. (Kingston); Foster, W. (Ottawa); Lundhal, S. (Boston, Mass.); Marcus, S. (Valhalla, N.Y.); Reid, R.L. (Kingston) 219 Use of ALA-PDT for Endometrial Ablation in the Treatment of

Menorrhagia. First Clinical Trials Brown, S.B; Gannon, M.J.; Holroyd, J.A.; Johnson, N.; Stringer, M.; Vernon, D.I. (Leeds)

227 Pharmacokinetics of ALA in Human Uterine Tissue Fehr, M.K.; Wyss, P. (Zu¨rich); Tadir, Y. (Irvine, Calif.)

Contents

VI

234 Photodynamic Endometrial Ablation: Morphological and Functional

Results Wyss, P.; Fehr, M.K. (Zu¨rich); Tadir, Y. (Irvine, Calif.); Hornung, R.; Haller, U. (Zu¨rich)

243 A Light Distributor for Photodynamic Endometrial Ablation Bays, R. (Lausanne/Ecublens); Woodtli, A. (Ecublens); Mosimann, L. (Lausanne); Wyss, P. (Zu¨rich); Wagnie`res, G. (Lausanne); Haller, U. (Zu¨rich); van den Bergh, H. (Lausanne) 246 Intrauterine Light Probe for Uterine Photodynamic Therapy Tadir, Y. (Irvine, Calif.); Hornung, R. (Irvine, Calif./Zu¨rich); Tromberg, B.J. (Irvine, Calif.)

Cervic/Vulva/Vagina 251 Selective Photosensitization in Vulvar Condyloma and PDT of

Vulvar Intraepithelial Neoplasia Fehr, M.K.; Wyss, P.; Dobler, D. (Zu¨rich); Chapman, C.F.; Krasieva, T.B.; Tromberg, B.J.; Berns, M.W.; Tadir, Y. (Irvine, Calif.); Schwarz, V.; Haller, U. (Zu¨rich)

265 Photodynamic Therapy for Cervical Dysplasia Schmidt, S. (Marburg); Spaniol, S. (Bonn) 270 Application and Characteristics of Photodynamic Therapy for

Cervical Cancer Muroya, T.; Kawasaki, K.; Kunugi, T.; Akiya, T.; Iwabuchi, H.; Sakunaga, H.; Sakamoto, M.; Sugishita, T.; Tenjin, Y. (Tokyo)

278 Photodynamic Therapy in Recurring Gynecologic Cancer Corti, L.; Boso, C. (Padova)

Ovaries and Peritoneal Cavity 285 Intraperitoneal PDT for the Treatment of Recurrent Ovarian Cancer Hahn, S.M.; Sindelar, W.F.; Delaney, T.F.; Rubin, S.C.; Fraker, D.; Hsi, R.A.; Glatstein, E. (Philadelphia, Pa.) 296 PDT for Cytoreduction in Cases of Ovarian Cancer Schmidt, S. (Marburg); Wagner, U. (Bonn) 302 Potential Application of PDT for Treatment of Endometriosis and

Ectopic Pregnancy: Animal Models Reid, R.L.; Yang, J.Z.; Krzemien, A.; Melchior, M.F.; Van Dijk, J.P.; Hahn, P.M.; Van Vugt, D.A.; Greer, P.A.; Dickson, E.; Pottier, R.H. (Kingston)

308 Photodynamic Diagnosis of Endometriosis in Patients Hillemanns, P.; Korell, M. (Munich)

Contents

VII

Breast Cancer 312 Introduction Wyss, P. (Zu¨rich); Schmidt, S. (Marburg) 316 Photodynamic Therapy in Breast Cancer Patients: Application of

SnET2 for Skin Metastases Schmidt, S. (Marburg)

322 Photodynamic Therapy of Recurrent Breast Cancer Wyss, P.; Fehr, M.K.; Hornung, R.; Schwarz, V.; Haller, U. (Zu¨rich)

Photon-Assisted Reproduction 326 Ten Years of Laser-Assisted Gametes and Embryo Manipulations Tadir, Y. (Irvine, Calif./Orange, Calif.) 340 PALMÔ Robot-MicroBeam for Laser-Assisted Fertilization, Embryo

Hatching and Single-Cell Prenatal Diagnosis. An Overview Clement-Sengewald, A.; Schu¨tze, K.; Sandow, S.; Nevinny, C.; Po¨sl, H. (Munich)

352 Diode Laser for Assisted Hatching Germond, M.; Primi, M.-P.; Senn, A.; Rink, K.; Descloux, L.; Delacre´taz, G. (Lausanne) 366 Er:YAG (2,940 lm) Laser for Assisted Hatching Obruca, A.; Feichtinger, W. (Vienna) 372 Author Index 374 Subject Index

Contents

VIII

Introduction Wyss P, Tadir Y, Tromberg BJ, Haller U (eds): Photomedicine in Gynecology and Reproduction. Basel, Karger, 2000, pp 1–3

............................

General Introduction U. Haller Department of Obsterics and Gynecology, University Hospital of Zu¨rich, Switzerland

Photomedicine – now at the end of the 20th century a promising new tool in medicine – was already known in antiquity. Research on photosensitizers, especially psoralens, started in Egypt, where the active ingredients were isolated from plants and soon commercialized for treatment of vitiligo. The modern history of photomedicine started in this century, when Raab discovered in 1901 that paramecia, having acridine acid incorporated, died in the sunlight. PDT (photodynamic therapy) is a typical example of a clinical procedure which was used before the underlying scientific processes were identified and understood. Although the principle of PDT is simple, it is based on complex physical, chemical and physiological interdependencies. As late as 1990 there were still many important preclinical and clinical questions open, for example: dark toxicity of different sensitizers; genotoxicity of sensitizers; distribution of the sensitizer in the tissue; location of the sensitizer in the cell; drug dosage and pharmacokinetic studies; dynamic curve of photosensitizer distribution in relation to the curve of photodynamic activity; photo-bleaching effects; application mode of the sensitizer to avoid skin photosensitization; construction of laser light-diffusing systems, and studies on light dosage and the kind of light focused on a target, e.g. influence of fractionated light. Many of these issues have been clarified within the last 10 years, so that photodynamic therapy has developed into a promising new clinical treatment and diagnostic modality. The success of this new technique is mainly due to interdisciplinary exchange and collaboration. All over the world, research centers have sprung up to investigate this phenomenon.

Fig. 1. World of PDT in gynecology. FIGO Congress 1997.

Fig. 2. Participants of the 1st World Congress of Photomedicine in Gynecology 1998.

Haller

2

In 1997 the organizers of the FIGO World Congress in Copenhagen provided an opportunity for a ‘Meet the Experts’ meeting on photodynamic therapy in gynecology (fig. 1). While it was the first chance to sensitize our colleagues worldwide about the development and clinical use of photomedicine in gynecology, Lancet printed an article in 1998 titled ‘Photodynamic Therapy Begins to Shine’. It was a privilege to welcome some 150 outstanding personalities from all over the world in Zu¨rich on the occasion of the 1st World Congress of Photomedicine in Gynecology in February 1998. All participating scientists and clinicians were pioneers of the first hour in basic research in this field that stresses the importance of such a scientific meeting for the exchange of ideas and experiences on a high level (fig. 2). This book presents an overview of the current status of PDT and of recent developments, including preclinical investigations and clinical aspects of photomedical diagnostics and therapeutics. It intends to offer the opportunity to review both the progress and the drawbacks and to contemplate future directions worldwide to establish photomedical techniques as a successful minimally invasive procedure in gynecology.

Prof. Dr. Urs Haller, Department of Obstetrics and Gynecology, University Hosptial of Zu¨rich, Frauenklinikstrasse 10, CH–8091 Zu¨rich (Switzerland) Tel. +41 1 255 52 00, Fax +41 1 255 44 33, E-Mail [email protected]

General Introduction

3


............................

History of Photomedicine P. Wyss Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Switzerland

As early as 1550 BC, procedures of photomedicine were already being described in the Ebers Papyrus [1] and India’s sacred ‘Atharva Veda’ [2]. The main indications for photomedical treatments were pigmentless areas of the skin considered to be leprosy, most of which were probably vitiligo (leukoderma). According to the ancient Indian medical literature [3] application of black seeds from the plant Bavachee or Vasuchika was followed by exposure to natural sunlight. Bavachee or Vasuchika was identified as Psoralea corylifolia (fig. 1). This plant contains psoralens (furanocumarins) which are anaerobic photosensitizers. The same plant is referred to for the treatment of vitiligo in Buddhist literature from about 200 AD [4] and in Chinese documents from the Sung period of the 10th century [5]. In his famous book (13th century) ‘Mofradat Al Adwiya’, Ibn El Bitar described the treatment of depigmented skin (vitiligo) with a tincture of honey and powdered seeds from a plant called ‘Aatrillal’, which was abundant in the Nile River valley [6]. Aatrillal has been identified as a plant called Ammi majus containing different psoralens. After systemic application of the powdered A. majus seeds through licking the honey mixture, a patient was advised to sit in strong sunlight for 1 or 2 h exposing the depigmented skin. The sunlight unspecifically delivered the needed spectrum of wavelength to activate these psoralens. The photodynamic reaction of the leukoderma resulted in vesiculation followed by reepithelization and repigmentation. In this century, intensive research on psoralens was undertaken in Egypt. Active ingredients were isolated from plants and chemically analyzed [7]. Soon thereafter the most important compound, 8-methoxypsoralen (fig. 2), was commercialized for treatment of psoriasis [8, 9]. Photosensitizing and photochemical reactions using psoralens can occur without involving oxygen.

Fig. 1. Psoralea corylifolia, from Matthew KM: Illustration of the Flora of Tamilnadu Carnatic; in: Rapinat Herbarium. India, Tiruchirapalli, 1982. [Because the publisher is unknown, all efforts to get written permission were unsuccessful.]

However, the oxygen-dependent photodynamic reaction was discovered and elucidated by Raab [10] in 1900 when studying the effects of light and dyes on ‘paramecia’ (fig. 3). Paramecia treated with acridine acid (dye) in the dark or those exposed to sunlight without acridine acid incubation survived. But the combination of acridine dye solution and sunlight exposure killed the paramecia. This observation raised two main questions: (1) does acridine dye change the light to a toxic medium, for example by absorption of wavelengths necessary for life? or (2) does the light change the acridine dye to a toxic compound? The first hypothesis was disproved by an experiment in which paramecia survived in usual solutions when the light was filtered separately by an acridine dye solution. Therefore acridine dye did not change light to

History of Photomedicine

5

Fig. 2. 8-Methoxypsoralen.

toxic radiation. However, a specific property of acridine dye is absorption and fluorescence. Since absorption does not reveal toxic light, fluorescence was considered as the main harmful parameter in the killing mechanism. Raab [10] suggested that fluorescent substrates like acridine dye transfer the energy of light to a ‘living chemical energy’ which causes the death of paramecia. This ‘living chemical energy’ lies at the heart of ‘photodynamic therapy’. Based upon the new photodynamic knowledge, Tappeiner and Jesionwk [11] performed the first photodynamic therapy of skin cancer using eosin as photosensitizer, and Tappeiner and Jodlbauer [12] introduced the term ‘photodynamic action’. The first report on the phototoxicity of hematoporphyrin was offered by Hausmann [13] in 1908. He found that hematoporphyrin is an effective sensitizer in paramecia and erythrocytes [13]. Around 1910 white mice, injected with hematoporphyrin and exposed to light, developed reactions which varied with the amount of sensitizer or the amount of light [14]. Hausmann suggested that the primary effect of photodynamic therapy was due to damage to the peripheral vessels. The first application of hematoporphyrin to human was performed in a sensational self-experiment. On October 14, 1912, Meyer-Betz [15] injected himself intravenously with 0.2 g of hematoporphyrin (fig. 4) and demonstrated solar photosensitivity associated with edema and hyperpigmentation that lasted for 2 months. The study proved that systemically applied hematoporphyrin causes intensive photosensitization. The diagnostic significance of hematoporphyrin-based fluorescence in neoplastic tissues was emphasized by Policard [16] in 1924. He suggested that red fluorescence induced in an experimental rat sarcoma by UV light may be attributed to accumulation of endogenous hematoporphyrin resulting from secondary infection by hemolytic bacteria [16]. Red fluorescence of neoplastic tissues could be increased by exogenously applied hematoporphyrin in another rat model [17]. Evidently, tumors accumulate hematoporphyrin more than normal tissues indicating a new diagnostic and therapeutic implication of sensitizers for cancers. In 1948 Figge et al. [18] demonstrated the increased affinity of highly proliferative tissues such as neoplastic, embryonic and regenerative tissue for porphyrins in an animal model, and mentioned the possibility of photodynamic

Wyss

6

Fig. 3. Paramecia aurelia, from Van Wagtendonk WJ: Paramecium. Amsterdam, Elsevier, 1974. [Because the author’s address is unknown, all efforts to get written permission were unsuccessful.]

cancer treatment with porphyrins. A series of 11 cancer patients was given injections of hematoporphyrin intravenously in dosages varying from 300 to 1,000 mg 12–72 h prior to surgery in 1954 [19]. Bright red fluorescence was visualized by a near-ultraviolet spotlight during surgery. Rassmussen et al. [19] concluded that the red fluorescence of hematoporphyrin and its tendency


7

Fig. 4. F. Meyer-Betz before and after intravenous application of 200 mg hematoporphyrin causing solar photosensitivity [15]. [Because the journal no longer exists, all efforts to get written permission were unsuccessful.]

to concentrate in tumors may be utilized to assist the surgeon to visualize and delineate neoplastic tissue during operations. The detection of small or obscure lymph nodes may also be facilitated by these photodynamic techniques. Since the hematoporphyrin used in past experiments was a crude mixture of numerous porphyrins [20], the phototoxicity of a particular hematoporphyrin derivative (HpD; acetic acid-sulfuric acid derivative) was tested in a mouse model [21]). The derivative was demonstrated to be twice as toxic as the crude preparation and had twice the photodynamic action. In 1960, Lipson and Baldes [21] injected intravenously 2 mg HpD/kg body weight in 15 cancer patients approximately 3 h before endoscopy. Fluorescence endoscopy revealed neither false-positive nor false-negative results when a sufficient amount of the activating light reached the region of involvement. They concluded that fluorescence endoscopy following systemic HpD application may be a valuable technique for the detection of malignant disease. Another study of 226 patients [22] injected intravenously with HpD showed a 75–85% correlation of fluorescence with positive biopsies of squamous and adenocarcinoma, but also a 23% false-positive rate in 53 benign lesions. While the primary purpose of the previous studies was photomedical tumor diagnosis, the first experiments for

Wyss

8

photodynamic tumor destruction started in 1966 [23]. Lipson et al. [23] realized the potential for selective destruction of tumors containing HpD by making use of its photodynamic properties. Objective evidence of response to photodynamic treatment was found in a patient with a large ulcerating recurrent breast cancer treated by multiple HpD injections and local exposure of the tumor to filtered light (spectrum not specified). In a further study, gliomas transplanted subcutaneously in rats were destroyed photodynamically by exposure to white light following injection with hematoporphyrin [24]. Kelly and Smell [25] implanted 11 human bladder carcinomas in immunosuppressed mice. The administration of HpD, followed 24 h later by exposure to white light, caused marked destruction of tumors. They suggested photodynamic therapy as applicable in the treatment of superficial transitional cell carcinoma of the bladder. One year later, they photodynamically treated a patient with a recurrent superficial anaplastic cancer of the bladder using HpD [25]. 48 h after treatment the recurrent carcinoma showed necrosis of several papillary tumors but only in the illuminated areas. In 1978, Dougherty et al. [26] reported complete or partial response to photodynamic therapy in 111 of 113 cutaneous or subscutaneous malignant lesions using HpD. Highly pigmented or larger tumors required higher HpD doses. Damage of normal adjacent skin was essentially limited by reducing the light doses or by extending the time interval between injection and light application. They initially mentioned laser light as an effective alternative to the arc lamp and already used a tunable argon dye laser system with fiberoptic delivery systems for photodynamic therapy of osteogenic sarcoma lesions in 1980 [27]. The major advantage to the laser, as they emphasized, is not the intensity or wavelength, but the flexibility of use. Consequently, endobronchial cancers were irradiated with the red light laser beam, delivered by a quartz fiber inserted through the instrumentation channel of the fiberoptic bronchoscope in 1982 [28]. Since photosensitivity is a major side effect of photodynamic therapy requiring patients to avoid bright light for more than 30 days, other efficient photosensitizers with shorter elimination time were requested. Several investigators observed a porphyrin of unknown structure which was found to be the substance primarily responsible for the photosensitizing activity of the HpD mixture [29–31]. The structure of the HpD-active component was isolated and analyzed by Dougherty et al. [32] in 1984. They designated the component found, dihematoporphyrin ether, and it has become a current sensitizer called Photofrin for clinical trials and was approved by the American Food and Drug Administration in 1984. The development of new photosensitizers and clinical studies in photodynamic therapy have increased exponentially in the meantime.


9

References 1 2 3 4 5 6 7

8 9

10 11 12 13 14 15

16 17 18

19 20 21 22 23 24 25 26

El-Mofy AM: Vitiligo and Psoralens. Oxford, Pergamon Press, 1968, p 147. Bloomfield M: Sacred Books of the East: Hymns of Atharva-Veda. Oxford, Clarendon Press, 1897, vol 42. Jolly J: Medicine. Strassburg, Tru¨bner, 1901. Benetto AV: The psoralens: An historical perspective. Cutis 1977;20:469–471. Fitzpatrick TB, Pathak MA: Historical aspects of methoxsalen and other furocoumarins. J Invest Dermatol 1959;31:229–231. Ibn El Bitar: Mofradat Al-Adwiya; cited by Fahmy IR: Pharmacognostical study and isolation of cristalline constituent, Ammoidin. Q J Pharm Pharmacol 1947;20:291. Fahmy IR, Abu-Shady H: Ammi majus linn: The isolation and properties of ammoidin, ammidin and majudin, and their effect in the treatment of leukoderma. Q J Pharm Pharmacol 1948;21: 499–503. Allyn B: Studies on phototoxicity in man and laboratory animals. 21st Annu Meet Am Acad Dermatol, Chicago, 1962. Tronnier H, Schule D: Zur dermatologischen Therapie von Dermatosen mit langwelligem UV nach Photosensibilisierung der Haut mit Methoxsalen. Erste Ergebnisse bei der Psoriasis vulgaris. Z Haut Geschlechtskr 1973;48:385–393. Raab O: Ueber die Wirkung fluorescirender Stoffe auf Infusorien. Z Biol 1900;39:524–546. Tappeiner H, Jesionwk X: Therapeutische Versuche mit fluoreszierenden Stoffen. Mu¨nch Med Wochenschr 1903;50:2042–2044. Tappeiner H, Jodlbauer A: Ueber die Wirkung der photodynamischen (fluoreszierenden) Stoffe auf Protozoen und Enzyme. Dtsch Arch Klin Med 1904;80:427–487. Hausmann WH: Die sensibilisierende Wirkung tierischer Farbstoffe und ihre physiologische Bedeutung. Wien Klin Wochenschr 1908;21:1527–1529. Hausmann WH: Die sensibilisierende Wirkung des Ha¨matoporphyrins. Biochem Z 1910;30:276– 316. Meyer-Betz F: Untersuchung u¨ber die biologische (photodynamische) Wirkung des Ha¨matoporphyrins und anderer Derivate des Blut- und Gallenfarbstoffs. Dtsch Arch Klin Med 1913;112: 476–503. Policard A: (1924) Etudes sur les aspects offerts par des tumeurs expe´rimentales examine´s a` la lumie`re de Wood. CR Soc Biol 1924;91:1423–1424. Auler H, Banzer G: Untersuchungen u¨ber die Rolle der Porphyrine bei geschwulstkranken Menschen und Tieren. Z Krebsforsch 1942;53:65–68. Figge FHJ, Weiland GS, Manganiello OJ: Cancer detection and therapy. Affinity of neoplastic, embryonic, and traumatized tissues for porphyrins and metalloporphyrins. Proc Soc Exp Biol Med 1948;68:640–641. Rassmussen DS, Ward GE, Figge FHJ: Fluorescence of human lymphatic and cancer tissues following high doses of intravenous hematoporphyrin. Cancer 1955;1:78–81. Schwartz S, Absolon K, Vermund H: Some relationships of porphyrins, X-rays and tumors. Bull Minn Univ School Med 1955;27:7–13. Lipson RL, Baldes EJ: The photodynamic properties of a particular hematoporphyrrin derivative. AMA Arch Dermatol 1960;82:508–516. Gregorie HB, Horger EO, Ward JL, Green JF, Richards T, Robertson HC, Stevenson TB: Hematoporphyrin-derivative fluorescence in malignant neoplasms. Ann Surg 1968;167:820–828. Lipson RL, Baldes EJ, Gray MJ: Hematoporphyrin derivative for detection and management of cancer. Cancer 1967;20:2255–2257. Diamond I, McDonagh AF, Wilson CB, Granelli SG, Nielsen S, Jaenicke R: Photodynamic therapy of malignant tumors. Lancet 1972;ii:1175–1177. Kelly JF, Snell ME: Hematoporphyrin derivative: A possible aid in the diagnosis and therapy of carcinoma of the bladder. J Urol 1976;115:150–151. Dougherty TJ, Kaufman JE, Goldfarb A, Weishaupt KR, Boyle D, Mittleman A: Photoradiation therapy for the treatment of malignant tumors. Cancer Res 1976;38:2628–2635.

Wyss

10

27

28 29 30 31 32

Dougherty TJ, Thoma RE, Boyle D, Weishaupt KR: Photoradiation therapy of malignant tumors: Role of the laser; in Pratesi R, Sacchi CA (eds): Laser in Photomedicne and Photobiology. New York, Springer, 1980, pp 67–75. Hayata Y, Kato H, Konaka C, Ono J, Takizawa N: Hematoporphyrin derivative and laser photoradiation in the treatment of lung cancer. Chest 1982;81:269–277. Moan J, Sommer SP: Fluorescence and absorption properties of the components of hematoporphyrin derivative. Photobiochem Photobiophys 1981;3:93–98. Berenbaum MC, Bonnett R, Scourides PA: In vivo biological activity of the components of hematoporphyrin derivative. Br J Cancer 1982;45:571–581. Kessel D, Chow T: Tumor-localizing components of the porphyrin preparation hematoporphyrin derivative. Cancer Res 1983;43:1994–1999. Dougherty TJ, Potter WR, Weishaupt KR: The structure of the active component of hematoporphyrin derivative; in Doiron RD, Gomer CJ (eds): Porphyrin Localisation and Treatment of Tumors. New York, Liss, 1984, pp 301–314.

PD Dr. Pius Wyss, Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Frauenklinikstrasse 10, CH–8091 Zu¨rich (Switzerland) Tel. +41 1 255 52 39, Fax +41 1 255 44 33, E-Mail [email protected]


11


............................

Features of Photodynamic Techniques: Diagnostics and Therapeutics P. Wyss Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Switzerland

Photodynamic technique involves the administration of a photosensitizer followed by application of light of a specific wavelength. Photodynamic diagnosis and therapeutics are each based on distinct physical concepts. When light interacts intracellulary with the photosensitizer, the photon energy absorbed by the sensitizing molecule generates mainly fluorescence or photo-oxidation (fig. 1). Whereas fluorescence is used for photomedical diagnosis of cancer cells, photo-oxidation leads to the cell-destroying process used for photodynamic therapy. Since an increased fluorescence following hematoporphyrin administration was discovered in neoplastic tissues compared to normal tissues [1, 2], photodynamic techniques were initially applied in humans for the purpose of cancer diagnosis [3, 4]. A ‘real-time’ system of tissue analysis would eliminate the dissociation of time and space and allow the examining endoscopist to identify the areas of greatest suspicion and to focus the biopsy sampling on those particularly worrisome areas. Any therapy, such as a procedure to ablate the tissue that is dysplastic or malignant, could also be linked to real-time analysis of the tissue in question for an accurate delineation of the extent of the lesion. Photodynamic therapy of tumors in humans was initiated by Lipson et al. [5] in 1966 and has become a main focus of recent studies in photomedicine. Since photosensitizers are mainly accumulated by cancerous tissues [6], photodynamic therapy provides a photo-oxidation-induced, minimal-invasive and selective cancer destruction, which may be performed in an ambulatory setting.

Fig. 1. Photodynamic mechanism: activation of an intracellulary accumulated photosensitizer by laser light of appropriate wavelength induces fluorescence for diagnosis or photooxidation processes for tissue destruction.

References 1 2 3

4 5 6

Policard A: Etudes sur les aspects offerts par des tumeurs expe´rimentales examine´s a la lumie`re de Wood. CR Soc Biol 1924;91:1423–1424. Auler H, Banzer G: Untersuchungen u¨ber die Rolle der Porphyrine bei geschwulstkranken Menschen und Tieren. Z Krebsforsch 1942;53:65–68. Figge FHJ, Weiland GS, Manganiello OJ: Cancer detection and therapy. Affinity of neoplastic, embryonic, and traumatized tissues for porphyrins and metalloporphyrins. Proc Soc Exp Biol Med 1948;68:640–641. Rassmussen DS, Ward GE, Figge FHJ: Fluorescence of human lymphatic and cancer tissues following high doses of intravenous hematoporphyrin. Cancer 1955;1:78–81. Lipson RL, Baldes EJ, Gray MJ: Hematoporphyrin derivative for detection and management of cancer. Cancer 1967;20:2255–2257. Dougherty TJ, Kaufman JE, Goldfarb A, Weishaupt KR, Boyle D, Mittleman A: Photoradiation therapy for the treatment of malignant tumors. Cancer Res 1978;38:2628–2635.


Features of Photodynamic Techniques: Diagnostics and Therapeutics

13

Basics Wyss P, Tadir Y, Tromberg BJ, Haller U (eds): Photomedicine in Gynecology and Reproduction. Basel, Karger, 2000, pp 14–38

............................

Characteristics of Different Photosensitizers Sol Kimel a, Arie Orenstein b, Gad Lavie c a b c

Department of Chemistry, Technion-Israel Institute of Technology, Haifa; Department of Plastic Surgery, and Institute of Hematology and Blood Transfusion Center, Sheba Medical Center, Tel Hashomer, Israel

Introduction Photodynamic therapy (PDT), also termed photochemotherapy, is an experimental modality for treatment of cancer [1–4]. It is based on systemic or topical administration of a photosensitizer which, over time, becomes preferentially retained in tumor tissue. When the ratio of photosensitizer accumulation in the tumor and surrounding normal tissue (tumor/tissue ratio, TTR) is optimal, the tumor area is exposed to an appropriate light dose at a wavelength selected to coincide with an absorption peak of the photosensitizer. Photoexcitation of a sensitizer molecule and intersystem crossing to its triplet state, followed by energy transfer to tissue oxygen, leads to generation of short-lived reactive oxygen species (ROS), including singlet molecular oxygen, 1O2. These species are cytotoxic and elicit tumor necrosis when the tolerance threshold is exceeded [5]. This may occur either directly by cell inactivation (‘cellular mechanism’) [6] and/or indirectly by destruction of microcirculation that supplies blood to the tumor, inflicting lethal damage to cancerous cells (‘vascular mechanism’) [7–10]. The synthesis, in the early 1960s, of the tumor-localizing hematoporphyrin derivative (HpD) preparation marks the beginning of modern PDT. Since then hundreds of articles have been published reporting on the chemical constituents of HpD and of PhotofrinÔ, a commercial porphyrin preparation enriched in tumor-localizing components possessing more potent photodynamic activity. However, the intense research efforts expended in analyzing the complex mixtures HpD and PhotofrinÔ have not led to a satisfactory understanding of

their composition, nor of their mode(s) of action in PDT. PhotofrinÔ has recently been approved in some countries for PDT of selected neoplastic diseases. Among the limitations of PhotofrinÔ we mention: (i) partially known heterogeneous composition; (ii) weak absorption in the therapeutic spectral region; (iii) poor preferential uptake in tumor tissue (TTR =2), and (iv) longlasting generalized cutaneous phototoxicity. Second-generation photosensitizers for improved PDT protocols are under development in many laboratories and are in various phases of preclinical [11] and clinical trials [4, 12]. Such photosensitizers should be chemically pure, nontoxic compounds, characterized by strong absorption in the visible red and near infrared spectral region 600–800 nm (where deeper tissue penetration of light occurs) and possess photophysical properties conducive for generating ROS. Ideally, they should display advantageous tumor-localizing features (TTR ?3) and be rapidly cleared from healthy tissue and plasma. A few dozen porphyrin analogs are currently being evaluated for their PDT potential. These include sulfonated metallo-phthalocyanines (MPcSn, n>0, 1, 2, 3, 4) [13–15], chlorins [16–20], meta-tetra(hydroxyphenyl)chlorin (m-THPC) [21–24], hydroxylated and sulfonated tetraphenyl porphines (THPP and TPPSn) [24–26], benzoporphyrin derivative mono-acid (BPD-MA) [27–29], purpurins [30], porphycenes (Pn) [31, 32], texaphyrins [33–35], and 5-aminolevulinic acid (ALA), a precursor of endogenous protoporphyrin IX [20, 36– 40]. Many of these will probably not pass the stringent requirements before approval for clinical use. Because of the large number of photosensitizers currently under active consideration, we focus on those in phase-II or phase-III trials. For the purpose of this review, we have enumerated clinical and preclinical studies reported in the past decade. Table 1 shows sustained research activity (as gauged by the article flux) mainly for groups 2–5 of second-generation photosensitizers (MPcSn, chlorins, TPP, and BPD). From table 1 we have excluded purely chemical and photophysical investigations and also in vitro studies, important as these may be for elucidating the PDT mechanism(s). About half a dozen novel compounds (including m-THPC, Pn, etiopurpurin, Lu-texaphyrin), and bacteriochlorophyll [41] are quite promising but still without an independent clinical evaluation. A special case is the highly successful and explosively growing development of ALA-mediated PDT in patients [37–40]. In the following sections we will review the general properties and structure-activity relationships (SAR) for the major groups of sensitizers. Their advantages will be discussed in comparison with PhotofrinÔ. This will be followed by a detailed discussion of SAR for two groups of compounds, Pn and AlPcSn. Finally, we will survey some nonporphyrin-based photosensitizers, such as hypericin (HY), that appear also to exert direct tumoricidal activity.

Characteristics of Different Photosensitizers

15

Table 1. Papers/year published for classical (group 1) and some secondgeneration photosensitizers (groups 2–8) used in human or animal PDT studies 1 Por

2 Pc

3 Chl

4a TPP

5 BPD

6 Pur

7 Pn

k/nm

630

675

665

630

690

660

640

e* 104

0.3

25

4

0.4

4.3

3

5

8 ALA

year =87 88 89 90 91 92 93 94 95 96 97

91 24 24 50 30 34 50 49 45 49 45

20 5 10 30 29 26 35 22 24 30 24

4 2 6 5 5 5 7 12 11 15 19

7 1 2 8 3 4 5 1 4 4 3

1 – – 2 1 2 7 6 – 5 2

2 1 3 1 1 – – – – 2 1

– – 1 – – – – 1 1 4 3

– – 1 3 1 9 15 20 44 55 47

Totalb

491

255

91

42

26

11

10

196

Also listed are k (nm) and e (MÖ1 cmÖ1) of relevant absorbance peaks. a Includes ‘TPPSn’ (n>0, 1, 2, 3, 4) and THPP. b Some totals for other photosensitizers: bilirubin, total>13, included in group 3; hypericin, total>8, discussed below; rose bengal, total>6.

Spectral Considerations This is one area in PDT where ‘molecular designing’ has been successful in optimizing sensitizers with desired spectral properties. By judicious modification of chemical structure (e.g. hydrogenation, insertion of double bonds, addition of ring moieties, peripheral substitution) spectral characteristics can be changed to order in a predetermined fashion [42]. Intensities The large range of absorbance values associated with various photosensitizers, as exemplified in figure 1, presents interesting challenges for light dosimetry. Often clinical comparisons do not take this into account. Comparison of photoactivity based on drug dosages should be normalized for absorbance

Kimel/Orenstein/Lavie

16

Fig. 1. Absorbance in the visible spectral region of three typical photosensitizers. Note that the Soret band of Por (at 400 nm) has been scaled down by a factor of five.

to obtain inherent values. On a practical level, a high absorbance allows using smaller drug dosages with concomitant lower toxicity. Spectral Range A characteristic spectrum shows a strong absorbance (Soret band) in the blue and several weak absorbances (Q bands) in the red spectral region (fig. 1). By chemical modification of structure, Q bands can be shifted further into the red to allow deeper tissue penetration of therapeutic light. However, occasionally it is advantageous to use a sensitizer with a less red-shifted Q band, to avoid PDT damage to deeper lying, healthy tissue.

Biodistribution and Pharmacokinetics This remains the single most important open question in PDT. Intensive research has been carried out to correlate biological activity, such as TTR, with chemical structure but the mechanisms remain largely unknown [43]. In the ‘cellular mechanism’, sensitizer molecules administered by intravenous


17

Fig. 2. Structure of TPPS4 (R>SOÖ 3 ) and pTHPP (R>OH).

route have to pass through endothelial cells and vascular membrane structures, diffuse through interstitial spaces before entering tumor cells. In the ‘vascular mechanism’ too, the route to endothelial cell uptake and destruction is arduous. We mention here four approaches to better understanding, on the molecular level, of uptake and retention of sensitizers in various biological compartments: (i) hydro/lipo/amphiphilicity [42–46]; (ii) conjugation to tumor-specific delivery systems [47, 48]; (iii) tissue-pH dependent uptake [49–51], and (iv) stereochemical factors [52]. (i) Cellular internalization of sensitizers by passive diffusion seems to correlate with lipophilicity. For hydrophilic and ionic compounds endocytosis becomes the dominant mode of cellular uptake. This has been studied systematically for series of increasingly sulfonated tetraphenyl porphines, TPPSn (fig. 2) [26] and MPcSn (fig. 3) [53–56]. However, whereas uptake in vitro correlates positively with lipophilicity, uptake in vivo shows a negative correlation [55]. Amphiphilic compounds, such as TPPS2a, AlPcS2a (sulfonated on adjacent subunits, with R1>R2>SOÖ 3 ; fig. 2, 3), or BPD, are taken up to a greater extent and show higher biological activity than otherwise similar compounds, presumably because they can enter cells by complementary uptake mechanisms [46]. (ii) Binding of sensitizers to lipoproteins [28, 47, 48] and monoclonal antibodies [48] or encapsulation in liposomes [29, 48] have been used to improve in vitro cellular uptake, in vivo tumor uptake (TTR), and PDT efficacy. Peripheral substitution with high-molecular weight polyethylene glycol chains influences TTR and the ‘drug-time window’ [57].


18

Fig. 3. Structure of MPcSn (Rn>SOÖ 3 and R4Ön>H).

(iii) Because of metabolic acidosis, the average pH of tumor tissue is one unit lower (pH 6.4) than normal tissue (pH 7.4). Porphyrins and phthalocyanines exist in a dynamic ionic equilibrium of various species, ranging from dicationic, through neutral, to dianionic, depending on the pH of the microenvironment. In extracellular space, at pH 6.4, neutral molecules cross a cell membrane by diffusion, but in the intracellular medium at pH 7.4, are transformed into ionic species and become ‘trapped’ [49–51]. Anionic species localize in lysosomes and cationic in mitochondria. (iv) The marked difference in the relative PDT efficacy of amphiphilic S2a and lipophilic S2o sensitizers, sulfonated on adjacent (R1>R2>SOÖ 3 ) or opposite (R1>R3>SOÖ 3 ) subunits, and of m-THPP versus o- and p-THPP (fig. 2), led us to a study of common geometrical features [52]. Structural modeling of these and related compounds was carried out which showed that Ö a pair of oxygen atoms (originating in SOÖ or OH substituents), 3 , COO separated by a ‘critical’ distance of 1.2 nm, confer increased photodynamic efficacy to a sensitizer. The more difficult question as to what receptor molecule in a target cell is responsible for preferential sensitizer binding remains to be resolved. Tubulin and calmodulin have been suggested [52].

Some Representative Groups of Sensitizers Porphycenes Porphycenes (Pn) are structural isomers of porphine, based on a 16membered macrocycle with two ethine bonds and two direct links between


19

pyrroles, instead of four methine bonds (as in porphine), so that the Pn macrocycle possesses rectangular symmetry rather than fourfold symmetry [58]. Lowering the symmetry increases the Q-band absorbance of Pn in the red spectral region, characterized by a molar absorption coefficient (e630V50,000 MÖ1 cmÖ1) which is about 15-fold larger than for porphyrins. Pns have become highly promising second-generation photochemotherapeutic agents. Some Pns were found to generate singlet molecular oxygen [31, 59] and to possess superior tumor-localizing properties, together with extremely fast clearance from plasma (on the minute scale) [60]. We have studied four derivatives of 2,7,12,17-tetrakis(methoxyethyl)porphycene (TMPn) [61], substituted in the 9 position of the macrocycle, which were synthesized by Aramendia et al. [31] and Vogel [32] and used as received without further purification (fig. 4). They differ in the type of substituent in the 9 position as follows: 9-glutaric acid amide TMPn (GlamTMPn); 9-nicotinic acid amide TMPn (NicamTMPn); 9-capronyloxy (CpoTMPn), and 9stearoyloxy TMPn (StoTMPn). The substituents, in the given order, impart an increasing degree of lipophilicity to the TMPn molecule, as demonstrated by increasing retention times (Rt) in reverse phase HPLC [62] (table 2). Intraliposomal Location Pns were encapsulated in dipalmitoylphosphatidylcholine (DPPC) liposomes; concentrations were determined from absorbance spectra. Fluorescence quenching by iodide was used to determine the influence of the substituent in the 9-position on Pn location within the liposome membrane [61]; data were analyzed using the Stern-Volmer equation: F0/F>1+K [Q].

Here F0 and F denote, respectively, fluorescence intensities in the absence and presence of the IÖ quencher; [Q] is the molar concentration of the quencher, and K is the quenching constant. Figure 5 depicts Stern-Volmer plots for fluorescence quenching of the four Pns in DPPC liposomes. Clearly, GlamTMPn, NicamTMPn and CpoTMPn were accessible to iodide ions roughly proportional to their hydrophilic nature, given in terms of measured Rt [62] (table 2, columns 2 and 3), whereas the lipophilic compound StoTMPn (for which Rt>21.6 min) is ‘buried’ inside the DPPC phospholipid bilayer and cannot be quenched by anions. Thus, fluorescence measurements provide insight on the intraliposomal location of sensitizer molecules in unilamellar DPPC vesicles. This information may be important for optimizing delivery of liposome-bound drugs to cultured cells and to tissue [29, 48, 63].


20

Fig. 4. Structures of four TMPn derivatives.

Photodynamic efficacy of Pns was tested using in vitro cultured epithelial MDCK cells, and the in vivo chorioallantoic membrane (CAM) model [63]. PDT Injury in vitro Cells were incubated for 4 h in a medium containing 1.2 lg/ml liposomebound Pn. A red-filtered halogen light source (delivering 8.7 mW/cm2 at k


21

Table 2. HPLC Rt, Stern-Volmer quenching constants, and PDT damage in vitro and in vivo for four TMPn derivatives TMPn

Rt mina

K MÖ1

Cell damageb

CAM injury c

Glam Nicam Cpo Sto

1.3 1.8 3.7 21.6

4.1 1.5 2.4 –0.2

1.1 1 0.8 0.05

0.3 1 0.1d 0.5

a

Segalla et al. [62]. 16 J cmÖ2. c 50 ng, 24 J cmÖ2. d Corrected for dark toxicity. b

?600 nm) was used to irradiate a tissue culture microplate through a frame, exposing 9 wells at a time, for 4, 8, 16 or 32 min. PDT efficacy was expressed as percentage cell death relative to ‘blank’, defined as number of cells in microplate wells that were incubated with porphycene but not exposed to red light. Cells were counted 24 h after PDT, using vital staining with methylene blue (MB). The absorbance of MB at 620 nm, which is related to the number of living cells, was measured for each well by a microplate photometer Elisa reader. Results are shown in figure 6. Porphycene Uptake in vivo The CAM model is based on the vascular system of the CAM of the chick embryo [63]. Liposome-bound porphycene at a concentration of 250 ng/ 20 ll was administered topically in the area demarcated by a 6-mm diameter teflon ring placed on the CAM. Fluorescence measurements during drug uptake into CAM tissue were performed using a bifurcated optic light guide, which transmits excitation light to the CAM surface and emission from CAMembedded porphycene to the detection system of a spectrofluorimeter. For each porphycene, the emission spectrum (kexcV370 nm; kfl>600–700 nm) was recorded at 10-min intervals, as a measure for the uptake dynamics in the CAM. Since initial fluorescence intensity F(t>0) differed for individual CAMs, the evolution of the fluorescence intensity F(t) was determined as F(t)/F(t>0). PDT Injury of CAM Blood Vessels Irradiation with HeNe laser light (633 nm, 14.3 mW) was started 30 min after Pn administration at 50, 125 or 250 ng/20 ll, when fluorescence


22

Fig. 5. Stern-Volmer plots for fluorescence quenching with IÖ of TMPn derivatives in DPPC liposomes.

measurements had indicated that efficient uptake in the vascular and extravascular compartments of the CAM had taken place. The laser was positioned such that the tissue-scattered beam filled the entire ring area, giving an irradiance of 50 mW/cm2. Three hours after irradiation, blood vessels in the irradiated area were inspected with a stereomicroscope in a doubleblind fashion. Damage was classified [63] as: 0>no damage; 1>slight damage, vasodilation/constriction; 2>moderate damage, hemostasis, clotting, CAM denaturation, 3>severe damage, widespread occlusion, hemorrhage. Results of light and drug dose-dependent damage are depicted in figure 7.


23

Fig. 6. Damage in cell cultures following Pn-mediated PDT. Percentage dead cells versus light dosage (every 8 min irradiation corresponds to 4 J/cm2).

SAR for TMPn Examination of the combined in vitro and in vivo results (table 2, columns 4 and 5) shows that the in vitro photodynamic mechanism is basically different from that in vivo: (i) the proximity of the Pns to the membrane outer surface was roughly in accordance with the hydrophilicity of the substituent in the 9 position: GlamTMPn?CpoTMPn?NicamTMPn9 StoTMPn; (ii) photodynamic damage to MDCK cell cultures, as assayed by vital staining with MB, could be correlated to hydrophilicity, i.e. to Pn location in a liposome membrane: Pns located close to the liposome outer surface caused larger cell damage, and (iii) in contrast to the in vitro case, Pn location inside liposome membranes was not relevant to CAM blood vessel injury: StoTMPn that was ‘buried’ inside the bilayer still caused efficient PDT in vivo. The model proposed for sensitizers encapsulated in liposomes assumes that for in vitro conditions (without serum) photodynamic damage occurred after contact-mediated transfer of Pn from liposomes, which remain distinct, to cell membranes in culture. Transfer is facilitated by the proximity of a Pn molecule to the outer surface of the liposome vesicle. In contrast, in the in vivo model the photosensitizer is delivered to blood vessels via fusion of liposomes and endothelial cell membranes. This model is supported by reports of increased liposome permeability in vivo compared to in vitro where no blood compartments are present.


24

Fig. 7. Vascular injury in the CAM following TMPn-mediated PDT, as a function of drug and light dosages (every 8 min irradiation corresponds to 24 J/cm2).

Phthalocyanines Phthalocyanines (Pcs) have received considerable interest for use in PDT, mainly because of their large absorption coefficient (e672>2.5 * 105 MÖ1 cmÖ1) in the spectral region where light penetration into tissue is optimal [42]. Sulfonated Pcs with a centrally chelated metal ion (MPcSn, M>Al, Zn) are particularly attractive PDT agents; they are void of local and systemic toxicity and their photosensitizing properties have been evaluated extensively in cell cultures and under in vivo conditions. A mixture of differently sulfonated, water-soluble AlPcSn (PhotosensÔ) is used routinely in Russia for clinical PDT [64]. The degree of sulfonation, the isomeric composition and the nature of the central metal ion affect the solubility and the extent of aggregation. These, in turn, affect cellular uptake and in vivo pharmacokinetics [44]. Uptake in vitro decreases with the degree of sulfonation [53–56]; however, phototoxicity


25

does not parallel uptake, possibly reflecting differences in intracellular distribution patterns among the sulfonated compounds [54]. Furthermore, in contrast with the in vitro behavior, tumor retention in vivo appears to increase with sulfonation [55, 56]. Such discrepancies point to complex relationships between structure and composition of MPcSn preparations and their biological behavior. Spectroscopic Studies Physical-chemical characteristics of purified compounds AlPcSn (n>2, 4) and of commercial preparations AlPcSnmix (n>0, 2, 3, 4) were evaluated spectroscopically [65] with regard to molecular symmetry, tendency to aggregate, and lipophilicity. These could be correlated with cell uptake and intracellular localization patterns. For AlPcSn we distinguish between ‘compounds’ (i.e. different n), ‘geometrical’ isomers (e.g. AlPcS2a and AlPcS2o sulfonated on adjacent and opposite phthalic subunits, respectively) and ‘regioisomers’ (e.g. sulfonate substituents on positions 4,4, 5,4 and 4,5 in adjacent phthalic subunits, as indicated in figure 2). Rt, obtained in reverse-phase HPLC, is a measure of hydrophobicity. In the case of differently sulfonated AlPcSn, Rt varies from 1.0 for AlPcS4 (taken as unity) to 2.4 for the hydrophobic AlPcS1. AlPcS3 isomers elute around Rt>1.3, whereas the disulfonated fractions elute from Rt>1.6 for AlPcS2o to Rt>1.9 for amphiphilic AlPcS2a [66]. HPLC analysis of the commercial preparations showed that AlPcS2mix contained both geometrical isomers, whereas AlPcS3mix and AlPcS4mix consisted of a mixture of different sulfonated and isomeric compounds. Aggregation All AlPcSn compounds have a similar absorbance spectrum in DMSO and exhibit a strong Q band around 672 nm, e672V2.5 * 105 cmÖ1 MÖ1, and a weak Q band at 606 nm. AlPcS3mix and AlPcS4mix dissolved easily in PBS and their spectra were similar as in DMSO. Under the same conditions, AlPcS2mix and AlPcS4 dissolved in PBS exhibited DMSO-type spectra, with an additional weak band at 644 nm, whereas for the pure compounds AlPcSn (n>0, 2a, 2o) the 644-nm band became the main feature. These compounds were less soluble in PBS than nonaggregated AlPcS3mix and AlPcS4mix, which did not show the 644-nm band. The increase in the 644-nm band was at the expense of the monomer bands at 672 and 606 nm, and became more pronounced with aggregation (table 3). Dimer formation was corroborated by the lower fluorescence measured at kex>644 nm, compared to kex>606 nm in spite of the fact that the absorbance (A) is larger; i.e. A644?A606. Since monomer fluorescence is more intense than that of dimers, and aggregates do not fluoresce at all [67, 68], we


26

Table 3. Effective absorption coefficients of AlPcSn monomer eeff,m(672) and dimer eeff,d(644) bands, aggregation in PBS solution, and intracellular uptake Sensitizer

eeff,m(672) * 105

eeff,d(644) * 105

aggreg %

AlPc AlPcS2a AlPcS2o AlPcS4

D0.33 0.40 0.61 1.6

0.38 0.98 1.4 0.92

85×5 85×5 75×5 37×5

– 8.7 1.7 1.4

2.0 2.6 2.5

0.48 – –

11×5 12×5 0

60 1.9 2.4

AlPcS2mix AlPcS4mix AlPcS3mix

uptake 108 molecules/cell

assigned the 644-nm band to dimers. For AlPcS4 and AlPcS2mix in PBS, dimers persisted even at 10Ö7 M. The degree of aggregation was defined as: Aggreg (%)>(A672 (DMSO)ÖA672 (PBS)) * 100/A672 (DMSO),

where A672 (DMSO) and A672 (PBS) denote, respectively, absorbance of a 5 * 10-7 M AlPcSn solution in DMSO and in PBS. Clearly, pure isomeric compounds show a higher degree of aggregation than any of the composite commercial preparations (table 3, column 4). The steric conformation of the dimers may also be deduced from spectral observations. The Soret bands of AlPcSn (n>2a, 2o, 4) at 360 nm were blue shifted by 1,500 cmÖ1 and broadened compared to the Soret bands of (nonaggregated) AlPcS4mix or AlPcS3mix, whereas the Q bands showed a bathochromic shift (100 cmÖ1). Similar features in porphyrins have been attributed to ‘sandwich’-type dimer formation [69, 70]. Such dimers are face-to-face complexes where the axial chlorides or hydroxyl groups of monomers are in opposite (outward) directions. Because of steric hindrance between SOÖ 3 groups, Pc skeletons in the dimer are staggered; i.e. rotated by 180º, 90º, and 45º for, respectively, n>2a, 2o, and 4. Aggregation of monomers and of sandwich-type dimers can occur for pure, symmetrical regioisomers where building blocks can be repeated with minimal steric perturbations. For AlPcS4, the decrease in the ratio A644/A672 with concentration implies formation of oligomers. Moreover, the bathochromic shift of the Soret band with increasing concentration suggests branched-chain formation [69]. The commercial preparations AlPcS4mix and AlPcS3mix do not form dimers and aggregates but, as expected, AlPcS2mix molecules do dimerize in ‘sandwich’-type structures.


27

For the pure isomeric compounds AlPcSn (n>2a, 2o, 4) the degree of aggregation in PBS increased with lipophilicity, as expected (table 3). The ‘effective’ absorption coefficients, eeff(k) [65], can be understood from the lipophilicity-symmetry relationship. The pure compounds AlPcSn showed monomer bands eeff,m(672) which increased with hydrophilicity; mixed isomeric compounds generally had higher eeff,m(672) because their low symmetry prevented aggregation. AlPcS2a and AlPcS2o behave more complexly [67]. AlPcS2a is more aggregated than AlPcS2o, yet its dimer band, eeff,d(644) is weaker (table 3). This indicates that amphiphilic AlPcS2a aggregates to higher oligomers, whereas AlPcS2o creates mostly dimers. AlPcS2mix contains various regioisomers within each geometrical isomer, AlPcS2a or AlPcS2o, which prevents dimerization. In general, in NaOH solutions, eeff,m(672) was higher than in PBS and eeff,d(644) was lower. Fluorescence is even more sensitive to aggregation. For example, the absorbance of AlPcS2a increased 3.5-fold due to dissociation of aggregates in NaOH, whereas the fluorescence intensity increased 8-fold [65]. SAR for AlPcSn Comparing photophysical and chemical properties of a series of structurally related photosensitizers in solution provides important information on their potential photodynamic applications. The monomeric (photoactive) form is favored for preparations consisting of mixtures of various regioisomers or differently substituted compounds. Although pure compounds are generally preferred as pharmaceuticals, our data indicate that mixed isomeric preparations may be favored as photodynamic agents compared to isomerically pure compounds. Mixed preparations are likely to be more efficient drugs for biomedical applications. They contain a higher monomer fraction, exhibit better uptake characteristics (table 3, column 5), and can attack different intracellular target molecules [67]. This may be the reason for the relatively high PDT efficacy of the ‘impure’ commercial preparations HpD, PhotofrinÔ and PhotosensÔ.

Polycyclic Aromatic Ketones Hypericin The best known nonporphyrinic polyaromatic ketone photosensitizer is hypericin (HY), 10,11-dimethyl-1,3,4,6,8,13-hexahydroxynaphthodianthrone (fig. 8). It has a perihydroxylated mesonaphthodianthrone backbone consisting of eight fused aromatic rings. HY was originally isolated from the medicinal plant Hypericum perforatum commonly known as St. John’s wort, together


28

Fig. 8. Hypericin.

with a congener pseudohypericin. It is now being synthesized by various methods that involve self condensation of emodin or emodin anthrone [71]. Based on pH-dependent absorption spectral analyses, HY is found to be an acidic compound. Its first pKa values are 1.7 and 1.2 in 80% EtOH and DMSO, respectively, and the second pKa value is 12.0 in both solvents [72]. At pH =4.5 HY exists in a free (acid) form. Free HY is insoluble in aqueous and most polar organic solvents. At pH values between 4.5 and 11, HY forms monobasic salts (ion pairs) with metal and nonmetal cations, including basic amines and amino acids [72]. Most biological studies, as well as those performed with plant-purified HY, involved HY-Na+. Above pH 11 a second proton is substituted to form dibasic salts. A single-crystal X-ray diffraction of pyridinium salt of HY shows that the molecule is distorted and displays a helical twist [73]. HY has two primary visible range absorption peaks at 545 and 590 nm. The compound is insoluble in aqueous media forming aggregates that quench its characteristic fluorescence at 640 nm. The aggregates dissociate in organic solvents or upon interaction with lipids and human serum albumin, and HY regains its fluorescence and spectral properties that are characteristic for the monomeric form. From a photophysical viewpoint, HY is a unique, amphielectronic compound due to the vicinity of electron-donating phenolic groups and acceptor quinones. HY generates a quantum yield of singlet oxygen in excess of 0.7 [74, 75]. Superoxide anions have also been noted to be generated by HY [76], as well as semiquinone radicals [77, 78]. Consequently, a variety of photodynamic, oxygen-dependent and -independent biological effects have been reported for HY. In addition, HY is a lipophilic compound that was also found to bind cell membrane phospholipids [79]. The powerful photosensitizing properties of HY cause cutaneous photosensitization in humans. The clinical syndrome of phototoxicity has been


29

known for many years among grazing animals that consume large amounts of HY-containing hypericum plants and remain in the sun all day. It has been termed ‘hypericism’ and refers to lesions of necrosis occurring in epithelial tissues not protected by fur, that under extreme situations may cause death of the animal. Studies aiming to utilize the combination of photodynamic and lipophilic properties of HY for medical uses unravelled multiple potential mechanisms that can be employed for a variety of applications. HY was found to possess antiviral and antiretroviral properties [80–82] that affected numerous phases of the retroviral infection and replication cycle. The compound was virucidal to extracellular virions in vitro [80], an activity that could be prevented by sodium azide and b-carotene, both singlet oxygen quenchers [79]. The affinity of HY to cell membranes has also been found to inactivate newly assembled virus particles as they bud from infected cells [81]. Cell bound HY was also protective from de novo infection with virus, possibly during the fusion process of viral and cell surface membranes that occurs during infection [83]. The potential for utilization of HY in PDT of tumors appears to entail a number of facets. HY binds well to tumor cell lines in vitro and the degrees of phototoxicity correlate with the cellular uptake of HY as determined by laser confocal microscopy [84]. HY elicits direct, light-dependent cytotoxicity to various tumor cell lines in vitro. The human MRC5 fibroblast cell line [85], EMT6 murine mammary carcinoma cells in vitro [86] and human neuroblastoma cells [87] were strongly damaged. Phototoxicity to the cells occurred via an oxygen-dependent mechanism that was insensitive to reduction in intracellular glutathione levels [84], quite likely a type-II singlet oxygen reaction. In vivo effects also appear to be mediated by a type-II reaction. The two main mechanisms of cell death induction are necrosis, which is inflicted externally, and apoptosis, in which an endogenous program of cell suicide is activated. Necrosis is associated with rupture of the cell membrane and leakage of cytosolic and lysosomal contents into the extracellular environment, which induce inflammation. Apoptosis is a programmed physiological mode of cell death initiated by mitochondrial damage that results in release of cytochrome c [88, 89]. This leads to degradation of cellular DNA in extranucleosomal spacings, resulting in a characteristic ladder pattern of oligonucleosomal-sized fragments of DNA. In addition, the cell nuclei are fragmented and the chromatin becomes condensed. Photodynamically induced cell death initially occurs via activation of a cellular pathway to apoptosis [87]; however, as the level of photosensitization with HY increases, cell death shifts to necrosis. Similar shifts have been reported in photodynamic cell killing with protoporphyrin IX [90] and with Pcs [91, 92].


30

There is also evidence for good association of HY with tumor cells in vivo. HY, administered systemically to nude mice bearing a human squamous cell carcinoma tumor, was retained within tumors at concentrations that exceed most other organs and for longer periods than in normal tissues [93]. The antitumoral effects of HY have been attributed to numerous mechanisms affecting a variety of cellular targets. In membranes, the photodynamic effects of HY cause lipid peroxidation and oxidation of amino acids in proteins. HY binds to mitochondrial membranes where it was found to elicit light-dependent reduction in membrane potential, to interfere with respiration [94] and to inhibit mitochondrial succinoxidase [75]. In addition, HY possesses inhibitory activities of cell proliferation, signal transduction pathways. It was shown to inhibit protein kinase C (PKC) [95]. Both membrane-associated and cytosolic PKC are irreversibly inhibited. However, inactivation of the membrane-associated enzyme appears to be independent of light [96]. HY is under evaluation as a tumoristatic agent in brain glioblastoma due to its PKC inhibitory activity. PKC appears to play an important role in signal transduction of glioblastoma cell proliferation [97]. HY acts as inhibitor of mitogen-activated protein kinase [98] and epidermal growth factor receptor tyrosine kinase [99], both are part of a signal transduction cascade that culminates in cell division. Inhibition of tyrosine kinase was evident only when the enzyme was in association with membranes. Solubilization of the membranes with a nonionic detergent prior to administration of HY eliminated the inhibitory activity of HY; however, it was not effective if treatment with HY preceded membrane solubilization. The association of HY with cell membranes maintains its monomeric photoactive form, which may explain the link between the inhibition of enzymatic activity with the need for membrane integrity. The inhibition of PKC and tyrosine kinase activities by HY suggests new potential antiviral inhibitory pathways. Tyrosine phosphorylation has been reported to occur after HIV interaction with its CD4 cellular receptor and to participate in the cytopathic activity of HIV [100]. The virus-receptor interaction during infection also induces PKC activation and phosphorylation of the CD4 cell surface receptor [101]. Another feature of the biological activity of HY, which may carry therapeutic potential, is inhibition of the lytic phase of the cytotoxicity reaction of CD8 lymphocytes [79]. Cytotoxic CD8 T cells (CTL) comprise the cellular effector arm of the immune system and mediate the destruction of cells targeted for elimination. These include allogeneically transplanted cells during immune rejection, targeted tumor cells or virus-infected cells. Target cell killing by sensitized CTL has been shown to be completely inhibited by HY in culture [79]. Some degrees of inhibition of graft versus host disease (GVHD) induced


31

in mice in vivo have been observed and resulted in some prolongation in survival of HY-treated animals in which GVHD has been induced [102]. Furthermore, the expression of CD4, a molecule responsible for the interaction between different subsets of immune competent cells via the class-II histocompatibility complex, is effectively inhibited in a dose-dependent manner after treatment of murine cells with HY in vitro [79]. This observation may help explain the suppression of GVHD by HY and suggests that other immunemediated inflammatory diseases such as psoriasis can be treated with HY. PKC-mediated signal transduction pathways that lead to inflammatory reactions affect epidermal cells in psoriasis. Thus, many photosensitizers, including HY and the related hypocrellins, which will be discussed later in this review, are under clinical evaluation in PDT of psoriasis. The types of ion pairs formed have been found to affect the antiretroviral activities and plasma protein-binding properties of HY. HY-Na+ binds well to plasma proteins, liposomes or cell membranes [72]. Binding was confirmed by fluorescence intensity changes, gel filtration chromatography and ultrafiltration profiles through Diaflo semipermeable membranes, and by measurements of circular dichroism spectra. Electrochemical and electron paramagnetic resonance (EPR) studies show that HY and its salts have electronaccepting as well as electron-donating properties and are, thus, both oxidizing and reducing agents [102]. This property enables HY to act as an electron scavenger from biological electron transfer reactions and may explain some biological activities of HY that are not strictly light-dependent, yet are enhanced by light [103]. Human serum albumin (HSA) appears to be the major binding protein of HY in plasma, as well as the low-density lipoproteins (LDL), high-density lipoproteins (HDL) and very LDL. Scatchard regression analyses revealed HY:HSA molar ratios of 1:1, indicating true binding. However, higher molar ratios in which HY associates with LDL and HDL reflect interactions with the lipid moiety of lipoproteins, most likely the phospholipid shell [72]. Some organic ion pairs of HY, such as the lysine or pyridine salts, do not associate well with HSA. The dispersion of HY-lysine or pyridine in plasma is not accompanied by dissociation of the colloids, and HY-lysine also failed to inactivate HIV in vitro [72]. Hypocrellins and Cercosporin Hypocrellins and cercosporin belong to a group of photosensitizers with a perylene quinone backbone that share a basic molecular chromophore, 4,9dihydroxy–3,10-perylenequinone, which generates singlet oxygen. The perylenequinones show marked similarities with the mesonaphthodianthrones (HYs), both structurally and in ability to act as electron donors as well as electron


32

acceptors [104]. The group includes a number of naturally produced pigments, mostly by fungi. Hypocrellins A and B are lipophilic perylenequinonoid products of Hypocrella bambuase, a parasitic fungus of western China, Tibet and Sri Lanka. Cercosporin, produced by the parasitic fungus Cercospora, appears to be utilized by the fungus to cause necrotic lesions to leaves that form substrate for the fungus. Photodynamically induced singlet oxygen reactions mediate the formation of these lesions at sites infected by the fungus. Hypocrellins appear to differ from HY in the Red/ox potential. They act as weaker electron donors [105] and as acceptors from ferrous ion, ascorbate or dithionite. These compounds could not act as donors for HY [106]. Hypocrellins have been used for many years in China to treat psoriasis. They are now being considered for PDT of tumor cells mainly for two reasons: extremely rapid kinetics of accumulation within tumors with less accumulations in skin, bone muscle and other organs [105], and in mice, they are cleared rapidly from plasma and are fully incorporated into tumors 2 h after intravenous administration, resulting in much less skin photosensitization than with porphyrins [105]. Hypocrellins have also been found to act as PKC inhibitors with activity directed towards the regulatory domain of the enzyme [107].

References 1 2 3 4

5 6

7

8 9

10

Henderson BW, Dougherty TJ: How does photodynamic therapy work? Photochem Photobiol 1992; 55:145–157. Moan J, Berg K: Photochemotherapy of cancer: Experimental research. Photochem Photobiol 1992; 55:931–948. Dougherty TJ: Photodynamic therapy. Photochem Photobiol 1993;58:895–900. Schuitmaker JJ, Baas P, van Leengoed HLLM, van der Meulen FW, Star WM, van Zandwijk N: Photodynamic therapy: A promising new modality for the treatment of cancer. J Photochem Photobiol B 1996;34:3–12. Ochsner M: Photophysical and photobiological processes in the photodynamic therapy of tumours. J Photochem Photobiol B 1997;39:1–18. Hilf R: Cellular targets of photodynamic therapy as a guide to mechanism; in Henderson BW, Dougherty TJ (eds): Photodynamic Therapy: Basic Principles and Clinical Applications. New York, Dekker 1992, pp 47–54. Star WM, Marijnissen JPA, van den Berg-Blok AE, Versteeg AAC, Franken NAP, Reinhold HS: Destruction of rat mammary tumor and normal tissue microcirculation by hematoporphyrin derivative photoradiation observed in vivo in sandwich observation chambers. Cancer Res 1986;46:2532– 2540. Nelson JS, Liaw LH, Orenstein A, Roberts WG, Berns MW: Mechanism of tumor destruction following photodynamic therapy. J Natl Cancer Inst 1988;80:1599–1605. Tromberg BJ, Orenstein A, Kimel S, Barker SJ, Hyatt J, Nelson JS, Berns MW: In vivo tumor oxygen tension measurements for the evaluation of the efficiency of photodynamic therapy. Photochem Photobiol 1990;52:375–385. Fingar VH, Wieman TJ, Wiehle SA, Cerrito PB: The role of microvascular damage in photodynamic therapy: The effect of treatment on vessel constriction, permeability, and leukocyte adhesion. Cancer Res 1992;52:4914–4921.


33

11 12 13 14 15

16 17

18

19 20

21

22

23

24

25 26 27 28 29

30 31

32 33

Gomer CJ: Preclinical examination of first and second generation photosensitizers used in photodynamic therapy. Photochem Photobiol 1991;54:1093–1107. Fisher AMR, Murphree AL, Gomer CJ: Clinical and preclinical photodynamic therapy. Lasers Surg Med 1995;17:2–31. Rosenthal I: Phthalocyanines as photodynamic sensitizers. Photochem Photobiol 1991;53:859–870. van Lier JE: Phthalocyanines as sensitizers for the PDT of cancer; in Kessel D (ed): Photodynamic Therapy of Neoplastic Disease. Boca Raton, CRC Press 1990, vol I, pp 279–291. Ben-Hur E: Basic photobiology and mechanisms of action of phthalocyanines; in Henderson BW, Dougherty TJ (eds): Photodynamic Therapy: Basic Principles and Clinical Applications. New York, Dekker 1992, pp 63–78. Spikes JD: New trends in photobiology. Chlorins as photosensitizers in biology and medicine. J Photochem Photobiol B 1990;6:259–274. Spikes JD, Bommer JC: Photosensitizing properties of mono-L-aspartyl chlorin e6 (Npe6): A candidate sensitizer for the photodynamic therapy of tumors. J Photochem Photobiol B 1993;17: 135–143. Kostenich GA, Zhuravkin IN, Furmanchuk AV, Zhavrid EA: Photodynamic therapy with chlorin e6. A morphological study of tumor damage efficiency in experiment. J Photochem Photobiol B 1991;11:307–318. Kostenich GA, Zhuravkin IN, Furmanchuk AV, Zhavrid EA: Sensitivity of different rat tumour strains to photodynamic treatment with chlorin e6. J Photochem Photobiol B 1993;17:187–194. Orenstein A, Kostenich G, Roitman L, Schechtman Y, Kopolovic Y, Ehrenberg B, Malik Z: A comparative study of tissue distribution and photodynamic therapy selectivity of chlorin e6, Photofrin II and ALA-induced protoporphyrin IX in a colon carcinoma model. Br J Cancer 1996;73:937–944. Morlet L, Vonarx-Coinsmann V, Lenz P, Foultier MT, de Brito LX, Stewart C, Patrice T: Correlation between meta(tetrahydroxyphenyl)chlorin (m-THPC) biodistribution and photodynamic effects in mice. J Photochem Photobiol B 1995;28:25–32. Ris HB, Altermatt HJ, Nachbur B, Stewart CM, Wang Q, Lim CK, Bonnett R, Althaus U: Intraoperative photodynamic therapy with m-tetrahydroxyphenylchlorin for chest malignancies. Lasers Surg Med 1993;18:39–45. Andrejevic-Blant S, Hadjur C, Ballini JP, Wagnieres G, Fontolliet C, van den Bergh H, Monnier P: Photodynamic therapy of early squamous cell carcinoma with tetra(m-hydroxyphenyl)chlorin: Optimal drug-light interval. Br J Cancer 1997;76:1021–1028. Peng, Q, Moan J, Ma LW, Nesland JM: Uptake, localization, and photodynamic effect of mesotetra(hydroxyphenyl)porphine and its corresponding chlorin in normal and tumor tissues of mice bearing mammary carcinoma. Cancer Res 1995;55:2620–2626. Bonnett R, White RD, Winfield UJ, Berenbaum MC: Hydroporphyrins of the meso-tetra(hydroxyphenyl) porphyrin series as tumor photosensitizers. Biochem J 1989;261:277–280. Berg K, Bommer JC, Winkelman JW, Moan J: Cellular uptake and relative efficiency in cell inactivation by photoactivated sulfonated meso-tetraphenylporphines. Photochem Photobiol 1990;52:775–781. Steiner RA, Tadir Y, Tromberg B, Wyss P, Walt H, Haller U: Benzoporphyrin derivative mono acid for photodynamic therapy of the endometrium. Geburtshilfe Frauenheilkd 1996;56:1–7. Allison BA, Pritchard PH, Levy JG: Evidence for low-density lipoprotein receptor-mediated uptake of benzoporphyrin derivative. Br J Cancer 1994;69:833–839. Richter AM, Waterfield E, Jain AK, Canaan AJ, Allison BA, Levy JG: Liposomal delivery of a photosensitizer, benzoporphyrin derivative monoacid ring A (BPD), to tumor tissue in a mouse tumor model. Photochem Photobiol 1993;57:1000–1006. Garbo GM: Purpurins and benzochlorins as sensitizers for photodynamic therapy. J Photochem Photobiol B 1996;34:109–116. Aramendia PF, Redmond RW, Nonell S, Schuster W, Braslavsky SE, Schaffner K, Vogel E: The photophysical properties of porphycenes: Potential photodynamic therapy agents. Photochem Photobiol 1986;44:555–559. Vogel E: Novel porphyrinoids. Pure Appl Chem 1990;62:557–564. Sessler JL, Hemmi G, Mody TD, Murai T, Burrell A, Young SW: Texaphyrins: Synthesis and applications. Acc Chem Res 1994;27:43–50.


34

34

35

36 37 38 39 40

41

42 43

44 45

46

47

48 49 50 51 52

53

54 55

Young SW, Woodburn KW, Wright M, Mody TD, Fan Q, Sessler JL, Dow WC, Miller RA: Lutetium texaphyrin (PCI-0123): A near-infrared, water-soluble photosensitizer. Photochem Photobiol 1996; 63:892–897. Kostenich G, Orenstein A, Roitman L, Malik Z, Ehrenberg B: In vivo photodynamic therapy with the new near-IR absorbing water soluble photosensitizer lutetium texaphyrin and a high intensity pulsed light delivery system. J Photochem Photobiol B 1997;39:36–42. Kennedy JC, Pottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 1992;14:275–292. Orenstein A, Kostenich G, Malik Z: The kinetics of protoporphyrin fluorescence during ALA-PDT in human malignant skin tumors. Cancer Lett 1997;120:229–234. Peng Q, Warloe T, Berg K, Moan J, Kongshaug M, Gierscksky KE, Nesland JM: 5-Aminolevulinic acid-based photodynamic therapy. Clinical research and future challenges. Cancer 1997;79:2282–2308. Szeimies RM, Calvazaro-Pinton PG, Karrer S, Ortel B, Landthaler M: Topical photodynamic therapy in dermatology. J Photochem Photobiol B 1996;36:213–219. Tope WD, Ross EV, Kollias N, Martin A, Gillies R, Anderson RR: Protoporphyrin IX fluorescence induced in basal cell carcinoma by oral d-aminolevulinic acid. Photochem Photobiol 1998;67: 249–255. Rosenbach-Belkin V, Chen L, Fiedor L, Tregub I, Paviotsky F, Brumfeld V, Salomon Y, Scherz A: Serine conjugates of chlorophyll and bacteriochlorophyll: Photocytotoxicity in vitro and tissue distribution in mice bearing melanoma tumors. Photochem Photobiol 1996;64:174–181. Jori G: Tumour photosensitizers: Approaches to enhance the selectivity and efficiency of photodynamic therapy. J Photochem Photobiol B 1996;36:87–93. Winkelman JW, Kimel S: The relationship between porphyrin transport and tissue uptake with photosensitization from 1O2; in Kessel D (ed): Photodynamic Therapy of Neoplastic Disease. Boca Raton, CRC Press, 1990, vol 2, pp 29–42. Boyle RW, Dolphin D: Structure and biodistribution relationships of photodynamic sensitizers. Photochem Photobiol 1996;64:469–485. Strauss WSL, Sailer R, Schneckenburger H, Akgu¨n N, Gottfried V, Chetwer L, Kimel S: Photodynamic efficacy of naturally occurring porphyrins in endothelial cells in-vitro and microvasculature in-vivo. J Photochem Photobiol B 1997;39:176–184. Paquette B, Ali H, Langlois R, van Lier JE: Biological activities of phthalocyanines. VIII. Cellular distribution in V-79 Chinese hamster cells and phototoxicity of selectively sulfonated aluminum phthalocyanines. Photochem Photobiol 1988;47:215–220. Mazie`re JC, Morlie`re P, Santus R: The role of the low density lipoprotein receptor pathway in the delivery of lipophilic photosensitizers in the photodynamic therapy of tumours. J Photochem Photobiol B 1991;8:351–360. Reddi E: Role of delivery vehicles for photosensitizers in the photodynamic therapy of tumours. J Photochem Photobiol B 1997;37:189–195. Pottier R, Kennedy JC: The possible role of ionic species in selective biodistribution of photochemotherapeutic agents toward neoplastic tissue. J Photochem Photobiol B 1990;8:1–16. Brault D: Physical chemistry of porphyrins and their interaction with membranes: The importance of pH. J Photochem Photobiol B 1990;6:79–86. Brault D, Vever-Bizet C, Kuzelova K: Interactions of dicarboxylic porphyrins with membranes in relation to their ionic state. J Photochem Photobiol B 1993;20:191–195. Winkelman JW, Arad D, Kimel S: Stereochemical factors in the transport and binding of photosensitizers in biological systems and in photodynamic therapy. J Photochem Photobiol B 1993;18: 181–189. Peng Q, Moan J: Correlation of distribution of sulphonated phthalocyanines with their photodynamic effect in tumour and skin of mice bearing CaD2 mammary carcinoma. Br J Cancer 1995; 72:565–574. Margaron P, Gregoir MJ, Scasnar V, Ali H, van Lier JE: Structure-photodynamic activity relationships of series of 4-substituted zinc phthalocyanines. Photochem Photobiol 1996;63:217–223. Chan WS, Marshall JF, Svensen M, Bedwell J, Hart IR: Effect of sulfonation on cell and tissue distribution of the photosensitizer aluminum phthalocyanine. Cancer Res 1990;50:4533–4538.


35

56

57

58 59

60 61 62

63 64

65 66

67 68

69 70 71

72 73

74 75 76

77

Van Leengoed HLLM, Van der Veen N, Versteeg AAC, Ouellet R, van Lier JE, Star WM: In vivo fluorescence kinetics of phthalocyanines in a skin-fold observation chamber model: Role of central metal ion and degree of sulfonation. Photochem Photobiol 1993;58:233–237. Hornung R, Jentsch B, Crompton NE, Haller U, Walt H: In vitro effects and localisation of the photosensitizers m-THPC and m-THPC MD on carcinoma cells of the human breast (MCF-7) and Chinese hamster fibroblasts (V-79). Lasers Surg Med 1997;20:443–450. Waluk J, Mu¨ller M, Swiderek P, Ko¨cher M, Vogel E, Hohlneicher G, Michl J: Electronic states of porphycenes. J Am Chem Soc 1991;113:5511–5527. Braslavsky SE, Mu¨ller M, Ma´rtire DO, Po¨rting S, Bertolotti SG, Chakravorti S, Koç-Weier G, Knipp B, Schaffner K: Photophysical properties of porphycene derivatives (18p porphyrinoids). J Photochem Photobiol B 1997;40:191–198. Abels C, Szeimies RM, Steinbach P, Richert C, Goetz AE: Targeting of the tumor microcirculation by photodynamic therapy with a synthetic porphycene. J Photochem Photobiol B 1997;40:305–312. Toledano H, Edrei R, Kimel S: Photodynamic damage by liposome-bound porphycenes: Comparison between in vitro and in vivo models. J Photochem Photobiol B 1998;42:20–27. Segalla A, Fedelli F, Reddi E, Jori G, Cross A: Effect of chemical structure and hydrophobicity on the pharmacokinetic properties of porphycenes in tumour-bearing mice. Int J Cancer 1997;72: 329–336. Gottfried V, Davidi R, Averbuj C, Kimel S: In-vivo damage to chorioallantoic membrane blood vessels by porphycene-induced photodynamic therapy. J Photochem Photobiol B 1995;30:115–121. Sokolov VV, Stranadko EF, Zharkova NN, Iakubovskaia RI, Filonenko EV, Astrakhankina TA: The photodynamic therapy of malignant tumors in basic sites with the preparations photohem and photosens (the results of 3 years of observations) (in Russian). Vopr Onkol 1995;41:134–138. Edrei R, Gottfried V, van Lier JE, Kimel S: Sulfonated phthalocyanines: Photophysical properties, in vitro cell uptake and structure activity relationships. J Porphyrins Phthalocyanines 1998;2:191–199. Ali H, Langlois R, Wagner JR, Brasseur N, Paquette B, van Lier JE: Biological activities of phthalocyanines. X. Syntheses and analyses of sulfonated phthalocyanines. Photochem Photobiol 1988;47:713–717. Erdrei R, Kimel S: Oxygen depletion during in vitro photodynamic therapy: Structure activity relationships of sulfonated Al phthalocyanines. J Photochem Photobiol B 1999;50:197–203. Dhami S, Rumbles G, MacRobert AJ, Phillips D: Comparative photophysical study of disulfonated aluminum phthalocyanine in unilamellar vesicles and leukemic K562 cells. Photochem Photobiol 1977;65:85–90. Streckyte G, Rotomskis R: Phototransformations of porphyrins in aqueous and micellar media. J Photochem Photobiol B 1993;18:259–263. Ribo JM, Crusats J, Farrera JA, Valero ML: Aggregation in water solutions of tetrasodium diprotonated meso-tetrakis(4-sulfonatophenyl)porphyrin. J Chem Soc Chem Commun 1994:681–682. Lavie D, Freeman D, Bock H, Fleischer J, Van Kranenburg K, lttah Y, Mazur Y, Lavie G, Liebes L, Meruelo D: Hypericin, a potential anti-AIDS drug. Trends in Medicinal Chemistry. Proc 11th Int Symp Med Chem, 1990, pp 321–327. Lavie G, Mazur Y, Lavie D, Prince AM, Pascual D, Liebes L, Levin B, Meruelo D: Hypericin as an inactivator of infectious viruses in blood products. Transfusion 1995;35:392–400. Etzlstorfer C, Falk H, Mueller N, Schmitzberger W, Wagner UG: Tautomerism and stereochemistry of hypericin: Force field, NMR and X-ray crystallographic investigation. Monatsh Chem 1993;124: 751–761. Racinet H, Jardon P, Guatron R: Formation d’oxyge`ne singulet, oxyge`ne photosensibiliseé par l’hypericine, e´tude cine´tique en milieu micellaire non ionique. J Chim Phys 1988;85:971–977. Thomas C, MacGill RS, Miller GC, Pardini RS: Photoactivation of hypericin generates singlet oxygen in mitochondria and inhibits succinoxidase. Photochem Photobiol 1992;55:47–53. Hadjur C, Jeunet A, Jardon P: Photosensitization by hypericin: ESR evidence for singlet oxygen and superoxide anion radicals formation in an in vitro model. J Photochem Photobiol B 1994;26: 67–74. Weiner L, Mazur Y: EPR studies of hypericin. Photogeneration of free radicals and superoxide. J Chem Soc Perkin Trans 1992;2:1439–1442.


36

78

79

80

81

82 83 84

85 86 87

88 89

90 91

92 93

94 95

96

97 98

Diwu Z, Lown JW: Photosensitization with anticancer agents. 17. EPR studies of photodynamic action of hypericin: Formation of semiquinone radical and activated oxygen species on illumination. Free Radical Biol Med 1993;14:209–215. Meruelo D, Degar S, Amari N, Mazur Y, Lavie D, Levin B, Lavie G: Mode of action of hypericin as an antiretroviral agent and other relevant findings; in Chu CK, Cutler H (eds): Natural Products as Antiviral Agents. New York, Plenum Press 1992, pp 91–119. Meruelo D, Lavie G, Lavie D: Therapeutic agents with dramatic antiretroviral activity and little toxicity at effective doses: Aromatic polycyclic diones hypericin and pseudohypericin. Proc Natl Acad Sci USA 1988;85:5230–5234. Lavie G, Valentine F, Levin B, Mazur Y, Gallo G, Lavie D, Weiner D, Meruelo D: Studies of the mechanisms of action of the antiretroviral agents hypericin and pseudohypericin. Proc Natl Acad Sci USA 1989;86:5963–5967. Tang J, Colacino JM, Larsen SH, Spitzer W: Virucidal activity of hypericin against enveloped and non-enveloped DNA and RNA viruses. Antiviral Res 1990;13:313–326. Degar S, Lavie G, Meruelo D: Photodynamic inactivation of radiation leukemia virus produced from hypericin treated cells. Virology 1993;197:796–800. Vandenbogaerde AL, Cuveele JF, Proot P, Himpens BE, Merlevede WJ, de Witte PA: Differential cytotoxic effects induced after photosensitization by hypericin. J Photochem Photobiol B 1997;38: 136–142. Hadjur C, Richard MJ, Parat MO, Favier A, Jardon P: Photodynamically induced cytotoxicity of hypericin dye on human fibroblast cell line MRC5. J Photochem Photobiol B 1995;27:139–146. Thomas C, Pardini RS: Oxygen dependence of hypericin-induced phototoxicity to EMT6 mouse mammary carcinoma cells. Photochem Photobiol 1992;55:831–837. Zhang W, Lawa RE, Hintona DR, Su Y, Couldwell WT: Growth inhibition and apoptosis in human neuroblastoma SK-N-SH cells induced by hypericin, a potent inhibitor of protein kinase C. Cancer Lett 1995;96:31–35. Kluck RM, Bossy-Wetzel E, Green DR, Newmeyer DD: The release of cytochrome c from mitochondria: A primary site for Bcl-2 regulation of apoptosis. Science 1997;275:1132–1136. Yang J, Liu X, Bhalla K, Kirn CN, Ibrado AM, Cai J, Peng TI, Jones DP, Wang X: Prevention of apoptosis by Bcl-2: Release of cytochrome c from mitochondria blocked. Science 1997;275: 1129–1132. Noodt BB, Berg K, Stokke T, Peng Q, Nesland JM: Apoptosis and necrosis induced with light and 5-aminolevulinic acid-derived protoporphyrin IX. Br J Cancer 1996;74:22–29. Zhou C, Shunji C, Jinsheng D, Junlin L, Jori G, Milanesi C: Apoptosis of mouse MS-2 fibrosarcoma cells induced by photodynamic therapy with Zn(II)-phthalocyanine. J Photochem Photobiol B 1996; 33:219–223. Luo Y, Kessel D: Initiation of apoptosis versus necrosis by photodynamic therapy with chloroaluminum phthalocyanine. Photochem Photobiol 1997;66:479–483. Chung PS, Saxton RE, Paiva MB, Rhee CK, Soudant J, Mathey A, Foote C, Castro DJ: Hypericin uptake in rabbits and nude mice transplanted with human squamous cell carcinomas: Study of a new sensitizer for laser phototherapy. Laryngoscope 1994;104:1471–1476. Utsumi T, Okuma M, Kanno T, Takehara Y, Yoshioka T, Fujita Y, Horton AA, Utsumi K: Effect of the antiretroviral agent hypericin on rat liver mitochondria. Biochem Pharmacol 1995;50:655–662. Takahashi I, Nakanishi S, Kobayashi E, Nakano H, Suzuki K, Tamaoki T: Hypericin and pseudohypericin specifically inhibit protein kinase C: Possible relation to their antiretroviral activity. Biochem Biophys Res Commun 1989;165:1207–1212. Gopalakrishna R, Chen Z, Gundimeda U: Light and dark reactions of calphostin C and hypericin induce irreversible inactivation of protein kinase C by site-specific oxidations. Free Radical Biol Med 1993;15:530. Anker L, Gopalakrishna R, Jones KD: Hypericin in adjuvant brain tumor therapy. Drugs Future 1995;20:511–517. Agostinis P, Vandenbogaerde AL, Donella-Deana A, Pinna LA, Lee K, Goris J, Merlevede W, Vandenheede JR, de Witte PA: Photosensitized inhibition of growth factor-regulated protein kinases by hypericin. Biochem Pharmacol 1995;49:1615–1622.


37

99

100 101 102 103 104 105 106 107

Agostinis P, Donella-Deana A, Cuveele JF, Vandenbogaerde AL, Sarna S, Merlevede WJ, De Witte PA: A comparative analysis of the photosensitized inhibition of growth-factor regulated protein kinases by hypericin derivatives. Biochem Biophys Res Commun 1996;220:613–617. Cohen DI, Tani Y, Tian H, Boone E, Samuelson LE, Lane HC: Participation of tyrosine phosphorylation in the cytopathic effect of human immunodeficiency virus-1. Science 1992;256:542–545. Fields AP, Bednarik DP, Hess A, May WS: Human immunodeficiency virus induces phosphorylation of its cell surface receptor. Nature 1988;333:278–280. Lavie G, Mazur Y, Lavie D, Meruelo D: The chemical and biological properties of hypericin – A compound with a broad spectrum of biological activities, Med Res Rev 1995;15:111–119. Hudson JB, Lopez-Bazzocchi I, Towers GHN: Antiviral activities of hypericin. Antiviral Res 1991; 15:101–112. Diwu Z, Lown JW: Hypocrellins and their uses in photosensitization. Photochem Photobiol 1990; 52:609–616. Diwu Z: Novel therapeutic and diagnostic applications of hypocrellins and hypericins. Photochem Photobiol 1995;61:529–539. Yamazaki T, Ohta N, Yamazaki I, Song PS: Excited-state properties of hypericin-electronic spectra and fluorescence decay kinetics. J Phys Chem 1993;97:7870–7875. Diwu Z, Zimmermann J, Meyer T, Lown JW: Design, synthesis and investigation of mechanisms of action of novel protein kinase C inhibitors: Perylenequinonoid pigments. Biochem Pharmacol 1994;47:373–385.

Prof. Dr. Sol Kimel, Department of Chemistry, Technion-Israel Institute of Technology, Haifa 32000 (Israel) Tel. +972 4 829 3946, Fax +972 4 823 3735, E-Mail [email protected]


38


............................

Localization and Dynamics of Photosensitizers Angelika Ru¨ck, Nermin Akgu¨n, Klaus Heckelsmiller, Gerd Beck, Karin Kunzi-Rapp Institut fu¨r Lasertechnologien in der Medizin und Messtechnik, Ulm, Deutschland

Introduction The medical interest in photodynamic therapy (PDT) of malignant and nonmalignant diseases has increased in the past 15 years [1]. The biochemical background of the photochemical reaction induced during PDT has, however, only partly been resolved. The biodynamics, which has been shown to be different for lipophilic and hydrophilic photosensitizers, influences the photodynamic efficiency. A primary phototoxic effect is observed when tumor cells are directly affected. In this case, sufficient oxygen supply is important. Therefore vessels should not be damaged. In contrast vascular damage leads to partial or complete occlusion of tumor vascular supply, inducing a secondary phototoxicity. Possible targets for vascular damage are endothelial cells as well as blood cells [2–4]. Primary and secondary phototoxicity is dependent on the accumulation of the photosensitizer in tumor tissue. These different aspects are summarized in figure 1. On a cellular level, the efficacy and mechanism of the therapy are influenced by uptake and retention of the sensitizing compound by the target cells and the pattern of its subsequent intracellular localization both before and during irradiation [5]. Relocation of sensitizers is observed not only during the incubation time but sometimes during PDT treatment in correlation with dynamic fluorescence changes [6]. While small lipophilic molecules like protoporphyrin penetrate the plasma membrane by diffusion and may be retained there, an inhomogeneous subcellular distribution in extranuclear granules is observed for most of the hydrophilic photosensitizers like meso-tetra(4-sulfonatophenyl) porphyrin and Al(III)phthalocyanine tetrasulfonate (AlPcS4) [6]

Fig. 1. Accumulation of photosensitizers in tumor tissue.

which are taken up by pinocytosis, and for lipophilic dyes which are administered via the LDL-endocytotic pathway. In cells of high mitotic activity exhibiting a high lysosomal proton concentration, lysosomes can act as a trap for drugs [8, 9] such as the cationic sensitizers Nile blue derivative or methylene blue (MB+) [10, 11]. In addition, cationic drugs, such as rhodamines, cyanines, triarylmethanes, oxazines and thiazines like MB+ have received attention because of their selective uptake by mitochondria of carcinoma cells [12]. Uptake and localization of the different types of photosensitizers are summarized in table 1. The cytotoxic effect during PDT after light activation of a photosensitizer is regarded as being partly related to the in situ generation of reactive oxygen species (ROS), e.g. singlet oxygen (1O2) [13] via energy transfer or superoxide anions [14], which could induce local changes in the redox state of the cell. Recent observations indicate that signal transduction pathways are involved in cellular damage and cellular rescue effects after PDT treatment [15]. An increase in intracellular-free calcium is thought to trigger some kind of cellular rescue response [15], whereas a transient increase in cytoplasmatic-free calcium

Ru¨ck/Akgu¨n/Heckelsmiller/Beck/Kunzi-Rapp

40

Table 1. Uptake and localization of sensitizers Sensitizer

Uptake

Localization

Hydrophilic Negatively charged

pinocytosis

extranuclear granules: endosomes, lysosomes ER, Golgi apparatus lysosomes mitochondria

Positively charged weak bases Hydrophobic

pinocytosis diffusion LDL endocytosis

membranes extranuclear granules

was observed to contribute to photodynamic cell killing [16]. It is suggested that prolonged high intranuclear calcium levels activate programmed cell death or apoptosis [17, 18]. In addition, ROS are thought to be involved under certain conditions in signalling pathways leading to either programmed cell death or stimulation [19, 20]. In this chapter, the biodynamics with respect to application time and irradiation of hydrophilic and lipophilic photosensitizers in vivo as well as in cell cultures will be summarized. Most of the work so far was done with clinically important natural porphyrins, hematoporphyrin and protoporphyrin. During light exposure these porphyrins are degradated, a phenomenon which is called photobleaching. During this, fluorescent photoproducts develop due to reaction of 1O2 with the porphyrin molecule [21, 22], which again can act as photosensitizers. In summary, the light-induced dynamics of photosensitizers can be quite complex. Relocation, photobleaching, photoproducts or formation of protein-binding complexes can be observed, depending on localization of the drug. It will be demonstrated that in some special cases the light-induced dynamics of the sensitizers was correlated with local changes in cellular signals – redox potential, calcium concentration or pH value – which were important findings for an understanding of the PDT mechanism.

Experimental Procedure Microscopic Techniques The localization and dynamics of photosensitizers on a cellular level as well as in complex in vivo systems can be studied by high resolution light microscopy as well as time-resolved microspectrofluorometry. Confocal techniques are useful when a subcellular resolution is demanded. In our case, a laser scanning microscope (LSM 410 invert, Zeiss,

Localization and Dynamics of Photosensitizers

41

Germany) gave promising results. Cells and tissue samples were irradiated with 220 lW at 633 nm (internal HeNe laser of the microscope). The fluorescence of the sensitizer was detected above 665 nm (dichroic mirror FT 655, long pass filter RG 665) and visualized in the red channel of the LSM (fluorescence detector PMT1, pinhole in front of PMT1 was set to 50). Synchronously, cells were observed in phase contrast in the green channel (transmitted-light detector). The time-dependent fluorescence intensity of the sensitizer during continuous illumination was analyzed with Zeiss standard evaluation software ‘time series’ and was observed within defined regions of interest. Images were acquired at 0.5-second intervals (frame rate 2 images/s, scanned field 160¶40 lm). In order to get the same irradiation dose for all independent experiments, the magnification (objective lens and zoom factor) and the scan field were kept constant (objective lens 40¶, NA 0.75, zoom factor 2¶, scanned field 160¶40 lm). In a separate experiment the kinetics of the photosensitizers were correlated with the dynamics of cellular signals. In this case, the fluorescence of the drugs was again excited at 633 nm and detected above 665 nm. Synchronously, the fluorescence of ion-sensitive probes (Fluo-3 to detect Ca2+ and BCECF-AM to measure the pH) was excited at 488 nm (internal Ar+ laser) and observed between 515 and 565 nm (interference filter). In addition, cells were observed in phase contrast (transmitted light detector), thus morphological changes could be correlated with the subcellular dynamics. Cell Cultures BKEz-7 endothelial cells derived from calf aorta or RR1022 endothelial cells from the rat were cultured in minimal essential medium (Sigma) or medium M199, respectively, supplemented with 10% fetal calf serum and antibiotics at 37 ºC and 5% CO2. For microscopic observation, the cells were seeded on microscope slides 36 h before incubation at a density of 100 cells/mm2 in the case of BKEz-7 cells and 25 cells/mm2 in the case of RR1022 cells. Cells were coincubated in the exponential growth phase with the photosensitizers and the Ca2+-sensitive probe Fluo-3 (Molecular Probes) or the H+-sensitive probe BCECF-AM (Molecular Probes) for pH measurements. The following photosensitizers were used: AlPcS4 (Cancer Research Institute, Moscow) which was incubated for 24 h at a concentration of 10 lM, zinc phthalocyanine (ZnPc) in a liposomal formulation (CGP 55847, Novartis, Basel, Switzerland) at a concentration of 1 lM for 1 h and MB+ from Merck (Germany) at a concentration of 1 lM and incubated for 4 h. In the case of Ca2+ detection, cells were also incubated for 20 min with 1 lM Fluo-3. Measurements of pH changes were performed by BCECF-AM, incubated at a concentration of 1 lM 30 min prior to irradiation. Prior to observation of light-induced modifications, cells were washed 3 times with isotonic NaCl solution. Every experiment was performed with a fresh microscopic slide. In vivo System The dynamic behavior of lipophilic and hydrophilic photosensitizers in endothelial cells, the lumen of the vessels and tumor cells was observed with an appropriate in vivo system. The chorioallantoic membrane (CAM) of fertilized eggs served as a substrate for tumor cells. The localization of the drugs in the lumen of the vessels, the endothelial cells and the tumor cells after different application times was observed with the confocal LSM. The method established by us has been described [23, 24]. Our technique enables online observation of photosensitizer localization in different cellular systems. On day 5 of hatching of fertilized


42

hen eggs, tumor cells were seeded on the CAM by dropping 25 ll of a cell suspension (5¶105 cells) into a small silicone ring placed on the CAM. 3 days later, the photosensitizers were applied intravesically and the fluorescence in the vessels and the tumor cells observed after different application times. In order to perform fluorescence microscopy, the CAM was prepared in a perfusion chamber, as previously described [24]. AY 27 cells from a carcinogeninduced bladder carcinoma of Fischer CDF rats were cultured in RPMI 1640 medium (Gibco BRL, Life Technologies, Eggenstein, Germany) supplemented with 10% fetal calf serum and antibiotics at 37 ºC and 5% CO2. 25 ll of a cell suspension containing 5¶105 cells were seeded on the CAM as described above. Hydrophilic AlPcS4 and lipophilic liposomally delivered ZnPc stock solutions of the sensitizers were made up in phosphate-buffered saline and sterilized by filtration using a 0.2-lm filter. AlPcS4 and ZnPc were injected into the veins of the CAM with a 30-gauge needle. AlPcS4 was given at a dose of 0.5 mg/ml with a total volume of 0.1 ml. This corresponds to a concentration of 1.0 mg/kg body weight. ZnPc was given at a dose of 0.1 mg/ml, again with a total volume of 0.1 ml, which corresponds to a concentration of 0.2 mg/kg body weight. The fluorescence of the photosensitizers was quantified after different application times with the LSM in defined areas. For every application time, 6 eggs were measured at least at 3 different regions of interest.

Results and Discussion Accumulation of Hydrophilic and Lipophilic Phthalocyanines in the Vascular System and Tumor Cell System The accumulation of hydrophilic and lipophilic phthalocyanines after various application times in the lumen of the vessels, the endothelial cells and the tumor cells was observed with a confocal LSM, using the prepared CAM system as described above. The time-dependent accumulation of hydrophilic AlPcS4 in the vessels and the tumor cells is demonstrated in figure 2. After intravenous application of the drug, the fluorescence in the vessels was high in the first 30 min, dropped down and led to a second maximum 4 h after drug application. Interestingly, the fluorescence in the tumor cells increased rapidly immediately after drug application with a maximum at around 10 min and a second maximum at 3 h. After a longer application time, the fluorescence in the vessels dropped, and 48 h after injection the vascular system showed only negligible fluorescence (data not shown). It has to be mentioned that the fluorescence in the vascular system was seen in the lumen of the vessels and probably not in the endothelial cells. Although a high fluorescence was observed in the tumor cells of the CAM system, comparable treatments in a rat bladder tumor model induced only partial necrosis during AlPcS4 PDT [6]. In the case of lipophilic ZnPc, significant tumor necrosis was observed with the same rat bladder model [6]. Therefore, we analyzed the time-dependent


43

Fig. 2. Pharmacokinetics of AlPcS4 in tumor cells and vessels after intravenous application.

Fig. 3. Pharmacokinetics of ZnPc in tumor cells and vessels after intravenous application.

localization of ZnPc in the different cellular systems of the CAM model after intravenous application. As demonstrated in figure 3, the fluorescence in the vessels was high all the time, even 24 h after injection. After a short incubation time, fluorescence was observed in the lumen of the vessels, mainly in erythrocytes. 2 h after application, ZnPc could be detected in the endothelial vessel wall and was maintained for at least 4 h. The fluorescence in the tumor cells slowly increased and reached a maximum at 24 h. Summarizing the pharmacodynamic studies of ZnPc, the accumulation in the vessels was high all the time. The main difference between ZnPc and AlPcS4 is that AlPcS4 is


44

cleared from the plasma more quickly than ZnPc. AlPcS4 seems to have a tendency to accumulate faster in the tumor cells, whereas ZnPc is found in a higher amount in the endothelial cells of the vessels. Our results confirm reports in the literature that tumor control by PDT with liposomal ZnPc requires vascular damage, the endothelium is probably a target as well as the tumor cells [25]. These two examples demonstrate in which way primary and secondary phototoxicity of hydrophilic and lipophilic photosensitizers can be modulated. An accumulation in the endothelial cells as well as tumor cells as observed for ZnPc explains the observed secondary and primary phototoxic effects. In the case of AlPcS4 however, direct tumor cell killing mechanisms are probably more important and should therefore be optimized. Light-Induced Subcellular Dynamics of Hydrophilic and Lipophilic Phthalocyanines Besides the observed differences concerning the biodynamics of AlPcS4 and ZnPc in the vascular system and tumor cell system the light-induced subcellular processes of the two photosensitizers differed in a significant way. As was reported recently by us, the fluorescence of AlPcS4 increased quickly in the beginning of the process in cell cultures as well as in vivo, whereas the fluorescence of ZnPc decreased exponentially [6]. Photobleaching of ZnPc was correlated with an efficient tumor response to PDT. It seems that photobleaching which involves singlet oxygen is the main process of lipophilic photosensitizers which accumulate in cellular membrane systems. Necrotic processes are probably the consequence, whereas hydrophilic photosensitizers act in a different way. This hypothesis was further supported by the different Ca2+ dynamics which was induced during PDT. In the case of AlPcS4, synchronous to the significant fluorescence increase in the photosensitizer, we observed nonlinear Ca2+ dynamics inside the cell nucleus during the light-induced process which was correlated with apoptotic processes (in preparation). The kinetics of this process, which was analyzed with LSM, is demonstrated in figure 4. Fluorescence changes in AlPcS4 were observed in the cytoplasm, located near the nucleus which probably coincides with the endoplasmatic reticulum. The fluorescence of AlPcS4 increased during irradiation in a sigmoid-shaped behavior. Synchronously, the Ca2+ concentration in the cell nucleus increased significantly in a nonlinear fashion, which was indicated by a fluorescence intensity increase by a factor of 3 for the Ca2+-sensitive probe Fluo-3. The onset of the Ca2+ increase started between 5 and 10 s before the AlPcS4 increase. Immediately afterwards, the Ca2+ concentration decreased followed by a second and perhaps a third increase in oscillatory behavior. Morphological changes in the cells at the time of fluorescence intensity increase


45

Fig. 4. Subcellular kinetics of AlPcS4 and Ca2+ in RR 1022 cells during irradiation with 633 nm.

were not observed. Damage at the plasma and nuclear membrane level could be observed during the ‘bleaching’ period. When we correlated the dose applied during LSM with the dose used to determine the quantitative phototoxicity, a light dose which led to the first Ca2+ increase stimulated the apoptotic processes (in preparation). In contrast to AlPcS4, the fluorescence of ZnPc decreased exponentially during irradiation, no fluorescence increase could be observed. A nonlinear Ca2+ increase, as in the case of AlPcS4, could not be detected. This is demonstrated in figure 5. Morphological changes in the cells during the ‘bleaching’ process were significant. This correlates with the fact that ZnPc was a very efficient photosensitizer and induced no stimulation of cell proliferation at low light doses [26]. The failing of the nuclear Ca2+ increase in a nonlinear way seems to be the key for the different behavior. Because nuclear Ca2+ is known to increase in some cases during apoptosis, the different dynamics of hydrophilic and lipophilic sensitizers reported here led to the hypothesis that apoptotic and necrotic processes contribute differently for the two types of drugs. The efficiency of PDT is therefore not only determined by the biodynamics in vivo as stated above but also the light-induced dynamics on a subcellular level. Nonlinear Dynamics of MB+ during PDT During PDT with MB+, again highly nonlinear reactions were observed. The cationic MB+ can act as a photosensitizer as well as a redox indicator, therefore enabling the observation of oxidation during the light-induced process. The redox potential of MB+ is dependent on pH, favoring the reduction of MB+ at low pH. A value of –0.12 V vs. NHE at pH 7 is reported for the


46

Fig. 5. Subcellular kinetics of ZnPc and Ca2+ in RR 1022 cells during irradiation with 633 nm.

one-electron reduction [27]. Redox potentials of 1.55 and 1.21 V vs. NHE were calculated for the one-electron reduction of excited singlet and triplet MB+ intercalated in polynucleotides [27]. Thus, the ability of MB+ to act as an oxidizing agent should be enhanced when excited with light. Moreover, as discussed by Atherton and Harriman [27], photo-induced electron transfer would become thermodynamically more favored if accompanied by proton transfer. This is important because MB+ is accumulated in a high amount in acidic organelles. MB+, which was successfully used as a photosensitizer in the intraluminal treatment of inoperable esophageal tumors [28] and in the topical PDT treatment of psoriasis [29], is a potential photonuclease, possibly due to rapid electron transfer from guanine and adenine to intercalated MB+ in its excited singlet state [27, 30]. However, it is also suggested that MB+ mediates cell cytotoxicity via the generation of hydroxyl radicals which change the intracellular Ca2+ homeostatic mechanisms [31]. Moreover, MB+ is well known to generate 1O2 [32]. Recently, the subcellular fluorescence of MB+ during irradiation was studied and analyzed by us using LSM [33]. The dynamic behavior, which appeared to be highly nonlinear, was correlated with the phototoxic response of the treated cells. The induction of spatiotemporal inhomogeneities suggests an abrupt oxidation of the cells during the light-induced process [33]. The fluorescence dynamics of MB+ during irradiation with 633 nm light was observed with subcellular resolution in BKEz-7 cells as described above.


47

Fig. 6. Intracellular fluorescence dynamics of MB+ in a defined region of interest during irradiation with 633 nm light. Curves represent three independent single cells.

Images were acquired at 0.5-second intervals. The fluorescence was observed in the red channel of the LSM. Synchronously, the phase contrast image was visualized with the green channel. Morphological changes could therefore be correlated with the dynamics of MB+. In addition, the light dose-dependent phototoxicity at 633 nm irradiation was determined by viable cell counting. The subcellular dynamics of MB+ [33] is shown for three independent cells in figure 6. After an induction period (phase I), fast fluorescent spikes could be observed in the whole cytoplasm, which decayed with a time constant of about 20 s (phase II), followed by a period of nearly constant fluorescence intensity (phase III) and exponential photobleaching (phase IV). Phase II exhibits highly nonlinear kinetics which probably correlates with a nonlinear ‘quantal’ production of ROS. Morphological cell changes were not observed during phase II. During phase III, a pycnotic cell nucleus developed. A gallery of images demonstrating these temporal and spatial inhomogeneities is shown in figure 7. Abrupt fluorescence changes like spikes were observed at different sites in the cytoplasm. Continued irradiation of the cell finally induced diffuse fluorescence enhancement and a bright MB+ fluorescence in the nucleus and the nucleoli. After the period of spikes (phase III, fig. 6) first a pycnotic cell nucleus and later blebs at the nuclear and plasma membranes developed (arrow


48

Fig. 7. Selected images of BKEz-7 cells incubated for 4 h with 1 lM MB+ after various irradiation times. Spatiotemporal inhomogeneities of MB+ were obtained during permanent illumination with 633 nm.

in 735.7 s, fig. 7). From the determination of viable cells we could conclude that a light dose applied within phase II was only sublethal in correlation with morphological observations. In a further experiment, we correlated the ‘quantal’ dynamics of MB+ with local pH changes. For this reason, the cells were coincubated with MB+ and the H+-sensitive probe BCECF-AM. Most interestingly, the pH in the cytoplasm decreased locally in a nonlinear ‘quantal’ behavior (pH spikes), which was synchronized with the MB+ spikes. The pH in the nucleus rapidly decreased and reached the same value as in the cytoplasm. After the last MB+ spike which correlated with a drastic decrease in the nucleolar pH, morphological changes were observed in the cell, indicating cell death (data not shown). From our experiments it is evident that light activation of MB+ induces local nonlinear changes (spikes) in the redox potential as well as changes in the pH value inside the cells. We could further show that these local changes did not kill the cells. However, a light dose which led to a permanent oxidation of the cells was correlated with ROS overproduction which finally killed the


49

cells. In summary, as discussed for AlPcS4 and ZnPc, nonlinear changes in photosensitizer fluorescence and cellular signals can be observed during light activation. It seems that these reactions are typical for hydrophilic photosensitizers which are accumulated inside cellular organelles.

Conclusion Accumulation and dynamics of photosensitizers determine the efficiency of PDT. In general, lipophilic drugs seem to induce primary as well as secondary phototoxic effects as a consequence of accumulation in the endothelial vessel wall as well as tumor cells. In contrast, accumulation of hydrophilic drugs in the endothelial vessel wall is negligible. For these types of sensitizers, primary phototoxicity must be optimized by recognizing the biodynamic behavior. Also the light-induced subcellular dynamics immediately during PDT treatment differ for hydrophilic and lipophilic drugs in a significant way. While photobleaching is observed for membrane-bound lipophilic molecules, the nonlinear dynamics of drugs and cellular signals are induced at the beginning of irradiation with hydrophilic photosensitizers which accumulate inside cytoplasmatic organelles. However, a light dose, which induces nonlinear reactions, did not kill the cells sufficiently. Moreover, it seems that photobleaching, which is normally correlated with singlet oxygen formation, is predominantly observed in cell membranes and leads to a more effective photodynamic treatment.

Acknowledgments This work was supported by BMBF grant 13N6293/1 and BMBF grant IZKF Ulm.

References 1 2 3 4 5

Fisher AMR, Murphree AL, Gomer CJ: Clinical and preclinical photodynamic therapy. Lasers Surg Med 1995;17:2–31. Milanesi C, Biolo R, Reddi E, Jori G: Ultrastructural studies on the mechanism of the photodynamic therapy of tumors. Photochem Photobiol 1987;46:675–681. He DP, Hampton JA, Keck R, Selman SH: Photodynamic therapy: Effect on the endothelial cell of the rat aorta. Photochem Photobiol 1991;54:801–804. Lev B, Hanania J, Malik Z: Morphological deformations of erythrocytes induced by hematoporphyrin and light. Lasers Life Sci 1993;5:219–230. Ru¨ck A, Diddens H: Uptake and subcellular distribution of photosensitizing drugs in malignant cells; in Ho¨nigsmann H, Jori G, Young AR (eds): The fundamental bases of phototherapy. Milan, OEMF, 1996, pp 209–227.


50

6

7

8

9 10 11

12 13 14 15 16 17 18 19 20 21 22

23 24 25

26 27 28 29

Ru¨ck A, Beck G, Bachor R, Akgu¨n N, Gschwend MH, Steiner R: Dynamic fluorescence changes during photodynamic therapy in vivo and in vitro of hydrophilic Al(III) phthalocyanine tetrasulfonate and lipophilic Zn(II) phthalocyanine administered in liposomes. J Photochem Photobiol B Biol 1996;36:127–133. Strauss WSL, Gschwend MH, Sailer R, Schneckenburger H, Steiner R, Ru¨ck A: Intracellular fluorescence behaviour of meso-tetra(4-sulphonatophenyl)porphyrin during photodynamic treatment at various growth phases of culltured cells. J Photochem Photobiol B Biol 1995;28:155– 161. Geze M, Morliere P, Maziere JC, Smith KM, Santus R: Lysosomes as key target of hydrophobic photosensitizers proposed for photochemotherapeutic applications. J Photochem Photobiol B Biol 1993;20:23–35. Berg K, Moan J: Lysosomes as photochemical targets. Int J Cancer 1994;59:1–9. Lin C-W, Shulok JR, Kirley SD, Cincotta L, Foley JW: Lysosomal localization and mechanism of uptake of Nile blue photosensitizers in human tumor cells. Cancer Res 1991;51:2710–2719. Ru¨ck A, Ko¨llner T, Dietrich A, Strauss W, Schneckenburger H: Fluorescence formation during photodynamic therapy in the nucleus of cells incubated with cationic and anionic water-soluble photosensitizers. J Photochem Photobiol B Biol 1992;12:403–412. Oseroff AR, Ohuoha D, Ara G, McAuliffe D, Foley J, Cincotta L: Intramitochondrial dyes allow selective in vitro photolysis of carcinoma cells. Proc Natl Acad Sci USA 1986;83:9729–9733. Weishaupt K, Gomer CJ, Dougherty T: Identification of singlet oxygen as the cytotoxic agent in photoinactivation of a murine tumor. Cancer Res 1976;36:2326–2329. Buettner GR, Need MJ: Hydrogen peroxide and hydroxyl free radical production by hematoporphyrin derivative, ascorbate and light. Cancer Lett 1985;25:297–304. Penning LC, Dubbelman TMAR: Fundamentals of photodynamic therapy: Cellular and biochemical aspects. Anticancer Drugs 1994;5:139–146. Ben-Hur E, Dubbelman TMAR: Cytoplasmic free calcium changes as a trigger mechanism in the response of cells to photosensitization. Photochem Photobiol 1993;6:890–894. Clapham DE: Calcium signaling. Cell 1995;80:259–268. Nicotera P, Zhivotovsky B, Orrenius S: Nuclear calcium transport and the role of calcium in apoptosis. Cell Calcium 1994;16:279–288. Jacobson MD: Reactive oxygen species and programmed cell death. Trends Biochem Sci 1996;21: 83–86. Friedmann H, Lubart R: Competition between activating and inhibitory processes in photobiology. SPIE 1995;2630:60–64. Ko¨nig K, Schneckenburger H, Ru¨ck R, Steiner R: In vivo photoproduct formation during PDT with ALA-induced endogenous porphyrins. J Photochem Photobiol B Biol 1993;18:287–290. Giniunas L, Rotomskis R, Smilgevicius V, Piskarskas A, Didziapetriene J, Bloznelyte L, Griciute L: Activity of haematoporphyrin derivative photoproduct in photodynamic therapy in vivo. Lasers Med Sci 1991;6:425–428. Kunzi-Rapp K, Akgu¨n N, Schneckenburger H, Ru¨ck A, Steiner R: In vivo uptake and biodistribution of lipophilic and hydrophilic photosensitizers. SPIE 1996;2924:176–180. Kunzi-Rapp K, Westphal-Fro¨sch C, Schneckenburger H: Test system for human tumor cell sensitivity to drugs on chicken chorioallantoic membranes. In vitro Cell Dev Biol 1992;28A:565–566. van Leengoed HLLM, Cuomo V, Versteeg AAC, van der Veen N, Jori G, Star WM: In vivo fluorescence and photodynamic activity of zinc phthalocyanine administered in liposomes. Br J Cancer 1994;69:840–845. Ru¨ck A, Akgu¨n N, Heckelsmiller K, Beck G, Kunzi-Rapp K: Dynamics of photosensitizers in cell cultures and non-animal in vivo systems. SPIE 1997;3191:10–14. Atherton SJ, Harriman A: Photochemistry of intercalated methylene blue: Photoinduced hydrogen atom abstraction from guanine and adenine. J Am Chem Soc 1993;115:1816–1822. Orth K, Ru¨ck A, Stanescu A, Beger HG: Intraluminal treatment of inoperable oesophageal tumours by intralesional photodynamic therapy with methylene blue. Lancet 1995;345:519–520. Schick E, Ru¨ck A, Boehncke W-H, Kaufmann R: Topical photodynamic therapy using methylene blue and 5-aminolevulinic acid in psoriasis. J Derm Treatment 1997;8:17–19.


51

30

31 32 33

Dunn DA, Vivian HL, Kochevar IE: The role of ground state complexation in the electron transfer quenching of methylene blue fluorescence by purine nucleotides. Photochem Photobiol 1991;53: 47–56. Lee YS, Wurster RD: Methylene blue induces cytotoxicity in human brain tumor cells. Cancer Lett 1995;88:141–145. Tuite EM, Kelly JM: Photochemical interactions of methylene blue and analogues with DNA and other biological substrates. J Photochem Photobiol 1993;21:103–124. Ru¨ck A, Heckelsmiller K, Akgu¨n N, Beck G, Kunzi-Rapp K, Schick E, Steiner R: Nonlinear dynamics of intracellular methylene blue during light activation of cell cultures. Photochem Photobiol 1997;66:837–841.

Angelika Ru¨ck, Institut fu¨r Lasertechnologien in der Medizin und Messtechnik, Helmholtzstrasse 12, D–89081 Ulm (Germany) Tel. +49 731 14 29 16, Fax +49 731 14 29 42, E-Mail [email protected]


52


............................

Light and Chemistry of Polyatomic Organic Molecules1 Paul Suppan † Institute of Physical Chemistry, University of Fribourg, Switzerland

Light, photons of electromagnetic radiations in the range of 200–1,000 nm (UV-VIS-NIR; fig. 1), is absorbed in photochemical reactions. It acts as a reactant, never as a catalyst. In a very general scheme (fig. 2) the energy of the photon becomes (at least partly) the energy of the excited molecule, M+hmaKM*. This molecule will then react either on its own (unimolecular processes M*KP) or with another molecule (M*+NKP). An excited molecule M* differs from the ground-state molecule M by the weaker bonds (e.g. free rotation of p bonds; fig. 3), by better electron acceptor and donor potentials, and by different charge distributions (fig. 4).

‘Dark’ Chemistry and Light-Induced Chemistry [1] Pathways of ‘Dark’ Reactions and Photochemical Reactions In the Arrhenius-Eyring model of a chemical reaction which takes place without the intervention of light, the reactant(s) R go over the product(s) P through a transition state (X) which determines the activation barrier Ea in the rate constant equation: k>A exp (–Ea /RT).

When the molecule is excited by light (fig. 5) to reach one of its electronically excited states (M*) it may undergo some other chemical reaction, leading 1

For a treatment of photochemical reactions see: Suppan P: Chemistry and Light. Cambridge, Royal Society of Chemistry, 1994.

Fig. 1. Energy levels of molecules and properties of electromagnetic radiation. The molecule M has low lying rotational states Mr and vibrational states Mm. Electronic excited states M* can be reached by absorption of light of wavelengths above 1,000 nm, in near infrared, visible (VIS) and ultraviolet (UV) spectra. At shorter wavelength ionization is the major process. A few examples of bond dissociation energies are given in kJ molÖ1.

to some high-energy product P through an activation barrier EaH. Both in the ground-state (dark) and excited-state (photoinduced) reactions the activation barriers (Ea and EaH, respectively) must be overcome from the thermal energy of the chemical system. For this reason, it may be somewhat misleading to define ground-state reactions as ‘thermal’ and excited-state reactions as ‘lightinduced’. The photochemical reaction is in fact the thermal reaction of the electronic excited state M* of the molecule M, while the ‘dark’ reaction of M is the thermal reaction of the ground state. Mistaken Concept of ‘Catalysis’ by Light In a photochemical reaction light always acts as a reactant, never as a catalyst. By definition, a catalyst in a chemical reaction must be recovered unchanged, and can in principle be reused again and again. In a photochemical reaction, light is absorbed and its energy is used to make electronically excited molecules; it will not come out and cannot be used after the reaction, so by definition light is not a catalyst and the concept of ‘light catalyzed’ reaction is fundamentally incorrect.

Suppan

54

Fig. 2. General reaction schemes of ground-state (M) and excited-state (M*) molecules. The ground-state M reacts through thermal activation (Ea ). The excited state(s) formed by absorption of light hma are finally the short-lived S1 state (Dns) and/or the long-lived T1 state (Dls) which can be formed only through intramolecular deactivation (d). In many cases the primary photochemical process (1) leads to high-energy products PPP (free radicals, ions, etc.) which lead to the final, stable products through secondary (dark) reactions (2).

Chemical reactions in thermal (‘dark’) chemistry originate from the ground state of a molecule, whereas photochemical reactions originate from its excited states. Photochemistry is therefore different from thermal chemistry because electronically excited molecules are different from ground-state molecules. (1) As a result of electronic excitation the molecule finds one of its electrons promoted to a higher energy orbital, leaving vacancy (a ‘positive hole’, so to speak) in the lower orbital. An excited molecule is therefore at the same time a stronger reducing agent (an electron donor) and a stronger oxidizing agent (an electron acceptor) than the ground-state molecule. Not surprisingly, we shall find that there are many redox photochemical processes which cannot occur in the dark, and the primary reaction of photosynthesis is one of these. (2) The promotion of an electron to an antibonding orbital means that one (or several) bonds are weakened; such bonds may dissociate, or may become very reactive. (3) The distribution of electrical charges can be very different (in particular as a result of charge transfer between various parts of the molecule) and this can result in quite different electrostatic activation barriers, e.g. towards ‘electrophilic’ reactions which involve the attack of negative-charge

Light and Chemistry of Polyatomic Organic Molecules

55

Fig. 3. Molecular orbital (MO) representations of ground- and excited-state molecules. Each bond is shown as an orbital occupied by two electrons, the corresponding antibonding orbital p* being vacant in M. In M* the p bond is broken and the molecule resembles a biradical. Oxidation M(*)KMf++eÖ and reduction M(*)+eÖKMfÖ (from an electrode at potential E>0 in this example, or from another molecule N) are also much easier in the excited states.

Fig. 4. Charge redistribution in an excited state. The n orbital is localized on the O atom, the p* orbital is delocalized over the C>O group. There is a transfer of 0.5 eÖ from O to C.

centers, or ‘nucleophilic’ reactions which involve the attack of positive-charge centers [2]. In a structural formula the bonds are of course localized between pairs of atoms, so that the corresponding bonding and antibonding orbitals are by implication localized in the same way (fig. 6). This picture of localized orbitals

Suppan

56

Fig. 5. Energy (E) – reaction coordinate (RC) diagram of ground-state (dark) and lightinduced chemical reactions.

Fig. 6. In a polyatomic molecule one-electron excitations are described by the labels of the orbitals involved in the transition from the ground state (G). When these orbitals are localized in different zones of space the transition implies a charge transfer (CT).

is adequate in general for the r orbitals of single bonds, and also for the p orbitals of isolated double bonds such as the one in formaldehyde. The important difference between diatomic and polyatomic molecules, which was alluded to above, arises when the molecule contains alternating single and double bonds. Such molecules are said to be conjugated and the orbitals become delocalized. Principle of Exclusion An orbital defines an energy level available to electrons; it can of course remain vacant, but it can hold only one or two electrons and no more. This results from the principle of exclusion as stated by Pauli (fig. 7). There is a rather obvious principle of exclusion in classical physics which states that no two material bodies can be found at the same time at the same place.


57

Fig. 7. The two possible arrangements of spin and orbital motions of an electron; for a given orbital angular momentum the spin can be a(+1/2) or b(–1/2).

Fig. 8. The interaction of two atomic orbitals (uA, uB) leads to the formation of bonding (rAB) and antibonding (rHAB) molecular orbitals of different energies.

Excited States of Molecules [3] Diatomic Molecules In a very simple molecule such as hydrogen, H2, two H atoms come together to form a chemical bond, and this bond is shown as a line in the structural formula, i.e. H–H. In a more detailed picture, the two atomic orbitals interact when they are close together, and this interaction can lead to two opposite situations, as shown for the general case of two different atoms A and B (fig. 8). A bonding molecular orbital is formed if the two electrons are found mostly in the space between the nuclei; they provide a shield between the two positively charged protons so that their electrostatic repulsion is lowered. An antibonding molecular orbital is formed if the two electrons are localized essentially outside the internuclear space.

Suppan

58

Fig. 9. The interaction of the carbon p orbital in conjugated molecules leads to interactions which result in the formation of delocalized orbitals.

Linear Conjugated Molecules: The Polyenes In such molecules the carbon p orbitals which form the double bonds in the structural formula constitute a continuous array of interacting orbitals (fig. 9). The p orbitals which would be drawn between localized pairs of atoms now form new orbitals which are delocalized over the entire conjugated system. The wave functions of these delocalized orbitals are easily drawn according to their nodal properties. Remember that a standing wave has boundaries defined by the size of the molecule, and that it can have 0, or 1, 2, etc. nodes in between, in the order of increasing energies. The molecule of ethene (ethylene) which has only one double bond can be considered as the first member of this series (fig. 10), although it is not conjugated. The two p orbitals of C1 and C2 interact to form one pair of molecular orbitals p and p*, the first one being a bonding orbital with no node between the carbon atoms, and the second one being antibonding with it one node. Butadiene contains two double bonds, i.e. four carbon atoms which constitute one p orbital each; there are then four delocalized molecular orbitals which are shown in figure 10 as having no node (W1), one node (W2), two nodes (W3) and finally three nodes (W4). The bonding or antibonding character of an orbital is obtained by counting the number of bonds and antibonds between the atoms: a node is an antibond, and absence of a node being a bond.


59

Fig. 10. The wave functions of delocalized p orbitals of linear conjugated molecules can be drawn following the number of nodes (open circles).

Orbital W2, is the highest occupied molecular orbital (HOMO) in the ground state. It corresponds to the structural formula of the molecule, with double bonds between C1 and C2, and between C3 and C4. Orbital W3 is likewise the lowest unoccupied molecular orbital (LUMO) in the ground state and corresponds to a biradical structure of the molecule with unpaired electrons on C1 and C4. Such a biradical structure can be a very simple but sometimes useful representation of the excited molecule (in states S1 or T1).

Quenching of Excited States Quenching is the nonradiative deactivation of an excited molecule M* by a molecule Q (the quencher), the excited state energy eventually becoming heat energy of the surroundings (e.g. liquid solvent or solid matrix): M*+QKM+Q(+energy).

This scheme shows only the overall process, and we shall see that there may be intermediate steps between these initial and final states. In the way it is written here, quenching could be defined as the catalytic deactivation of excited molecules without chemical reaction. It is obvious that a chemical reaction also results in the disappearance of the excited molecule M* but this is a photochemical process. We shall use here the definition of ‘quenching’ in

Suppan

60

Fig. 11. The Dexter mechanism of energy transfer through simultaneous electron exchange. The HOMO and LUMO of the quencher must fall between the HOMO and LUMO of the energy donor.

the restricted sense of a photophysical process in which the molecule M* is resorted unchanged to its ground state M. Energy Transfer If the quencher molecule Q has an excited state Q* lower than M*, the excitation energy can be transferred according to: M*+QKM+Q*.

Radiative or nonradiative deactivation of Q* and Q then completes the quenching process. There are two major energy transfer processes. (1) Electron exchange: The Dexter mechanism shows a simple orbital diagram of an electron exchange between an excited molecule M* and a ground-state molecule Q (fig. 11). The first requirement is of course that the excitation energies should be in order E(MÖM*)qE(QÖQ*); the second requirement is that the orbitals should allow exergonic electron transfer from H uM to uQH and from uQ to uM . The two electrons transfers are simultaneous; both M and Q remain neutral species throughout the electron exchange and no recognizable ions are formed as intermediates. This double electron transfer requires the spatial overlap of the orbitals so that the molecules M* and Q must be in close contact (van der Waals or hard sphere contact). Although the overall spin quantum number of the system M*/Q must be kept in electron exchange, quenching can take place between singlet and triplet states in any combination:


61

Fig. 12. The Fo¨rster mechanism of energy transfer through transition dipole interaction. Deactivation of the excited molecule M* creates an electric field T/r3 which promotes the !. excitation of Q through its transition dipole R

1

M*+QKM+1Q*;

3

M*+QKM+3Q*; etc.

(2) Dipole – dipole interaction: The Fo¨rster mechanism. This is in fact the interaction of the transition moments of the excitation QKQ* and the deactivation M*KM (fig. 12). As the excited electron of M* falls to the lower orbital of M there is a change in dipole moment which produces an ! and to electric field; this field is proportional to the transition moment M the inverse cube of the distance. An electron in the molecule Q therefore ! /r3, and as it moves towards a higher experiences a force proportional to M orbital it produces its own electric field which results in a force being applied on the electron in molecule M*. In this way the ‘downward’ motion of the electron in M* and the ‘upward’ motion of the electron in Q are couple by their electric fields. Heavy Atom Effect The heavy atom effect can rely on the presence of an atom of high atomic number either within the molecule itself (the internal heavy atom effect) or in the solvent (the external heavy atom effect). In both cases the fluorescence quantum yields decrease and the triplet yields increase, but the triplet lifetimes (e.g. phosphorescence) decrease as well. Paramagnetic Quenching External magnetic fields can also enhance spin-orbit coupling, so that molecules which have a permanent magnetic moment have an action similar to the heavy atom effect. Molecular oxygen 3O2 is a paramagnetic species with a triplet ground state and is the best known paramagnetic quencher. It must however, be noted that its quenching action can be quite complex, for it can take part in energy and electron transfer reactions as well.

Suppan

62

Fig. 13. Quenching of an excited molecule through sequential electron transfer. Here an ion pair is formed in the first place, within which back electron transfer leads to overall deactivation of the excited energy donor.

Quenching by Electron Transfer In an electron transfer reaction two neutral molecules, for example, will form an ion pair (fig. 13); if one of the neutrals is electronically excited it will be deactivated, and the overall process may appear as a quenching if the ion pair recombines to return to the neutral ground-state partners: M*+QKMf++QfÖKM+Q (+energy).

We arrive here at the limit quenching considered as a purely photophysical process, involving no permanent chemical change. Electron transfer is in fact a chemical reaction which leads to new, distinct species Mf+ and QfÖ; these may separate and react to form new molecules, in which case we enter the realm of photochemistry [4]. Photoinduced electron transfer must be considered to be a photochemical reaction, but in some cases it may appear to be a quenching reaction when the reactants are restored to their initial state.

Intramolecular and Intermolecular Electron Transfer [5] Electron transfer between two molecules, e.g. M* and N is one of the most fundamental and widespread of all photoinduced chemical reactions. It


63

is at the basis of photosynthesis in nature, and of photography in industrial applications. M*+NKMf++NfÖ or M*+NKMfÖ+Nf+.

In the first case, M* is the electron donor and N is the electron acceptor, these roles being reversed in the second case. These properties of ‘donor’ and ‘acceptor’ are relative, the same molecule M* being a donor towards some species N and an acceptor towards other partners N. It will be seen presently that the direction of the electron transfer depends simply on the energy balance of the reactions. The loss of an electron by M, MKMf++eÖ , is the process of oxidation in electrochemistry. The electron is then accepted by an electrode of well-defined potential, so that the oxidation potential (Eox) is the free energy of the reaction. Similarly the reduction potential (Ered) is the energy of the reduction, e.g. N+eÖKNfÖ. By definition the molecule, which is oxidized, is the donor (M in this case), and the molecule, which is reduced, is the acceptor. The electron transfer from M to N is therefore equivalent to the combined oxidation of the donor and reduction of the acceptor, so that the energy balance is: DG>Eox (D)ÖEred (A)+CÖE*.

Here E* is the energy of the excited state involved in the light-induced process. It has been noted that both the oxidation and the reduction potentials of an excited molecule are lowered by the excited-state energy, compared with the ground-state molecule. In the present example the relevant excited-state energy is that of the reactive state of M*, which can be in general either the lowest singlet state S1 or the lowest triplet state T1, in the case of an organic molecule. The additional term C in the ‘Rehm-Weller’ equation is called the ‘Coulomb term’ and represents the electrostatic energy gained when the two product ions Mf+, NfÖ, are brought from ‘infinite separation’ to the actual encounter distance in electron transfer (usually Van der Waals or ‘hard sphere’ contact).

Monophotonic and Multiphotonic Processes When a photochemical reaction is written as: M*KP (unimolecular) or M*+NKP (bimolecular),

it is implied that the excited state M* has been formed by the absorption of one photon; the entire process is then said to be monophotonic. There are, however, processes which depend on the absorption of two photons (biphotonic processes), and in principle there could be reactions resulting from the absorption of any number of photons. Such multiphotonic processes, however, become increasingly improbable as the number of photons required increases, and it

Suppan

64

is sufficient to consider the biphotonic case of which there are indeed many important examples. There are two distinct ways in which a biphotonic process can take place. (1) By the sequential absorption of two photons by the same molecule, according to: hm

hm

M CB M* CB M**,

whereby the molecule M is promoted to an excited state of very high energy. (2) By the encounter of two excited molecules: M*+ M*KP.

The emission of delayed fluorescence through triplet-triplet annihilation can be taken as an example of this type of biphotonic process. Biphotonic processes depend greatly on light intensity, and on the lifetimes of the excited states M*. Their importance increases markedly in conditions of very intense, pulsed light excitation, e.g. in laser beams or in flash light.

Primary and Secondary Photochemical Processes These take place in chemical reactions in which the excited molecule M* takes part in the ‘primary’ photochemical process. This may lead directly to the final products (e.g. in isomerizations), or more often to unstable or reactive chemical species (e.g. free radicals or radical ions) which then react further in ‘secondary’ processes through dark reactions which lead ultimately to the final photoproducts. The sequence of a photochemical reaction can therefore be given as a succession of steps. (1) (2) (3) (4)

Absorption of light Fluorescence Nonradiative deactivation Intersystem crossing

(5) Primary reaction Unimolecular Bimolecular (6) Secondary processes

M+hmKM* M*KM+hmF M*KM (+heat) M*K3M* light reaction

M* CCCB P light reaction

M*+N CCCB P dark reaction

P CCCB F (final products)

P is the ‘primary photochemical product(s)’; these can then react further in a variety of dark reactions.


65

In the case of the closed-shell organic molecules M* can be an excited singlet or triplet state. M* can react on its own in unimolecular reactions (dissociations, isomerizations) or it can react with another (ground-state) molecule; N in bimolecular processes (e.g. additions, substitutions, etc.). For a reaction originating from the (lowest) triplet state the radiative and nonradiative deactivations of the molecule (3M*) must also be considered: (2a) Phosphorescence (3a) Nonradiative deactivation

3

M*KM+hmP M*KM (+heat)

3

Yield, Quantum Yield and Rate Constant of Photochemical Reactions The overall efficiency of a thermal reaction is given by its ‘yield’, that is the fraction of reactants convertible into products; this is usually expressed in percent. In the case of a photochemical process the overall efficiency is given by the ‘quantum yield’ which is defined as the number of molecules of photoproducts formed for each photon absorbed: u>

number of product molecules obtained . number of quanta of light absorbed

It is important to bear in mind the fact that the intrinsic reactivity of an excited state is given by the rate constant, not by the quantum yield. An excited state can be very ‘reactive’ in terms of the rate constant, yet the reaction quantum yield can be very low if it is of very short lifetime as a result of other deactivation or quenching processes. The relationships between rate constants and quantum yields can be derived from a Jablonski diagram (fig. 14). Let us assume that absorption of the photon promotes the molecule to some high excited state M** (this could be a higher vibrational level of S2, for example), but the chemical reaction itself takes place only from the lowest triplet state T1. After excitation the molecule will undergo very fast internal conversion to S1; here we assume that intersystem crossing between the higher states (e.g. S2 to T2) can be neglected compared with internal conversion. The population of S1 states is then unity, but now the triplet state T1 is formed in competition with other deactivation paths of S1 such as fluorescence (rate constant kF) and internal conversion S1KS0 (rate constant kd) to which bimolecular quenching processes could be added as well. In a general form the yield of T1 will be given by the branching ratio of these rate constants: UT>kisc /(kisc+kF+kd ).

The molecules which have reached T1 will now react with a rate constant kr (unimolecular reaction) or krE [N] (bimolecular reaction with a ground-state

Suppan

66

Fig. 14. Jablonski diagram of electronic transitions in a polyatomic molecule which reacts chemically from the triplet state T.

partner N) in competition with radiative (phosphorescence of rate constant kp), nonradiative (kdE ) deactivations as well as quenching processes kqE [Q]) so that the final reaction quantum yield of the primary process is: Ur>UT {krE [N] / (krE [N]+kP+kdE +kqE [Q])}.

Note that all the rate constants are unimolecular or pseudo-unimolecular (in the case of bimolecular processes like quenching or chemical reaction M*+NKP). When the donor and the acceptor are linked together (fig. 15) by covalent bonds so that they are formally part of a single molecule, the process of intramolecular electron transfer can be observed. Some rigid spacers made up of saturated bonds seem to act somewhat like electrical conductors and ˚ ) as a electron transfer can then take place over quite large distances (20 A result of the ‘through-bond’ interaction. When a quinone is linked to a porphyrin, the rate of electron transfer depends on the nature of the linkage, even if it is too short to allow a folding back to give direct contact between the chromophores. This is strong evidence for the through-bond character of the electron transfer, since through-space interaction between identical pairs of chromophores should be independent of the linkage.


67

a

b

Fig. 15. Examples of bichromophoric molecules used for the study of intramolecular electron transfer. a Dimethoxynaphthalene electron donor, dicyanoethylene electron acceptor with rigid spacers. b Porphyrin donor and quinone acceptors separated by flexible spacers.

Unimolecular Reactions An isolated molecule can only undergo two types of chemical change: (1) the rearrangement of its atoms into a new molecule which keeps the general formula unchanged (the product is then an isomer of the reactant, so that the term isomerization is equivalent to that of rearrangement, and (2) the dissociation into two or more ‘fragments’ which can be individual atoms, free radicals, radical ions, closed-shell ions, biradicals, smaller molecules, protons, electrons, etc. A distinction must be made between truly isolated molecules which react in the absence of any collision with other molecules, as in the gas phase at very low pressures or in molecular beams, and molecules in liquid or solid environments. A condensed phase medium, liquid or solid, imposes a ‘cage effect’ which can prevent large geometrical changes in rearrangement reactions, and the separation of fragments in dissociation reactions (fig. 16). Dissociation reactions. A glance at the list of fragments which can be produced from the dissociation of a molecule shows that these apparently simple reactions can in fact follow several mechanisms. Of these, two are treated separately: the loss of an electron, which is the process of photoionization; and the loss of a proton, which is one side of the acid-base equilibria. Reactions of rearrangement (isomerization). In reactions of isomerization (or ‘rearrangement’) the molecule keeps its overall formula but undergoes a structural change which either keeps the bonding pattern of the atoms unchanged but modifies the geometry of the molecule (stereo-isomerism), or produces a new bonding pattern (valence isomerization).

Suppan

68

a

b

c

Fig. 16. a-Dissociation of carbonyl compounds. a Aromatic ketone or benzaldehyde. b Acetone. c Ester.

Fig. 17. Potential energy diagram of the ground and excited states of the cis and trans isomers of an olefin. The energy is shown as a function of the bond angle.

Stereo-isomerizations are quite common photochemical processes with unsaturated organic molecules (the primary photochemical reaction of vision is of this type). A substituted ethylene molecule (fig. 17) can have two stereo-isomers in the ground state, which can be labeled as the cis and the trans forms. Rotation around the double bond is practically impossible and the stereo-isomers do not interconvert thermally. In the excited states, S1 or T1, the promotion of an electron from the bonding p orbital to the antibonding p* orbital reduces


69

Fig. 18. Valence isomerization of cyclobutene and butadiene.

the p bond order to near zero. The molecule can be represented in the form of a biradical (this is a highly simplified view of an excited state, but reasonably accurate in this case), the central C atoms being linked by a single bond which allows free rotation (subject only to steric hindrance of the substituents). The photochemical process can be pictured as the excitation of the molecule without change in geometry, to the excited state S* potential energy surface (this could be the S1 or the T1 state). Now the molecular geometry is unstable in this state, and relaxation brings it down to the most stable excited state geometry, in which the two unpaired electrons have the least possible interaction; this corresponds to the twisted form, the orbitals of the two unpaired electrons being orthogonal. This molecular shape, however, is also that of the least stable ground-state geometry, so here the ground- and excited-state potential energy surfaces come very close together; this is described as a ‘funnel’, through which deactivation can take place by nonradiative crossing from S* to S0. Another important class of cycloaddition reactions is the formation of oxetane rings between a photoexcited carbonyl compound and an unsaturated molecule. These reactions also occur probably through an exciplex although these exciplexes are nonfluorescent as they are formed from the triplet state of the ketone or aldehyde. The formation of the four-membered oxetane ring is an interesting example of a typical photochemical reaction with no obvious equivalent in the ground state. There are indeed other examples of formation of small, strained rings through photochemical processes, when such rings are quite difficult to form by dark reactions. Valence isomerizations are rearrangement reactions in which, the ‘connectivity’ of atoms is changed (fig. 18): for example, two atoms that were linked together in the starting molecule become separated in the product molecule. There are indeed many photochemical processes of valence isomerization, and these may follow quite different mechanisms which we shall consider in turn, with the help of actual examples. A first general question must now be discussed, concerning the mechanism of any chemical reaction in which a new bond (or several new bonds) is made while other bonds are broken. Concerted and unconcerted reaction pathways: intermediates in chemical reactions. When the making and breaking of bonds take place simultaneously,

Suppan

70

Fig. 19. Charge transfer in excited states of hydroxy and amino aromatics.

the process is ‘concerted’; there is then no actual ‘intermediate’ in the reaction, which simply passes through a ‘transition state’ which is the point of highest energy in an energy-reaction coordinate diagram. However, there is in principle another pathway for the same reaction, if first the AB bond is broken but the new bonds are not yet formed. Protolytic Equilibria (Acid-Base Reactions) [6] Acid-base equilibria are important chemical processes both in thermal and in photochemical reactions, since many molecules have an ‘acid’ form AH, and a ‘base’ form A, which are connected by the transfer of a proton: AH+ ½ A+H+.

The equilibrium constant of this reversible proton transfer reactions is: K>[A][H+] / [AH+],

and is usually given by the ‘pK’ of the reaction, which is the pH of the solution in which the concentrations of the acid and base forms are equal (remembering that the ‘pH’ of a solution is defined as, pH>–log [H+]). These acid-base equilibria are important since the acid and base forms of many molecules have quite different physical and chemical properties (fig. 19). An equilibrium constant K is related to the change in free energy G between the products and reactants, DG>–RT ln(K). DG here is the chemical free energy of the reaction, but it can be equated with the reaction enthalpy, DH, if the change of entropy, DS, can be neglected, or if, as in the present case, it can be assumed to be the same for the ground-state (thermal) and excited-state (photochemical) equilibria. The enthalpy change, DH, is then related to the ground-state acid-base equilibrium, and DH* is similarly related to the equilibrium in the excited state. The species A/A*, on the one hand,


71

a

b Fig. 20. a Potential energy diagram of the Fo¨rster cycle. DH are the enthalpy changes between the acid and base forms. b Absorption and fluorescence spectra of a naphthol and a naphtholate.

and AH+/AH+*, on the other hand, are connected through the excitation energies hmOH and hmOÖ in this example, so that the energy cycle becomes: DH+hmOH>DH*+hmOÖ .

The Fo¨rster cycle therefore permits the calculation of the excited-state pK* (from DH*) simply from spectroscopic data, assuming of course that the ground-state pK is known, from usual titration measurements (fig. 20). Bimolecular Reactions All these reactions result from the interaction of an electronically excited molecule M* with a ground-state molecule N. The most important processes of this type can be classified as:

Suppan

72

a

b

Fig. 21. Examples of photoinduced addition reactions of double bonds. a Acyclic addition of methanol. b Cycloaddition of an olefin.

Fig. 22. Cycloaddition of anthracene to form a photodimer.

Additions Reductions Oxygenations Substitutions

M*+NKMN M*+ZHKMHf+Zf M*+O2KMO2 (or other products) (XY)*+NKXN+Y

In addition reactions an excited, unsaturated molecule uses its weakened p bond to form two new r bonds; when these r bonds form a new ring in the molecule MN the reaction is one of ‘cycloaddition’. Thus alkenes can add photochemically to alcohols (fig. 21a) to form a noncyclic adduct; or (fig. 21b) to other alkenes in cycloaddition processes. Anthracene undergoes a photochemical 9,10,9,10-cycloaddition (fig. 22) which goes through the excimer as intermediate. Many aromatic molecules follow similar cycloaddition paths. The close approach of the molecules in the excimer is essential for bond formation, and steric hindrance can prevent the reaction; unsubstituted anthracene dimerizes so fast that no excimer fluorescence can be detected. 9,10-dimethylanthracene shows both excimer fluorescence and photodimerization, but 9,10-diphenylanthracene shows neither excimer emission nor photodimerization.


73

Fig. 23. Mechanism of the photoreduction of carbonyl compounds with hydrogen atom donors, ZH. Ketyl radicals and donor radicals are formed in the primary photochemical process.

Photoreduction occurs by hydrogen atom abstraction or by electron transfer. The first process is a common photochemical reaction of carbonyl derivatives and other unsaturated molecules in the presence of suitable hydrogen atom donors (which can be alcohols, paraffins, ethers, etc., that is, almost any molecule with a not-too-strong C–H bond). In figure 23 the hydrogen donor is noted simply as ZH. The primary photochemical process leads to a pair of radicals which can undergo various secondary processes leading to the final reduction product. The reactivity of the ketone or aldehyde depends greatly on the nature of the lowest excited states S1 and mainly T1. The n–p* states are the most reactive and this can be explained by the fact that a half-filled n orbital on the oxygen atom acts as an electron acceptor; p–p* states are less reactive since a C>O bond is less weakened when the antibonding orbital is delocalized over the p system, and the reactivity diminishes as the charge-transfer character of p–p* state increases with stronger electron donor substituents. Photo-oxydation and photo-oxygenation reactions are the additions of O atoms or O2 molecules to some reactant R. Here the photophysical properties of the oxygen molecule O2 are the most important, since many of these reactions involve the attack of ‘excited singlet oxygen 1O2H on the ground-state reactant molecule: R+1OH2 KRO2,

where the products are usually peroxides or hydroperoxides. The molecule of oxygen O2 differs from closed-shell chemical species in having a triplet ground state (fig. 24); it has therefore triplet and singlet excited

Suppan

74

Fig. 24. Jablonski diagram of the oxygen molecule. The energy is in kcal molÖ1.

states as shown above, and the first excited state is a singlet 1O2H (denoted 1Dg in spectroscopy, but known simply as ‘singlet oxygen’ in photochemistry). This is a very reactive form of molecular oxygen which is at the basis of most photo-oxidation processes. The energy of this exited state 1O2H is very low. The transition 3O2K1O2H corresponds to an absorption in the near IR region. Singlet oxygen can therefore be formed through energy transfer from many excited organic molecules according to the following schemes: 1

M*+3O2K3M*+1OH2 , M*+3O2KM+1OH2 .

3

Acknowledgment This work was supported by the Swiss National Science Foundation through project No. 20-53568-98.

References 1 2 3 4 5 6

Pitts JN Jr, Wilkinson F, Hammond GS: The vocabulary of photochemistry. Adv Photochem 1963. Coulson CA, O’Leary B, Mallion RB: Huckel Theory for Organic Chemists. London, Academic Press, 1978. Murrell JN: The Theory of the Electronic Spectra of Organic Molecules. New York, Wiley, 1963. Gilbert A, Baggott J: Essentials of Molecular Photochemistry. London, Blackwell, 1991. Fox MA, Chanon M: Photoinduced Electron Transfer. Amsterdam, Elsevier, 1998. Ireland JF, Wyatt PAH. Adv Phys Org Chem 1976;12:131.


75


............................

Photobleaching and Photodynamic Effect of Protoporphyrin IX Marie-The´re`se Wyss-Desserich a, Chung H. Sun b, Pius Wyss a, Lars O. Svaasand b, c a b

c

Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Switzerland; Beckman Laser Institute and Medical Clinic, University of California, Irvine, Calif., USA, and Norwegian University of Science and Technology, Trondheim, Norway

Introduction Light exposure of any photosensitizer will tend to destroy the sensitizer itself. Therefore the concentration of the nondegraded sensitizer participating in the photo-oxidation process decreases continuously during illumination [1, 2]. The efficacy of the sensitizer is thus maximal at the beginning of illumination, and decreases to zero when the sensitizer has been completely photodegraded. The change in 5-aminolevulinic acid (ALA)-induced protoporphyrin IX (PpIX) fluorescence in two gynecological cell lines such as Chinese hamster ovary (CHO) cells and human endometrial adenocarcinoma-1A (HEC-1A) cells by monitoring fluorescence following irradiation is presented. Additionally, a possible relationship between photobleaching of the sensitizer and photodynamic therapy (PDT) response was examined.

Materials and Methods Cell Lines and Cultivation CHO cells (Cricetulus griseus; ATCC CCL 61) were routinely cultured in minimum essential medium with Earle’s salts and nonessential amino acid (Gibco 410-1500EL) supplemented with 10% fetal bovine serum (FBS). Well-differentiated HEC-1A (ATCC HTB 112) were obtained as an established culture system from the American Type Culture Collection. The cells were grown in complete culture medium 199 with Earle’s salts (Irvine Scientific, Irvine Calif., USA) supplemented with 10 mM L-glutamine (Gibco), 100 lg/ml streptomycin, 50 lg/ml gentamycin (Gibco) and 10% FBS.

Both CHO and HEC-1A cells were subcultured with 0.25% trypsin/1 mM EDTA on a routine basis. Incubation of 5-Aminolevulinic Acid ALA hydrochloride (Sigma Chemicals, St. Louis, Mo., USA) was prepared according to the concentrations needed in the incubation medium. The pH was adjusted to 7.4. Cells were incubated with ALA in exponential growth stage. Uptake Studies. Concentrations of 0.01, 0.1, 1 and 10 mg ALA/ml phenol red free growth medium (with 10% serum) were used for CHO and 0.01, 0.1, 1, 2 and 5 mg ALA/ ml medium (phenol red free, 10% serum) for HEC-1A. Cells were incubated with ALA for 3 and 6 h. At the end of the incubation time, the medium was removed, the cells were washed with PBS twice and detached with 0.25% typsin/1 mM EDTA from the culture dishes. The relative fluorescence intensity of PpIX was determined by flow cytometry. Bleaching Studies. CHO were incubated with 0.025 mg ALA/ml phenol red free growth medium (with 10% serum) and HEC-1A with 0.5 mg ALA/ml medium (phenol red free, 10% serum) for 6 h. At the end of the incubation time, both cell lines were treated with laser light. Afterwards fluorescence intensities of the detached cells (in 0.25% trypsin/mM EDTA) were measured and the bleaching parameter were determined. The bleaching parameter of a photosensitizer is defined as the light dose (J/cm2) used to reduce the intensity of photosensitizer to 1/e>0.37 8 37% of the initial value [3]. Laser Treatment All experiments were conducted with an argon-dye pumped laser Innova 200 and Dye Laser 599 containing 4-dicyanomethylene-2-methyl-6(p-dimethylamino-styryl)-4H-pyran (DCM) dye. The laser system was set to emit laser light at a wavelength of 630 nm. The wavelength was verified with a Jobin Yvon No. 51354 UV monochrometer (Longjuneau, France). Photodynamic treatment was performed with light doses between 10 and 120 J/ cm2 produced by a constant fluence rate of 100 mW/cm2 at different duration (s). Flow Cytometer System Fluorescence measurements were performed on the Epics V flow cytometer from Coulter Electronics, Inc., equipped with coherent Innova 90 Argon laser. The samples were excited with 488 nm wavelength of light. A long path filter of 610 nm was used for emission. Fluorescence intensities were measured on log integrated red fluorescence (LIRF) channel for 10,000 cells/sample. Cells without ALA incubation were used as control. The LIRF channel numbers of each sample were converted to a linear scale according to the 3.00 decade channel conversion table provided by Coulter Electronics. Thereafter, control values (autofluorescence) were substracted from ALA-treated samples. Mean values and standard errors were calculated for fluorescence data. Each study was repeated three times with exception of the bleaching studies in HEC-1A cells, which were studied only twice. Each measurement is carried out with two different culture dishes for each incubation time point, each ALA concentration and each light dose (J/cm2). Trypan Blue Test and MTT Assay In order to determine cell viability of the treated and untreated group of CHO and HEC-1A, the trypan blue exclusion test was used 24 h after irradiation (100 mW/cm2, 630 nm)

Photobleaching and Photodynamic Effect of Protoporphyrin IX

77

Fig. 1. ALA-induced relative fluorescence of PpIX in Chinese hamster ovary cells (CHO) and well-differentiated endometrial adenocarcinoma cells (HEC-1A) determinated by flow cytometry. Cells were measured after incubation with different concentrations of ALA (0.01, 0.1, 1, 2, 5 and 10 mg/ml medium) during 3 and 6 h. Some error bars are not visible due to very small values. at different light doses (J/cm2). The colorimetric MTT cytotoxicity assay adopted from Mosmann [4] was performed on CHO cells 24 h after laser treatment to measure the activity of the mitochondrial enzyme succinate dehydrogenase.

Results Fluorescence of Two Different Cell Lines The relative fluorescence intensity of ALA-induced PpIX in CHO and HEC-1A cells incubated with different concentrations of ALA during 3 and 6 h is shown in figure 1. A significant increase in fluorescence was revealed between 0.01 and 0.1 mg/ml medium for CHO and between 0.1 and 1 mg/ml medium for HEC-1A at both 3- and 6-hour incubation times. This increase in fluorescence was more pronounced following incubation for 6 compared to 3 h in both cell types. CHO exhibited significantly higher fluorescence compared to HEC-1A over all concentrations. A substantial decrease in fluorescence intensity was detected at ALA concentrations of ?2 mg/ml medium.

Wyss-Desserich/Sun/Wyss/Svaasand

78

Fig. 2. Light dose-dependent decay of PpIX fluorescence of CHO and HEC-1A due to photobleaching. With 0.025 mg/ml medium in CHO and 0.5 mg/ml in HEC-1A for 6 h before light treatment, similar relative fluorescence intensities of PpIX were induced. This initial fluorescence is marked with 100%. Autofluorescence (control) was substracted from fluorescence values of ALA-treated samples. The power density of the dye laser was 100 mW/cm2 at 630 nm. Some error bars are not visible due to very small values.

Decay in Fluorescence The light-dose-dependent photobleaching effect of ALA-induced PpIX in HEC-1A and CHO cells is shown in figure 2. HEC-1A were labelled for 6 h with 0.5 mg ALA/ml medium and CHO with 0.025 mg ALA/ml medium exhibiting similar initial fluorescence intensity (fig. 1). In our setup the cell fluorescence is composed of two parts: a bleachable PpIX fraction together with a nonbleachable autofluorescence. The fluorescent intensity of PpIX may therefore be expressed as: F>a1eÖ¶ //+a2,

where ¶ and / are the optical fluence and the bleaching parameter (bleaching fluence), respectively. The quantities a1 and a2 are dependent on the unbleached PpIX contribution and the unbleachable autofluorescence. The curve in figure 2 represents fluorescence values of ALA-treated samples with substracted autofluorescence (control values). The intensity of cell fluorescence measured by flow cytometry decreased exponentially with the duration of illumination (fig. 2). Thus, the fluorescence decay was highest at the beginning with low light doses (10 J/cm2) and became low and asymptotic at higher light doses (80–120 J/cm2).


79

Fig. 3. Vitality curves for light-exposed CHO and HEC-1A mediated by ALA-induced PpIX. Incubation conditions: With 0.025 mg ALA/ml medium in CHO and 0.5 mg ALA/ ml in HEC-1A for 6 h before light treatment, a similar relative fluorescence intensity of PpIX was induced. The power density of the dye laser was 100 mW/cm2 at 630 nm. Twenty-four hours after irradiation the trypan blue test was used. There was neither dark toxicity (ALA alone) nor light alone toxicity. Some error bars are not visible due to very small values.

ALA-induced PpIX is photobleached to 37% of the inital value by a light dose of about 20 J/cm2 in HEC-1A and 40 J/cm2 in CHO at 630 nm light. The bleaching parameter of PpIX in HEC-1A (20 J/cm2) is thus lower than in CHO (40 J/cm2). A lower bleaching parameter means stronger bleaching. Exposing cells in the absence of ALA-induced PpIX to light of different energy densities did not result in changes in fluorescence intensity. Decay of Cell and Mitochondrial Vitality The phototoxicity of PpIX on HEC-1A and CHO was evaluated for a photosensitizer incubation time of 6 h at various energy densities (fig. 3). The evaluation was based upon the trypan blue dye exclusion method 24 h after irradiation. In the absence of light no cells with an incubation of 0.5 mg ALA/ ml (HEC-1A) or 0.025 mg ALA/ml medium (CHO) for 6 h showed dark toxicity. Additionally there was no light alone toxicity. HEC-1A labeled with 0.5 mg/ml medium administered ALA showed about the same content of the


80

Fig. 4. Light dose-dependent decrease in mitochondrial activity in CHO sensitized with ALA-induced PpIX. After incubation with 0.025 mg ALA/ml medium for 6 h and additional laser treatment (100 mW, 630 nm), the MTT cytotoxicity assay was performed. The decrease in mitochondrial activity of light-treated cells is calculated in comparison with the untreated cells (0 J/cm2).

photosensitizer PpIX as CHO labeled with 0.025 mg ALA/ml (fig. 1). Inducing similar cellular content of PpIX in both cell lines, HEC-1A were much more sensitive on illumination with laser light than the CHO. All HEC-1A were killed using a light dose of 5 J/cm2, whereas 20% of CHO survived even at a light dose of 40 J/cm2. The mitochondrial activity of CHO decreases as the laser dose increases (fig. 4). The rate of mitochondrial activity decreases more when compared to the rate of cell death using the same light dose (fig. 3).

Discussion Evaluation of photobleaching characteristics of the photosensitizer PpIX during illumination is of interest because bleaching affects singlet oxygen generation [2, 5]. Singlet oxygen is the main cytotoxic agent responsible for the photodynamic effect. Since ALA-induced PpIX has been widely used in experimental and clinical studies [6–10], this study focused on the effect of ALA-induced PpIX fluorescence and its bleaching behavior on PDT.


81

Fluorescence Intensity of the Two Cell Lines The evaluation of intracellular PpIX concentrations not causing darktoxicity effects was crucial for the bleaching studies. An appropriate amount of intracellular PpIX depends on the ALA concentration in the medium and incubation time. The influence of different ALA concentrations on the intracellular PpIX content was evaluated in CHO and HEC-1A cells. These dose-response studies indicate significant concentration-dependent variations in PpIX content. Fluorescence intensity was substantially increased to between 0.01 and 0.1 mg ALA/ml medium in CHO and between 0.1 and 1 mg/ml medium in HEC-1A to obtain transient saturation values as described in the study of Steinbach et al. [11]. At concentrations of ?1 mg ALA/ml medium, fluorescence intensities decreased in both CHO and HEC-1A cells due to cytotoxic effects causing cell death. Additionally, nonfluorescent aggregates of highly concentrated PpIX may contribute to the decrease in fluorescence. Porphyrins are known for their tendency to produce aggregates of very poor fluorophores [12] in solution and cells. The duration of cellular exposure to ALA affected the amounts of PpIX fluorescence intensity as well. An ALA incubation time of 6 h revealed higher fluorescence intensity than 3 h especially at ALA concentrations of between 0.1 and 1 mg/ml. ALA as a precursor of PpIX in heme synthesis must be metabolized to the active photosensitizer PpIX. This process may explain the delayed fluorescence elevation. The photosensitizer concentration was highest in CHO cells which exhibited a faster doubling time (15 h) [13], whereas HEC-1A showed less fluorescence corresponding to a longer doubling time (22.5 h) [14]. Rebeiz et al. [15] suggested that cells with rapid turnover produced more PpIX. Iinuma et al. [16] evaluated different cell lines of varied origin (excluded endometrial cells) and compared the doubling time to PpIX cellular contents after ALA administration. Cells with high proliferative rates often, but not always synthesized more PpIX. Additionally, they observed that cells which synthesized more PpIX also effluxed it more quickly. These data suggest that cellular PpIX accumulation is a dynamic process which is determined both by enzyme activities in the heme biosynthesis pathway and the efflux of PpIX from the cells. Decay of Fluorescence (Photobleaching) Many porphyrins in simple solution are photobleached on illumination, i.e. they are converted into products that do not absorb visible light appreciably [17]. In this study intracellular, nonmaximum PpIX contents at a similar fluorescence level were induced in both CHO and HEC-1A cells to evaluate the photobleaching characteristics of PpIX. The bleaching was monitored by flow cytometry indicating a decrease in cell fluorescence during irradiation.


82

The rapid decay in fluorescent level in both cell lines indicates a low bleaching parameter in the range of 20 to 40 J/cm2 using a constant fluence rate of 100 mW/cm2. Svaasand et al. [18] also found low bleaching parameter values (35 and 50 J/cm2) of PpIX for basal cell carcinomas. Another study [19] shows that the bleaching light dose (a dose that achieves a 1/e [D37%] reduction in sensitizer signal) varies with different fluence rates. Ru¨ck et al. [20] described that the change in fluorescence intensity during irradiation correlates with the chemical modification of the sensitizer and depends on the environment of the porphyrin molecules, migration of the dye destruction, deaggregation and formation of photoproducts. Photoinduced aggregates are of very poor fluorescence [12] and photochemically stable [21]. Subsequent deaggregation may yield photolabile monomers which are photodynamically active. Possible formation of photoproducts whose emission bands overlap the parent compound may influence the decay of fluorescence [21, 22]. Additionally, different degradation rates appear to be related to different types of binding sites. During light exposure, PpIX molecules appear to move to different binding sites, evidently sites that are more vital for cell survival [23]. Photodynamic Effect Our results showed a clear tendency to a stronger bleaching effect with a higher PDT sensitivity (fig. 3) for the tumor cells (HEC-1A) and a smaller photobleaching with a lower cell toxicity for normal cells (CHO). This correlation between photobleaching and PDT efficacy may primarly reflect the production of highly reactive singlet oxygen during photobleaching of the photosensitizer resulting in photodestruction of essential cellular components. Free radicals may also be involved in the process of PDT [20]. The cytotoxic effect of PDT was different in CHO and HEC-1A cells, both with a similar intracellular induced PpIX content. Sensitivity to PDT was significantly stronger in the tumor cells (HEC-1A) compared to the normal cells (CHO). The efficacy of PDT is thus dependent on the cell type. Lin et al. [13] reported that cultured CHO cells can effectively repair and recover from sublethal damage caused by laser treatment. PDT may induce reversible nuclear and cytoplasmic effects even at sublethal doses. Additionally, some protective singlet oxygen-quenching agent may exist in normal compared to tumor tissue [24]. This may represent a difference in the capacity to destroy the tumor preferentially to normal tissue. Using a light dose of 10 J/cm2, in our study PDT of CHO with ALAinduced PpIX did not cause membrane defects or cell death (trypan blue test). However, the MTT test indicating mitochondrial activity revealed damage of mitochondrial function in 60% of the CHO at the same light dose. Using electron micrographs of cells photosensitized with ALA-induced PpIX, Iinuma


83

et al. [16] confirmed that the primary site of phototoxic damage is the mitochondria, followed by membrane damage with delayed expression. Photodamage to mitochondrial, nuclear and celluar membranes in vitro have been well described [25–28] and lead to increased permeability, reduced transport mechanisms, inhibition of mitochondrial enzymes and cell death [28–31]. It is concluded that the results indicate a correlation between photobleaching and the photodynamic effect of the photosensitizer PpIX. The decay of fluorescence and the cytotoxic effect of PDT are dependent on the cell type. The fact that PpIX degradation depends on multiple biochemical, physical and cellular parameters demonstrates the complexicity of photobleaching mechanisms and effects. References 1 2

3 4 5 6

7

8 9

10 11

12 13 14

Moan J: Effect of bleaching of porphyrin sensitizers during photodynamic therapy. Cancer Lett 1986;33:45–53. Svaasand LO, Potter WR: The implications of photobleaching for photodynamic therapy; in Henderson B, Dougherty TJ (eds): Photodynamic Therapy: Basic Principles and Clinical Aspects. New York, Dekker, 1992, pp 369–385. Svaasand LO, Gomer CJ, Morinelli E: On the physical rationale of photodynamic therapy. SPIE Inst ADV Opt Technol 1990;6:233–248. Mosmann T: Rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays. J Immunol Methods 1983;65:55–63. Ko¨nig K, Schneckenburger H, Ru¨ck A, Steiner R: In vivo photoproduct formation during PDT with ALA-induced endogenous porphyrins. J Photochem Photobiol B 1993;18:287–290. Yang JZ, van Vugt DA, Kennedy JC, Reid RL: Evidence of lasting functional destruction of rat endometrium after 5-aminolevulinic acid-induced photodynamic ablation: Prevention of implantation. Am J Obstet Gynecol 1993;168:995–1001. Wyss P, Tromberg BJ, Wyss MT, Krasieva T, Schell M, Tadir Y, Berns MW: Photodynamic destruction of endometrial tissue using topical 5-aminolevulinic acid (5-ALA) in rats and rabbits. Am J Obstet Gynecol 1994;171:1176–1183. Kennedy JC, Pottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 1992;14:275–292. Wyss P, Steiner R, Liaw LH, Wyss MT, Ghazarians A, Berns MW, Tromberg BJ, Tadir Y: Regeneration processes in the rabbit endometrium: A photodynamic therapy (PDT) model. Hum Reprod 1996;11:1992–1997. Judd MD, Bedwell J, MacRobert AJ: Comparison of the distribution of phthalocyanine and ALAinduced porphyrin sensitizers within the rabbit uterus. Lasers Med Sci 1992;7:203–211. Steinbach P, Weingandt H, Baumgartner R, Kriegmair M, Hofsta¨dter F, Knu¨chel R: Cellular fluorescence of the endogenous photosensitizer protoporphyrin IX following exposure to 5-aminolevulinic acid. Photochem Photobiol 1995;62:887–895. Schneckenburger H, Ru¨ck A, Bartos B, Steiner R: Intracellular distribution of photosensitizing porphyrins measured by video-enhanced fluorescence microscopy. J Photochem Photobiol B 1988;2:355–363. Lin GS, Al-Dakan AA, Gibson DP: Inhibition of DNA and protein cells. Br J Cancer 1986;53: 265–269. Wyss-Desserich MT, Wyss P, Sun CH, Kurlawalla CS, Haller U, Berns MW, Tadir Y: Accumulation of 5-aminolevulinic acid-induced protoporphyrin IX in normal and neoplastic human endometrial epithelial cells. Biochem Biophys Res Commun 1996;224:818–824.


84

15

16 17 18 19

20

21

22 23 24 25

26 27 28

29

30

31

Rebeiz N, Rebeiz CR, Arkins S, Kelley KW, Rebeiz CA: Photodestruction of tumor cells by induction of endogenous accumulation of protoporphyrin IX: Enhancement by 1,10-phenanthroline. Photochem Photobiol 1992;55:431–435. Iinuma S, Farshi SS, Ortel B, Hasan T: A mechanistic study of cellular photodestruction with 5aminolaevulinic acid-induced porphyrin. Br J Cancer 1994;70:21–28. Spikes JD: Quantum yields and kinetics of the photobleaching of hematoporphyrin, photofrin II, tetra (4-sulfonatophenyl)-porphine and uroporphyrin. Photochem Photobiol 1992;55:797–808. Svaasand LO, Wyss P, Wyss MT, Tadir Y, Tromberg BJ, Berns MW: Dosimetry model for photodynamic therapy with topically administered photosensitizers. Lasers Surg Med 1996;18:139–149. Robinson J, de Bruijn HS, van der Veen N, Stringer MR, Brown SB, Star WM: Fluorescence photobleaching of ALA-induced protoporphyrin IX during photodynamic therapy of normal hairless mouse skin: The effect of light dose and irradiance and the resulting biological effect. Photochem Photobiol 1998;67:140–149. Ru¨ck A, Hildebrandt C, Ko¨llner T, Schneckenburger H: Competition between photobleaching and fluorescence increase of photosensitizing porphyrins and tetrasulphonated chloroaluminiumphthalocyanine. J Photochem Photobiol B 1990;5:311–319. Bezdetnaya L, Zeghari N, Belitchenko I, Barberi-Heyob M, Merlin JL, Potapenko A, Guillemin F: Spectroscopic and biological testing of photobleaching of porphyrins in solutions. Photochem Photobiol 1996;64:382–386. Gudgin Dickson EF, Pottier RH: On the role of protoporphyrin IX photoproducts in photodynamic therapy. J Photochem Photobiol B 1995;29:91–93. Moan J, Streckyte G, Bagdonas S, Bech Ø, Berg K: Photobleaching of protoporphyrin IX in cells incubated with 5-aminolevulinic acid. Int J Cancer 1997;70:90–97. Mang TS, Wieman TJ: Photodynamic therapy in the treatment of pancreatic carcinoma: Dihematoporphyrin ether uptake and photobleaching kinetics. Photochem Photobiol 1987;46:853–858. Cozzani I, Spikes JD: Photodamage and photokilling of malignant human cells in vitro by hematoporphyrin-visible light: Molecular basis and possible mechanisms of phototoxicity. Med Biol Environ 1982;10:271–276. Moan J, Christensen T: Porphyrin sensitized photoinactivation of human cells in vitro. Am J Pathol 1982;109:184–192. Kessel D: Yearly review. Hematoporphyrin and HPD: Photophysics, photochemistry and phototherapy. Photochem Photobiol 1984;39:851–859. Perlin DS, Murant RS, Gibson SL, Hilf R: Effects of photosensitization by hematoporphyrin derivative on mitochondrial adenosine triphosphatase-mediated proton transport and membrane integrity of R3230AC mammary adenocarcinoma. Cancer Res 1985;45:653–658. Hilf R, Smail DB, Murant RS, Leakey PB, Gibson SL: Hematoporphyrin derivative-induced photosensitivity of mitochondrial succinate dehydrogenase and selected cytosolic enzymes of R3230AC mammary adenocarcinomas of rats. Cancer Res 1984;44:1483–1488. Hilf R, Murant RS, Narayana N, Gibson SL: Relationship of mitochondrial function and cellular adenosine triphosphate levels to hematoporphyrin derivative-induced photosensitization in R3230AC mammary tumors. Cancer Res 1986;46:211–217. Sandberg S, Romslo I: Porphyrin induced photodamage at the cellular and the subcellular level as related to the solubility of the porphyrin. Clin Chim Acta 1981;109:193–201.

Dr. Marie-The´re`se Wyss-Desserich, Department of Obsterics and Gynecology, University Hospital of Zu¨rich, Frauenklinikstrasse 10, CH–8091 Zu¨rich (Switzerland) Tel. +41 1 255 52 02, Fax +41 1 255 44 33, E-Mail [email protected]


85


............................

Photoproduct Formation during Porphyrin Photodynamic Therapy Karsten Ko¨nig Institute of Anatomy II, Friedrich Schiller University Jena, Germany

Introduction As seen in figure 1, light excitation of porphyrin sensitizers (PS) results in a population of electronic singlet states (1PS). The first electronic state (S1) serves as the initial state for the radiant transition into the ground state (fluorescence, quantum yield =0.1) as well as for the conversion into the triplet state (3PS) by intersystem crossing (ISC). Energy transfer to molecular oxygen (type-II photooxidation) and, to a less extent, also charge and hydrogen transfer (type-I photooxidation) result in the formation of cytotoxic singlet oxygen and oxygen radicals. After the final step of photooxidation the photosensitizer is considered to revert to the ground state (S0, 0PS): 0

PS CCCCC PS CCCCC 3 PS+O2 CCC 1

ht CCCCCCCCB ISC CCCCCCCCB energy transfer CCCB

1

PS PS 0 PS+1O2 3

According to this photophysical scheme, the photosensitizer molecule should be ready again for the next excitation process. It should work like a ‘catalyst’ for singlet oxygen production. However, the photosensitizer does not remain unaffected by the formation of the cytotoxic species. There is an additional photodestruction process directed against the sensitizer molecules leading to bleaching and photoproduct formation [1–8]. Already in 1969, Inhoffen et al. [1] described the oxygen-mediated formation of photoproducts from protoporphyrin (PP). At the end of the 1980s, photoproducts from hematoporphyrin (Hp) and the porphyrin mixture (Hp derivative, HpD) were found [2–4]. Ko¨nig et al. [7, 9] and Rotomskis et al. [8] described photoproducts from naturally

Fig. 1. UV and short wavelength visible light exposure radiation results in the population of higher excited electronic states followed by the transition into the first singlet state. Red light exposure at B630 nm leads to a direct population of the S1 state. For metal-free porphyrin photosensitizers, about 90% of excitation events result in the population of the first triplet state. The remaining 10% induce heat and fluorescence. Energy transfer induces singlet oxygen formation which causes cell and photosensitizer destruction.

occurring porphyrins in human skin and from aminolevulinic acid (ALA)induced endogenous porphyrins in the tumor. In this article we present a review of the studies related to porphyrin photoproducts.

Materials and Methods Measurements were performed on solutions of PP IX, Hp (Porphyrin Products Inc. Logan, Utah), tetramethyl-Hp (SeeLab, Germany), and various HpDs (Photofrin from Ipsen Biotech, Photosan from Seelab, and Halle University). The PP IX and coproporphyrinproducing bacterium Propionibacterium acnes were used without further porphyrin incubation. Cell experiments were conducted using Chinese hamster ovary cells, Ehrlich carcinoma cells, and RR1022 cells. A variety of tumor-bearing nude mice (e.g. xenotransplanted G3 bladder tumor) were injected with HpD or ALA (Sigma). Fluorescence detection during photodynamic therapy (PDT) was also performed on ALA- and Photosan-incubated patients suffering from psoriasis and mycosis fungoides. PDT fiber-guided radiation at 633 nm (100 mW/cm2) was provided by dye lasers or a 50-mW He-Ne laser. PDT was interrupted for 1 s (mechanical shutter) for fluorescence measurements during PDT. Fluorescence detection was performed with a home-build fiberoptical sensor (central 200-lm quartz excitation fiber and 8 peripheral detection fibers) with a slit-shaped distal end in combination with a polychromator equipped with long-pass filters (remitted light rejection) and an optical multichannel analyzer. Photosensitizer fluorescence was excited with the spectrally purified (prism pair) 407 nm output of a krypton ion laser to match the major porphyrin absorption band (Soret band).

Photoproduct Formation during Porphyrin PDT

87

a

b Fig. 2. Excitation of electronic transitions of porphyrin solutions (a Hp in water; b PP IX in DMSO), results in photobleaching and formation of fluorescent photoproducts.

Ko¨nig

88

3a

3b


89

3c

3d

Fig. 3. Photoproduct formation depends on singlet oxygen. D2O (long 1O2 lifetime) increases the photoconversion rate, whereas the 1O2 quencher NaN3 avoids sensitizer photodestruction.

Ko¨nig

90

a

b Fig. 4. PDT-induced photoproduct formation in ALA-incubated G2 bladder tumors (a) and non-labeled Propionibacteria acnes (b).

Results Porphyrin excitation results in irreversible spectral modifications due to photobleaching and photoproduct formation such as fluorescence decrease at the major fluorescence bands and fluorescence increase at 642 nm (Hp, HpD) and 670 nm (PP IX; fig. 2). The conversion into fluorescent photoproducts needs singlet oxygen (fig. 3).


91

a

b Fig. 5. Fluorescence modifications during PDT of ALA-incubated psoriasis patients (a) and Photosan-incubated mycosis fungoides patients (b).

ALA-incubated cells and tumor-bearing mice [7] as well as nonincubated, porphyrin-producing bacteria [9] reveal similar PDT-induced PP IX modifications as in hydrophobic solution (fig. 4). However, not always was an absolute fluorescence increase in the region between the major PP IX/HpD fluorescence peaks detectable. Nevertheless, the bleaching rates exhibited a clear dependence

Ko¨nig

92

Fig. 6. Singlet-oxygen associated PP photoconversion into chlorin-type porphyrin isomers [6].

on emission wavelength due to photoproduct formation. As seen in figure 5, the fluorescence decrease at 635 nm from an ALA-incubated psoriasis patient [10] by 65% (fourth curve) differed significantly from the behavior at 670 nm (=15% decrease). The Photosan-incubated patient revealed a fluorescence intensity increase due to photoproduct formation already at 5 J/cm2 fluence. Photoproducts are chlorin-type agents, their formation may occur by singlet oxygen involved vinyl group oxidation and hydrogen rearrangement [1] (fig. 6). Chlorin-type photoproducts possess relatively strong electronic transitions in the far red, the spectral range of high light penetration depth in biological tissues. Indeed, the PDT-induced modifications in fluorescence


93

Fig. 7. Two-step PDT (633 nm, 75 J/cm2 exposure followed by 665–670 nm, 75 J/cm2 irradiation) led to worse results compared to conventional one step 633 nm treatment (150 J/cm2).

correlate with absorption changes. A new absorption band at 640 (Hp/HpD in aqueous solution) and 665 nm (PP IX in DMSO) occurs. In order to prove the photodynamic activity of these photoproducts, a two-step PDT on xenotransplanted tumors of ALA-incubated nude mice was performed. In a first step, tumors were exposed to 633 nm radiation at 75 J/cm2 fluence. This phototreatment resulted in the formation of fluorescent photoproducts. Later on the dye laser was tuned to 665–670 nm. Another 75-J/cm2 phototreatment was performed. The tumor response to this 2-step treatment was compared to conventional 1-step PDT (633 nm, 150 J/cm2). As indicated in figure 7, better therapeutical effects were obtained in the case of one-step 633-nm PDT. Also the incubation of pre-exposed HpD solution and PP IX solution with a high amount of photoproducts in combination with broadband (600–680 nm) exposure did not result in improved PDT efficiency. Photoproducts therefore appear to be agents with a low photodynamic activity [11].

Discussion Photobleaching and the photo-induced formation of new absorption bands as well as fluorescence bands are the result of singlet oxygen action resulting in the photoconversion of porphyrin photosensitizers. Far-red PDT

Ko¨nig

94

based on photoproduct excitation does not improve the treatment efficiency compared to conventional B630 nm porphyrin PDT. However, photoproduct detection can be used as an indication of photo-induced formation of cytotoxic singlet oxygen. On-line fluorescence measurements during PDT may be a useful tool to gain information on the PDT efficiency.

References 1 2 3 4

5 6 7 8 9

10

11

Inhoffen HH, Brockmann H, Bliesener, KL: Photoprotoporphyrine und ihre Umwandlung in Spirographis- sowie Isospirographis-porphyrin. Liebigs Ann Chem 1969;730:173–185. Moan J, Kessel D: Photoproducts formed from photofrin II in cells. J Photochem Photobiol B 1988;1:429–436. Valat P, Reinhartt GD, Jameson DM: Application of time-resolved fluorometry to the resolution of porphyrin-photoproduct mixtures. Photochem Photobiol 1988;47:787–790. Rotomskiene J, Kapociute R, Rotomskis R, Jonusauskas G, Szito T, Nishnik A: Light-induced transformations of hematoporphyrin-diacetate and hematophorphyrin. J Photochem Photobiol B 1988;2:373–379. Ko¨nig K, Wabnitz H, Dietel W: Variation in fluorescence decay properties of hematoporphyrin derivative during its conversion to photoproducts. J Photochem Photobiol B 1990;8:103–111. Dietel W, Ko¨nig K, Zenkevich E: Photobleaching of HpD fluorescence and formation of photoproduct in vivo and in solution. Lasers Life Sci 1990;3:197–203. Ko¨nig K, Schneckenburger H, Rueck A, Steiner R: In vivo photoproduct formation during PDT with ALA-induced endogenous porphyrins. J Photochem Photobiol B 1993;18:287–290. Rotomskis R, Bagdonas S, Streckyte G: Spectroscopic studies of photobleaching and photoproduct formation of porphyrins used in tumour therapy. J Photochem Photobiol B 1996;33:61–67. Ko¨nig K, Meyer H, Schneckenburger H, Ru¨ck A: The study of endogenous porphyrins in human skin and their potential for photodynamic therapy by laser-induced fluorescence spectroscopy. Lasers Med Sci 1993;8:127–132. Boehnke WH, Ko¨nig K, Kaufmann R, Scheffold W, Pru¨mer O, Sterry W: Photodynamic therapy in psoriasis: Suppression of cytokine production in vitro and recording of fluorescence modifications during treatment in vivo. Arch Dermatol Res 1994;286:300–303. Ko¨nig K, Felsmann A, Dietel W, Boschmann M: Photodynamic activity of the HpD photoproducts. Stud Biophys 1990;138:219–228.

Dr. Karsten Ko¨nig, Institute of Anatomy II, Friedrich Schiller University Jena, Teichgraben 7, D–07743 Jena (Germany) Tel. +49 3641 938560, Fax +49 3641 938552, E-Mail [email protected]


95


............................

Optical Dosimetry for Photodynamic Therapy L.O. Svaasand a

b

a, b

, P. Wyss b, M.K. Fehr b, A.L. Major b, Y. Tadir b

Department of Physical Electronics, Norwegian University of Science and Technology, Trondheim, Norway, and Beckman Laser Institute and Medical Clinic, University of California, Irvine, Calif., USA

Light as Electromagnetic Interaction Electromagnetic radiation occurs when the state of movement of an electric charge is changed; emission takes place during the acceleration process. The energy of radiation propagates into the surrounding space as electromagnetic waves. These waves consist of two mutually orthogonal fields; the electric field E (V/m) and the magnetic field B (T). Both fields are orthogonal to the direction of propagation and they propagate in free space at the velocity of light, i.e. c>3 · 108 m/s. The frequency of the wave is dependent on the time scale of the acceleration, e.g. if a charged particle is moved harmonically back and forth during an interval of time, T, it will emit waves at a frequency of f>1/T (Hz). The interval in space between neighbor positions corresponding to exactly the same state of movement of the emitting charge will therefore be: k>cT>c/f (m). The waves can also be represented by a flux of electromagnetic particles; the photons. Each photon has an energy e>hf (J) and a momentum p>h/k (kgm/s) where h is Planck’s constant, h>6.6 · 10Ö34 Js. Emission of electromagnetic energy also takes place when bound electrons are moved from a higher state of energy to a lower one. The energy of the emitted photon is then equal to the energy difference, De, between two states, i.e. the frequency of radiation is f>De/h. The acceleration of a harmonically oscillating dipole at frequency, f, is proportional to f 2. Thus, the radiated electric and magnetic fields, which are

Fig. 1. Electric field radiating in horizontal direction from a positive charge oscillating vertically around a negative charge. The graph shows the first 3 wavelengths of the field.

both determined by the acceleration, are proportional to the second power of the frequency. The emitted power, which is proportional to the product of the fields, is determined by f 4, corresponding to a wavelength dependence of 1/k4 (fig. 1). Medical applications of the electromagnetic energy cover a broad spectrum from electrical nerve stimulation around 1–103 Hz to gamma rays at 1020–1021 Hz, e.g. positron emission tomography (PET scan) is based on gamma rays emitted at 2.5 · 1020 Hz during positron-electron recombination. The visible spectrum covers the wavelength spectrum from violet light at 400 nm to deep red light at about 700 nm, corresponding to frequencies of 7.5 · 1014 and 4.3 · 1014 Hz, respectively. The photons can also be characterized by the energy. The single photon energies for 400 and 700 nm radiation are 5 · 10Ö19 J (3 eV) and 2.8 · 10Ö19 J (1.8 eV), respectively (fig. 2). The medical application of light ranges from ultraviolet radiation at 194 nm for cornea reshaping with ArF excimer lasers to infrared laser surgery at 10,600 nm with CO2 gas lasers. Other typical examples are PUVA (psoralen UVA) treatment of psoriasis with UV radiation close to 400 nm, pulsed dye laser treatment of port wine stains with yellow light at 585 nm, and photodynamic therapy (PDT) treatment of tumors with photosensitizers and red light.

Optical Dosimetry for Photodynamic Therapy

97

Fig. 2. Medical applications of electromagnetic energy. Columns from left to right: photon energy, wavelength, and frequency.

Solar irradiation is very powerful: the maximum power density of solar irradiation at the surface of the earth is about 1 kW/m2. This corresponds to an exposure of the human body to about 1 kW of power, which is 10 times larger than the metabolic heat generation rate of about 100 W. The electric and magnetic fields of the solar radiation are about E>600 V/m and B>2 lT, respectively. These fields are of the same order of magnitude as the corresponding 50-Hz (or 60-Hz) fields at ground level from a highvoltage power transmission line. The electric field is, however, much less than electric fields at ground level due to atmospheric activity, and the magnetic field is much less than the earth magnetic field of about 70 lT. The earth magnetic field is, on the other hand, significantly smaller than the DC magnetic fields of B>1–10 T used in NMR techniques. The radiation pressure of solar irradiation, i.e. the mechanical pressure exerted on an object by the light, is very weak. The maximum pressure is about 3 lPa which is only three parts per million of the atmospheric pressure of 100 kPa. However, the force from highly focused

Svaasand/Wyss/Fehr/Major/Tadir

98

laser beams, where the power density at the focal spot is in the range of 1 GW/m2, is strong enough to trap and manipulate swimming spermatozoa.

Propagation of Light in Matter The electric field of an electromagnetic wave will interact with any electrical charge or neutral molecule. The field will exert a force on every charge; in a neutral molecule the positive part of the molecule will be pulled along the field, whereas the negative part will be pulled in the opposite direction. The acceleration of the charges thus varies harmonically with the frequency of the incoming field; the molecule itself becomes a secondary emitting oscillating dipole. The power of the reemitted wave, i.e. the scattered light, is isotropic in the plane, normal to the direction of the incident field and proportional to f 4$1/k4 (fig. 1). This type of scattering, initiating from either a single oscillating molecule or from a cluster of molecules much smaller than the optical wavelength, is called Rayleigh scattering. The fields emitted from the various single molecules start to interfere strongly when the dimension of the molecular cluster becomes comparable to the wavelength. The resulting emitted field will then be directed more forward, and the strong frequency dependence will be reduced. This type of scattering is called Mie scattering. When the density of the molecules in the scattering medium is uniformly distributed, and the dimension of the medium is much larger than the wavelength, the reemitted waves will interfere constructively only in two directions corresponding to a refracted wave and a reflected wave. The refracted wave will propagate in matter at the reduced velocity of cm>c/n, where n is the index of refraction (the index of refraction is in reality a measure of the dipole density). The reflected wave, which originates from constructive interference of the oscillating dipoles at the surface layer, propagates back on free space with the same velocity of the incident wave. These types of scattering are illustrated in figure 3. The phenomena are also easily observed in nature. The blue color of the sky is due to scattering of solar light from randomly distributed molecules in the atmosphere. The wavelength of blue light is in the range of 450–500 nm whereas red light corresponds to 610–700 nm. The amount of scattered blue light is therefore about factor of 4 larger than the red light. The solar rays themselves become more red as the blue light is scattered out of the beam; this phenomenon is most pronounced at sunset when the distance traveled by the light rays through the atmosphere is at its maximum.


99

Fig. 3. Scattering and refraction of light.

The white color of the clouds is due to Mie scattering from water droplets. These droplets are comparable or larger than the optical wavelength, and the scattering is therefore almost wavelength-independent. Refraction and reflection from a very large optical medium are observed when light rays hit a water pool; some of the light is specularly reflected and the rest of the energy is transmitted into the water as refracted rays. The angle between the refracted wave in water in water and the normal surface is smaller than the corresponding angle for the incident beam. On the other hand, if the incident ray arrives at the water-air interface from the water side at an angle equal or larger than the angle corresponding to 90º in air, the energy of the ray will be totally reflected.

Light Propagation in Tissue The optical structures of biological media are very complex. The optical properties of subcellular organelles, the membrane and the extracellular


100

matrix vary strongly. The index of refraction varies from that of water, i.e. n>1.33 to about n>1.45 for proteins. The geometric dimensions range from typically 10–20 lm for cells and down to the 1 lm region for subcellular organelles such as the mitochondria. The large density of the scattering object with dimensions comparable and larger than the optical wavelength gives rise to a strong Mie-type scattering. This scattering is, as discussed above, only weakly dependent on the frequency or wavelength. Propagation of light in tissue thus resembles propagation of light in random media such as dense fog or clouds. An important exception from this rule is ocular media, i.e. the cornea,the lens and the vitreous body. The transparency of the lens is due to the regular alignment of the lens fibers; if this structure is destroyed the lens becomes highly scattering and opaque such as in the case of cataract. Parts of the eye also exhibit a strong wavelength-dependent scattering; the color of blue eyes is caused by light scattered in submicron-sized scatterers situated above the melanin-containing layer of the iris. Blue light is therefore strongly backscattered whereas light at longer wavelengths penetrates deep enough to be absorbed in the melanin layer. The decrease in amount of unscattered photons in a beam of light can be expressed by [1]: eÖls x

(1)

where ls is the scattering coefficient, x is the distance traveled by the beam, and e is the base number of the natural logarithm system, i.e. e>2.71. The inverse scattering coefficient corresponds to the distance traveled until only 1/e>0.37 of the photons are left unscattered. The wavelength dependence of scattering coefficient can be approximated by [2]: ls, k > ls, kr

AB kr k

N

,

(2)

where ls, k is the scattering coefficient at some arbitrary wavelength k; ls, kr is the coefficient at a reference wavelength kr , and N is a number which characterizes the wavelength dependence. The value is in the range of NB1 for most tissues, i.e. much smaller than the value of N>4 for Rayleigh scattering. The scattering is usually a strongly forward-directed Mie scattering as shown in figure 3. Typical average cosine of the scattering angle is in the range of g>0.8–0.9, corresponding to an average scattering angle in the range of 26–37º. However, some cells such as erythrocytes, exhibit an extraordinary strongly forward scattering [1]. The average scattering angle is only 6º and the corresponding g factor is 0.995.


101

Fig. 4. Light propagation in absorption-dominated and in scattering-dominated tissues. Small solid and open circles represent absorbers and scatterers, respectively. Larger open circles represent target molecules.

In a highly scattering medium the photons will undergo multiple scattering, and the light becomes almost isotropic. The effective scattering coefficient for isotropic scattering, i.e. the redued scattering coefficient, is given by lsE>ls (1Ög). Thus, on average it is required with 5–10 scattering events before the distribution is isotropic. The decrease in the number of photons due to absorption in a nonscattering medium can be expressed by Beer’s law, eÖla x,

(3)

where la is the absorption coefficient. The inverse absorption coefficient corresponds to the distance traveled until 1/e>0.37 of the photons are left. The propagation of light in tissue is visualized in figure 4. Figure 4a characterizes the absorption-dominated case. Most of the photons are then directly absorbed, and only a few are scattered out of the beam. Figure 4b gives the highly scattering case where the photons propagate in zigzag-formed


102

paths. There are several important consequences of this path: the probability for being absorbed increases even though the scattering process itself is lossless, backscattering enhances the light level in the tissue proximal surface, and photons are diffusely scattered back into the air. The light in tissue is not collimated as in the absorption-dominated case, but it reaches a target molecule from all directions. An adequate measure of light intensity in cases where the photons arrive from all directions is the radiant energy fluence rate, u (W/m2). This quantity is defined by the radiant power incident on a small sphere, divided by the cross-sectional area of that sphere. The corresponding energy, the radiant energy fluence, W (J/m2), is defined by the fluence rate multiplied with the time of exposure [3]. The light distribution in the absorption-dominated case can be expressed as: u>(1Öcs )IeÖ(la+ls )x,

(4)

where I is the power density (W/m2) of the incoming beam and cS is the specular reflection coefficient. The specular reflection, which is caused by the difference in index of refraction between tissue and air (fig. 3), is about 3%. The optical absoption in tissue is very high in the ultraviolet region, i.e. for a wavelength shorter than 400 nm, and in the near-infrared region for a wavelength larger than about 1,500 nm. The short wavelength limit is determined primarily by protein absorption and the long wavelength limit by water absorption (fig. 5). The absorption coefficients for these wavelength regions are much larger than the scattering coefficient; typical values for the reduced scattering coefficient are in the range of lsE>1,000–2,000 mÖ1. Light propagation is therefore absorption-dominated [2]. The absorption coefficient in the 420- to 1,200-nm region is, on the contrary, usually much smaller than the scattering coefficient. This is in particular true for wavelengths in the 610- to 1,200-nm region, where blood absorption is about two orders of magnitudes smaller than for the 420- to 600-nm region. This region of maximum penetration is sometimes referred to as the ‘transmission window of tissue’. Light propagation in the 420- to 1,200 nm wavelength region is therefore in general scattering-dominated, and this is in particular true for the red/nearinfrared ‘transmission window’. The propagation of light in highly scattering tissue is very composite. In particular this is true for the region proximal to the air-tissue interface where light from the incident collimated beam is scattered to a more isotropic distribution [1, 2, 4]. However, the distribution in deeper layers exponentially decays with depth. If the incident light is assumed to be scattered in an infinite thin surface layer, the distribution can be approximated by:


103

Fig. 5. Absorption coefficient of tissue constituents: hemoglobin (Hb), oxygenated hemoglobin (HbO2, 1% blood in tissue, hematocrit 0.4), melanin (average epidermal melanin content in north European skin), and water.

J

u>(1Öc) 3

lsE Ö)3l lE x a s >k IeÖx/d, Ie p la

(5)

where the factor

J

kp > (1Öc) 3

lsE la

accounts for the phenomenon that the fluence rate close to the surface is larger than the power density of the incoming beam. The total reflection coefficient, c>cs+cd , is given by the sum of the specular coefficient, cs , and the diffuse reflection coefficient, cd . The penetration depth d is defined as the inverse of the effective attenuation coefficient, i.e. Ö1

d>( J3la lsE ) .

The penetration depth is equal to a distance corresponding to a decrease in the fluence rate to 1/e>0.37 of the initial value. The fluence rate close to the surface is larger than the irradiation both due to backscattering and because a fraction of the diffusely backscattered light is totally reflected when reaching the tissue-air interface. The presence of backscattering thus enhances the light level close to the surface, but the


104

decay with distance is increased since the zigzag path of the photons increases the probability for absorption. Scattering also enhances the reflection; light is not only specularly reflected from the surface but also diffusely reflected from within the tissue. The diffuse reflected light is quite significant, e.g. the diffuse reflection from Caucasian skin in the red/near-infrared region is in the range of 50–60%, and the corresponding value for the blue/green/yellow region is about 20–30%. The corresponding specular reflection is only 3% for both regions. The light distribution in the scattering-dominated case is characterized in figure 6, where the graphs give the fluence rate versus distance from an irradiated surface. The curves are calculated from a mathematical model which also takes the details close to the surface into account, but the tails of distributions correspond to the expression given in equation 5. The results given in figure 6a, b correspond to the red/near-infrared region where the penetration depths are in the range of d>2–4 mm. Figure 6a corresponds to irradiation of a free tissue surface in air, whereas figure 6b corresponds to the case where the index of refraction of the tissue is matched to the outer medium, e.g. with a transparent gel of the same index of refraction as the tissue. The fluence rate is then lower because of missing internal reflection from the surface. The fluence rate at the surface region is about 3–4 times larger than the incident power density for the free surface case, and about 1.5–2 times for the matched case. However, the fluence rate remains higher than the incident irradiation to about the same depths for the two cases, i.e. to about 3–7 mm, dependent on the penetration depth. Figure 6c shows the corresponding results for blue/green/yellow region. The tissues are assumed to be the same as in figure 6a with a blood fraction of 2%. The high blood absorption in the blue/green/yellow region results in an absorption coefficient that is completely dominated by the blood fraction. The tissues, which had penetration depths of d>2, 3 and 4 mm, respectively, at 630-nm wavelength, have almost the same penetration depths at 514 nm, i.e. d>0.9–1 mm. The enhancement factor of 3–4 for the free tissue surface in the red/near-infrared region drops down to about 2, and the fluence rate becomes smaller than the irradiated power density at about 1.1–1.2 mm from the surface. The lower curve in figure 6c gives the light distribution in a 2% solution of blood with no scattering. The enhancement of the light level close to the surface is lost, but the level is higher than in the presence of scattering for depths larger than about 2 mm. Propagation of blue/green/yellow light on whole blood will be absorption-dominated, and the penetration depth will be in the range of d>25–75 lm [2]. The highest penetration depths are found in tissues with low scattering and absorption; typical examples of such tissues are retinoblastoma and neonatal


105

6a

6b


106

6c Fig. 6. Fluence rate versus distance from irradiated surface. Power density of incident beam normalized to unity, i.e. I>1 W/m2. Scattering properties: ls>10 mmÖ1, g>0.9. a Air/ tissue interface (630 nm). Penetration depth from lower to upper curve: d>2, 3 and 4 mm. b Optically matched interface (630 nm). Penetration depth from lower to upper curve: d>2, 3 and 4 mm. c Air/tissue interface (515 nm). Blood volume fraction 2%. Upper three curves correspond (from lower to upper) to tissues with penetration depths of d>2, 3 and 4 mm at 630 mm. The lower curve corresponds to a 2% blood solution with no scattering.

human brain. The minimum penetration depths are found in melanotic melanomas where the penetration depth for red light is d>0.5 mm, i.e. about 15–25% of the value found for most other tissues (table 1).

Light Distribution in Body Cavities Light can easily be coupled into tissue by use of optical fibers, and the fibers can be inserted either directly into tissue or into cavities of the body. Figure 7 gives the various geometries of the body cavities. In the planar case in figure 7 light is launched from a planar source with an emission, Ppl per unit area. The direct irradiated power density per unit area of the tissue surface is then I>Ppl /2. The reflected light will bounce back and forth in the cavity, building up a diffuse light distribution within the cavity. The light distribution in tissue can then be expressed (see Appendix):


107

Table 1. Optical penetration depth (mm) in tissue Tissue

515 nm

633 nm

660 nm

1,064 nm

Adlt brain, human (ex vivo) Neonatal brain, human (ex vivo) Astrocytoma, human (surgically resected) Glioblastoma multiforme, human (surgically resected) Brain metastasis, OAT cell from lung, human (ex vivo) Epithelium carcinoma, bladder, human (ex vivo) Squamous cell carcinoma of the lung, human (ex vivo) Human retinoblastoma in athymic mouse (in vivo) Murine mammary carcinoma, C3H mouse strain (in vivo) Murine melanotic melanoma, C57 mouse strain (in vivo)

0.4–0.6 1.1–1.7 0.5–1.3 1.4 0.6 – – 1.6 1.1 –

– – – – – 2.2–2.3 1.6 3.3 2.0 0.5

1.2–1.6 3.7–5.4 2.0–3.0 – 1.3 – – 3.6 2.3 0.5

3.2–4.3 7.1–8.8 3.0–6.3 6.6 2.8 5.2 – 7.5 3.7 2.0

According to Svaasand and Gomer [8].

J

u> 3

lsE Ppl Öx/d e >kpl IeÖx/d, la 2

(6)

where kpl is a coupling coefficient given by:

J

kpl > 3

lsE . la

(7)

The coupling coefficient kpl is a factor of (1Öc)Ö1 larger than the one for a free surface given in equation 5. The reason is that the reflected light here is not lost, but will be returned after multiple reflections in the cavity. Corresponding coefficients kcyl and ksph can be introduced for cylindrical and spherical geometry, respectively. However, in these cases the values are not only dependent on tissue properties, but also on the ratio between cavity radius, a, and penetration depth, i.e. on a/d. This dependence is visualized in figure 8 (equations A5, A8). The coupling coefficients for the cylindrical and spherical system are both smaller than for the planar case, but they approach that level for aAd. Light distributions for different cavity geometries are shown in figure 9 (equations 6, A3, A6). The tissue parameters are the same as for the surface irradiated case given by the lower curve in figure 6b, i.e. ls>10 mmÖ1,

g>0.9 and d>2 mm.

The coupling coefficients in figure 9 vary from kpl>6 for the planar case and kcylB4–5 for the cylindrical case, to ksph>3–4 for the spherical case.


108

Fig. 7. Various cavity geometries.

The fluence rate for the planar case decays exponentially with distance, whereas the decreases for the cylindrical and spherical cases are stronger due to geometrical factors. The behavior for the cylindrical and spherical case will, however, approach that of the planar one when the dimension of the cavity is much larger than the penetration depth (see appendix for a more detailed discussion). However, it should be noted that the enhancement of the fluence rate due to multiple scattering in the cavity requires a clear, loss-less fluid or air in the cavity. The absorption coefficient of the liquid, la, f must satisfy the condition 2a la, f@1. Blood should therefore be prevented from seeping into the cavity fluid; a 1% blood contamination in a 10-cm diameter cavity will give a value of 2a la,fB0.1 at 630-nm wavelength.


109

Fig. 8. Ratio between coupling coefficient for cylindrical and spherical cavities versus ratio between cavity radius and optical penetration depth.

Photochemical Reactions The rationale of photodynamic therapy is based on the cytotoxic action of singlet oxygen, 1O2. Singlet oxygen is an excited state of oxygen which differs from the ordinary ground-state triplet oxygen, 3O2, in energy and in type of electron binding. The difference in energy corresponds to the near-infrared optical wavelength of 1,270 nm; thus the photon energy at all wavelengths of visible light is high enough to furnish the required energy of excitation. However, photons do not have the ability to make the required change in type of electron binding, i.e. to change the elecronic spin. Therefore, the triplet to singlet, or singlet to triplet, transition is optically forbidden. This is where the photosensitizer comes in; the photosensitizer mediates the transition in such a manner that the excitation of singlet oxygen becomes acceptable. The details of the process are shown in figure 10. The excitation of the photosensitizer molecule is shown on the left-hand side of figure 10. The photon energy excites the singlet dye molecule to an excited singlet state. Some of the excited molecules return to the ground state giving off energy as fluorescent light, whereas others will interact with surrounding molecules. During an interaction process, which is referred to as intersystem crossings, the electron spin is changed and the photosensitizer


110

a

b Fig. 9. Fluence rate versus depth in tissue. Direct irradiated power density normalized to unity, i.e. I>1 W/m2. Tissue properties: ls>10 mmÖ1, g>0.1, d>2 mm. a Radius of cavity a>5 mm. b Radius of cavity a>2 mm.


111

Fig. 10. Excitation of photosensitizer and singlet oxygen generation.

enters its excited triplet state. The sensitizer molecule becomes optically trapped in the triplet state because return to the singlet ground state is optically forbidden. However, the triplet-state molecule has both the energy and the spin which is required to excite ground-state triplet oxygen to the singlet state. The energy and spin are therefore delivered to the oxygen molecules, and the photosensitizer returns to its singlet ground state. Although the excited oxygen molecules are optically trapped in the singlet state, they are chemically highly reactive. The chemical lifetime of generated singlet oxygen is very short compared to the time scale of diffusion. A proper cytotoxic action therefore requires the simultaneous in situ presence of free oxygen, photosensitizer, and light. There are several processes that might quench the production of singlet oxygen. The process itself consumes free oxygen, and it will easily deplete the reservoir of oxygen. Further on, singlet oxygen can react directly with the photosensitizer and decompose it [5]. The degradation process is very composite, but a useful concept is to characterize the photodegradation of the sensitizer with a bleaching fluence [6, 7]. The bleaching fluence corresponds to a value where 1/e>0.37 of the initial concentration is photodegraded, and a fluence of two times the bleaching value reduces the amount of active sensitizer to (1/e)2B0.1. Most photosensitizers have bleaching fluences of less than 50–60 J/cm2, and any increase in the in situ optical dose above 100–120 J/cm2 is therefore very inefficient.


112

Dosimetry The cytotoxic reaction is dependent on the in situ light dose, i.e. on the fluence w (J/m2 or J/cm2). This fluence is, as discussed previously, not only dependent on the irradiated dose, but also on the optical properties of the tissue and on the geometry of the light application system. All these parameters must therefore be considered when determining the proper treatment protocol. The fluence rate in an in vitro cell suspension with negligible scattering is very close to the irradiated power density, whereas the fluence in the upper layer of a surfaceirradiated tissue is about 3–4 times larger. Further on, the surface fluence can be up to 4–6 times larger if the irradiated surface is a part of a closed body cavity with linear dimensions much larger than the optical penetration depth. Thus, application of the same irradiant dose to an in vitro cell suspension as to a tissue surface yields an up to 3–4 times larger in situ dose in the surface layer of the tissue. In fact, the dose in tissue can be higher than the cell suspension dose down to depths of 3–7 mm (fig. 6b). When relating optical doses for the free surface-irradiated case to tissue surfaces of a body cavity, the dose in the cavity case can be twice as large as for the free surface case. However, the enhancement of the optical fluence in the cavity case is dependent on multiple transmission through the liquid, and the dose might easily be less than for the surface-irradiated case if the liquid in the cavity is absorbing. The maximum acceptable irradiated fluence rate is limited by temperature rise. A safe fluence value for an uncooled tissue surface is about 100 mW/cm2, which corresponds to solar irradiation with the sun in zenith. The fluence can usually be increased to 150 mW/cm2, but a value of ?200 mW/cm2 will easily result in a tissue temperature of more than 45 ºC. An irradiated dose of 100 mW/cm2 over an exposure period of 15 min will administer an irradiant dose of 90 J/cm2. This corresponds to a typical PDT protocol with light at some wavelength in the 630- to 670-nm region, dependent on the specific photosensitizer used. The in situ dose will then be above 90 J/cm2 down to 3–7 mm, dependent on the optical properties of the tissue (fig. 6b). The total optical power from a commercially available source is dependent on the wavelength, and the maximum power in the 630- to 635-nm band is about 5 W. This power enables simultaneous delivery of an irradiant dose of 90 J/cm2 over 50 cm2 during 15-min exposure, but the area can be enlarged to about 0.5 m2 for 24 h of exposure. A large portion of the total inner surface, the human abdominal cavity and its organs (2 m2) can in principle be treated simultaneously if an exposure time of up to several hours is acceptable. The singlet oxygen generation efficiency is wavelength-dependent, e.g. the excitation efficiency for a hematoporphyrin derivative is a factor of 2–3 larger


113

at 515 nm than at 630 nm. The optical coupling coefficient at 515 nm is, as discussed previously, about a factor of 2 smaller than for 630 nm radiation. The excitation efficiency per unit irradiated optical dose is therefore expected to be about the same. The penetration depth is, however, only about d>1 mm, and the cytotoxic action will be limited to a depth of =1 mm from the surface. The maximum available power at 515 nm is, however, typically about 20 W, and the total inner surface of the abdomen can therefore be exposed to an irradiant dose of 90 J/cm2 for 24 h of exposure.

Appendix Light distribution in highly scattering media can be characterized by the optical diffusion approximation. The flux of diffuse photons is then proportional to the gradient of the diffuse photon density. This condition can be expressed as [1, 2]: !j>Ö 1 gradu, 3lsE

(A1)

where !j is the diffuse photon flux vector. The conservation of energy can be expressed as: divj!>Öla u+q ,

(A2)

where q is the source density of diffuse photons. The calculations will be based on a simplifed model where all light is assumed to be scattered to an almost isotropic distribute in the surface layer itself. This source density in equation A2 is then equal to zero, and the source is taken into account by setting the normal component of the flux vector at the surface equal to the incident power density, i.e. jn>I. The distribution in the cylindrical case, where the emission comes from a cylindrical source can then be expressed by (equations A1, A2):

J

u> 3

lsE Pcyl K0((a+x)/d) K0((a+x)/d) >kcyl I , la 2pa K1(a/d) K0(a/d)

(A3)

where Pcyl (W/m) is the power per unit length of a line source along the cylinder axis. The corresponding direct irradiated power density at the cavity wall is I>

Pcyl (W/m2), 2pa

where a is the radius. The coupling coefficient kcyl is given by,

J

kcyl > 3

lsE K0(a/d) K0(a/d) >kpl . la K1(a/d) K1(a/d)

(A5)

The functions K0 and K1 are modified Bessel functions of the zero’th and first order, respectively.


114

In the spherical case a total amount of power Psph (W) is emitted from the source at the center, and the corresponding direct irradiated power density at the wall is I>

Psph . 4pa2

The fluence rate becomes,

J

u> 3

lsE a a/d Psph a eÖx/d>ksph I eÖx/d, a+x la (1+a/d) 4pa2 a+x

(A6)

where the coupling ksp is given by,

J

ksph > 3

lsE a/d a/d >kpl . la 1+a/d 1+a/d

(A7)

References 1 2

3 4 5 6 7 8

9 10 11

Ishimaru A: Wave Propagation and Scattering in Random Media. New York, Academic Press, 1978. Svaasand LO, Norvang LT, Fiskestrand EJ, Stopps EKS, Berns MW, Nelson JS: Tissue parameters determining the visual appearance of normal skin and port-wine stains. Lasers Med Sci 1995;10: 55–65. International Standard, ISO 31/6-1980. Keijzer M, Star WM, Storch PRM: Optical diffusion in layered media. Appl Optics 1988;27: 1820–1824. Georgakourdi I, Nichols MG, Foster TH: The mechanism of Photofrin photobleaching and its consequences for photodynamic therapy. Photochem Photobiol 1997;65:134–144. Potter WR, Mang TS, Dougherty TJ: The theory of photodynamic therapy dosimetry: Consequences of photodestruction of sensitizer. Photochem Photobiol 1987;46:97–101. Svaasand LO, Wyss P, Wyss M-T, Tadir Y, Tromberg BJ, Berns MW: Dosimetry model for photodynamic therapy with topically administered photosensitizers. Lasers Surg Med 1996;18:139–149. Svaasand LO, Gomer C: Optics of tissue; in Mu¨ller GJ, Sliney DH (eds): Dosimetry of Laser Radiation in Medicine and Biology. Bellingham, SPIE Optical Engineering Press, 1989, vol IS5, pp 114–132. Dougherty TJ: Photodynamic therapy (PDT) of malignant tumors. CRC Crit Rev Oncol Hematol 1984;2:83–116. Kennedy JC, Pottier RH, Pross DC: Photodynamic therapy with endogenous porphyrin. IX: Basic principles and present clinical results. J Photochem Photobiol B 1990;6:143–148. Bhatta N, Anderson RR, Flotte T, Schiff I, Hasan T, Nishioka NS: Endometrial ablation by means of photodynamic therapy with Photofrin II. Am J Obstet Gynecol 1992;167:1856–1863.

Lars O. Svaasand, Department of Physical Electronics, Norwegian University of Science and Technology, N–7034 Trondheim (Norway) Tel. +47 73 594421, Fax +47 73 591441, E-Mail [email protected]


115


............................

Noninvasive Characterization of Tissue Optical Properties Using Frequency Domain Photon Migration Bruce J. Tromberg a, Olivier Coquoz b, Joshua B. Fishkin a, John Butler c, Lars O. Svaasand d a

Laser Microbeam and Medical Program, Beckman Laser Institute and Medical Clinic, University of California, Irvine, Calif., USA; b Institut d’Optique Appliqueé, EPFL, Lausanne, Switzerland; c Department of Surgery, University of California, Irvine, Calif., USA, and d Department of Physical Electronics, Norwegian University of Science and Technology, Trondheim, Norway

Introduction Light launched into tissue is efficiently scattered out of an incident collimated beam into an almost isotropic distribution within a few millimeters from the source. Until recently, this multiple-scattering behavior implied that lasers and conventional sources were virtually interchangeable for most low-power optical diagnostics. Accordingly, it is not surprising that relatively simple, inexpensive optical methods using incoherent sources have flourished. Consider, for example, the pulse oximeter, a diode-based instrument for determining arterial hemoglobin oxygen saturation. Pulse oximeters are used on nearly every patient admitted to hospitals, as well as in many outpatient procedures. Similarly, optical transillumination techniques are employed to image the neonatal brain [1], locate abnormalities in breast tissue [2], and pinpoint blood vessels. Over the past few years, remarkable gains in our understanding of tissue optical properties have been realized by interrogating tissues with the unique temporal (i.e. short pulse) properties of lasers [3–5]. In contrast to continuous illumination techniques [6, 7], pulse propagation methods provide information about the distribution of scatterers and absorbers in a single measurement [8, 9]. These optical properties may be used in a variety of therapeutic and

diagnostic techniques, including: imaging tissue structure [10–12] and cortical activity [13]; monitoring physiology [14–16] and drug concentration [17]; and predicting optical dosimetry for laser-based procedures [18]. The conceptual basis for the time-domain approach generally involves solutions to the radiative transfer equation [19, 20] using Monte Carlo simulation [21–23] and diffusion theory approximations [5, 24]. Diffusion-based methods provide relatively straightforward analytical expressions that describe the shape of a diffusely reflected or transmitted pulse in terms of the optical properties of the medium [8]. Thus, the observed temporal broadening of ultra short pulses can be mathematically related to the large number of optical paths available in multiple-scattering media. Since the introduction of losses (absorbers) reduces the average path length, absorber-dependent changes in pulse propagation time can be used to calculate absorption coefficients [8]. Frequency-domain optical methods can be adapted to diffusion theory models in a similar manner. Fishkin and Gratton [25] first suggested that amplitude-modulated light propagates through homogeneous multiple-scattering media as diffuse waves with a coherent front. These photon density waves can be characterized by a phase velocity (Vp) and modulation wavelength (km) that are primarily functions of media optical properties. Measurements are performed by launching intensity-modulated light into tissue and recording the phase delay (/) and demodulation amplitude (m; with respect to the source) at a fixed distance from the launch site. Optical properties (the absorption coefficient la, and the reduced scattering coefficient l) s are calculated from measured frequency- and distance-dependent behavior by fitting the experimental results to an appropriate theoretical framework. Thus frequencydomain photon migration (FDPM) can rapidly and quantitatively assess tissue optical properties in a single noninvasive measurement. Since all diagnostic and therapeutic methods capitalize on variations in tissue optical properties, density wave techniques are expected to have a broad impact on the rapidly growing field of biomedical optics.

Light Propagation in Tissues Spectral Dependence Light propagation in biological tissues is a complex function of scattering and absorption, which, in turn, are dependent on cellular structure and molecular composition. Figure 1 illustrates that within the visible (300–700 nm) water window, hemoglobin and melanin are the dominant endogenous absorbers [26]. Although melanin absorption coefficients are, on average, greater than those of hemoglobin, the overall influence of melanin on light propagation is

Noninvasive Characterization of Tissue Optical Properties

117

Fig. 1. Special dependence of tissue chromophores. Adapted from Boulnois [26].

generally less substantial. This is due to the fact that melanin is typically confined to a thin, superficial skin layer. In contrast, hemoglobin is widely distributed throughout most tissues at a volume fraction that ranges from 1 to 5%. Exceptions to this include avascular structures and tissues with abnormally high melanin content, such as melanotic melanomas. Scattering originates from inhomogeneities in tissue structure, which, in turn, are determined by refractive index discontinuities occurring both between and within cells. In the region between 600 and 700 nm, hemoglobin absorptivity diminishes by approximately 20 dB and scattering eclipses absorption. This condition persists until water absorption resumes as the primary attenuation mechanism at approximately 1.3 lm. Thus, the spectral region between 0.6 and 1.3 lm is considered to be the tissue optical ‘window’ since both absorption and scattering losses are minimal throughout this interval. Table 1 provides several practical examples of wavelength-dependent optical penetration depths, where the optical penetration depth (d) corresponds to the distance at which the optical fluence rate is reduced to 1/e>0.37 of the initial value [27]. These data highlight the dynamic interplay between absorption and scattering. For example, since human retinoblastomas contain no melanin, low levels of hemoglobin, and are only minimally scattering, penetration depths are relatively large, ranging from 2.9 mm at 600 nm to 5.1 mm at 1.06 lm. Light penetrance is commensurately reduced in brain and hand tissue due to their high-scattering/low-absorption and moderate-scattering/ moderate-absorption properties, respectively. In contrast, the high melanin

Tromberg/Coquoz/Fishkin/Butler/Svaasand

118

Table 1. Optical penetration depth (mm) as a function of wavelength Tissue

Wavelength, nm 600

650

700

750

800

850

900

1,064

Human retinoblastoma (in vitro) 2.9 Porcine brain (in vitro) 1.8 Human hand (in vivo) 1.4 Melanotic melanoma in – athymic mice

3.8 2.4 2.0 0.28

4.0 2.9 2.6 0.34

4.0 3.0 2.7 0.41

4.1 3.3 3.0 0.50

4.2 3.5 3.0 0.56

4.3 3.7 3.0 0.64

5.1 4.0 – 1.4

From Svaasand [27].

content of melanotic melanomas strongly attenuates light throughout the visible region and maximum light penetrance is only 1.4 mm at 1.064 lm. These examples indicate that, in general, d for nonmelanotic tissues follows hemoglobin absorption. Thus, from 480 to 590 nm, d ranges for 0.5 to 1.5 mm. A substantial penetration depth increase from about 1 to 3 mm occurs between 600 and 650 nm, and a more gradual increase is observed from 650 to 750 nm (d>2–4 mm). Between 750 and 900 nm, values tend to plateau, and maximum light penetrance of 3–6 mm generally occurs in the 1,000- to 1,100-nm region. Multiple Scattering Perhaps the single most important conclusion supported by these and many other measurements [28] is that tissue optical properties vary substantially according to cellular structure and metabolic activity. For example, on a cellular level, there are clear, visible differences between healthy and cancerous tissue. These changes may include, among others, multinucleation, structural distortion, hyperpigmentation, and membrane and mitochondrial variations. In addition, there are distinct structural (e.g. vasculature) and biochemical variations in tumors. Pathologists utilize these visible, microscopic anomalies in cell/tissue architecture to formulate diagnoses. Thus, variations in bulk tissue optical penetration depths are expected given the large body of supporting histologic evidence. These fundamental optical property differences provide both diagnostic contrast and therapeutic selectivity. Unfortunately, multiple scattering degrades the information content of light that propagates through tissue. In practical terms, optical imaging methods that cannot distinguish between scattering and absorption are incapable of providing a complete diagnostic picture. This is illustrated conceptually in figure 2.


119

Fig. 2. Effect of turbid medium.

In homogeneous, nonscattering media, light propagation is simply a function of the absorption coefficient, la, and distance, z: I>I0eÖlaz,

(1)

where la>1/1ab and 1ab is the absorption length, or average distance between absorption events. The absorption coefficient can be expressed in terms of molecular concentration via the Lambert-Beer law (i.e. A>log I/I0>ebC, where A, e, b, and C are absorbance units, molar extinction coefficient, path length and molar concentration, respectively): la>2.3eC.

(2)

Thus measurements of light penetrance in tissues can, in principle, provide quantitative information regarding the concentration of physiologically relevant absorbers. However, in scattering media, both scattering and absorption contribute to the distance-dependent light attenuation. The total attenuation coefficient, lt, is the sum of absorption and scattering parameters (lt>la+ls).Here ls is the scattering parameter, or the reciprocal of the scattering length (lsc). In the special case of multiple scattering, the incident light intensity diminishes according to: I>I0eÖleff z,

(3)

1/2 where the effective attenuation coefficient leff>1/d>[3la(la+l)] is the recips rocal of the mean penetration depth (d). In order to account for multiple scattering, the reduced scattering parameter, l>l s s(1–g) is defined in terms of ls and the angular dependence of scattering, g, where g>=cosh? (i.e. the average cosine of the scattering angle). In most tissues, scattering is highly forward-directed and g values range from 0.7 to 0.9. In practical terms, large


120

g values delay the distance photons must travel before optical energy is isotropically distributed. Generally, this ‘transport mean free path’ (tmfp) is on the order of 1 mm in most tissues and can be approximated by the reciprocal of the effective scattering coefficient, i.e. tmfpB1/l>1/[(1–g)l s s]. Diffusion Approximation In order to probe large volumes of tissues with acceptable signal-to-noise levels, measurements are generally conducted in spectral regions of greatest transparency, i.e. the tissue optical window (600–1,300 nm). Under these conditions, scattering dominates absorption and light propagation can be described by diffusion approximation [19]. Since scatterers and absorbers are distributed in tissues on dramatically different spatial scales (i.e. l9l s a), an additional descriptive domain, i.e. time or frequency, is required to resolve these phenomena. Thus, when combined with the appropriate conceptual framework, measurements of the propagation characteristics of short pulses or amplitudemodulated waves can yield tissue absorption and scattering properties. Optical power in tissues is typically characterized by the quantities u, the radiant energy fluence rate, and L, the radiance. The radiant energy fluence rate, 4p

u>

P

LdX,

X>0

is defined as the optical energy flux incident on an infinitesimally small sphere divided by the cross-sectional area of that sphere. Since the integration is taken over all solid angles, X, fluence rate is a measure of the total optical flux. The radiance, L, is the optical energy flux in some direction per unit solid angle per unit area orthogonal to that direction. In a completely isotropic light field, u L> . 4p

As light propagates through tissue, the flux of diffuse photons from a region of high fluence rate, u, to a region of low fluence rate can be described by a diffusion equation [29].

Instrumentation In FDPM, the intensity of light incident on an optically turbid sample is modulated at high frequencies (e.g. hundreds of MHz), and the diffusely reflected, transmitted, or reemitted (e.g. fluorescent) signal is measured with


121

Fig. 3. Phase and modulation measurements.

a phase-sensitive detector. Intensity-modulated light propagates through multiple-scattering media with a coherent front, forming photon density waves (PDWs). PDW dispersion is highly dependent on the optical properties of the medium. Exact absorption (la) and reduced scattering (l) s parameters are calculated by comparing the measured frequency- or distance-dependent PDW phase and amplitude behavior to analytically derived nonlinear model functions. Model functions are, in turn, derived from solutions to the photon diffusion equation which yield expressions for phase (/) and amplitude (A) as a function of modulation frequency (x) and tissue optical properties under various boundary conditions [30]. The values /, A, and x, illustrated in figure 3, are defined in the usual manner, where />the measured phase lag between the source (reference) and sample response; A>ACsample/ACsource, and x>2pf is the angular modulation frequency. When optical properties (ls and la) are recovered for various source wavelengths, the spectral dependence of absorption can be used to calculate physiologically relevant parameters, such as oxygenated, deoxygenated, and total hemoglobin concentration; oxygen saturation; drug concentration; blood volume fraction, and water concentration. We have constructed a 1-GHz, portable FDPM device that employs an avalanche photodiode (APD) detector and direct modulation of 4 near-infrared diode lasers [31]. Modulation is swept from 300 kHz to 1 GHz in less than 1 s, providing rapid, multi-wavelength (~670–960 nm) characterization of most tissues in a single measurement. The instrument is compact and can easily be transported to operating rooms and bedridden patients. By incorporating several near-infrared laser diodes, a number of wavelength-dependent physiological parameters are determined. Four diode laser sources are currently used


122

Fig. 4. 1-GHz bandwidth FDPM apparatus.

in our FDPM instrument: 674, 811, 849, and 956 nm. Source modulation frequencies typically range form 300 kHz to 1 GHz, with FDPM data recorded in 5-MHz increments. A schematic is shown in figure 4. The core component of the FDPM apparatus is a network analyzer (Hewlett Packard, model 8753C), which is used to produce modulation swept from 300 kHz to 1 GHz (20 dBm RF output). RF from the network analyzer is serially superimposed (via the AC switch) on the direct current of four different diode lasers (SDL, Inc. models 7421, 5420, 5421 and 6321 at 674, 811, 849 and 956 nm, respectively) using individual bias tees (model 5575A, Picosecond Pulse labs) and an RF switch (model 8768K, Hewlett Packard). Four, 100-lm-diameter gradient-index fibers are used to couple each light source to an 8¶8 optical multiplexer (model GP700, DiCon Instruments). The currently employed 8¶8 optical multiplexer allows up to eight different laser diode light sources and eight different optical fiber light sources in our FDPM instrument. Light is launched onto the tissue (or test object) using the optical switch and one source fiber. An APD (Hamamatsu, model C5658) is used to detect the diffuse optical signal that propagates through the biological tissue. Both the APD and probe end of the source optical fiber are in direct contact with the patient (i.e. a ‘semi-infinite medium’ measurement geometry). The optical


123

power coupled into the tissue averages approximately 10–30 mW, roughly a factor of 10 below thermal damage threshold levels for fiber-delivered red/ NIR light. Measurement time depends on the precision required, the number of sweeps performed, and RF/optical switch times. For human subject studies, approximately 0.1 s is used to sweep over the entire 1-GHz band of modulation frequencies. However, total elapsed time for 4 diodes (typically 12–16 sweeps/ diode), data transfer, display and source switching is approximately 40 s. Most components, including the network analyzer, RF/optical switches, diode power supplies and temperature of diode mounts are controlled by computer using virtual instrument software (LabView, National Instruments). Phase and amplitude data (represented by / and A, respectively) obtained from FDPM tissue measurements are fit to a semi-infinite model to extract the absolute optical absorption coefficient, la, and the absolute optical reduced scattering coefficient, ls at a given k and source-detector separation q. Minimization of the v2 surface is achieved for simultaneous, error-weighted fitting of the / and A versus frequency data using a Marquardt-Levenberg algorithm. Typical la, and ls uncertainties, determined from the v2 distribution of phase and amplitude fits, range from 0.5 to 5% of the mean value.

Application to Breast Tumor Characterization The sensitivity and specificity of commonly-used breast-screening methods are less than optimal. The practical impact of increasing diagnostic specificity would be twofold: (1) enhanced early detection of relatively small (0.5 to 1-cm-diameter tumors) is likely to lead to mortality improvements (mammographic specificity is only about 70% for 1-cm-diameter tumors), and (2) more specific characterization of benign vs. malignant lesions would lead to reductions in the large number of unnecessary surgical biopsies. As a result, there is considerable room for the development of new, noninvasive optical methods for characterizing breast tissue. FDPM is appealing because it utilizes nonionizing radiation, is noninvasive, provides quantitative information, and is technologically simple and fast. It can generate tissue functional maps with contrast that is not achievable using other more established devices. For example, X-rays and ultrasound are excellent tools for identifying structural features but, unlike optical methods, cannot be used to characterize biological molecules such as oxy- and deoxyhemoglobin, water, and fat. In addition, optical methods are sensitive to cellular/organelle size, shape, and structure factors. Since these parameters appear to vary with tissue type (e.g. tumor vs. normal), optical properties are promising indicators of malignant transformation.


124

In order to demonstrate the clinical potential of FDPM, we initiated breast tumor studies in 30 human subjects [32]. Our goal is to carefully define the level of accuracy and precision required from our measurement techniques so that malignant lesions could be identified and resolved from normal structures with a reasonable level of confidence. In the following discussion we present the results from 2 individuals. Experiments were performed under the guidelines of UC Irvine IRBapproved protocol No. 95-563. Patient 1 was a 56-year-old postmenopausal female with a single distinct lesion buried approximately 1×0.5 cm beneath the skin surface in the lateral region of the right breast. Histological examination following surgical biopsy revealed a 1.5¶2¶1 cm fibroadenoma with ductal hyperplasia (benign tumor). Patient 2 was a 27-year-old, premenopausal, lactating patient with a similarly sized lesion in the upper outer quadrant of the right breast. Subsequent needle biopsy revealed the presence of a benign fluidfilled cyst, consistent with fibrocystic changes. Measurements were performed on each patient by gently placing the FDPM probe on both the normal and tumor-containing breast. Just enough pressure was applied to ensure optical contact between tissue and probe. Sequential scans of the same location following probe removal and replacement revealed no significant variations in optical properties. Tumor data were acquired for each patient by positioning the probe so that the source and detector (separation q>2.2 cm) bracketed the tumor. Normal tissue measurements were acquired from a symmetric site on the opposite, uninvolved breast. Four diode laser sources were used: 674, 811, 849 and 956 nm. Source modulation frequencies ranged from 300 kHz to 1 GHz, and FDPM data were recorded in 5-MHz increments. Phase and amplitude data (represented by / and A, respectively) obtained from FDPM tissue measurements were fit to semi-infinite model functions to extract the absolute optical absorption coefficient, la, and the absolute optical reduced scattering coefficient l,s at a given k and source-detector separation q. We assume that the chromophores contributing to la at a given wavelength in the human subject are principally oxy- and deoxyhemoglobin, water, and fat (fig. 5). The tissue concentration of 3 of these components (oxy- and deoxyhemoglobin, water) is determined from la measurements at a minimum of three different wavelengths. We were not able to calculate the fat content in the current study due to our poor sensitivity to the peak fat absorption feature (about 920 nm) with the wavelengths employed. The calculation performed to extract the chromophore concentration is described in detail elsewhere [33]. Results of 811-nm FDPM measurements are shown for patients 1 and 2 in figures 6 and 7, respectively. Raw data reveal clear differences between


125

Fig. 5. Absorption spectra of tissue with composition similar to human postmenopausal breast: 10 lM Hb, 15 lM HbO2, fat (80% volume fraction), and water (15% volume fraction).

a

b

Fig. 6. FDPM measurements of phase (a) and amplitude (b) vs. source modulation frequency obtained from normal and tumor sites in patient 1. Source-detector distance> 2.2 cm; k>811 nm. Solid lines are best simultaneous fits to phase and amplitude data from tumor (X) and normal (T) data points, respectively.


126

a

b

Fig. 7. FDPM measurements of phase (a) and amplitude (b) vs. source modulation frequency obtained from normal and tumor sites in patient 2. Source-detector distance> 2.2 cm; k>811 nm. Solid lines are best simultaneous fits to phase and amplitude data from tumor (X) and normal (T) data points, respectively.

normal and tumor sites. Normal tissue scans were obtained from the symmetric position on the opposite breast. In patient 1, tissue differences are more obvious in amplitude (fig. 6b), rather than phase (fig. 6a), data. For this reason, we have employed a technique for simultaneously fitting phase and amplitude to FDPM nonlinear model functions. Optical properties obtained from these fits are shown as a function of wavelength in figures 8 and 9. Figures 8a and 9a show that absorption values obtained from tumorcontaining tissue are consistently higher than normal for each patient. Although the magnitude of the absorption change is greater for patient 2, it is impossible to determine whether this is a consequence of intrinsic lesion properties or differences in size and depth. In patient 1, tumor absorption increases slightly from 674 to 811 nm, then drops at 849 nm. This behavior is reversed for normal and tumor tissue in patient 2. Physiological differences between patients are further amplified by results of reduced scattering coefficient, l,s calculations. Figures 8b and 9b show that ls decreases gradually with increasing wavelength. Tumor values are consistently higher for patient 1 than normal, while the opposite is observed for patient 2. Histology revealed a fibroadenoma (approximately 1¶1.5¶2 cm and 1 cm deep) with mild ductal hyperplasia, and an acellular, fluid-filled cyst for patients 1 and 2, respectively. Scattering values are compatible with histology reports since lower ls is expected for fluid-filled structures (i.e. fluid-filled cyst), while increased scattering should occur in the case of fibrotic tissue (i.e. fibroadenoma). Normal tissue scattering is 1.3- to 1.4-fold higher for the pre-


127

a

b

Fig. 8. Absorption coefficient, la (a) and reduced scattering coefficient, ls (b) versus source wavelength for normal and malignant breast tissue in patient 1. Values calculated from best simultaneous fits of model functions to phase and amplitude data.

a

b

Fig. 9. Absorption coefficient, la (a) and reduced scattering coefficient, ls (b) versus source wavelength for normal and malignant breast tissue in patient 2. Values calculated from best simultaneous fits of model functions to phase and amplitude data.

vs. postmenopausal patient. It is likely that these differences are a consequence of the fact that premenopausal, highly glandular tissue has substantial structural complexity while postmenopausal breast is dominated by low-water content adipose. Physiological properties reflected by oxyhemoglobin, deoxyhemoglobin, and water may be extracted from la values obtained from at least three wave-


128

Fig. 10. Hemoglobin (lM), deoxy-, oxy-, and total) and water (M) concentrations for normal and tumor breasts in patients 1 and 2, calculated from wavelength-dependent la values.

lengths. These properties have been measured for the above patients and are summarized in figure 10. Total and oxyhemoglobin levels are approximately two-fold higher in tumor vs. normal tissue for both patients. Deoxyhemoglobin contrast persists for patient 2 but does not appear for patient 1. As a result, differences in normal vs. tumor tissue oxygen saturation (SaO2) are greater for patient 1 (68 vs. 79%, respectively) than patient 2 (66 vs. 74%). SaO2 values are larger in tumor-containing tissue, perhaps due to higher blood flow and lower oxygen extraction in regions probed by FDPM measurements. This observation is consistent with the view that benign fibrotic and cystic tumors are not likely to display marked hypoxic zones, a characteristic that is expected to be unique to malignant transformations. Normal SaO2 values are comparable for both patients (68 and 66% for patients 1 and 2, respectively). Tissue water concentration is also displayed in figure 10. It is difficult to confirm the accuracy of these values since they are based on pure water extinction coefficients (as opposed to protein-bound forms) at 25 ºC and our calculations do not take into account the contribution of fat to the 956-nm signal. Nevertheless, all results fall within the 11.4–30.5% range given in the literature for percentage water in human fatty adipose tissue. In addition, our data make sense in the context of what is known about


129

patient physiology. For example, premenopausal, lactating breast tissue has substantially higher water content than postmenopausal, adipose tissue (17 vs. 12%; where 100% D55 M). The presence of a fluid-filled tumor elevates the percentage water from 17 to 21% in patient 2, while the fibroadenoma does not affect patient 1. Histologic examination of the fibroadenoma shows dense, fibrotic zones surrounding shrunken glandular structures. Further confirmation is provided by our observation of elevated fibroadenoma ls values (fig. 8b). Conclusions FDPM is a noninvasive optical technique that utilizes near-infrared light to monitor physiology in bulk tissues. Optical properties derived from FDPM measurements can be used to construct low-resolution functional images and, consequently, provide a relatively low-cost adjunct to many conventional diagnostic tools. Substantial work remains to accurately describe heterogeneous tissues that vary in geometry, structure and composition. Despite these limitations, several unique applications of photon migration measurements are currently underway, particularly in breast, brain, and tumor characterization. Since these techniques are compatible with high-bandwidth, near-infrared semiconductor lasers, we anticipate future development will incorporate these technologies into compact, durable, and inexpensive bedside monitoring devices.

Acknowledgements This work was made possible, in part, through access to the Laser Microbeam and Medical Program (LAMMP) and the Chao Cancer Center at the University of California, Irvine. These facilities are supported by the National Institutes of Health under grants RR01192 and CA-62203, respectively. Support was also provided by the Department of Energy (DOE No. DE-FG03-91ER61227), the NIH Institute of General Medical Sciences (GM50958), the Swiss Fund for Cancer Research, and the Beckman Foundation.

References 1

2 3

Wyatt JS, Cope M, Delpy DT, Wray S, Reynolds EOR: Quantitation of cerebral oxygenation and haemodynamics in sick newborn infants by near infrared spectrophotometry. Lancet 1986;ii: 1063–1066. Barlow CH, Burns DH, Callis JB: Breast biopsy analysis by spectroscopic imaging; in Chance B (ed): Photon Migration in Tissues. New York, Plenum Press, 1990, pp 111–119. Chance B, Leigh J, Miyake H, Smith D, Nioka S, Greenfield R, Finlander M, Kaufmann K, Levy W, Yound M, Cohen P, Yodshioka H, Boretsky R: Comparison of time-resolved and -unresolved measurements of deoxyhemoglobin in brain. Proc Natl Acad Sci USA 1988;85:4971–4975.


130

4 5 6 7 8 9 10 11 12 13 14 15

16 17 18 19 20 21 22

23 24 25 26 27

28 29

Delpy DT, Cope M, van de Zee P, Arridge S, Wray S, Wyatt J: Estimation of optical pathlength through tissue from direct time of flight measurement. Phys Med Biol 1988;33:1433–1442. Wilson BC, Sevick EM, Patterson MS, Chance B: Time-dependent optical spectroscopy and imaging for biomedical applications. Proc IEEE 1992;80:918–930. Bonner RF, Nossal R, Havlin S, Weiss GH: Model for photon migration in turbid biological media. J Opt Soc Am [A] 1987;4:423–432. Groenhuis RAJ, Ferwerda HA, Ten Bosch JJ: Scattering and absorption of turbid materials determined by reflection measurements. 1: Theory. Appl Opt 1983;22:2456–2462. Patterson MS, Chance B, Wilson BC: Time-resolved reflectance and transmittance for the noninvasive measurement of tissue optical properties. Appl Opt 1989;28:2331–2336. Yoo KM, Alfano RR: Determination of the scattering and absorption lengths from the temporal profile of a backscattered pulse. Opt Lett 1990;15:276–278. Singer JR, Grunbaum FA, Kohn P, Zubelli JP: Image reconstruction of the interior of bodies that diffuse radiation. Science 1990;248:990–993. Barbour RL, Graber HL, Aronson R, Lubowsky J: Imaging of subsurface regions of random media by remote sensing. Proc SPIE 1991;1431:192–203. Benaron D, Lennox MA, Stevenson DK: Two-dimensional and 3-D images of thick tissue using time-constrained time-of-flight spectrophotometry. Proc SPIE 1991;1641:35–45. Ts’o DY, Frostig RD, Lieke EE, Grinvald A: Functional organization of primate visual cortex revealed by high resolution optical imaging. Science 1990;249:417–420. Sevick EM, Chance B, Leigh J, Nioka S, Maris M: Quantitation of time and frequency-resolved optical spectra for the determination of tissue oxygenation. Anal Biochem 1991;195:330–351. Chance B, Nioka S, Kent J, McCully K, Fountain M, Greenfeld R, Holtom G: Time-resolved spectroscopy of hemoglobin and myoglobin in resting and ischemic muscle. Anal Biochem 1998; 174:698. Schmitt JM, Zhou G-X: Measurement of blood hematocrit by dual-wavelength near-IR photoplethysmography. Proc SPIE 1992;1641: in press. Patterson MS, Wilson BC, Feather JW, Burns DM, Pushka W: The measurement of dihematoporphyrin ether by reflectance spectrophotometry. Photochem Photobiol 1987;46:337–343. Jacques SL, Prahl SA: Modeling optical and thermal distributions in tissue during laser irradiation. Lasers Surg Med 1987;6:494–503. Furutsu K: Diffusion equation derived from space-time transport equation. J Opt Soc Am [A] 1980;70:360. Patterson MS, Wilson BC, Wyman DR: The propagation of optical radiation in tissue. I. Models of radiation transport and their application. Lasers Med Sci 1991;6:155–168. Jacques SL: Time resolved propagation of ultrashort laser pulses within turbid tissues. Appl Opt 1989;28:2223–2229. Flock ST, Patterson MS, Wilson BC, Wyman DR: Monte Carlo modeling of light propagation in scattering tissues. I. Model prediction and comparison with diffusion theory. IEEE Trans Biomed Eng 1989;36:1162–1168. Hasegawa Y, Yamada Y, Tamura M, Nomura Y: Monte Carlo simulation of light transmission through living tissues. Appl Opt 1991;31:4515–4520. Ishimaru A: Diffusion of light in turbid materials. Appl Opt 1989;28:2210–2215. Fishkin J, Gratton E, VandeVen MJ, Mantullin WW: Diffusion of intensity modulated near-infrared light in turbid media. Proc SPIE 1991;1431:122–135. Boulnois JL: Photophysical processes in recent medical laser developments: A Review. Lasers Med Sci 1986;1:47–66. Svaasand L: On the physical rationale of photodynamic therapy; in Gomer CJ (ed): Future Directions and Application in Photodynamic Therapy. SPIE Inst Ser Adv Opt Technol. Los Angeles, SPIE, 1990, pp 233–248. Cheong WF, Prahl SA, Welch AJ: A review of the optical properties of biological tissues. IEEE JQE 1990;26:2166–2185. Tromberg BJ, Svaasand LO, Tsay T-T, Haskell RC: Properties of photon density waves in multiplescattering media. Appl Opt 1993;32:607–616.


131

30 31 32

33

Haskell RC, Svaasand LO, Tsay T-T, Feng T-C, McAdams MS. Tromberg BJ: Boundary conditions for the diffusion equation in radiative transfer. J Opt Soc Am [A] 1994;11:2727–2741. Madsen SJ, Anderson ER, Haskell RC, Tromberg BJ: A portable, high-bandwidth frequency-domain photon migration instrument for tissue spectroscopy. Opt Lett 1994;19:1934–1936. Tromberg BJ, Coquoz O, Fishkin JB, Pham T, Anderson ER, Butler J, Cahn M, Gross JD, Venugopalan V, Pham D: Non-invasive measurements of breast tissue optical properties using frequencydomain photon migration. Philos Trans R Soc Lond [B] 1997;352:661–668. Fishkin JB, Coquoz O, Anderson EA, Brenner M, Tromberg BJ: Frequency-domain photon migration measurements of normal and malignant tissue optical properties in a human subject. Appl Opt 1997;36:10–20.

Bruce J. Tromberg, Beckman Laser Institute and Medical Clinic, 1002 Health Sciences Road East, Irvine, CA 92612 (USA) Tel. +1 949 824 8705, Fax +1 949 824 8413, E-mail [email protected]


132


............................

Fluorescence Microscopy for Pharmacokinetics Tatiana B. Krasieva Beckman Laser Institute, University of California, Irvine, Calif., USA

Introduction Fluorescence microscopy is of great use in photodynamic therapy (PDT) for assessment of accumulation, localization and decaying of photosensitizers. The efficacy of PDT depends on three ‘rights’: right place, right time, and right amount. Specifically, the concentration of a photosensitizer should be sufficient for inflicting a desired level of tissue damage during light application at the exact moment in time when the best tumor-to-normal tissue drug ratio is observed. This clearly shows that PDT is a multiparametrical process and requires an optimization of several parameters: photosensitizer type and concentration, method of delivery (intravenous, intrauterine, topical, oral, etc.), light dose and time of light application. Usually, the first step in this highly iterative process would be an assessment of the pharmacokinetics and the biodistribution of a chosen PDT reagent. Biodistribution would provide information about selectivity (or nonselectivity) of a PDT compound toward targeted tissue and/or subcellular localization, and pharmacokinetics will show its time dependence. Absorption of light transfers the photosensitizer from the ground into the short-lived excited singlet state. Following relaxation, processes are fast and competitive. Radiationless intersystem crossing (ISC) yields into the first excited triplet state T1 (FD) and is ultimately responsible for triggering biophysical and biochemical mechanisms which cause tumor necrosis [1]. Energy transfer competes with a radiation relaxation process well known as fluorescene (Ff). The following scheme of initial photophysical and photochemical events shows that although fluorescence competes with the energy transfer process, on the other hand, it makes it a perfect tool to observe

the biodistribution and pharmacokinetics of the photosensitizers. Roughly Ff+FDV1 +h)ex

energy transfer

S (sensitizer) CCB 1S* CB 3S* CCCCB 1O2 , |+hz B S fl

and further.

Fluorescence and Digital Imaging Fluorescence has been used in biological and medicobiological studies for a long time. The introduction of fluorochromes into fluorescence microscopy by von Provazek in 1914 marked this new era in cytology and experimental biology as a whole (an excellent historical review of the early steps of fluorescence microscopy is given by Kasten [2]). Fluorescence methods have become methods of choice in modern biology and medicine due to recent developments. The most significant ones were the progress in instrumentation, use of lasers as light delivery sources, significant improvements in fluorescence detecting (detectors with better sensitivity at a lower cost), and analysis of the fluorescent signals (both software and hardware). Modern fluorescence methods include spectroscopy, cell sorting, and a variety of the imaging techniques such as epifluorescence, confocal, multiphoton microscopy, in vivo video monitoring, and digital imaging. Until today, the most complete and comprehensive book on video microscopy is that by Inoue [3]. All methods are well known, highly developed (except for multiphoton microscopy), very sensitive, and could be used for qualitative, semi- or fully quantitative measurements. Here we will discuss the use of quantitative fluorescence microscopy applied to the study of biodistribution and pharmacokinetics of photodynamic compounds in vitro and ex vivo (tissue biopsies, frozen sections). A modern fluorescence quantitative microscopy system consists of four main elements. The basic one is an optical high-tech microscope (Zeiss, Leika, Olympus, etc.) equipped with the excitation light source. There are a variety of them: xenon or mercury high-pressure lamps, lasers, super-luminescent diodes or alike coupled to a microscope via ‘traditional’ optical path – epifluorescence port of a microscope or through light-conducting fibers. The heart of the system, a detector, is mainly an intensified or slow-scanning charge-coupled device (CCD) or a high-sensitivity photo-multiplier tube (PMT) with an at least 12bit/pixel dynamic range, although a range of 14 or 16 bit/pixel is more desirable. The fourth element is a computer interfaced with a detector (usually via PCI

Krasieva

134

Fig. 1. Schematic of a low-light level imaging system.

or IEEE) enabling visualization of the acquired image, its post-processing and all measurements. A typical scheme is depicted in figure 1. The speed of accumulation of compounds relative to PDT is not very high (usually in the order of minutes, hours and days). Thus, researchers can fully utilize the advantages of modern cooled CCD cameras that are readily available for use with light microscopes. Slow-scanning cooled CCD cameras provide the user with exceptional spatial resolution up to several million pixels, ultra-high intrascene dynamic range up to 16 bit/pixel (which translates into more than 65,000:1), and exceptional sensitivity with the ability to integrate for hours in low-light level imaging. In many articles devoted to fluorescence imaging, one can find a phrase stating that ‘all fluorescence images were processed by the following algorithm (algorithm follows) to correct for nonuniform illumination, variations in the

Fluorescence Microscopy for Pharmacokinetics

135

excitation source luminosity, and dark noise contributed during the exposure time’. Let us briefly examine the process of an image formation, sources of artifacts, principal limitations, and practical algorithms, sampling two uniform fluorescent layers with concentrations of 5 and 10 (arbitrary units, mg/ml, mol/1, %, etc.). Let us also assume for the sake of simplicity that quantum yield of fluorescence for this hypothetical dye is equal to one (quantum yield is the ratio between the number of emitted fluorescence photons to the absorbed excitation photons). The resulting information we are to obtain is a two-dimensional image. For our purposes we will define an image as a two-dimensional matrix I (row)¶J (column). I and J depend on the type and size of a detector. Numerical data represent an intensity of fluorescence (gray scale values) per pixel which is proportional to the concentration of the fluorophore (PDT compound). Let us consider our image as a small 4¶4 matrix of data representing arbitrary ‘counts per pixel’ units (fig. 2). Let us also establish that ‘a mean value’ of an image is a number calculated as sum of gray scale values (R) of all pixels divided by the number of all pixels (n). First, all electronic devices are not ideal and will produce some noise (thermal electrons). However, this noise is quite low and Gaussian for modern devices (fig. 2a). Second, our light source cannot be distributed perfectly even over the whole field of view (fig. 2b). It means that where there is a lower amount of light received there is a lower amount of emission detected, while the real object contains the same amount of the fluorophore. Which brings us to our imaginary result: we will not have a uniform image (fig. 2c, d). In addition, the ratio between two images is not equal to 2: the mean value (m.v.) of figure 2c is 6 and the mean value of figure 2d is 11. Thus the ratio between figure 2d/2c>11:6>1.83. Is it possible to restore the initial uniformity and obtain proper relative concentrations? The answer is ‘yes’, and such an algorithm with some variations is known as a ‘flat field’ procedure (fig. 3). First, we will subtract the amount of dark signal from the raw image. Second, an obtained result (image) will be divided by excitation light intensity (image). When we are operating on two images it means that we apply the suggested operation (subtraction, for example) to the correspondent pixels of the first and second image on a pixel-by-pixel basic Nc (i>n, j>m)ÖNa (i>n, j>m). Third, we will multiply all points in the resulting image by the mean value of ‘light distribution’ of the image in order to bring the overall measured value of the raw image to the same amount. The arithmetic operation between image and number (i.e. mean value) means that each pixel of an image will be multiplied

Krasieva

136

a

b

c

d

Fig. 2. a Dark noise. m.v.>16. b Light source distribution. m.v.>80. c Acquired (raw) image (concentration equals 5). m.v.>96. d Acquired (raw) image (concentration equals 10). m.v.>176. n>16 in all cases.

(subtracted, divided) by a chosen number. The last step could be dropped because we are operating with arbitrary units and the coefficient between ‘counts per pixel’ and the concentration of the fluorophore generally is not known. Instead, some algorithms will substitute this coefficient with the ratio between ‘standard’ light distribution image (LDI) and current experiment LDI. This will take care of the day-to-day variations in the excitation source luminosity. (Raw fluorescence image – DNI) ¶ k (mean value); (Background image – DNI)

Corrected fluoresence image>

where DNI is a dark noise image and k is a mean value of a LDI.


137

Fig. 3. ‘Flat field’ algorithm application.

After the ‘flat field’ operation has been done on the second image as well, its mean value equals 10. Thus, we restored the proper ratio and uniformity of our imaginary samples. What do we need in practical terms to implement this procedure in a real experiment? First, the parameters of image acquisition should be set. This includes an optimized acquisition time, a filter set and a microscope objective. If the PDT compound is endogenously converted from a non-photoreactive predecessor, the positive control should be used first. Ideally, it should be a sample with a maximum expected fluorescence level. Exposure time should be chosen so fluorescence of the test sample would not saturate the detector but leave enough room to detect less intensive signals on the linear portion of the detector response. Next, the ‘service images’ – dark noise of the detector (DNI) and excitation source light distribution (LDI) – should be obtained at the chosen parameters (fig. 4). DNI is obtained when the source of excitation is turned off. The uniformity of the light distribution can be tested by different means. The most accurate reading may be achieved by imaging the solution of the studied fluorophore. This is not always an easy task, especially if the PDT compound is converted from a non-photoactive form in vivo or in vitro. Another way is by imaging some standard with stable characteristics and a wide excitation-emission range, like uranium glass. Usually, fluorescence levels or uranium glass are incomparably higher than observed in biological objects and cannot be acquired at the same parameters (exposition time, excitation source intensity) even with the greatest dynamic range detectors now available. A more common way to execute this procedure is simply to image an empty surface of the glass slide or a Petri dish at the settings close to experimental focal plane. Some additional and undetected problems should be mentioned

Krasieva

138

a

b Fig. 4. Service data. a Dark noise distribution image. b Light source distribution image.

in conclusion. That is, a saturation of the source or the detector, and nonlinear response of the detector. Saturation of the source when it fails to produce enough photons for the sample excitation is unlikely to be a problem in relevant applications. Saturation or nonlinear response of the detector is much more probable, especially when a sensitive detection system is tested on a sample


139

with lower fluorescence yield and subsequent samples contain much higher fluorophore quantities. Testing for this problem is simple. At the same experimental settings, install a neutral density filter with known optical density (0.1, 0.3, 0.01) in front of the excitation source and calculate if the level of the recorded fluorescence changes linearly with the changes in the excitation.

Fluorescence and PDT Usually, the first step in the design of a new treatment would be the choice of photosensitizer. The second step is defining the drug delivery system (intravenous, topical, oral or other). The third would be to find where the drug is localized and when it reaches its maximum concentration or best contrast, which represents a target tissue-to-normal tissue (TTN) ratio where the target tissue is very often a tumor (fig. 5). Unlike many other cases where the drug needs to be specially ‘tagged’ with a fluorescent marker, Mother Nature made all PDT drugs fluorescent ones. Absorption of a quantum of light by a sensitizer triggers a chain of events leading to a singlet oxygen formation and, ultimately, to cellular death and tumor necrosis. However, part of the absorbed energy could be used in a different process, fluorescence, where an excited molecule subsequently decays to a lower energy level and emits the rest of the absorbed energy as a visible light. Roughly, these two processes are responsible for all absorbed energy conversion by the photosensitizer. In the best cases (Zn-phthalocyanine, for example) about 70% of energy is used for the triplet state formation (and, ultimately, singlet oxygen and PDT) and 30% for the fluorescence emission. Such property makes PDT drugs perfect candidates for fluorescence monitoring. The third and highly unwanted process of energy ‘spending’ is a photochemical modification of a photosensitizing molecule leading to a loss of its photodynamic function known as ‘photobleaching’ or drug decomposition. What kind of information may or may not be extracted from data provided by fluorescence imaging? Generally, quantitative fluorescence does not allow the direct measurement of how much photosensitizer has accumulated in the samples. Usually the coefficient of proportionality between the absolute concentration of a fluorophore and the measured intensity of its fluorescence is unknown. This is mainly due to unknown quantum yields of fluorescence for compounds (not only PDT drugs) in tissues and cellular matrices. Generally, noble efforts to obtain such data through traditional methods of drug extraction or radioactive labeling do not result in absolute numbers either [4, 5]. Also a lack of proportionality between the concentration of administered

Krasieva

140

a

b Fig. 5. ‘Raw’ (a) and corrected (b) image of the frozen section. The biopsy of a peritoneum with tumor, rat ovarian cancer model, photosensitizer ALA. Notice the disappearance of the ‘hot spot’ in the center of the image and the dark spot in the top right corner.

drugs (both exogenous and endogenous) and recorded fluorescence is sometimes observed. It may be explained by several different factors. The quantum yield of fluorescence may be ‘environmentally’ sensitive (pH, binding to proteins) or fluorescence could be quenched, thus lowering the detectable signal, but this will not affect the actual concentration of the sensitizer.


141

Fig. 6. a Transversial section of rat colon after HE staining. b Fluorescence image of the same section. 30 mg/kg m-THPC MD, rat ovarian cancer model, courtesy of Dr. Rene Hornung.

Fortunately, fluorescence imaging may provide the researcher with precise and useful information. This includes exact localization of the drug, its relative concentration in different tissue types or cellular organelles (biodistribution), and the dynamics of its accumulation and disappearance in different loci (pharmacokinetics) which perfectly serves the stated goal of a PDT treatment optimization. Figures 6–8 illustrate the practical uses of the fluorescence imaging. As mentioned before, not only overall drug accumulation but also its distribution between the target and surrounding normal tissues play a very important role in establishing the proper time for the light application (fig. 8). It is interesting to note, for example, that the mean value of the fluorescence signal does not depend of the initial aminolevulinic acid (ALA) concentration. In contrast, the TTN ratio shows a relative difference between the two drug concentrations (fig. 8c). In the case of ALA, which is a nonfluorescent precursor of a fluorescent protoporphyrin IX, the independence from the ALA concentration should be attributed to the effect of saturation, since the conversion into photosensitizer is limited by metabolic activity of the tissue. Sur-

Krasieva

142

Fig. 7. From Steiner et al. [6]. a, b Bright-field image (a) and fluorescence micrograph (b) of the rat uterus 1.5 h following topical application of photofrin with 4% azone. CE> Columnar epithelium. c Rate of accumulation of photofrin for endometrium (excluding CE) and myomertium.


143

Fig. 8. a, b Fluorescence photomicrographs showing fluorescence of condyloma acuminum after 1.5 (a) and 24 h (b) after 20% ALA-containing cream application. At the early time point, fluorescence is primarily in the basal epithelial layer; at the later time, fluorescence has shifted to superficial epidermal layers. c Measured mean values (left y axis) of fluorescence of epithelium of the condyloma and the TTN (epithelial condyloma to adjacent skin) ratio (right y axis).

Krasieva

144

rounding normal tissue is less permeable to ALA and the mean fluorescence correlates with initial concentration thus dramatically changing the TTN ratio. An interesting problem is the calculation of the ratios themselves. While the TTN ratio is a simple arithmetical operation, difficulty arises in comparing the photosensitizer fluorescence to the background. Fluorescence of the drug at the initial time point of the drug delivery is fluorescence of the tissue with no drug accumulation, which is essentially a zero. Division of anything by a zero produces infinity. Comparison with autofluorescence is also not productive because autofluorescence is a fluorescence of endogenous compounds of cells and/or tissues and, in the majority of cases, is different from the photosensitizer. Direct fluorescence measurements do not allow us to compare concentrations of two different fluorophores by the same reasoning that we do not know the concentration of an individual compound. In addition, different absorption spectra would require different excitation/emission filters. Thus, according to the spectral characteristics of a source’s emission and filter’s bandwidth and transmission, different compounds will generally receive different amounts of excitation energy, and will experience different limitations on the fluorescence emission. Knowledge of absolute concentrations on its own will not be an ultimate judge of the drug efficacy. As mentioned above, the characteristics of drug performance may include: (a) the ratio of quantum yield of the triplet state to the quantum yield of the fluorescence; (b) quantum yield of photo decay (photobleaching); (c) biological mechanism of action; (d) TTN value; (e) photo- and dark toxicity; (f) metabolic rate, and so on. This ‘makes the comparison of these agents interesting’ [8], and leaves the opportunity to the researchers for more or less comprehensive testing descriptions. Measurements of autofluorescence are very important for producing both quantitative and qualitative data. It ensures that the intensity of the signal and its localization belongs to the compound under study and not to something different from it. In spite of all limitations, fluorescence imaging of PDT drugs remains increasingly popular. New-generation photosensitizers are scrutinized by all available methods. Confocal microscopy brings new clarity into the drugs’ localization, including 3-dimensional reconstruction of fluorescence in tissues and cells. Two-photon microscopy helps to overcome the limitations of the ‘traditional’ laser scanning confocal microscopy [8]. Two (or multi-)-photon microscopy allows the tunability of the wavelength of the probing laser, which diminishes the negative effects of photobleaching the fluorophore above and below the plane of observation and exceeds the depth of scanning in opaque media (tissue) [9]. Fluorescence is a valuable tool in PDT treatment optimization, but is not the absolute one. Fluorescence helps to visualize the drug, and may suggest


145

the best time or concentration of a fluorophore but it should not be forgotten that this process is a competitor to the main process leading to the PDT effect. The ‘ultimate’ PDT drug should have no fluorescence at all. Thus, the ultimate judge of a PDT drug is its PDT performance. Fluorescence methods are a powerful modality in the PDT research area and will further provide valuable information given that they are used with a good sense of judgement and accuracy.

References 1 2

3 4

5 6

7

8

9

Ochsner M: Photophysical and photobiological processes in the photodynamic therapy of tumors. J Photochem Photobiol 1997;39:1–18. Kasten FH: Introduction to fluorescent probes: Properties, history and applications; in Mason WT (ed): Fluorescent and Luminescent Probes for Biological Activity. New York, Academic Press, 1993, pp 12–33. Inoue S: Video Microscopy. New York, Plenum Press, 1986. Forssen EA, Male-Brune R, Adler-Moore JP, Lee MJA, Schmidt PG, Krasieva TB, Shimizu S, Tromberg BJ: Fluorescence imaging studies for the disposition of daunorubicin liposomes (DaunoXome) with tumor tissue. J Cancer Res 1996;56:2066–2075. Bhatta N, Anderson RR, Flotte T, Schiff I, Hasan T, Nishioka NS: Endometrial ablation by means of photodynamic therapy with photofrin II. Am J Obstet Gynecol 1992;167:1856–1863. Steiner RA, Tromberg BJ, Wyss P, Krasieva TB, Chandanani N, McCullough J, Berns MW, Tadir Y: Rat reproductive performance following photodynamic therapy with topically administered Photofrin. Hum Reprod 1995;1:227–233. Fehr MK, Chapman CF, Krasieva TB, Tromberg BJ, McCullough JL, Berns MW, Tadir Y: Selective photosensitizer distribution in vulvar condyloma acuminatum after topical application of 5-aminolevulinic acid. J Obstet Gynecol 1996;174:951–957. Orenstien A, Kostenich G, Roitman L, Shechtman Y, Kopolovic Y, Ehrenberg B, Malik Z: A comparative study of tissue distribution and photodynamic therapy selectivity of chlorin e6, Photofrin II and ALA-induced protoporphyrin IX in a colon carcinoma model. Br J Cancer 1996; 73:937–944. So PC, Kim H, Kochevar IE: Two-photon deep tissue ex vivo imaging of mouse dermal and subcutaneous structures. Opt Express 1998;3:339–350.

Dr. Tatiana B. Krasieva, PhD, Beckman Laser Institute, University of California, Irvine, 1002 Health Sciences Road East, Irvine, CA 92612 (USA) Tel. +1 949 824 3664, Fax +1 949 824 8413, E-Mail [email protected]

Krasieva

146


............................

Correlation of Tissue Fluorescence and Photodynamic Effect Michael F. Grahn, Janet K. Ansell, Martin L. de Jode Academic Department of Surgery, Queen Mary and Westfield College, Royal London Hospital, London, UK

Introduction That fluorescence and the photodynamic properties of sensitisers for photodynamic therapy (PDT) are closely linked has long been recognised. At the close of the last century, Prof H. von Tappeiner was reported as saying that ‘only fluorescent compounds have the possibility of photodynamic action’ [1]. In more recent times photosensitiser fluorescence has become prominent in both the fluorescence diagnosis of lesions and in the non-invasive measurement of photosensitiser concentrations. The aim of the present study was to determine whether the non-invasive technique of tissue surface fluorescence measurement could be used to determine the optimum time for PDT treatment.

Mouse Model for the Study of PDT Response In order to study the interrelationships between tissue surface fluorescence and the biological effects of PDT following irradiation, it is necessary to have a model in which the responses to PDT follow the applied ‘PDT dose’ in a reproducible way. The mouse model system employed in this study was designed, not to closely mimic any particular clinical situation, but in order to provide a simple model for investigating the role of the many variables influencing the PDT response in a quantitative manner [2]. Briefly, female BALB/c mice (Charles River (UK) Ltd) were implanted with the syngeneic tumour cell line colo26. The tumours were used after 12–16 days when they had reached an average diameter of 8–10 mm. Photosensitiser The photosensitiser mTHPC (Temoporfin, FoscanÔ ) was obtained from Scotia QuantaNova, Guildford. The drug was prepared on the day of use in its clinical dilution

vehicle (1 g ethanol and 1.5 g polyethylene glycol 400 made up to 5 ml with water for injection). The concentration was adjusted so that the correct dose for a 20-gram mouse was contained in 0.1 ml. The quantity injected was then adjusted according to the weight of each mouse. Injection was intraperitoneal using a 0.25-ml graduated glass syringe equipped with a 25-gauge needle. Care was taken to avoid exposure of the photosensitiser to light during preparation and injection. Tissue Fluorescence Measurement At set periods after photosensitiser injection the mice were anaesthetised and the skin overlying the tumour and the superficial group of thigh muscles was removed. The fluorescence intensity was measured at three points on the surface of both the tumour and the gastrocnemius muscle [3]. The mouse was then killed and tissues sampled and stored for subsequent assay. Tissue mTHPC content was measured by extraction and reverse phase high performance liquid chromatography [4]. Irradiation and Assessment of Acute PDT Effect An Oxford Lasers CU15 copper vapour laser pumping a DL20 dye laser was used containing DCM (Exciton, Dayton, Ohio) as the laser dye and tuned to an output of 652 nm. The laser was coupled to a 0.4-mm quartz optical fibre fitted with a microlens (Quadra Logic Technologies Inc., Vancouver, B.C., Canada). The hair on the skin overlying the tumours and the upper portion of the contralateral leg were depilated using a proprietary cream. The tumour and the superficial group of calf muscles (gastrocnemius, plantaris and soleus) were illuminated at a surface irradiance of 100 mW cm–2. During lasering the mice were sedated with Hypnorm (Janssen Pharmaceutical Ltd.). Acute tumour necrosis and muscle swelling were determined 24 h after irradiation by direct measurements of the depth of tumour necrosis and muscle swelling [2]. Briefly, the tumours were cut open parallel to the incident light and the maximum visible depth of any PDT necrosis (in mm) was measured from the tumour surface using a micrometer. Muscle oedema was quantified by carefully dissecting and weighing the gastrocnemius, plantaris and soleus muscles from both legs 24 h after irradiation. Muscle oedema was expressed as the percentage difference in weight between the control and treated muscles. The acute responses to PDT (measured 24 h after irradiation of the target tissues) are used as indicators of effect rather than the final outcome (such as tumour ‘cure’). The depth of necrosis is used as the indicator of tumour response. Oedema rather than necrosis is used as the indicator of response in skeletal muscle. This is used because, with the majority of photosensitisers, muscle damage requires a considerably greater ‘PDT dose’ than tumour damage and so would not provide as sensitive an indicator of damage in the dose range normally employed for tumour PDT. The interrelationship between oedema and necrosis in mouse skeletal muscle has been described by Chevretton et al. [5].

Characterisation of Acute PDT Responses That the acute PDT damage in both tumour and muscle is closely dependent upon the ‘PDT dose’ is shown by figures 1 and 2 which show tumour necrosis and muscle swelling over a range of both photosensitiser and light

Grahn/Ansell/de Jode

148

a

b

Fig. 1. a The influence of photosensitiser dose on tumour necrosis. The points show the mean×SEM (n>6) of tumour necrosis measured 24 h after irradiation with 5 J cm–2 652 nm light at 100 mW cm–2 over a range of photosensitiser doses and a drug-light interval of 24 hours. b The influence of light dose on tumour necrosis. The points show the mean×95% CI (n>8–16) of tumour necrosis measured 24 h after irradiation with 0–10 J cm–2 652 nm light at 100 mW cm–2 at 0.6 mg kg–1 mTHPC and a drug-light interval of 24 h.

a

b

Fig. 2. a The influence of photosensitiser dose on muscle swelling. The points show the mean×SEM (n>18) of muscle swelling measured 24 h after irradiation with 5 J cm–2 652 nm light at 100 mW cm–2 over a range of photosensitiser doses and a drug-light interval of 24 h. b The influence of light dose on muscle swelling. The points show the mean×95% CI (n>8–16) of muscle swelling measured 24 h after irradiation with 1–10 J cm–2 652 nm light at 100 mW cm–2 at 0.6 mg kg–1 mTHPC and a drug-light interval of 24 h.

Tissue Fluorescence and PDT Bioactivity

149

Fig. 3. The effect of the drug-light interval on PDT bioactivity using a log time scale. The points show the mean×SEM (n>12) of tumour necrosis and muscle swelling measured 24 h after PDT using 5 J cm–2 652 nm light at 100 mW cm–2 given between 1 and 240 h following the injection of 0.6 mg kg–1 mTHPC.

doses at a drug-light interval of 24 h. Whilst the two factors of photosensitiser and light dose are primarily responsible for determining the overall magnitude of the PDT response, the drug-light interval is most important in determining the relative magnitude of the response between different tissues. Whilst it is arguable whether the selective destruction of tumour can be achieved routinely in clinical practice in all but the most shallow of lesions, it is clear that the kinetics of both photosensitiser accumulation and of PDT effect differs between many tumours and non-tumour tissues. The combination of implanted tumour and skeletal muscle used in this model was chosen partly in order to allow the effects of this ‘selective photosensitiser retention’ to be studied. The Drug-Light Interval and PDT Response This differential in the time course of the magnitude of the PDT effect obtained for a given photosensitiser and light dose is shown in figure 3. Note that a log scale has been used for the time after drug injection in order to clarify the changes at early times. All the animals were given a single dose of photosensitiser on day 0, and then groups were irradiated with the same light dose at a number of intervals following drug administration. It can be seen that skeletal muscle damage reaches a maximum between 6 and 8 h following


150

photosensitiser injection after which the muscle sensitivity to PDT damage falls away. The tumour, on the other hand, does not reach maximum sensitivity until 24–48 h after photosensitiser injection, after which sensitivity also declines. It is clear that if, in this model system, the intention were to obtain maximal tumour damage whilst sparing skeletal muscle, then PDT should be given at a drug-light interval of 48 h. Whilst it is clear that such clear-cut differentials will rarely exist between the tumour and its surrounding tissues in clinical cases, the principle is the same and there will be an optimal time for effective PDT treatment in every case. Remittance Fluorescence Spectrophotometry A simple remittance fluorescence device was constructed by adapting a commercial spectrofluorimeter [3]. Excitation light from a 100-watt xenon lamp is passed through a monochromator into one limb of a random bifurcated quartz fibre bundle. The second limb of the bundle carries the returned scattered light and fluorescence and is directed into the collecting lens of the fluorimeter which passes this light through a second monochromator onto an extended-range photomultiplier. Tissue Fluorescence, Photosensitiser Content and PDT Activity It first had to be shown that the signals obtained using this device were proportional to the photosensitiser content of the tissue. This was tested by taking measurements from groups of animals 24 h after the administration of a range of photosensitiser concentrations. After fluorescence measurements the animals were killed and tissue samples were assayed for photosensitiser by extraction and HPLC. The results are shown in figures 4 and 5 for muscle and tumour, respectively. It can be seen that in either tissue the fluorescence signal is directly proportional to both the injected dose and tissue photosensitiser concentrations. It should, however, be noted that the fluorescence signal obtained for a given photosensitiser content differs markedly between the two tissues. In muscle a signal of 1 unit is obtained at about 10 ng kg–1 mTHPC whilst in the tumour a similar fluorescence signal of 1 unit requires a tissue content of over 30 ng kg–1. This difference in response is most likely due to differences in the optical properties of the two tissues and illustrates the danger in comparing raw fluorescence signals without taking tissue optical properties into account. The Drug-Light Interval, Tissue Fluorescence and PDT Bioactivity Having shown that the fluorescence signal is proportional to photosensitiser content a further experiment was carried out in which the fluorescence signals were compared with the PDT effect under the same conditions of a


151

Fig. 4. Correlation of muscle fluorescence and photosensitiser content. The points show the mean×SEM (n>6) of muscle surface fluorescence and mTHPC content by HPLC measured 24 h after the administration of 0–1.2 mg kg–1 mTHPC.

Fig. 5. Correlation of tumour fluorescence and photosensitiser content. The points show the mean×SEM (n>6) of tumour surface fluorescence and mTHPC content by HPLC measured 24 h after the administration of 0–1.2 mg kg–1 mTHPC.

range of photosensitiser doses given 24 h before fluorescence measurement and tissue irradiation. The results are shown in figure 6. As might be expected, there is a close correspondence between the tissue fluorescence signal and the biological effect in both tumour and muscle. Since it has been shown that tissue surface fluorescence was closely correlated with the magnitude of the PDT response at a drug-light interval of 24 h,


152

Fig. 6. The correlation of tissue surface fluorescence and PDT bioactivity. The points show the mean×SEM of tissue surface fluorescence (n>6) plotted against the mean×95% CI of PDT biological effect (n>16–46). Mice were injected with 0–1.2 mg kg–1 mTHPC 24 h after either the measurement of tissue surface fluorescence or irradiation with 5 J cm–2 of 652 nm light at 100 mW cm–2.

it might be expected that fluorescence would serve to predict the PDT response over the entire range of possible drug-light intervals. However, when the experiment was done (figures 7 and 8 for tumour and muscle, respectively), marked differences in the kinetics of the responses were uncovered. In the case of the tumour the peak fluorescence signal was not seen until 72 h after drug administration, 24 h after the peak of biological activity. In muscle the differences were even more marked, with peak fluorescence also occurring at 72 h even though the peak biological activity was seen at 6–8 h after drug administration. Further analysis showed that, whilst fluorescence was proportional to bioactivity at all times tested, the proportionality constant changed over the range of drug-light intervals. In general, the longer the interval after photosensitiser administration, the greater the bioactivity associated with any given tissue fluorescence signal. So that at early times a small fluorescence signal could be associated with a relatively large PDT response, whilst at later (?48 h) times a relatively large fluorescence signal was associated with only modest biological activity. Similar discrepancies between PDT activity and tissue fluorescence to those shown here [6] have been independently noted with mTHPC in the nude mouse [7]. The reasons for this difference in the sensitivity of tissue surface fluorescence over time after photosensitiser administration remain unclear. It is


153

Fig. 7. The correlation of tumour surface fluorescence and PDT bioactivity over a range of drug-light intervals. The points show the mean×SEM of tumour fluorescence (n>6) and depth of necrosis (n>12) following PDT between 1 and 240 h following drug injection. See the legend to figure 3 for details.

Fig. 8. The correlation of muscle surface fluorescence and PDT bioactivity over a range of drug-light intervals. The points show the mean×SEM of muscle fluorescence (n>6) and depth of necrosis (n>12) following PDT between 1 and 240 h following drug injection. See the legend to figure 3 for details.


154

unlikely to be associated with a change in tissue optical properties, since an influence of photosensitiser absorption on the overall tissue absorption should also have been seen on the photosensitiser dose-response curves shown in figures 4 and 6. It is possible that these phenomena are an artefact of the very limited sampling depth of tissue surface fluorescence probes using blue light excitation. In most tissues the excitation light could be expected to have an effective penetration depth of =1 mm. It is conceivable that, at early times after photosensitiser administration, the surface layer of tissues is relatively depleted in photosensitiser, leading to a relative under-sampling by the probe. Such a phenomenon could be detected in further experiments using fluorescence microscopy of tissue sections. Finally, this alteration in fluorescence sensitivity over time may be indicative of more subtle changes in the localisation or aggregation of the photosensitiser. Again studies using fluorescence microscopy and microspectrophotometry might be of utility in resolving these matters.

Conclusions The practical conclusion from this work is that, whilst measurements of surface fluorescence can be of great use in predicting tissue photosensitiser concentrations and PDT bioactivity in well-characterised tissues at a given time after drug administration, great caution must be employed when comparing measurements taken at different times after photosensitiser injection. The time of peak fluorescence does not necessarily indicate the time of peak PDT response.

Acknowledgments We thank Scotia QuantaNova and the Cancer Research Campaign for grants supporting this work.

References 1 2

3

Raab O: Ueber die Wirkung fluorescirender Stoffe auf Infusorien. Z Biol 1900;39:524–546. Ansell JK, de Jode ML, Grahn MF: Characterisation of a murine model for the rapid assessment of the acute photodynamic response in tumour and muscle. Lasers Med Sci 1997;12:336– 341. Grahn MF, de Jode ML, Dilkes ML, et al: Tissue photosensitizer detection by low-power remittance fluorimetry. Lasers Med Sci 1997;12:245–252.


155

4 5 6 7

Wang Q, Altermatt HJ, Ris HB, et al: Determination of 5,10,15,20-tetra-(m-hydroxyphenyl)chlorin in tissues by high performance liquid chromatography. Biomed Chromatogr 1993;7:155–157. Chevretton EB, Berenbaum MC, Bonnett R: The effect of photodynamic therapy on normal skeletal muscle in an animal model. Lasers Med Sci 1992;7:101–110. Ansell JK, de Jode ML, Grahn MF, Maudsley J, Williams NS: Correlation of photodynamic effect with surface fluorescence and drug assay. Lasers Surg Med 1995;7(suppl):41. Morlet L, Vonarx-Coinsmann V, Lenz P, et al: Correlation between meta(tetrahydroxyphenyl)chlorin (m-THPC) biodistribution and photodynamic effects in mice. J Photochem Photobiol B 1995;28: 25–22.

Dr. M.F. Grahn, Academic Department of Surgery, Queen Mary and Westfield College, Royal London Hospital, Whitechapel, London E1 1BB (UK) Tel. +44 207 377 7000 ext 3156, Fax +44 207 377 7283, E-Mail [email protected]


156


............................

Mitochondria as Targets for the Induction of Apoptosis in Photodynamic Therapy Christoph Richter Institute of Biochemistry, Swiss Federal Institute of Technology (ETH), Zu¨rich, Switzerland

Introduction Apoptosis is an evolutionarily conserved form of physiological cell death important for tissue development and homeostasis. The causes and execution mechanisms of apoptosis are incompletely understood. Mitochondria are important mediators of the execution of apoptosis. They provide ATP to meet the energy demands of this highly regulated form of cell death, and they provide at least two proteins (a serine protease and cytochrome c) which probably trigger a cascade of protease activities (caspases) important for apoptosis. A promising strategy to kill tumor cells is the induction of apoptosis in these cells. This may be achieved with mitochondrial poisons. Carcinoma cell mitochondria have a higher membrane potential (DW) than mitochondria in normal cells. Since DW drives lipophilic cations into mitochondria, the use of lipophilic cationic dyes in photodynamic therapy is suggested to selectively kill tumor cells through destabilization of mitochondria. Damage inflicted on mitochondria has been shown to result in apoptosis. Thus, we have found that oxidative stress-induced release of Ca2+ from mitochondria followed by Ca2+ reuptake (Ca2+ cycling) causes destabilization of mitochondria and apoptosis. We have also found that nitric oxide and its congeners can induce Ca2+ release from mitochondria. Thus, nitrogen monoxide binds to cytochrome oxidase, blocks respiration, and thereby causes mitochondrial deenergization and Ca2+ release. Peroxynitrite (ONOOÖ), on the other hand, causes Ca2+ release from mitochondria by stimulating a specific

Ca2+ release pathway. These findings suggest that nitric oxide (NO) and its congeners can induce apoptosis by destabilizing mitochondria via deenergization and/or by inducing a specific Ca2+ release followed by Ca2+ cycling. We have also found that tetra(m-hydroxyphenyl)chlorin (mTHPC), a compound used clinically to kill tumor cells, damages mitochondria in a light-dependent fashion. The action of mTHPC is manyfold, comprising modifications of the mitochondrial respiratory chain, Ca2+ handling, and lipid and protein oxidation. Collectively, these findings suggest a novel strategy, the use of lightactivatable lipophilic cations, in tumor cell killing.

Mitochondria and Cell Death Ca2+-Induced Necrotic and Apoptotic Cell Death Apoptosis is an evolutionarily conserved form of physiological cell death important for tissue development and homeostasis. Its hallmarks are distinct morphological alterations such as nuclear condensation, cell shrinkage, and bleb formation, and the absence of inflammatory responses of the affected tissue. Deranged apoptosis plays a major role in diseases such as cancer, acquired immune deficiency syndrome, autoimmune diseases, and neurodegeneration [1]. Necrosis and apoptosis are distinct forms of cell death, the former being a passive process typified by gross damage and spillage of the intracellular content, the latter being a highly regulated and controlled process that avoids inflammation and damage to the surrounding tissue [2]. Attention was drawn to Ca2+-induced cell death many years ago [3, 4]. Excessive intracellular Ca2+ is thought to contribute to a final common pathway of cytotoxic events leading to formation of reactive oxygen species (ROS), and necrosis or apoptosis. These events include overactivation of protein kinase C, Ca2+/calmodulindependent protein kinase II, phospholipases, proteases, protein phosphatases, xanthine oxidase, endonucleases, and NO synthase (NOS). Although the exact role of Ca2+ in cell killing is unclear, a disturbance of mitochondrial Ca2+ handling can be fatal [5, 6]. The normal Ca2+ uptake and release (‘cycling’) across the inner mitochondrial membrane requires little energy [7]. However, when the Ca2+ release pathway is stimulated by prooxidants, ‘cycling’ may become excessive and lead to loss of DW, general leakiness of the inner mitochondrial membrane, inhibition of ATP synthesis, mitochondrial damage, and cell death [8] (see also below). Mediators of Apoptosis Several conditions, molecules, or organelles such as oxidative stress, Ca2+, proteases, nucleases, or mitochondria are considered participants of apoptosis,

Richter

158

but at present it is not always clear whether they are required for or are the consequence of apoptosis. There is ample evidence that apoptosis is accompanied by oxidative stress [9]. A valuable tool used to elucidate the importance of oxidative stress is the protooncogen bcl-2 (see also below), which stimulates an antioxidative response in cells and prevents apoptosis [10, 11]. Also the requirement of Ca2+ for apoptosis is controversial [12]. Early reports suggested that a rise in intracellular Ca2+ leads to apoptosis via endonuclease activation, and more recent work indicated that apoptosis is accompanied by shifts of Ca2+ between various intracellular pools. It is worth noting that cellular Ca2+ handling and ROS production are related. Thus, increased mitochondrial Ca2+ release followed by reuptake driven by DW (Ca2+ ‘cycling’) stimulates ROS production. Mitochondria in Apoptosis Although the causes and execution mechanisms of apoptosis are not clearly understood oxidative stress, NO (note that here NO indicates nitric oxide independent of its redox state while NO·, NO+ and NOÖ [see below] refer to the nitrogen monoxide radical, nitrosonium ion, and nitroxyl anion, respectively) and its congeners, Ca2+, proteases, nucleases, and mitochondria, are important mediators of apoptosis. Recently, particularly the role of mitochondria in apoptosis was scrutinized [13, 14]. At present their importance and exact role are elusive but it is clear that mitochondria are both the target and the source of oxidative stress, NO, and Ca2+. During apoptosis DW, which is the driving force for mitochondrial ATP synthesis, declines, and maintenance of DW prevents apoptosis. Since apoptosis is highly regulated and involves the activity of hydrolytic enzymes, chromatin condensation and vesicle formation, apoptosis is likely to have a high-energy demand. Indeed, it was proposed that apoptosis induced by intracellular Ca2+ overload in neurons requires active mitochondria [15]. We proposed [14] that the cellular ATP level is an important determinant of cell death. Another line of evidence also puts mitochondria on the center stage of apoptosis: When they are destabilized, for example by Ca2+ ‘cycling’, they release proteins, some of which induce apoptosis. One is cytochrome c, which acts together with cytosolic factors to induce nuclear apoptosis [16]. The other is a 50-kD protease that by itself suffices to cause nuclear apoptosis [17]. The present knowledge allows the suggestion that mitochondria function as a cellular sensor of stress into which very different apoptosis induction pathways converge, and that mitochondria act as ‘central apoptotic executioners’ [18].

Photodynamic Destruction of Mitochondria

159

Nitric Oxide in Mitochondria Biology of NO NO is presently receiving enormous attention. It mediates beneficial responses such as the maintenance of blood pressure, inhibition of platelet aggregation, tumoricidal activities, or destruction of foreign invaders in the immune response, and is probably of major importance in long-term memory. The dichotomy of NO is in part due to a broad array of redox species with distinctive properties and reactivities: NO+ (nitrosonium), NO·, and NOÖ [19, 20], and the ability of NO· to combine with superoxide (OÖ 2 ) to yield peroxynitrite (ONOOÖ) [21]. NO in Mitochondria NO· and the Regulation of Cytochrome Oxidase The most cited and best understood physiological target of NO· is guanyl cyclase. NO· binds to and stimulates it and thus controls cell functions via cGMP, cGMP-gated channels, cGMP-dependent protein kinases, and phosphodiesterases. However, NO· also binds to cytochrome oxidase and reversibly inhibits respiration as seen with the isolated enzyme, submitochondrial particles, mitochondria, hepatocytes, brain nerve terminals, and astrocytes [22–28]. Cytochrome oxidase inhibition is competitive with oxygen due to binding of NO to the oxygen binding site of the reduced enzyme [29, 30]. Why the inhibition is transient is not clear at the moment but several findings point to consumption of NO as the underlying reason [31]. Thus, cytochrome oxidase can reduce NO· [32], NO· can combine with O2 to form NOx, and with OÖ 2 to form ONOOÖ. Concentrations of NO· measured in a range of biological systems are similar to those shown to inhibit cytochrome oxidase and mitochondrial respiration, and inhibition of NO· synthesis results in a stimulation of respiration in many systems. Therefore, it was recently proposed that NO· exerts a good part of its physiological and pathological effects on cells by inhibiting cytochrome oxidase [33]. Presence of NOS in Mitochondria NOS is present within mitochondria [34]. This knowledge offers exciting new insights into the biology of NO. For example, since the enzyme is stimulated by Ca2+ and located in the matrix or at the inner side of the inner mitochondrial membrane, this may provide a self-regulating system for mitochondrial Ca2+ homeostasis in which Ca2+ uptake by mitochondria would

Richter

160

lead to NO· formation. NO· could promote Ca2+ release (see below) by collapsing DW via inhibition of cytochrome oxidase.

Reactive Oxygen and Nitrogen Species as Regulators of Mitochondrial Ca2+ Homeostasis Mitochondria and Cellular CA2+ Homeostasis Intracellular Ca2+ regulates many processes. Its concentration is adjusted by binding to nonmembranous proteins, by mitochondria, and by membranebound Ca2+-ATPases located primarily in the plasma, nuclear, and endoplasmic reticular membrane [7]. Mitochondria contain Ca2+-sensitive targets regulated by moderate Ca2+ transients. These organelles are also able to take up large amounts of Ca2+ and buffer the cytosolic Ca2+. They thereby act as safety devices against potentially toxic increases in cytosolic Ca2+ [8]. Mitochondria take up and release Ca2+ by separate routes. As a consequence, Ca2+ is ‘cycled’ across their inner membrane [7]. Mitochondria are of central importance for physiological Ca2+ handling: they act as a reservoir for Ca2+, provide much of the ATP used by Ca2+ATPases, and Ca2+ regulates the activity of intramitochondrial dehydrogenases as well as nucleic acid and protein synthesis [35]. The importance of mitochondria as short-term modulators of cytosolic Ca2+ under physiological conditions was, until recently, considered minor. However, there is now compelling evidence [36, 37] that during physiological cell stimulation mitochondrial Ca2+ transport directly participates in the modulation and maintenance of cellular Ca2+ homeostasis. Several reports have additionally documented that physiological cytosolic Ca2+ pulses are relayed into mitochondria of brain, liver, and Xenopus laevis oocytes [38–40]. Mitochondrial Ca2+ Release In principle, Ca2+ can leave mitochondria in three ways: by nonspecific leakage through the inner membrane, by reversal of the influx carrier, and by an Na+-dependent or independent release pathway [7, 41]. Only the latter two are physiologically relevant because they operate when DW is high. The Na+dependent pathway predominates in mitochondria of the heart, brain, skeletal muscle, adrenal cortex, brown fat, and most tumor tissue. The Na+-independent pathway is important in the liver, kidney, lung, and smooth muscle mitochondria, probably exchanges Ca2+ with H+, and is linked to the redox state of mitochondrial pyridine nucleotides. Compounds causing their oxidation (and hydrolysis) promote Ca2+ release from intact mitochondria. This release has recently been reviewed [8, 35, 42].


161

Prooxidant-Induced, NAD+-Linked Ca2+ Release NAD+ Hydrolysis Is Required for Ca2+ Release from Intact Mitochondria Hydrogen peroxide can stimulate a specific Ca2+ release pathway from intact mitochondria by oxidizing mitochondrial pyridine nucleotides through the activities of glutathione peroxidase, glutathione reductase, and the energylinked transhydrogenase. Other prooxidants such as menadione, alloxan, or divicine also stimulate the specific Ca2+ release, because they furnish NAD+. The specific Ca2+ release requires for its activation the hydrolysis of intramitochondrial NAD+ to ADPribose and nicotinamide, and is prevented by inhibitors of NAD+ hydrolysis and protein monoADPribosylation. Recent experiments reveal that NAD+ hydrolysis and therefore Ca2+ release is regulated by vicinal thiols in mitochondria. When reduced or alkylated, the thiols prevent hydrolysis, but when they are cross-linked hydrolysis takes place. Cyclosporine A (CSA), which also prevents NAD+ hydrolysis, acts distal to these vicinal thiols [43]. NAD+ Hydrolysis Is under the Control of Vicinal Thiols NAD+ hydrolysis, and therefore ADPribosylation and Ca2+ release, are under the control of vicinal thiols: Phenylarsine oxide, which reversibly forms a five-membered ring with vicinal thiols, promotes the Ca2+-dependent intramitochondrial NAD+ hydrolysis and thereby the specific Ca2+ release [44]. Gliotoxin, a fungal metabolite carrying a disulfide moiety also promotes the Ca2+-dependent intramitochondrial NAD+ hydrolysis and thereby the specific Ca2+ release, but is inactive when its sulfurs are reduced or methylated [45]. Thus, intramitochondrial, Ca2+-dependent NAD+ hydrolysis is prevented when some vicinal thiols are in the reduced SH form, and occurs when they are connected, either by a cross-linking reagent or by oxidation to the disulfide form. Peroxynitrite (ONOOÖ) Stimulates the Specific Mitochondrial Ca2+ Release Pathway Since ONOOÖ oxidizes thiols, and since vicinal thiols control the specific mitochondrial Ca2+ release pathway, it was tested whether ONOOÖ is able to activate it [46]. ONOOÖ indeed induces Ca2+ release from rat liver mitochondria. This release occurs: (i) with preservation of DW; (ii) when mitochondrial pyridine nucleotides are oxidized but not when they are reduced; (iii) parallel to NAD+ hydrolysis; (iv) in a CSA-inhibitable manner; (v) without inhibition of respiration, and (vi) without entry of extra-mitochondrial solutes such as sucrose into mitochondria. These findings convincingly show that ONOOÖ

Richter

162

can mobilize mitochondrial Ca2+ by stimulating the specific. ADPribose-dependent release pathway. NO· Inhibits Cytochrome Oxidase and Causes Ca2+ Release from Mitochondria NO· at submicromolar, physiologically relevant concentrations potently deenergizes isolated mitochondria [24]. Deenergization is observed when mitochondria utilize respiratory substrates such as pyruvate plus malate, succinate, or ascorbate plus tetramethylphenylenediamine, but not when mitochondria are energized with ATP, and is due to a transient inhibition of cytochrome oxidase. The extent and duration of deenergization is determined by the concentration of NO· and oxygen, and the kind of respiratory substrate. The NO·-induced changes in the mitochondrial energy state are transient, and are paralleled by release and reuptake of mitochondrial Ca2+. Importantly, cytochrome oxidase is particularly sensitive to NO· at oxygen concentrations below 30 lM [47], i.e., at intracellular oxygen tensions. These findings reveal a direct action of NO· on the mitochondrial respiratory chain and suggest that NO· exerts some of its physiological and pathological effects by deenergizing mitochondria. Also in freshly prepared hepatocytes NO· deenergizes mitochondria [25]. Deenergization is reversible at low, but longer lasting at higher NO· concentrations. The drop and the recovery of DW are accompanied by a rise and fall in cytosolic Ca2+ levels. NO at higher concentrations, provided by nitrosoglutathione in combination with dithiothreitol (GSNO/DTT), kills hepatocytes apoptotically. Killing is reduced when the cytosolic Ca2+ is chelated, or when Ca2+ ‘cycling’ by mitochondria is prevented by CSA. Apparently NO can kill cells by releasing Ca2+ from mitochondria and thereby flooding the cytosol with Ca2+. Bcl-2 Links Oxidative Stress, Ca2+, and the Mitochondrial Membrane Potential to Apoptosis Given that bcl-2 elicits an antioxidative response in cells, what are the biochemical mechanism(s) by which bcl-2 prevents apoptosis? It was shown (see above) [48, 49] that one mechanism is the prevention of ROS-induced mitochondrial Ca2+ cycling, a process which results in a collapse of DW and in cellular ATP depletion. Thus, bcl-2 prevents disturbances in the cellular Ca2+ homeostasis and ROS production at the mitochondrial level. Based on these and other findings it was suggested [6] that a prooxidant-induced Ca2+ release from mitochondria, followed by Ca2+ ‘cycling’ and ATP depletion, is a common cause of apoptosis. Accordingly, maintenance of DW stabilizes mitochondria and thereby prevents apoptosis. Bcl-2 thus provides the link


163

between the antioxidant defense system, Ca2+, and DW [50]. In this context it is interesting to recall that many carcinoma cells have an increased DW (see below). Since prevention of apoptosis seems to contribute to carcinogenesis, it is conceivable that DW contributes to the decision between life and death of a cell. Mitochondrial Membrane Potential in Cancer Cells In pioneering work [51–54] it was revealed that carcinoma cell mitochondria have an increased DW. This finding corroborates the notion of increased DW in bcl-2-overexpressing cells [49]. It is conceivable that many cancer cells survive because of the increased DW, i.e. because their mitochondria are superstable and therefore do not liberate the serine protease, cytochrome c, or other yet to be identified factors important for the execution of apoptosis. The increased DW may be used to specifically target cancer cells [51–54]. NO and Peroxynitrite in Apoptosis It was recently shown in several systems that NO can cause apoptosis [55–58]. NO-induced apoptosis can be the consequence of DNA damage and subsequent expression of the tumor-suppressor gene p53 [58]. However, according to Mannik et al. [59] NO inhibits apoptosis in lymphocytes. Very recently it was found that also ONOOÖ induces apoptosis in a time- and concentration-dependent manner [23], and that depending on the concentration of ONOOÖ cells die either by apoptosis or necrosis [60]. Whether the NO congeners cause apoptosis due to interference with mitochondrial respiration or Ca2+ handling remains to be seen but it should be noted that the GSNO/ DTT-induced killing of hepatocytes [25] appears to engage mitochondrial Ca2+ cycling. Mitochondria Are Damaged by a Dye Successfully Used in Photodynamic Therapy mTHPC is used as a photosensitizer in photodynamic therapy (PDT), a novel modality for cancer treatment. Since little is known about mTHPCmediated damage in vitro, we chose isolated rat liver mitochondria as a model system to study its photodynamic effects [61]. Incubation of isolated mitochondria with mTHPC plus irradiation with light of a wavelength of 652 nm resulted in protein oxidation and lipid peroxidation, as measured by the mitochondrial content of carbonyl groups and thiobarbituric acid-reactive substances, respectively. Type-I and type-II photochemical reactions contribute to this oxidative damage as shown by the use of scavengers. Photodynamically treated mitochondria had a reduced membrane potential, and their Ca2+ uptake was impaired. Oxygen consumption of complex I or the respiratory chain was

Richter

164

stimulated at a low concentration of mTHPC plus irradiation, but decreased at higher concentrations, whereas oxygen consumption at complex II and IV decreased with all mTHPC concentrations offered. No mitochondrial changes were seen with mTHPC in the absence of irradiation. These results confirm the sensitivity of mitochondria to PDT and may help to understand the mechanisms by which PDT using mTHPC kills cells. Novel Strategy to Selectively Kill Tumor Cells As mentioned above carcinoma cell mitochondria are characterized by a high DW. This property has already been exploited to kill carcinoma cells in vivo by the use of the lipophilic di-cationic compound dequalinium [51]. The compounds of choice in future studies are di- or tri-cations since they can be expected to be four or nine times more effectively accumulated by mitochondria than mono-cations [51]. Since DW is also indirectly the driving force of anion uptake into mitochondria the use of anions also bears some promise.

Acknowledgments The work done in the author’s laboratory was generously supported over the years by the Schweizerische Nationalfonds, and partly by the Eidgeno¨ssische Technische Hochschule Zu¨rich. Drs. Gogvadze, Schlege, Schweizer, and Suter were also supported by the European Science Foundation, by the Schweizerische Krebsliga, and by an anonymous sponsor. S.D. Klein was supported by Leica and Avena Foundation, Switzerland.

References 1 2 3 4 5 6 7 8 9 10 11

Thompson CB: Apoptosis in the pathogenesis and treatment of diseases. Science 1995;267:1456– 1462. Steller H: Mechanisms and genes of cellular suicide. Science 1995;267:1445–1449. Schanne FA, Kane A, Young E, Farber J: Calcium dependence of toxic cell death: A final common pathway. Science 1979;206:699–700. Trump BF, Berezesky IK: Calcium-mediated cell injury and cell death. FASEB J 1995;9:219–228. Nicotera P, Bellomo G, Orrenius S: Calcium-mediated mechanisms in chemically induced cell death. Annu Rev Pharmacol Toxicol 1992;32:449–470. Richter C: Prooxidants and mitochondrial Ca2+: Their relationship to apoptosis and oncogenesis. FEBS Lett 1993;325:104–107. Carafoli E: Intracellular calcium homeostasis. Annu Rev Biochem 1987;56:395–433. Richter C, Kass GEN: Oxidative stress in mitochondria: Its relationship to cellular Ca2+ homeostasis, cell death, proliferation, and differentiation. Chem Biol Interact 1991;77:1–23. Buttke TM, Sandstrom PA: Oxidative stress as a mediator of apoptosis. Immunol Today 1994;15: 7–10. Steinman HM: The Bcl-2 oncoprotein functions as a prooxidant. J Biol Chem 1995;270:3487–3490. Garcia I, Martinou I, Tsujimoto Y, Martinou J-C: Prevention of programmed cell death of sympathetic neurons by the bcl-2 proto-oncogene. Science 1992;258:302–304.


165

12 13 14 15

16 17

18

19 20 21

22 23

24 25 26

27 28 29 30 31

32 33 34

Beaver JP, Waring P: Lack of correlation between early intracellular calcium ion rises and the onset of apoptosis in thymocytes. Immunol Cell Biol 1994;72:489–499. Kroemer G, Petit P, Zamzami N, Vayssie`re JL, Mignotte B: The biochemistry of programmed cell death. FASEB J 1995;9:1277–1287. Richter C, Schweizer M, Cossarizza A, Franceschi C: Hypothesis. Control of apoptosis by the cellular ATP level. FEBS Lett 1996;378:107–110. Ankarcrona M, Dypbukt JM, Bonfoco E, Zhivotovsky B, Orrenius S, Lipton SA, Nicotera P: Glutamate-induced neuronal death: A succession of necrosis or apoptosis depending on mitochondrial function. Neuron 1995;15:961–973. Liu X, Kim CN, Yang J, Jemmerson R, Wang X: Induction of apoptotic program in cell-free extracts: Requirement for dATP and cytochrome c. Cell 1996;86:147–157. Susin SA, Zamzami N, Castedo M, Hirsch T, Marchetti P, Macho A, Daugas E, Geuskens M, Kroemer G: Bcl-2 inhibits the mitochondrial release of an apoptotic protease. J Exp Med 1996; 184:1331–1342. Susin SA, Zamzami N, Castedo M, Daugas E, Wang H-G, Geley S, Fassy F, Reed JC, Kroemer G: The central executioner of apoptosis: Multiple connections between protease activation and mitochondria in Fas/APO-1/CD95- and ceramide-induced apoptosis. J Exp Med 1997;186:25–37. Stamler JS, Singel DJ, Loscalzo J: Biochemistry of nitric oxide and its redox-activated forms. Science 1992;258:1898–1902. Stamler JS: Redox signalling: Nitrosylation and related target interactions of nitric oxide. Cell 1994; 78:931–936. Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA: Apparent hydroxyl radical production by peroxynitrite: Implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci USA 1990;87:1620–1624. Brown GC, Cooper CE: Nanomolar concentrations of nitric oxide reversibly inhibit synaptosomal respiration by competing with oxygen at cytochrome oxidase. FEBS Lett 1994;356:295–298. Carr GJ, Ferguson SJ: Nitric oxide formed by nitrite reductase of Paracoccus denitrificans is sufficiently stable to inhibit cytochrome oxidase activity and is reduced by its reductase under aerobic conditions. Biochim Biophys Acta 1990;1017:57–62. Schweizer M, Richter C: Nitric oxide potently and reversibly deenergizes mitochondria at low oxygen tension. Biochem Biophys Res Commun 1994;204:169–175. Richter C, Gogvadze V, Schlapbach R, Schweizer M, Schlegel J: Nitric oxide kills hepatocytes by mobilizing mitochondrial calcium. Biochem Biophys Res Commun 1994;205:1143–1150. Cleeter MWJ, Cooper JM, Darley-Usmar VM, Moncada S, Schapira AHV: Reversible inhibition of cytochrome c oxidase, the terminal enzyme of the mitochondrial respiratory chain, by nitric oxide. Implications for neurodegenerative diseases. FEBS Lett 1994;345:50–54. Brown GC, Bolanõs JP, Heales SJR, Clark JB: Nitric oxide produced by activated astrocytes rapidly and reversibly inhibits cellular respiration. Neurosci Lett 1995;193:201–204. Takehara Y, Kanno T, Yoshioka T, Inoue M, Utsumi K: Oxygen-dependent regulation of mitochondrial energy metabolism by nitric oxide. Arch Biochem Biophys 1995;323:27–32. Brudvig OW, Stevens OH, Chan OI: Reactions of nitric oxide with cytochrome oxidase. Biochemistry 1980;19:5275–5285. Torres J, Darley-Usmar V, Wilson MT: Inhibition of cytochrome c oxidase in turnover by nitric oxide: Mechanism and implications for control of respiration. Biochem J 1995;312:169–173. Clarkson RB, Norby SW, Smirnov A, Boyer S, Vahidi N, Nims RW, Wink DA: Direct measurement of the accumulation and mitochondrial conversion of nitric oxide within Chinese hamster ovary cells using an intracellular electron paramagnetic resonance technique. Biochim Biophys Acta 1995; 1243:496–502. Zhao XJ, Sampath V, Caughey WS: Cytochrome c oxidase catalysis of the reduction of nitric oxide to nitrous oxide. Biochem Biophys Res Commun 1995;212:1054–1060. Brown GC: Nitric oxide regulates mitochondrial respiration and cell functions by inhibiting cytochrome oxidase. FEBS Lett 1995;369:136–139. Ghafourifar P, Richter C: Nitric oxide synthase activity in mitochondria. FEBS Lett 1997;418: 291–296.

Richter

166

35 36 37 38 39

40 41 42 43 44 45 46 47

48 49

50

51

52 53 54

55 56

57 58

Richter C: Mitochondrial calcium transport; in Ernster L (ed): Molecular Mechanisms in Bioenergetics. New Comprehensive Biochemistry. Amsterdam, Elsevier, 1992, pp 349–358. Rizzuto R, Bastianutto C, Brini M, Murgia M, Pozzan T: Mitochondrial Ca2+ homeostasis in intact cells. J Cell Biol 1994;126:1183–1194. Hajnoćzky G, Robb-Gaspers LD, Seitz M, Thomas AP: Decoding of cytosolic calcium oscillations in the mitochondria. Cell 1995;82:415–424. Loew LM, Carrington W, Tuft RA, FAy FS: Physiological cytosolic Ca2+ transients evoke concurrent mitochondrial depolarization. Proc Natl Acad Sci USA 1994;91:4340–4344. Sparagna GC, Gunter KK, Sheu S-S, Gunter TE: Mitochondrial calcium uptake from physiologicaltype pulses of calcium. A description of the rapid uptake mode. J Biol Chem 1995;270:27510– 27515. Jouaville LS, Ichas F, Holmuhamedov EL, Camacho P, Lechleiter JD: Synchronization of calcium waves by mitochondrial substrates in Xenopus laevis oocytes. Nature 1995;377:438–441. Crompton M: The regulation of mitochondrial calcium transport in heart. Curr Top Membr Transport 1985;25:231–276. Richter C, Frei B: Ca2+ release from mitochondria induced by prooxidants. Free Radic Biol Med 1988;4:365–375. Richter C: Control of the pro-oxidant-dependent calcium release from intact liver mitochondria. Redox Report 1996;2:217–221. Schweizer M, Durrer P, Richter C: Phenylarsine oxide stimulates the pyridine nucleotide-linked Ca2+ release from rat liver mitochondria. Biochem Pharmacol 1994;48:967–973. Schweizer M, Richter C: Gliotoxin stimulates Ca2+ release from intact rat liver mitochondria. Biochemistry 1994;33:13401–13405. Schweizer M, Richter C: Peroxynitrite stimulates the pyridine nucleotide-linked Ca2+ release from intact rat liver mitochondria. Biochemistry 1996;35:4524–4528. Richter C, Gogvadze V, Laffranchi R, Schlapbach R, Schweizer M, Suter M, Walter P, Yaffee M: Oxidants in mitochondria: From physiology to diseases. Biochim Biophys Acta 1995;1271:67– 74. Hennet T, Richter C, Peterhans E: Tumour necrosis factor-a induces superoxide anion generation in mitochondria of L929 cells. Biochem J 1993;289:587–592. Hennet T, Peterhans E, Richter C, Bertoni G: Expression of BCL-2 protein enhances the survival of mouse fibrosarcoid cells in tumour necrosis factor-mediated cytotoxicity. Cancer Res 1993;53: 1456–1460. Bornkamm GW, Richter C: A link between the antioxidant defense system and calcium: A proposal for the biochemical function of Bcl-2; in Potter M, Melchers F (eds): Mechanisms in B-Cell Neoplasia. Current Topics in Microbiology and Immunology. Berlin, Springer, 1994, vol 194, pp 323–330. Weiss MJ, Wong JR, Ha CS, Bleday R, Salem RR, Steele GD Jr, Chen LB: Dequalinium, a topical antimicrobial agent, displays anticarcinoma activity based on selective mitochondrial accumulation. Proc Natl Acad Sci USA 1987;84:5444–5448. Chen LB: Mitochondrial membrane potential in living cells. Annu Rev Cell Biol 1988;4:155–181. Sun X, Wong JR, Hu J, Garlid KD, Chen LB: AA1, a newly synthesized monovalent lipophilic cation, expresses potent in vivo antitumor activity. Cancer Res 1994;54:1465–1471. Koya K, Li Y, Wang H, Ukai T, Tatsuta N, Kawakami M, Shishido T, Chen LB: MKT-077, a novel rhodacyanine dye in clinical trials, exhibits anticarcinoma activity in preclinical studies based on selective mitochondrial accumulation. Cancer Res 1996;56:538–543. Albina JE, Cui S, Mateo RB, Reichner JS: Nitric oxide-mediated apoptosis in murine peritoneal macrophages. J Immunol 1993;150:5080–5085. Cui S, Reichner JS, Mateo RB, Albina JE: Activated murine macrophages induce apoptosis in tumor cells through nitric oxide-dependent and -independent mechanisms. Cancer Res 1993;54: 2462–2467. Ankarcrona M, Dypbukt JM, Bru¨ne B, Nicotera P: Interleukin 1-b-induced nitric oxide production activates apoptosis in pancreatic RINm5F cells. Exp Cell Res 1994;213:172–177. Messmer UK, Ankarcrona M, Nicotera P, Bru¨ne B: p53 expression in nitric oxide-induced apoptosis. FEBS Lett 1994;355:23–26.


167

59 60

61

Mannik JB, Asano K, Izumi K, Kieff E, Stamler JS: Nitric oxide produced by human B lymphocytes inhibits apoptosis and Epstein-Barr virus reactivation. Cell 1994;79:1137–1146. Bonfoco E, Krainc D, Ankarcrona M, Nicotera P, Lipton SA: Apoptosis and necrosis: Two distinct events induced, respectively, by mild and intense insults with N-methyl-D-aspartate or nitric oxide/ superoxide in cortical cell neurons. Proc Natl Acad Sci USA 1995;92:7162–7166. Klein SD, Walt H, Richter C: Photosensitization of isolated rat liver mitochondria by tetra(mhydroxyphenyl)chlorin. Arch Biochem Biophys 1997;348:313–319.

Dr. C. Richter, PhD, Institute of Biochemistry, Swiss Federal Institute of Technology (ETH), Universita¨tstrasse 16, CH–8092 Zu¨rich (Switzerland) Tel. +41 1 632 31 36 or +41 1 632 30 21, Fax +41 1 632 11 21, E-Mail [email protected]

Richter

168


............................

Cellular Effects of Photodynamic Therapy with Clinical Relevance T. Patrice, N. Rousset, S. Eleouet, J. Carre´, V. Vonarx, Y. Lajat Laser Department, Neurosurgery, Laennec Hospital, Nantes, France

Whatever the sensitizer used, photodynamic therapy (PDT) will alter several cell sites and organelles according to the cell sensitizer distribution, and the level of destruction for each site will depend on the sensitizer concentration. Some damage is not lethal and will be repaired, while some other damage will lead to cell death after various delays. However, the sensitizer concentration varies according to the incubation time with the sensitizer. Light illumination will thus have different effects according to time (fig. 1). This shows the tremendous importance of cell kinetics in sensitizer uptake [1, 2]. This is even more so in solid tumors that are made up of many cell types, some of them being malignant with various kinetics and biochemical patterns although arising from a single mother cell. Determining an optimal moment for irradiation is a challenge that can be approached through experimental procedures. Cell uptake monitored by fluorescence determination of a given sensitizer shows that the maximal fluorescence varies from one cancer cell type to another (fig. 2). Such a heterogeneity is clearly a general drawback of any anticancer chemotherapy including PDT. This is even worse when using a prodrug (ALA) which is said to be more likely metabolized into a photosensitizer (PPIX) in cancer cells than in normal cells. In this case cancer cells exhibit various cell kinetics at the mitotic level but also differences in their enzymatic equipment. If we consider leukemic cells obtained from 1 patient incubated with ALA, we can observe huge standard errors indicating a strong heterogeneity in ALA metabolism [3, 4]. This is also true for normal lymphocytes (fig. 2, 3). Therefore, when is the best moment to irradiate cells? This heterogeneity is also a problem when we are trying to find evidence of selective cancer cell destruction. Years ago, we demonstrated that L1210

Fig. 1. Influence of the cell line studied on the toxicity (0 J/cm2) or phototoxicity (25 J/cm ) of mTHPC (5 lg/ml) added for 3 h to cells before 514-nm irradiation. B16, C6, HT29, REGb, PROb and L1210 cell lines have been studied. 2

Fig. 2. ALA-induced (50 lg/ml) PpIX fluorescence in B (X) or T (T) lymphocytes from healthy subjects. Fluorescence had been measured every hour in living cells through a confocal microspectrofluorimeter allowing spectral fluorescence recording of an about 2-lm2 area.

murine leukemic cells were more sensitive to hematoporphyrin derivative PDT (HPD-PDT) than the semi-syngenic corresponding hemoprogenitors, and this was found whatever the conditions of incubation or irradiation [5]. It is probable that selectivity exists but in some cases just for a limited period of time, the determination of which is of course important in each patient.

Patrice/Rousset/Eleouet/Carre´/Vonarx/Lajat

170

a

b Fig. 3. ALA-induced (50 lg/ml) PpIX fluorescence in lymphocytes from 2 patients (a, b) with chronic lymphocytic leukemia. Fluorescence was measured every hour in living cells through a confocal microspectrofluorimeter allowing spectral fluorescence recording of an about 2-lm2 area.

It is also of interest to analyze the fluorescence distribution within a given cell. Most anticancer agents are aimed at stopping or at least modulating cell replication. DNA is a proeminent target. Since these treatments are obviously very poorly selective they are also powerfully genotoxic for normal cells. But as they are not able by themselves to cure patients, this is only of relative importance. It is quite different with PDT. PDT is aimed at treating patients with a curative purpose. Tumors are small, poorly invading, rather well differentiated and nonmetastatic at the time of diagnosis. It is thus necessary to demonstrate the safety of PDT on normal cell genes. Fluorescence distribution analysis will provide very important data concerning sensitizer nuclear uptake. Most sensitizers, for example mTHPC, only have a very weak affinity for the nucleus (fig. 4). This has to be correlated with DNA-induced lesions observed after PDT [6]. By analyzing the DNA content of gastrointestinal tumors in humans before and after PDT we could observe the emergence of aneuploid clones that were not detected before

Cellular Effects of Photodynamic Therapy with Clinical Relevance

171

Fig. 4. 3D mTHPC (10 mg/ml) fluorescence distribution in C6 glioma cell line after 3 h of incubation. Fluorescence was measured every hour in living cells through a confocal microspectrofluorimeter allowing spectral fluorescence recording of an about 2-lm2 area. The Z axis reflects the fluorescence intensity detected at 650 nm. The nuclear area is clearly nonfluorescent.

treatment. By considering the short delay between PDT and DNA determination we can estimate that DNA abnormalities existed before treatment and were not PDT-induced but were rather allowed to grow as more PDT-sensitive clones disappeared, reducing the self-growth inhibition. Among cell targets, membranes have been shown to be major. The word ‘membrane’ has to be understood in a wider sense than just cytoplasmic membrane. It has been found that PDT could alter almost all membranes within a cell with different consequences according to the organelle involved. Lysosomal destruction induces enzymatic destruction of cell structures by the released enzymes themselves. Mitochondrial destruction alters the differences in ion concentration between the intracellular compartment and the external one through a decrease in energy production required by ionic exchangers at the cytoplasmic level. Calcium exchanges in particular are strongly modified and a rise in intracellular calcium was found. This rise can in turn induce a cascade of biochemical events including lipid degradation in the membranes, eventually at a distance from the initial PDT-induced lesion. These changes, shown by the rise in malondialdehyde content (a final product of lipid degradation) in PDT-treated tissues [7], are greater in successfully treated tumors in man (fig. 5). It is also possible that calcium changes could explain the prostaglandin production observed after PDT [8]. Cytoplasmic membranes are of course the first barrier encountered by a sensitizer administered to a patient. If light illuminates membranes when in


172

a

b Fig. 5. Evolution of malondialdehyde (MDA) content of biopsies taken in patients with gastrointestinal cancers before or after HPD-PDT. MDA was determined by thiobarbituric fluorescence assay. a MDA evolution in 17 patients (complete responding or nonresponding to PDT). b MDA evolution according to the initial histology.

Fig. 6. Effect of Photofrin (incubation 2 h) alone or followed by irradiation on cytolitic activity (Cr51 cytolitic assay after 24 h) of anti-P511 myeloma cell lymphocytes (effector/ target ratio 100/1).

contact with a sensitizer it induces lipid peroxydations and direct changes in ion equilibrium due to ‘holes’. However, besides this rather trivial damage, PDT can alter all kinds of receptors or antigenic determinants and is able to induce modifications in cell behavior if they survive to PDT. It has been shown that PDT is likely to induce damage in estrogenic receptors [9], estrogen being known as a growth factor for many types of cell. We also noticed that HPDPDT could modify receptors to epidermal growth factors (EGF) [10]. This destruction could influence cell recovery after PDT, both in cancer and normal cells. It could be possible to modify this influence by giving EGF analogs after PDT.


173

a

b

Fig. 7. Variation of adhesion molecules and lectines expression on: a BPD-PDT-treated PROb cells (50 ng/ml, 1 h of incubation, 10 J/cm2 at 457 nm), or (b) Photofrin-PDT treated cells (0.5lg/ml, 2 h of incubation, 25 J/cm2 at 514 nm) as compared to untreated control cells.

PDT of cancer cells alters the immune response. It has been shown for years that PUVA or, more generally, light illumination strongly induced immune suppression. Some authors even believe that it could play a role as at least a cofactor in some acquired immune suppressions. PDT induces a decrease in nonspecific as well as specific immune responses (fig. 6) to cancer [11]. Theoretically, this is a rather negative effect of PDT within the cancer field, although the immune tolerance of cancer antigens already occurred long before PDT. It is also possible that this could open new fields for PDT application. Several years ago we also noticed that HPD-PDT decreased cancer cell adhesiveness to cocultured endothelial cells. We correlated this decrease to a reduced rate of metastasis when PDT-treated colonic cancer cells were reinjected to syngeneic rats [12]. When nonmetastatic cells were PDT treated they did not acquire the metastazing phenotype. More recently we noticed that the benzoporphyrin derivative or Photofrin (B) PDT-treated cells had reduced


174

expression of CD44V antigens (fig. 7), a molecule known to be overexpressed in metastatic cells [13]. This CD44V is also involved in the development of apoptotic processes occurring after cell treatment. We could not observe any recovery of CD44V decrease for at last 24 h in vitro, which is a pretty long time for treated cells. Analyzing expression of such antigens is not just for scientific fun. The best indications for PDT are early cancers, this meaning that tumors are not metastatic at the time of diagnosis. It is thus of importance to characterize PDT safety to make it sure that we do not impair the prognosis of the disease we are treating.

References 1

2

3 4 5

6

7

8 9

10

11 12 13

Morlet L, Vonarx V, Lenz P, Foultier MT, Xavier de Brito L, Stewart C, Patrice T: Correlation between m-THPC biodistribution and photodynamic effects in mice. J Photochem Photobiol 1995; 28:25–32. Blais J, Amirand C, Ballini JP, Debey P, Foultier MT, Patrice T: Photofrin induced fluorescence in progressive and regressive cancer cells: Correlation with cell photosensitivity. J Photochem Photobiol B 1995;27:225–231. Eleouet S, Carre J, Vonarx V, Heyman D, Lajat Y, Patrice T: Delta amino levulinic acid-induced fluorescence in normal human lymphocytes. J Photochem Photobiol B 1997;41:22–29. Eleouet S, Carre´ J, Rousset N, Vonarx V, Lajat Y, Patrice T: Heterogeneity of PpIX ALA induced in some human cells. Proc 7th Biennial Congr Int Photodynamic Assoc, CD ROM, p 34. Foultier MT, Patrice T, Praloran V, Robillard H, Le Bodic L: Influence of hematoporphyrin derivative concentration, incubation time, temperature during incubation and laser dose fractionation on photosensitivity of normal hemopoietic progenitors or leukemic cells. Biochimie 1989;71:819–825. Foultier MT, Vonarx-Coinsman V, Xavier de Brito L, Morlet L, Robillard N, Patrice T: DNA or cell kinetics flow cytometry analysis of 33 small gastrointestinal cancers treated by photodynamic therapy. Cancer 1994;73:1595–1607. Perret C, Foultier MT, Vonarx-Coinsman V, Quancard O, Morlet L, Patrice T: Malondialdehyde dosimetry in laser irradiated tissues sensitized by hematoporphyrin derivative. J Pharmacol Exp Med 1994;269:787–791. Fingar VH, Wieman J, Doak KW: Role of thromboxane and prostaglandin release on photodynamic therapy induced tumor destruction. Cancer Res 1990;50:2599–2603. Lanzafame RJ, Rogers DW, Naim JO, Herrera HR, Hinshaw R: Effect of hematoporphyrin derivative on oestogen receptors in the dimethylbenzanthracen-mammary tumor model. Lasers Surg Med 1987;6:543–545. Fanuel-Barret D, Patrice T, Foultier MT, Vonarx V, Robillard N, Lajat Y: Influence of epidermal growth factor on photodynamic therapy of glioblastomas cells in vitro. Res Exp Med (Berl) 1997; 197:219–233. Vonarx V, Anasagasti L, Lajat Y, Patrice T: Photodynamic effect on the specific antitumor immune activity. Int J Immunopharmacol 1997;19:101–110. Vonarx V, Foultier MT, Xavier de Brito L, Morlet L, Patrice T: Photodynamic therapy decreased cancer colonic adhesiveness and metastatic potential. Res Exp Med 1995;195:101–116. Rousset N, Vonarx V, Eleouet S, Carre J, Lependu J, Lajat Y, Patrice T: Photodynamic effect on cell adhesion molecules. J Photochem Photobiol B 1999, in press.

Dr. T. Patrice, Laser Department, Neurosurgery, Laennec Hospital, F–44093 Nantes (France) Tel. +33 240 16 56 75, Fax +33 240 16 59 35, E-Mail [email protected]


175

Endometrium Wyss P, Tadir Y, Tromberg BJ, Haller U (eds): Photomedicine in Gynecology and Reproduction. Basel, Karger, 2000, pp 176–182

............................

Introduction P. Wyss a, R.A. Steiner a, M.J. Gannon b, R.L. Reid c, Y. Tadir d a

Department of Gynecology and Obstetrics, University Hospital of Zu¨rich, Switzerland; b Department of Gynecology and Obstetrics, The General Infirmary, Leeds, UK; c Department of Obstetrics and Gynecology, Queens University, Kingston, Ont., Canada, and d Beckman Laser Institute and Medical Clinic, University of California, Irvine, Calif., USA

Approximately 4% of all woman of reproductive age consult their physician on an annual basis because of menorrhagia [1]. Menorrhagia, defined as excessive uterine bleeding, may be a source of social embarrassment, physical distress, and economic hardship for the afflicted woman [2]. Menorrhagia has been treated medically by induction of ovulation, oral contraceptives, cyclic or continuous progestin therapy, danazol, prostglandin synthetase inhibitors, antifibrinolytics such as tranexamic acid, or medical ovarian suppression with LHRH agonists. Each of these medications may be associated with significant expense and side effects. Often medical means to control menstrual bleeding provide only temporary relief and many of these women will ultimately submit to hysterectomy in order to achieve a more permanent solution [3, 4]. It is estimated that approximately 40% of the 700,000 hysterectomies performed annually in North America are for treatment of menorrhagia [1]. Hysterectomy may be associated with a variety of complications, and the cost to society from both the procedure and the loss of productivity it entails are staggering. Endometrial ablation has long been recognized as an alternative to hysterectomy for treating menorrhagia. Minimally invasive hysteroscopic ablative surgery such as Nd:YAG laser coagulation [5] and electrocautery with resectoscope [6] or rollerball [7] have become popular [8–11]. These procedures require specialized training and equipment. They have been associated with a variety of complications, including uterine perforation, electrocautery damage to adjacent structures, intravasation of uterine distending medium with cardiovascular overload, embolization and death, and may be followed by continued bleeding

or sequestration of endometrial tissue due to incomplete endometrial destruction [4, 12–17]. More recently, a number of nonhysteroscopic techniques have been used. These techniques include radiofrequency [18] and microwave devices [19], thermal [20] or multielectrode balloons [21], and hot water instillation [22]. But these approaches are based on temperature elevations up to 90 ºC requiring some form of anesthesia and may not result in ablation of the whole endometrium. Additionally, there is no intrinsic specificity for endometrium over myometrium and adjacent organs. In contrast to endometrial ablation by heat, photodynamic therapy (PDT) provides photooxidation-induced, selective endometrial tissue destruction. PDT has been developed largely for cancer treatment, many thousands of patients have now been treated worldwide and the approach has been approved by the FDA and other national drug-licensing authorities, albeit so far for a limited number of applications. The human endometrium exhibits several features that meet the requirements for PDT: it is easily accessible, only a few millimeters thick, and it is surrounded by a thick muscle layer that may act as protective shield for the intra-abdominal organs. Moreover, it is a vascular, neoproliferative tissue highly selective to hormonal stimulation. The substantial effect PDT has on tumor microvasculature [23, 24] suggests that it should be highly effective in the well-vascularized endometrium. As such, endometrial tissue may serve as an experimental model for the photodynamic effects, and endometrial disease may potentially be treated by this therapeutical concept. The only drug currently approved is Photofrin, a complex mixture of porphyrin monomers and oligomers, although many new and potentially improved photosensitizers are currently under study, some of them in clinical trials. One of the disadvantages of systemic photosensitizer application is that it has a long residence time in the body and, in particular, it enters the skin where it causes a prolonged skin photosensitivity. Local intrauterine photosensitizer administration may provide lower drug doses and avoidance of skin photosensitivity. However, several complex, interrelated factors including, photosensitizer type, concentration and delivery mode (systemic vs. topical), as well as light dose, light delivery, and timing ultimately determine the success or failure of endometrial PDT. These general parameters must be well characterized in model systems in order to optimize the probability of success in humans.

Chronology of Photomedical Endometrial Studies (table 1) 1978: Dougherty et al. [25] reported complete or partial response in 111 of 113 malignant lesions using a hematoporphyrin derivative and exposure

Endometrium: Introduction

177

Table 1. Photomedical endometrial studies Reference

Model

Dose mg/kg

Laser nm

Dougherty et al. [25], 1978 Human endometrium HpD i.v. cancer

5

630–700 Endometrium cancer metastasis destruction

Rettenmaier et al. [26], 1984

Human endometrium HpD i.v. cancer

3

630

Endometrium cancer metastasis destruction

Schneider et al. [27], 1988

Rat endometrium

DHE i.v.

7

630

Endometrial ablation

Schneider et al. [28], 1988

Rat endometrium

DHE i.v.

7

630

Effect of estrogen

Raab et al. [29], 1990

Human endometrial cancer

Porphyrin in vitro

0–10 lg/ml 630 medium

In vitro response of endometrial cancer to PDT

Bhatta et al. [30], 1992

Rabbit endometrium

DHE i.v.

1, 2, 5, 10

630


Judd et al. [31], 1992

Rabbit endometrium

ALA i.v.

200

–

Compare tissue fluorescenceuptake

Kennedy and Pottier [32], 1992

Mouse uterus

ALA i.v.

N/A

–


Chapman et al. [33], 1993

Rat endometrium

DHE i.v., i.p., i.u.

7/0.7

–


Steiner et al. [34, 35], 1993 Rat endometrium

Photosensitizer

Aim

DHE i.u. 0.7 (plus Azon) BPD i.u. 1 ALA i.u. 58

630

Fluorescence uptake

690 630

Reproductive performance Endometrial ablation

Yang et al. [36], 1993

Rat endometrium

ALA i.u.

40, 80, 160 mg/ml

red light Endometrial ablation

Wyss et al. [37], 1994

Rabbit endometrium

BPD i.u.

2 mg/ml

690

Fluorescence uptake Endometrial ablation

Wyss et al. [38], 1994

Rabbit endometrium

ALA i.u.

200 mg/ml

630

Fluorescence uptake Endometrial ablation

Gannon et al. [39], 1995

Human endometrium ALA i.u.

25–900 mg

–

Pharmacokinetics of ALA

Wyss-Desserich et al. [40], 1996

Human endometrial cancer

ALA in vitro

1 mg/ml medium

630

Pharmacokinetics of ALA in normal and neoplastic human endometrial cells

Wyss et al. [40], 1996

Rabbit endometrium

ALA i.u.

200 mg/ml

630

Regeneration processes after lowlight doses

Yang et al. [42], 1996

Monkey endometrium ALA i.u., i.v.

100, 250 – mg/ml (i.u.) 15, 150, 250 mg/kg (i.v.)

Pharmacokinetics and toxicology of ALA

Fehr et al. [43], 1996


400 mg/ml

–

Pharmacokinetics of ALA

Roy et al. [44], 1997

Rat endometrium

ALA i.u.

1–50 mg/ 100 ll

–

Dose-response relation

Roy et al. [45], 1997

Rat endometrium

ALA i.u.

10, 25 mg

–

Dose-response, application mode

Gannon et al. [46], 1997


400 mg/ml

630


Wyss et al. [47], 1998


400 mg/ml

630


Wyss/Steiner/Gannon/Reid/Tadir

178

times of 20 min at 100 mW/cm2. Two of the tumors with complete response to PDT were endometrium carcinoma metastases. A specific description of the procedure in these 2 cases is not evident in the study. 1984: Rettenmaier et al. [26] reported PDT of endometrial cancer recurrent to vulva and groin. A hematoporphyrin derivative was given as an intravenous bolus at a dose of 3 mg/kg. Seventy-two hours later, the lesions were exposed to light of 630 mm provided by an argon ion-pumped dye laser using light dose of 40 J/cm2. Vulvar metastases were destroyed whereas groin metastases were persistent. Aside from the usual treatment photosensitivity, no side effects were noted. 1988: Schneider et al. [27, 28] studied the potential use of PDT for selective endometrial ablation in rat uteri. The same group further evaluated the influence of estrogen on the uptake and localization of dihematoporphyrine ether (DHE) in the uterus of ovariectomized rats. The photosensitizer was concentrated in the endometrial tissue and estrogen treatment significantly increased the uterine uptake but had no effect on other organs. 1990: The response of human gynecological carcinoma cell-lines HEC1-A (endometrial carcinoma) to PDT in vitro was studied by Raab et al. [29]. Cell lines did not survive PDT. Complete endometrial cell death was observed after application of irradiation doses of 10 J/cm2 combined with porphyrin concentrations of 5 lg/ml at fixed incubation of 48 h. 1992: Endometrial ablation by means of PDT with DHE in the rabbit was described by Bhatta et al. [30]. The drug was injected i.v. (1, 2, 5 and 10 mg/kg) and 24 h later, intrauterine laser illumination at 630 mm was administered. The authors concluded that endometrial ablation can be effectively achieved in rabbits by means of PDT. Judd et al. [31] studied the fluorescence of the uterine layers following i.v. injection of aminolevulinic acid (ALA) and phthalocyanine. The endometrium showed a peak fluorescence at 2 and 3 h with ALA and phthalocyanine, respectively. When using ALA, the endometrial layer showed fluorescence levels 5 times higher than the myometrium. Kennedy and Pottier [32] studied the fluorescence of ALA in the mouse uterus. Following i.v. injection, the endometrium became strongly fluorescent whereas the myometrium did not. 1993: Since photosensitivity signifies a major side effect following systemic sensitizer application, pharmacokinetic studies using topical drug administration were performed by Chapman et al. [33]. Intrauterine (i.u.) delivery of the photosensitizer in rats appeared to allow more selective retention within the surface endometrial cells although Photofrin concentrations were 10 times less for i.u. compared to intraperitoneal or intravenous application. Finally, despite a 10-fold reduction in dose, i.u. application yielded a significant increase


179

in extracted Photofrin of the endometrium, lending support to the hypothesis that the site-specific delivery of the photosensitizer can achieve selective retention of the drug at a much reduced dose. In a subsequent experiment, all rats were administered to the photosensitizer by i.u. application. The main question was focused on the effect of estrogen on uptake and retention of the photosensitizer within the uterine layers. Estrogen stimulation resulted in a significantly higher DHE uptake in the endometrial epithelial cells compared to the epithelium of nonstimulated rats. No significant difference was measured in the endometrial stroma and myometrium. Steiner et al. [34, 35] were the first to study PDT in the uterus after local i.u. application of both the photosensitizer as well as the laser light. Using different photosensitizers they were able to demonstrate selective destruction of the endometrium resulting in a greatly reduced reproductive performance. Yang et al. [36] injected ALA at doses of 40–160 mg in 1 ml of saline into the rat uterine horn. Red light was applied from the serosal side at the dose of 150 J/cm2. The rate of implantation significantly decreased in the ALAtreated uterine horns of rats bred 10 and 60 days after photodynamic therapy. Histologic studies revealed complete destruction of endometrium even 60 days after treatment. Pharmacokinetic studies and results of photodynamic endometrial ablation on animal models and patients published since 1994 [37–47] are presented in table 1.

References 1 2 3 4 5 6 7 8 9 10

Pokras R, Hufnagel VG: Hysterectomy in the United States, 1965–84. Am J Public Health 1988; 78:852–853. Boyd ME: Dysfunctional uterine bleeding. Can J Surg 1986;29:305–307. Unger JB, Meeks GR: Hysterectomy after endometrial ablation. Am J Obstet Gynecol 1996;175: 1432–1436. Molloy D, Taylor PT: Gynaecological surgery after endometrial ablation (see comments). Med J Aust 1994;l61:604–606. Goldrath MH, Fuller TA, Segal S: Laser photovaporization of endometrium for the treatment of menorrhagia. Am J Obstet Gynecol 1981;140:14–19. DeCherney A, Polan ML: Hysteroscopic management of intrauterine lesions and intractable uterine bleeding. Obstet Gynecol 1983;61:392–397. McLucas B: Endometrial ablation with the roller ball electrode. J Reprod Med 1990; 35:1055–1058. Daniell JF, Kurtz BR, Ke RW: Hysteroscopic endometrial ablation using the rollerball electrode (see comments). Obstet Gynecol 1992;80:329–332. Nisolle M, Donnez J: Alternative techniques of hysterectomy (letter). N Engl J Med 1997;336: 291–292. Donnez J, Polet R, Mathieu PE, Konwitz E, Nisolle M, Casanas-Roux F: Endometrial laser interstitial hyperthermy: A potential modality for endometrial ablation. Obstet Gynecol 1996:87: 459–464.


180

11 12 13 14 15 16

17

18 19 20 21

22 23

24

25 26 27

28 29

30 31 32 33

Neuwirth RS, Duran AA, Singer A, MacDonald R, Bolduc L: The endometrial ablator: A new instrument. Obstet Gynecol 1994;83:792–796. Holt EM, Gilmer MD: Endometrial resection (review). Baillie`res Clin Obstet Gynaecol 1995;9: 279–297. Nathanson MH, Ezeh U: Carbon dioxide embolism following diagnostic hysteroscopy (letter, comment). Br J Obstet Gynaecol 1995;102:505–506. Witz CA, Silverberg KM, Burns WN, Schenken RS, Olive DL: Complications associated with the absorption of hysteroscopic fluid media (review). Fertil Steril 1993;60:754–756. Brooks PG: Complications of operative hysteroscopy: How safe is it? (review). Clin Obstet Gynecol 1992;35:256–261. Jedeikin R, Olsfanger D, Kessler I: Disseminated intravascular coagulopathy and adult respiratory distress syndrome: Life-threatening complications of hysteroscopy (see comments). Am J Obstet Gynecol 1990;162:44–45. Baggish MS, Daniell JF: Catastrophic injury secondary to the use of coaxial gas-cooled fibers and artificial sapphire tips for intrauterine surgery: A report of five cases. Lasers Surg Med 1989;9: 581–584. Phipps JH, Lewis BV, Roberts T, Prior MV, Hand JW, Elder M, Field SB: Treatment of functional menorrhagia by radio-frequency-induced thermal ablation. Lancet 1990;335:374–376. Sharp NC, Cronin N, Feldberg I, Evans M, Hodgson D, Ellis S: Microwaves for menorrhagia: A new fast technique for endometrial ablation. Lancet 1995;346:1003–1004. Vilos GA, Vilos EC, Pendley L: Endometrial ablation with a thermal balloon for the treatment of menorrhagia. J Am Assoc Gynecol Laparosc 1996;3:383–387. Soderstrom RM, Brooks PG, Corson SL, Dequesne J, Gallinat A, Garza-Leal JG, Iglesias-Benavides JL, Indman PD, Liu J, van der Pas H, Stern RA, Sutton C, Vancaillie TG, Wamsteker K: Endometrial ablation using a distensible multielectrode balloon. J Am Assoc Gynecol Laparosc 1996;3:403–407. Baggish M, Paraiso M, Breznock EM, Griffey S: A computer-controlled continuously circulating, hot irrigating system for endometrial ablatation, Am J Obstet Gynecol 1995;173:1842–1848. Chauduri K, Keck RW, Selman SH: Morphological changes of tumor microvasculature following hematoporphyrin derivative sensitized photodynamic therapy. Photochem Photobiol 1987;46:823– 827. Nelson JS, Liaw LH, Orenstein A, Roberts WG, Berns MW: Mechanisms of tumor destruction following photodynamic therapy with hematoporphyrin derivative, chlorin, and phtalocyanine. J Natl Cancer Inst 1988;80:1599–1605. Dougherty TJ, Kaufman JE, Goldfarb A, Weishaupt KR, Boyle D, Mittleman A: Photoradiation therapy for the treatment of malignant tumors. Cancer Res 1978;38:2628–2635. Rettenmaier M, Berman M, DiSaia P, Burns RG, Weinstein GD, McCullough JL, Berns MW: Gynecologic uses of photoradiation therapy. Prog Clin Biol Res 1984;170:767–775. Schneider D, Schellhas HF, Wessler TA, Chen IW, Moulton BC: Endometrial ablation by DHE photoradiation therapy in estrogen treated ovariectomized rats. Colposc Gynecol Laser Surg 1998; 4:73–77. Scheider D, Schellhas HF, Wessler TA, Moulton BC: Hematoporphyrin derivative uptake in uteri of estrogen treated ovariectomized rats. Colposc Gynecol Laser Surg 1988;4:67–71. Raab GH, Schneider AF, Eirmann W, Gottschalk-Deponte H, Baumgartner R, Beyer W: Response of human endometrium and ovarian carcinoma cell-lines to photodynamic therapy. Arch Gynecol Obstet 1990;248:13–20. Bhatta N, Anderson R, Flotte T, Schiff I, Hasan T, Nishioka NS: Endometrial ablation by means of photodynamic therapy with Photofrin II. Am J Obstet Gynecol 1992;167:1856–1863. Judd MD, Bedwell J, MacRobert AJ: Comparison of the distribution of phthalocyanine and ALAinduced porphyrin sensitizers within the rabbit uterus. Lasers Med Sci 1992;7:203. Kennedy JC, Pottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 1992;14:275–292. Chapman JA, Tadir Y, Tromberg BJ, Yu K, Manetta A, Sun CH, Berns MW: Effect of administration route and estrogen manipulation on endometrial uptake of Photofrin. Am J Obstet Gynecol 1993; 168:685–692.


181

34 35

36

37

38

39

40

41

42

43

44

45

46 47

Steiner RA: Photodynamische Therapie des Endometriums: Pharmakokinetische, morphologische und funktionelle Aspekte; Habilitationsschrift, Medizinische, Fakulta¨t, Universita¨t Zu¨rich, 1993. Steiner RA, Krasieva T, Tadir Y, Tromberg B, Yuan YD, et al: Photodynamische Endometriumablation durch lokale intrauterine Applikation von Photosensibilisator und Laserlicht. Arch Gynecol Obstet 1993;253:158. Yang JZ, Van Vught DA, Kennedy JC, Reid RL: Evidence of lasting functional destruction of rat endometrium after 5-aminolevulinic acid induced photodynamic ablation: Prevention of implantation. Am J Obstet Gynecol 1993;168:995–1001. Wyss P, Tadir Y, Tromberg BJ, Liaw L, Krasieva T, Berns MW: Benzoporphyrin derivative: A potent photosensitizer for photodynamic destruction of the rabbit endometrium. Obstet Gynecol 1994;84: 409–414. Wyss P, Tromberg BJ, Wyss MTh, Krasieva T, Schell M, Berns MW, Tadir Y: Photodynamic destruction of endometrial tissue using topical 5-aminolevulinic acid (ALA) in rats and rabbits. Am J Obstet Gynecol 1994;171:1176–1183. Gannon MJ, Johnson N, Roberts DJH, Holroyd JA, Vernon DI, Brown SB, Lilford RJ: Photosensitization of the endometrium with topical 5-aminolevulinic acid. Am J Obstet Gynecol 1995;173: 1826–1828. Wyss-Desserich MT, Sun CH, Wyss P, Kurlawalla CS, Haller U, Berns MW, Tadir Y: Accumulation of 5-aminolevulinic acid-induced protoporphyrin IX in normal and neoplastic human endometrial epithelial cells. Biochem Biophys Res Commun 1996;44:819–824. Wyss P, Steiner R, Liaw LH, Wyss MT, Ghazarians A, Berns MW, Tromberg BJ, Tadir Y: Regeneration process in the rabbit endometrium: A photodynamic therapy (PDT) model. Hum Reprod 1996; 11:1992–1997. Yang JZ, Van Vugt DA, Roy BN, Kennedy JC, Foster WG, Reid RL: Intrauterine 5-aminolevulinic acid induces selective endometrial fluorescence in the rhesus and cynomolgus monkey. J Soc Gynecol Invest 1996;3:152–157. Fehr M. Wyss P, Tromberg BJ, Krasieva T, DiSaia PJ, Lin F, Tadir Y: Selective photosensitizer localization in the human endometrium after intrauterine application of 5-aminolevulinic acid. Am J Obstet Gynecol 1996;175:1253–1259. Roy BN, Van Vugt DA, Weagle GE, Pottier RH, Reid RL: Effect of 5-aminolevulinic acid dose and estrogen on protoporphyrin IX concentrations in the rat uterus. J Soc Gynecol Invest 1997;4: 40–46. Roy BN, Van Vugt DA, Weagle GE, Pottier RH, Reid RL: Effect of continuous and multiple doses of 5-aminolevulinic acid on protoporphyrin IX concentrations in the rat uterus. J Photochem Photobiol B 1997;41:122–127. Gannon MJ, Vernon DI, Holroyd JA, Stringer M, Johnson N, Brown SB: PDT of the endometrium using ALA. SPIE 1997;2972:2–13. Wyss P, Fehr M, Van den Bergh H, Haller U: Feasibility of photodynamic endometrial ablation without anesthesia. Int J Gynaecol Obstet 1998;60:287–288.

PD Dr. Pius Wyss, Department of Gynecology and Obstetrics, University Hospital of Zu¨rich, Frauenklinikstrasse 10, CH–8091 Zu¨rich (Switzerland) Tel. +41 1 255 52 39, Fax +41 1 255 44 33, E-Mail [email protected]


182


............................

Preclinical Studies: Photodynamic Therapy in the Rat and Rabbit Endometrium with Various Photosensitizers Pharmacokinetic, Histological and Reproductive Studies R.A. Steiner a, P. Wyss b, M.-T. Wyss-Desserich b, Y. Tadir c, B.J. Tromberg c, T.B. Krasieva c, M.W. Berns c, U. Haller b a b

c

Kantonales Frauenspital Fontana, Chur; Department of Gynecology and Obstetrics, University Hospital of Zu¨rich, Switzerland, and Beckman Laser Institute and Medical Clinic, University of California, Irvine, Calif., USA

Introduction It was the aim of this study to evaluate the potential use of different photosensitizers for selective endometrial destruction. We have systematically analyzed the pharmacokinetic behavior of these topically administered photosensitizers and the morphologic characteristics of photodynamic therapy (PDT) in the rat and rabbit model. Physiologic changes were studied as measured by the reproductive performance in PDT-treated rats. To the best of our knowledge this was the first time that the impact of PDT on the endometrium has been studied after local intrauterine application of both the photosensitizer and the laser light [1, 2].

Material and Methods 5-Aminolevulinic acid (ALA) is a precursor of protoporphyrin IX (Pp IX) in the biosynthetic pathway of heme. The administration of exogenous ALA induces the accumulation of Pp IX, which is known to be a strong photosensitizer [3–5]. Benzoporphyrin derivative mono acid (BPD-MA) is a hydrophobic molecule with a high affinity for lipoproteins [6] that was modified from di-hematoporphyrin ether, a com-

monly used photosensitizer for investigations of PDT of the endometrium in animals [3, 7–9]. BPD-MA is 10–70 times more phototoxic than hematoporphyrin towards various cell lines [10]. Due to its absorption peak at 690 nm, BPD-MA qualifies as a drug for well-vascularized tissues, since hemoglobin does not absorb a significant amount of light at this wavelength [11]. The hematoporphyrin derivative (HPD) Photofrin, is a mixture of porphyrins which tend to be selectively retained in tumors [12]. The purified fraction of HPD is thought to be primarily composed of a mixture of di-hematoporphyrin esters and ethers (DHE), and its use has allowed the therapeutic dose of HPD to be reduced by almost one half [11, 13]. The estrous cycle of mature female Sprague-Dawley rats was monitored by frequent vaginal smears in order to synchronize the treatment to the day of diestrous [14]. In female New Zealand white rabbits no estrous monitoring was required as rabbits are induced ovulators. The animals were divided into different groups for complimentary studies: (1) determination of uptake, distribution and clearance of photosensitizers in the uterine layers following topical application by fluorescence microscopy including analysis of the relationship between drug concentration, pH and relative fluorescence in rabbits; (2) morphological changes following intrauterine drug and light application (PDT) in the rat and rabbit model, and (3) reproductive performance following PDT with various photosensitizers in the rat model. Pharmacokinetics The animals were sensitized by topical administration of following photosensitizers (rats 0.15 ml, rabbits 1.2 ml) into each uterine horn. 32 rats were treated with 58 mg/kg of ALA (Deprenyl USA Inc., Parsippany, N.J.) and 12 rabbits with 100, 200, 300 or 400 mg/ml ALA in Hyskon (Kabi Pharmacia, Piscataway, N.J.). It is well known that low-viscosity (e.g. aqueous) fluids pass easily through the fallopian tubes into the abdominal cavity. To minimize this effect before human application, we studied the pharmacokinetics of ALA in dextran 70 in the rabbit model (Hyskon). Dextran 70 is a viscous, hydrophilic, branched polysaccharide used for uterine distention during hysteroscopy. 40 rats were treated with 1 mg/kg BPD-MA (Quadra Logic Technologies, Inc., Vancouver, B.C., Canada). 36 rats were treated with 0.7 mg/kg Photofrin (Porfimer sodium, Quadra Logic Technologies, Vancouver, B.C., Canada), and 32 rats were treated with a mixture of Photofrin (0.7 mg/kg) and 4% Azone (1-dodecylazacycloheptane-2-one, Whitby Research, Inc., Richmond, Va.). Azone is a clear colorless liquid which has been used in the past to increase HPD penetrance in normal skin [15]. 24 rabbits were added to study the relationship between drug concentration, pH and relative fluorescence. A dose-response study was conducted for 100, 200, 300 or 400 mg/ml ALA-Hyskon at 3 h, the time of maximum Pp IX fluorescence as determined from the pharmacokinetic study. 3 animals were studied for each time point and drug dose. To determine the effects of pH adjustment, identical dose-response studies were conducted in rabbits with untitrated ALA-Hyskon, the pH ranging from 1.6 to 2.2 [16]. The animals were sacrificed at 0.5, 1, 1.5, 2, 3, 6, 12 and 24 h after drug administration. Data were obtained from at least 4 animals at each time point. 4 rats receiving no drug were assigned as controls to determine background noise and autofluorescence levels.

Steiner/Wyss/Wyss-Desserich/Tadir/Tromberg/Krasieva/Berns/Haller

184

Shortly after sacrifice, 6 uterine specimens of 3–4 mm each were embedded in medium for frozen sections (OCT media, Miles, Elkhart, Ind.). The blocks were rapidly frozen on dry ice and stored at Ö70 ºC in the dark. Low-light level tissue fluorescence imaging was performed with a slow-scan, thermoelectrically cooled CCD camera (384¶576 pixel) system (Princeton Instruments, Trenton, N.J.) with a 16-bit-per-pixel dynamic range. A Zeiss Axiovert 10 inverted microscope was used with a 10¶objective Zeiss Plan-Neofluar (NA>0.3) to visualize bright field and fluorescence images of the frozen tissue sections. A 100-watt mercury lamp filtered through a bandpass filter (405 nm, band width 20 nm) provided the excitation light that was reflected onto the sample through a dichroic mirror (Zeiss FT 420, Oberkochen, Germany). Emission light was isolated with a 635-nm broad bandpass filter. Instrument control, image accquisition and processing were performed with a Macintosh IIfx computer and IPlab software (Signal Analytics Corp., Vienna, Va.). In order to estimate light distribution, background images were acquired from blank slides using identical parameters. All fluorescence images were normalized in order to correct for nonuniform illumination: Normalized fluorescence image >

mean (background Ö dark noise) ¶ image fluorescence Ö dark noise. image background Ö dark noise

Both fluorescence and background images were corrected for dark noise contributed during the exposure time. The cross-section of the rat uterus was divided into different layers for comparative analysis (endometrial glands, endometrial stroma and myometrium) [17–19]. Histology Study 24 rats [18] and 6 rabbits [16] were photosensitized with ALA as described above. 3 h after drug administration the animals were reanesthetized and the left uterine horn carefully exposed. A few millimeters distal to the uterine bifurcation the uterine wall was punctured with a 20-gauge needle and a 400-lm diameter quartz optical fiber with a 2.0-cm (3.0 cm for rabbits) long cylindrical diffusing tip (PDT Systems, Buellton, Calif.) was inserted into the cavity through the hole formed by the needle (fig. 1). All rats received a total light dose of 80 J/cm2 (rabbits 80–160 J/cm2, depending on geometry) at 630 nm. At 24, 48 h, 4, 7, 14 and 21 days after PDT 4 rats were sacrificed at each time point (rabbits at days 3 and 7 with 3 animals each time point). 6 control rats received 80 J/cm2 of light dose without drug and were sacrificed on days 2, 4 and 7. The entire genital tract was retrieved and fixed in 10% neutral-buffered formalin after photo documentation of the situs and the excised genital tract. Cylindrical samples were taken from the treated area of the left horn and from the control side and further processed for hematoxylin and eosin staining. Sections (4–6 lm) were microscopically examined and documented. Scanning electron microscopy specimens of the rabbit uteri were fixed as above and further processed in 10% osminum tetroxide, dehydrated in graded acetone, critical point dried (Ladd Critical Point Dryer, Ladd Research Industries, Burlington, Calif.), and sputter coated with gold (Pelco PAC-1 evaporating system, Ted Pella, Redding, Calif.). Micrographs were then taken on scanning electron microscopy (SEM 513, Philips Electronic Instrument Company, Mahwake, N.J.) [16].

Photodynamic Therapy in the Rat and Rabbit Endometrium

185

Fig. 1. Fiber with the cylindrical diffusing tip inserted in the left uterine horn for PDT. Circles indicate areas of subthreshold irradiation, resulting in insufficient photodynamic action.

Reproductive Performance 9 rats were photodynamically treated with ALA as described in the previous section, 8 rats with BPD-MA (wave length 690 nm), 24 rats with Photofrin alone and 8 rats with Photofrin+Azone. The drug-light interval in the Photofrin-treated animals was 3, 24 and 72 h (8 rats each) and in the Photofrin+Azone group only 3 h. 3–4 weeks following PDT of the left uterine horn the animals were bred with mature male Sprague-Dawley rats. The female rats were sacrificed in the second trimester of pregnancy. Location and number of implantation sacs in the treated uterine horn and the control side were noted and photographed. Specimens of the treated uterine horns were also retrieved for histologic studies. A control group of 6 animals received the same amount of each photosensitizer intrauterine, but without laser illumination and another 6 control rats were treated only with laser in the same way as the others, but received no drug [17–19]. Skin Photosensitivity All rats in the reproduction performance study group were also light-treated in a shaved area of the upper right quadrant of the abdomen with a light dose of 100 J/cm2 and a spot size of 1 cm2 (160 mW in 8 min 20 s at a power density of 200 mW/cm2). The treated skin area was checked immediately after light application and during recovery at 1-hour intervals, and then daily until the animals were sacrificed [17–19]. Laser Light Delivery System An argon-pumped dye laser at 630 nm (Argon Laser model 171, and Dye Laser model 375, Spectra Physics, Mountain View, Calif.) was transmitted through the 400-lm fiber and diffused towards the walls. A Clinical Hartridge Reversion spectroscope (Ealing ElectroOptics, South Natick, Mass.) was used to verify the accuracy of laser wavelength to ×1 nm. For irradiation of the skin, the laser beam was coupled into a 400-lm fused, silica fiber optic using a Spectra-Physics (Mountain View, Calif.) model 316 fiber optic coupler. The


186

output end of the fiber terminated with a microlens that focused the laser radiation into a circular field of uniform light intensity with a spot size of 1 cm. Laser irradiation emanating from the diffusing fiber tip and the microlens, respectively, was monitored with a power meter (model 210, Coherent Corp., Palo Alto, Calif.) before and after treatment. Temperature Measurements To demonstrate the dynamics of temperature changes during PDT the surface temperature of the treated uterine horn was measured in 2 rats, before, during and after photodynamic treatment using ALA. An infrared thermal camera (model 600, Inframetrics, Bedford, Mass.) with oscillating mirrors to scan the field of view horizontally and vertically produced images at a rate of 30 frames/s. It employed a mercury-cadmium-telluride (Hg/Cd/Te) detector which was sensitive over the 8- to 12-lm spectral range in the infrared. The detector’s specified response time was 0.5–1 ls, allowing sampling rates up to 2.01 MHz with a minimum detectable temperature difference of 0.1 ºC [17–19]. Statistical Analysis The reproductive performance outcome between the treated area of the left uterine horn and the corresponding area of the control horn was analyzed with the paired 2-tailed t test and the paired Wilcoxon signed rank test. Significance was considered at p=0.05. Cited results refer to the Wilcoxon signed rank test.

Results Fluorescence Study ALA Analysis of fluorescence following topical application of ALA revealed quite similar patterns in both the rat and rabbit uterus. Peak values were found at 3–6 h with a significantly higher concentration in the endometrial glands (fig. 2). There was a gradual decrease at 6–12 h with concomitant narrowing of the gap between the various layers. The difference in fluorescence between glands and stroma decreased significant with time. At 24 h uptake of fluorescence was similar to that measured before injection. Figure 3 demonstrates the high uptake of ALA converted to Pp IX in the endometrial glands after as early as 1.5 h in the rat uterus. High magnification of the rat endometrial glands revealed that the fluorescenting Pp IX is located in the cytoplasm and/or membranes of cylindrical epithelial cells, but not in the nucleus of these cells [18]. In the rabbits no significant dose-dependent fluctuations were observed in glandular, stromal, and myometrial fluorescence when the concentration of ALA was increased from 100 to 400 mg/ml. However, glandular uptake was significantly higher than that of the other structures regardless of drug concentration (p>0.0001). Fluorescence contrast between glands and stroma-myomet-


187

Fig. 2. Kinetic curve of the fluorescence (means × standard error) in the rat uterine tissue after intrauterine injection of ALA (58 mg/kg in 0.15 ml sterile H2O).

Fig. 3. Fluorescence micrograph of the rat uterus 1.5 h following topical application of ALA (58 mg/kg in 0.15 ml sterile H2O intrauterine). Note the high fluorescence in the endometrial glands as compared with the myometrium, suggestive of potential selective destruction by PDT.


188

Fig. 4. Kinetic curve of the fluorescence (mean × standard error) in the rat uterine tissue after intrauterine injection of BPD-MA (1 mg/kg in 0.15 ml sterile H2O).

rium varied significantly with dose (p>0.0042), ranging from a peak difference of 1.00 for glands-myometrium at 200 mg/ml to 0.23 at 300 mg/ml. Dose-response data for untitrated and pH-adjusted ALA-Hyskon clearly demonstrate that there were no significant fluorescence variations within the uterine layers after pH adjustment. Concentration-dependent pH variations for ALA-Hyskon were observed to range from pH 2.2 to 1.6 [16]. BPD-MA Throughout the experiments relative fluorescence was highest in the epithelial lining of the endometrial glands resulting in a positive ratio glands/ stroma/myometrium (fig. 4). A first peak after 1 h was followed by even higher peak fluorescence values at 3 h after intrauterine injection of BPD-MA. Then a fast decrease of activity occurred within the next 6 h leaving only minimal fluorescence in all layers after 12 h [19]. Photofrin Within 90 min after intrauterine injection of Photofrin as a mono substance, relative fluorescence in the uterine tissue showed a steep rise to the first and highest, though very short maximum. Values in the endometrial glands and endometrial stroma were almost identical whereas fluorescence in the circular muscle layer was on a lower but more or less parallel level (fig. 5). Fluorescence increased again for a longer period of time after 3 h and then dropped continuously to reach minimal values at 24 h [17].


189

Fig. 5. Kinetic curve of the fluorescence (means × standard error) in the rat uterine tissue after intrauterine injection of Photofrin (0.7 mg/kg in 0.15 ml sterile H2O).

Fig. 6. Kinetic curve of the fluorescence (means × standard error) in the uterine tissue after intrauterine injection of a mixture of Photofrin (0.7 mg/kg) with 4% Azone (1-dodecylazacycloheptane-2-one).

Photofrin+Azone The kinetic curves of the fluorescence of Photofrin+Azone (fig. 6) were very similar to the curves of Photofrin alone (fig. 5). There was a positive ratio endometrium/myometrium at 1 and 3 h and minimal fluorescence 6 h after topical injection [17].


190

Fig. 7. Edematous thickening and rigidity of the left uterine horn 4 days after PDT with ALA in the excised rat uterus (58 mg/kg in 0.15 ml sterile H2O intrauterine, 80 J/cm2).

Morphological Study Macroscopic Changes Immediately after PDT with ALA significant blanching in the treated area of the rat uterus was observed. Edematous thickening and rigidity of the uterine horn increased up to the 4th day after treatment (fig. 7). Thereafter these signs diminished and at the 2–3 week follow-up the diameter of the left uterine horn gradually decreased in the area of treatment. Long-term appearance of the treated horn (7–10 weeks) revealed thinning and atrophy (fig. 8).


191

Fig. 8. Long-term appearance of the left rat uterine horn after PDT with ALA (58 mg/ kg in 0.15 ml sterile H2O intrauterine, 80 J/cm2).

Histological Changes 24 h after treatment the left uterine horn showed necrosis of the epithelial cells lining the lumen and the tissue in the lamina propria. Fibrinous debris were often found in the lumen. The endometrial glands appeared in various stages of disintegration which ranged from early degenerative change to complete liquefaction with only the glandular outlines remaining visible. In some animals the tunica muscularis was also affected, with lesions ranging from vacuolar degeneration to edema of the interstitial tissue and to complete necrosis of the muscle cells. 84 h after treatment there were areas of progressed necrosis. However, early regeneration of the epithelium lining the lumen could also be observed at this stage. Vascular collapse with fibrinous debris of either embolus or thrombi, were present in areas of severe tissue necrosis. Arteries were more often affected than veins. 7 days after PDT complete regeneration of the cylindrical epithelium lining the lumen was present. Many arteries presented with a complete endothelial lining, although necrotic debris still persisted beneath the endothelium. The regeneration process continued during the next 2–3 weeks. The arteries then appeared as if returned to normal, whereas the tunica muscularis still seemed to be somewhat disorganized. The lamina propria was found to be less cellular with an increased amount of


192

Fig. 9. Cross-section of the left uterine horn 9 weeks after PDT with topically applied ALA (58 mg/kg in 0.15 ml sterile H2O intrauterine) and 80 J/cm2. The endometrium has been destroyed by PDT, whereas the longitudinal and circular myometrial layers remained intact.

collagen. The final appearance in most animals was destruction of the endometrium leaving the myometrial layers intact (fig. 9). Interestingly the cells in the zona compacta were less responsive to the cyclic changes than the untreated control uterine horns. 7–10 weeks after PDT the treated area typically showed a marked atrophy of the endometrium even following exposure to estrogen stimulation that resulted from pregnancy in the nontreated uterine horn (fig. 10) [18]. Uneven damage was occasionally observed, regardless of the time point, and may have been the result of nonhomogeneous light delivery to the tissue. Occasionally PDT resulted in destruction of the entire endometrium leaving substantial scar tissue formation which effectively replaced the lumen (fig. 11) [18]. In the rabbit uterus low-magnification light microscopy images of untreated and treated uterine horns clearly showed destruction of luminal and glandular epithelial structures 7 days after photodynamic therapy. The endometrial stroma remained largely intact with moderate scarring. Minimal reepithelialization was occasionally observed in some sections, perhaps because of uneven light dosimetry. High-magnification images revealed dramatic changes in the columnar epithelium and the structural integrity of glandular


193

Fig. 10. Excised uterus 8 weeks after PDT with intrauterine application of ALA (58 mg/kg in 0.15 ml sterile H2O intrauterine) and 80 J/cm2. The functional integrity of the right uterine horn is demonstrated by 10 intact implantations, whereas there are no gestational sacs in the PDT-treated left uterine horn.

invaginations. In addition, the postirradiation luminal surface was composed of elongated, flattened fibroblasts and extracellular matrix components, whereas the lumen contained fibrinous and cellular debris. 3-day postirradiation findings were comparable and, in some cases, more acute with indications of hyperemia. Scanning electron micrographs of the untreated uterine horn (fig. 12a) showed ciliated cells surrounded by nonciliated microvillous cells. In contrast, the treated horn (fig. 12b) exhibited pronounced superficial damage. No evidence of the normal epithelial structure remained, and the surface appeared to be replaced by a collagen network resembling scar tissue or, perhaps, basal membrane [16]. There were no macroscopic or microscopic changes in the control uterine horns, nor in the tissue exposed to laser only or to the sensitizer alone. Reproductive Performance after PDT In all study groups there was a significant difference between the number of nidations per rat in the treated area of left uterine horn and the corresponding, untreated area of the control side (fig. 10, 13) [17–19].


194

Fig. 11. Destruction of the entire endometrium leaving substantial scar tissue formation which effectively replaces the lumen 21 days after PDT with Photofrin (0.7 mg/kg) with 4% Azone.

ALA There were 0.4×0.3 implanatation sacs in the PDT area of the left uterine horn and 8.9×1.0 in the corresponding area of the untreated right side (p=0.01). Controls exhibited no significant difference in the number of implantations (8.2×1.0 versus 6.8×0.9 and 8.3×0.7 versus 7.2×0.6). BPD-MA Only 0.44 implantations per rat occurred in the treated area of left horn after PDT, whereas 8.9 implantations were found in the corresponding area of the right horn. This difference was highly significant (p=0.001). Photofrin The mean number of gestational sacs per rat in the group with the short (3 h) interval was 2.38×0.92 in the treated area and 5.12×0.74 on the untreated right side (p>0.04). In the 24-hour interval group the numbers were 1.5×0.68 for the left side and 5.9×0.88 for the right side (p>0.02) and in the 72-hour interval group 0.88×0.52 on the left and 8.1×1.12 on the right (p>0.01). The differences between the 3-hour group and the 72-hour group were statistically significant (p=0.02).


195

a

b Fig. 12. Scanning electron micrographs of untreated (a) and treated (b ) uterine horns 7 days after photodynamic therapy. Original magnification ¶5,000.

Photofrin+Azone The Photofrin+Azone combination was more effective in reducing reproductive performance than Photofrin alone in the short interval (p>0.02). The mean number of gestational sacs was 0.38×0.26 per rat in the treated and 7.5×1.07 in the untreated area (p>0.01). The difference in efficacy to the


196

Fig. 13. Reproductive performance after PDT with different photosensitizers and laser light of 80 J/cm2 (630 nm, for BPD-MA 690 nm). Differences between implantations per rat in the treated area of the left uterine horn and in the untreated right horn are significant for all study groups. * p=0.05; ** p=0.001.

Fig. 14. Reproductive performance in the control groups treated either only with the different photosensitizers or only with laser light (630 nm; 690 nm for BPD-MA; 100 J/cm2). There was no significant difference between the right and the left horn. * n.s.


197

Fig. 15. Heat generation during PDT of the left uterine horn with ALA (58 mg/kg in 0.15 ml sterile H2O intrauterine) and laser light (630 nm; 80 J/cm2).

group treated with Photofrin alone and the same incubation time (3 h) was significant (p>0.02). In the control groups in which the animals received a photosensitizer but no light, or only light (80 J/cm2) without any drug, there was a tendency for fewer implantations in the manipulated, compared to the nonmanipulated horn, though the difference was not statistically significant (fig. 14). Skin photosensitivity did not occur in any of the PDT-treated animals. PDT was tolerated well by all with no adverse effects [17–19]. Heat Generation during PDT Illumination of uteri with 80 J/cm2 induced a temperature rise of 6.5 ºC which was noted within 2 min and leveled off thereafter. Switching off the light source was followed by an equally rapid temperature drop to the pretreatment level. The highest recorded temperature was 31.9 ºC in 1 rat and 33.6 ºC in the other. Moistening the tissue with saline during treatment resulted in a significant cooling of the treated area for a period of 2–3 min (fig. 15) [18].

Discussion The aim of this study was to evaluate the potential use of 3 different photosensitizers topically applied for selective endometrial destruction by PDT


198

in the rat and rabbit model. Endometrial ablation is a viable alternative to hysterectomy, and a minimally invasive approach holds significant benefits for the patients as well as for the medical services [20, 21]. We conducted 3 complimentary studies: (1) determination of uptake, distribution and clearance of ALA, BPD-MA, Photofrin and a mixture of Photofrin and Azone in the uterine layers by fluorescence microscopy; (2) morphological changes after PDT with ALA, and (3) reproductive performance following PDT with ALA, BPD-MA, Photofrin and Photo- frin+Azone. We further measured the heat generation during PDT and analyzed the extent of skin photosensitivity after topical application of the photosensitizers in the uterus horn. Determination of fluorescent activity as a marker for drug uptake in the tissue revealed a high concentration of all tested photosensitizers in the endometrium 3–6 h following intrauterine administration. Interestingly, drug uptake as demonstrated by digital fluorescence microscopy showed a significantly higher concentration in the endometrium than in the myometrium (fig. 2–6). This observation provides the background for selective destruction of the endometrium following light-induced photochemical effects. The pharmacokinetic curve suggests that the optimal window for illumination is at 3–6 h for the tested photosensitizers [17–19]. It has already been demonstrated that ALA is a precursor of Pp IX which may act as a photosensitizer for targeting the endometrium [22]. It is also known that ALA-Pp IX conversion in the biosynthetic pathway of heme occurs in the mitochondria. The fluorescent images suggested that Pp IX accumulated in the cytoplasm and around the membrane which may be the source of the multilocular and complex destructive mechanism observed in the histology studies [18]. The influence of drug dose on Pp IX synthesis was evaluated by several different ALA concentrations (100–400 mg/ml) in the rabbit model. These dose-response studies indicate that, in the case of pH 5.5 ALA-Hyskon, there are no significant concentration-dependent variations in Pp IX production. Glandular fluorescence is uniformly elevated regardless of ALA concentration. However, substantial differences between glandular stroma, and myometrial fluorescence appear at the 200-mg/ml dose. Accordingly for PDT, we conducted all irradiation studies 3 h after topical application of 200 mg/ml ALA-Hyskon [16]. In the course of the dose-response studies we observed concentrationdependent pH variations ranging from pH 2.2 at 100 mg/ml ALA-Hyskon to pH 1.6 at 400 mg/ml. Because the pH of uterine fluid in humans during the various phases of the menstrual cycle ranges from 5.9 to 7.3 [23], we were concerned that acidic ALA solutions applied to the uterine cavity might be


199

toxic to the endometrium. Low pH exposure results in endogenous prostaglandin F2a release [24], which, in turn can cause painful uterine contractions, spillage of photosensitizer, and a decrease in uterine blood flow. Low blood flow can reduce tissue oxygen saturation and, as a result, influence singlet oxygen generation during photodynamic therapy [25]. In contrast, alkaline solutions (pHq7.4) did not cause contractions in rats [26]. It is interesting to speculate that the highly acidic ALA solutions used in previous studies may, in part, account for the slight reduction in the number of implanted embryos observed in nonirradiated control uteri [18]. Because of the potentially deleterious effects of low pH, all photodynamic treatments in the rabbit model were performed with ALA-Hyskon adjusted to pH 5.5. This value was selected because titration to pHq6.0 leads to turbidity and, ultimately, an intense yellow solution that may have a different physiologic behavior from the more highly protonated acid. Results of a comparison study indicate that pH adjustment does not alter fluorescence levels or Pp IX synthesis. Hyskon is a clear, viscous, hydrophilic, branched polysaccharide used for uterine cavity distention in hysteroscopy. It was selected in an effort to minimize retrograde spillage through the cervix and passage through the fallopian tubes into the abdominal cavity. Although we have no conclusive evidence that Hyskon enhances ALA availability, there is no indication that Hyskon adversely affects ALA delivery to the target tissue [16]. In previous work [3], we demonstrated that intrauterine administration of Photofrin is feasible; however, it resulted in relatively high uptake and retention by the superficial columnar epithelium, and slow penetration to deeper uterine layers. In one part of these studies, we attempted to expand our earlier findings by improving Photofrin tissue distribution and photodynamic efficacy with Azone, a penetration-enhancing agent. Azone has been used with both hydrophobic and hydrophilic molecules. Optimum concentration can vary, but 1–5% Azone solutions appear to be appropriate for most formulations [15]. Pharmacokinetic studies of the two drug formulations (Photofrin and Photofrin+Azone) showed similar fluorescence patterns (fig. 5, 6). The Photofrin+Azone combination may have had slightly greater endometrial selectivity [17]. Nevertheless, due to the extremely high columnar epithelium fluorescence, it was difficult to determine the precise impact of Azone on Photofrin penetrance. Fluorescence microscopy-based pharmacokinetic studies provide visual evidence of drug localization but are not necessarily predictive of structural damage during PDT. This is particularly true for Photofrin, since it is a mixture of compounds which vary in photodynamic efficacy, subcellular localization and fluorescence quantum yield. In addition, Photofrin’s most photoactive components are probably least fluorescent.


200

We conclude that the penetration-enhancing drug Azone can be used to modulate the efficacy of topically applied Photofrin during uterine PDT. The mechanism of action is not clear solely from the results of fluorescence-imaging, pharmacokinetic studies. However, when these observations are considered in conjunction with tissue histology and postirradiation pregnancy tests, we surmise that Azone facilitates the delivery of photoactive Photofrin components to critical cellular structures [17]. Histological studies revealed clear signs of destruction throughout the endometrial layer after PDT and in some animals even the myometrial layers showed distinct areas of necrosis. These findings are consistent with our observations on drug distribution in the myometrium but contrasts with the work of others [22] who described a selective destruction of the endometrium in the rat using the same drug and the same time interval between photosensitization and light application but different light source and delivery. Interestingly we observed some inconsistency in the pattern of destruction which we considered to be mainly caused by the nonhomogeneous light distribution of the diffusing fiber tip as was documented in the sensitive thermographic imaging. Furthermore we observed leakage of the photosensitizing drug out of the injection site in the uterus, possibly resulting in sensitization of the animals with variable amounts of drug. Choosing appropriate parameters of treatment is extremely important in order to achieve the desired selectivity of destruction. The lack of damage in the control uteri and the minimal thermal fluctuations under the threshold of potential effects documented in this study are indicative of photochemical changes [17–19]. Myometrial damage observed in our study may not limit the conversion of this technology to man, since the combination of myometrial thickness and limited light penetration will protect undesired deep damage. The degree of destruction in the rabbit uterus after PDT with ALA was consistent between animals evaluated 3 days after irradiation. However, it is important to point out that, after 7 days, epithelial regeneration varied with anatomic location. We selected a 7-day end point because regeneration after mechanical destruction of the endometrium in rabbits is complete after 3 days [27]. Although our 7-day observations generally revealed substantial, persistent destruction, regional variations in reepithelization are a clear indication that maintaining the optical dose at or above the photodynamic threshold is critical to obtaining irreversible damage. Drug distribution may also play a role in these findings; however, our dose-response data indicate that there is no real practical value to increasing the drug dose beyond the 200-mg/ml level used for photodynamic therapy. As a result, it is expected that photodynamic therapy after topical application of ALA can be used for highly effective endometrial destruction. We have explored a variety of factors intended to optimize


201

this procedure [16]; however, the prospect of long-term tissue regeneration is a possibility and should be evaluated in more rigorous light-dose and lightdelivery studies. The observation of visible endometrial changes following PDT was also confirmed by functional assay. The reproductive performance study demonstrated a significant implantation failure in the treated uterine horns as compared to various controls [17–19]. Substantially impaired reproductive performance was also demonstrated for Photofrin PDT at all time points (drug-light intervals 3, 24 and 72 h), as well as for 3-hour Photofrin+Azone. However, implantations decreased significantly when the Photofrin drug-light interval was increased from 3 to 72 h, indicating that treatment efficacy increased with incubation time. Interestingly, 72-hour Photofrin results were comparable to those obtained for rats in the 3-hour Photofrin+Azone group. These observations suggest that the primary consequence of combining Photofrin with Azone may be to enhance the localization of ‘active’ Photofrin components in essential, oxidizable cellular structures. Without Azone, the systemic contribution of residual recirculating Photofrin at 72 h may play a role in enhancing efficacy. In addition, Photofrin fluorescence at 24 and 72 h may be derived from more photodynamically active components (compared to 3 h), since the spatial distribution of fluorescence appears to be the same for both formulations regardless of time. Evaluation of long-term structural effects shows similar damage patterns for 3-hour Photofrin+Azone and 72-hour Photofrin. Endometrial stroma was clearly destroyed using both formulations. Only columnar epithelium, muscle and peritoneal serosa remained after irradiation. Both formulations resulted in substantial scar tissue formation which, in some cases of Photofrin+Azone, replaced the lumen. However, our occasional observation of uneven damage, regardless of time point, underscores the importance of uniform light delivery. Endometrium can easily regenerate if the photodynamic damage threshold of the tissue is not exceeded. Regenerating tissue can originate in a single location and eventually may proliferate throughout the entire uterine cavity [28]. This observation is further confirmed by pregnancy study results using 2-cm-long diffusing fiber tips. Since the uteri were somewhat longer than 2 cm, regions proximal and/or distal to the tip were exposed to light at a reduced intensity (fig. 1, 16). These area of subthreshold irradiation exhibited more pregnancies as compared with the fully irradiated uterine regions [17]. The advantage of BPD-MA as a photosensitizer is the fact that light of 690 nm (maximum absorption) penetrates deeper into tissue and shows less absorption by hemoglobin. In BPD-MA-treated animals we found substantial atrophy of the uterine horn and more often and more dense adhesions than


202

Fig. 16. Eight gestational sacs are visible in the control horn and one sac appeared in the distal end of the photodynamically treated left horn. The single pregnancy on the treated side (Photofrin + Azone, 3 h) was located in an area that was exposed to an insufficient light intensity. This artifact of fiber geometry and placement was presumed to result in the delivery of a subthreshold light dose for irreversible photodynamic damage.

after treatment with the other sensitizers. We speculate that light of 690 nm penetrated the thin rat uterus more efficiently and had a photodynamic action on the surrounding intra-abdominal structures which were sensitized by leakage of the intrauterine-applied sensitizer though the hole in which the fiber was inserted. It is of note that the pregnancy rate was assessed at 3–4 weeks following the studied PDT protocol, a phase at which in some animals endometrial regeneration was demonstrated histologically. This is suggestive of nonvisible functional damage. It is not clear if a similar protocol in man will cause similar effects; however, further studies in our facility are specifically targeting the mechanisms of endometrial proliferation and optimizing the PDT protocol. The assumption that topical application of the photosensitizer for endometrium targeting may eliminate skin photosensitivity was proved in our studies. The skin of all animals in the reproductive performance study was exposed to 100 J/cm2 of 630 nm light after endometrial PDT. None of the animals showed any alterations in the light-treated skin area.


203

Minimal thermal fluctuations during light exposure, that was measured with a sensitive thermal camera, suggested that heating may not be a limiting factor once this technology is applied in the human.

References 1 2

3

4 5 6

7 8

9 10 11

12 13 14 15

16

17

18

Steiner RA: Photodynamische Therapie des Endometriums: Pharmakokinetische, morphologische und funktionelle Aspekte; Habilitationsschrift, Medizinische Fakulta¨t, Universita¨t Zu¨rich, 1993. Steiner RA, Krasieva T, Tadir Y, Tromberg B, Yuan Y-D, Walt H, Wight E, Ko¨chli O, Haller U, Bern MW: Photodynamische Endometriumablation durch lokale intrauterine Applikation von Photosensibilisator und Laserlicht. Arch Gynecol Obstet 1993;253:158. Chapman JA, Tadir Y, Tromberg BJ, Yu K, Manetta A, Sun CH, Berns MW: Effect of administration route and estrogen manipulation on endometrial uptake of Photofrin porfimer sodium. Am J Obstet Gynecol 1993;168:685–692. Kennedy JC, Pottier RH, Pross DC: Photodynamic therapy with endogenous protoporphyrin. IX: Basic principles and present clinical experience. J Photochem Photobiol B 1990;6:143–148. Kennedy JC, Pottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 1992;14:275–292. Allison BA, Pritchard H, Richter AM, Levy JG: The plasma distribution of benzoporphyrin derivative and the effects of plasma lipoproteins on its distribution. Photochem Photobiol 1990; 52:501–517. Bhatta N, Anderson R, Flotte T, Schiff I, Hasan T, Nishioka NS: Endometrial ablation by means of photodynamic therapy with Photofrin II. Am J Obstet Gynecol 1992;67:1856–1863. Raab GH, Schneider AF, Eirmann W, Gottschalk-Deponte H, Baumgartner R, Beyer W: Response of human endometrium and ovarian carcinoma cell-lines to photodynamic therapy. Arch Gynecol Obstet 1990;248:13–20. Schneider DF, Schellhas HF, Wesseler TA, Moulton BC: Endometrial ablation by DHE photoradiation therapy in estrogen-treated ovariectomized rats. Colposcopy Gynecol Laser Surg 1988;4:73–85. Richter AM, Kelly B, Chow J, Liu J, Towers GHN, Dolphin D, Levy JG: Preliminary studies on a more effective phototoxic agent than hematoporphyrin. J Natl Cancer Inst 1987;79:1327–1332. Richter AM, Cerruti-Sola S, Sternberg ED, Dolphin D, Levy JG: Biodistribution of tritiated benzoporphyrin derivative (3H-BPDD-MA), a new potent photosensitizer, in normal and tumorbearing mice. J Photochem Photobiol B 1990;5:231–244. Lipson RL, Baldes EJ, Olsen AM: The use of derivative hematoporphyrin in tumor detection. J Natl Cancer Inst 1961;26:76–81. Kessel D: Proposed structure of the tumor-localizing fraction of di-haematoporphyrin. J Photochem Photobiol 1990;44:193–196. Yuan YD: Female reproductive system; in Haschel W (ed): Handbook of Toxicologic Pathology. New York, Academic Press, 1991, pp 891–935. Chow DSL, Kaka I, Wang TI: Concentration-dependent enhancement of 1-dodecylazacycloheptan2-one on the percutaneous penetration kinetics of triamcinolone acetonide. J Pharm Sci 1984;73: 1794–1799. Wyss P, Tromberg BJ, Wyss MT, Krasieva T, Schell M, Berns MW, Tadir Y: Photodynamic destruction of endometrial tissue with topical 5-aminolevulinic acid in rats and rabbits. Am J Obstet Gynecol 1994;171:1176–1183. Steiner RA, Tromberg BJ, Wyss P, Krasieva T, Chandanani N, McCullough J, Berns MW, Tadir Y: Rat reproductive performance following photodynamic therapy with topically administered Photofrin. Hum Reprod 1995;10:227–233. Steiner RA, Tadir Y, Tromberg BJ, Krasieva T, Ghazains AT, Pyss P, Berns MW: Photosensitization of the rat endometrium following 5-aminolevulinic acid induced photodynamic therapy. Lasers Surg Med 1996;18:301–308.


204

19 20 21 22

23 24

25 26 27 28

Steiner RA, Tadir Y, Tromberg BJ, Wyss P, Walt H, Haller U: Benzoporphyrinderivat-Monosaüre zur Photodynamischen Therapie des Endometriums. Geburtsh Frauenheilk 1996;56:1–7. Garry R, Erian J, Grochmal S: A multicentre collaborative study into the treatment of menorrhagia by Nd:YAG laser ablation of the endometrium. Br J Obstet Gynaecol 1991;98:357–362. Magos AL, Baumann R, Lockwood GM, Turnbull AC: Experience with the first 250 endometrial resections for menorrhagia. Lancet 1991;337:1074–1078. Yang JZ, Van Vugt DA, Kennedy JC, Reid RL: Evidence of lasting functional destruction of the rat endometrium after 5-aminolevulinic acid induced photodynamic ablation: Prevention of implantation. Am J Obstet Gynecol 1993;168:995–1000. Sedlis A, Kandemir E, Stone ML: Intrauterine pH of women using stainless steel contraceptive device. Obstet Gynecol 1967;30:114–117. Pascoe DR, Stabenfeld CH, Hughes JP, Kindahl H: Endogenous prostaglandin F2 alpha release induced by physiologic saline solution infusion in utero in the mare: Effect of temperature, osmolarity, and pH. Am J Vet Res 1989;50:1080–1083. Tromberg BJ, Orenstein A, Kimel S, et al: In vivo tumor oxygen tension measurements for the evaluation of the efficiency of photodynamic therapy. Photochem Photobiol 1990;52:375–385. Heaton RC, Taggart MJ, Wray S: The effects of intracellular and extracellular alkalinization on contractions of the isolated rat uterus. Pflu¨gers Arch 1992;422:24–30. Schenker JG, Sacks MI, Polishuk WZ: Regeneration of rabbit endometrium following curettage. Am J Obstet Gynecol 1971;111:970–978. Wyss P, Steiner R, Liaw LH, Wyss MT, Ghazarians A, Berns MW, Tromberg BJ, Tadir Y: Regeneration processes in rabbit endometrium: A photodynamic therapy model. Hum Reprod 1996;11: 1992–1996.

Rolf A. Steiner, MD, Kantonales Frauenspital Fontana, CH–7000 Chur (Switzerland) Tel. +41 81 254 81 11, Fax +41 81 254 81 30, E-Mail [email protected]


205


............................

Pharmacology and Toxicology of 5-Aminolevulinic Acid and Protoporphyrin IX Used for Photodynamic Endometrial Ablation in Primates and Nonprimates P. Wyss Department of Gynecology and Obstetrics, University Hospital of Zu¨rich, Switzerland

The approach to photodynamic therapy of human endometrium utilizes 5-aminolevulinic acid (ALA) (fig. 1), a precursor of protoporphyrin IX (Pp IX) (fig. 2) in the heme biosynthetic pathway [1]. Heme biosynthesis is essential to life and occurs in all aerobic cells. It is formed when iron, as ferrous ion, is inserted into the protoporphyrin ring. In photosynthetic organisms only, magnesium can be inserted into the protoporphyrin ring from Mg protoporphyrin, which is converted in a series of steps to chlorophyll. The slowest step in heme synthesis is the conversion of Pp IX to heme. Therefore, the exogenous administration of photodynamically inactive ALA induces the production and accumulation of Pp IX, a potent photosensitizer (fig. 3). Heme, a ferric protoporphyrin of hemoglobin, contains a coordinated ferric ion which shortens the lifetime of the porphyrin-excited state to such an extent that it is no longer an effective photosensitizer [2]. Unlike heme and chlorophyll, free Pp IX does not appear to have intrinsic biological activity; however, it is a potent natural nontoxic photosensitizer and its illumination with certain wavelengths of ultraviolet or visible light can lead to significant photodynamic effects on tissues, cells, subcellular elements, and macromolecules via production of singlet oxygen [3–7]. Since only certain types of cells have a capacity to sensitize substantial amounts of Pp IX, the use of ALA can provide an additional element of selectivity to photodynamic therapy.

Fig. 1. Chemical structure of 5-aminolevulinic acid.

Fig. 2. Chemical structure of protoporphyrin IX.

Fig. 3. Mechanism of ALA-induced Pp IX accumulation. Due to the slow conversion of Pp IX to heme, large excess of exogenous ALA application causes Pp IX accumulation.

Photodynamic Endometrial Ablation in Primates and Nonprimates

207

Biosynthesis Mitochondria and the cytosol are the cellular locations in which biosynthesis of ALA and Pp IX occurs. In the mitochondria, glycine is coupled to the cofactor pyridoxal phosphate, which allows its condensation with succinyl CoA to form ALA [8]. This reaction is catalyzed by ALA synthase, the rate-limiting enzyme of the chemical pathway of Pp IX and heme. ALA then diffuses out of the mitochondria into the cytosol where, in the presence of the porphobilinogen synthase, two molecules of ALA condense to form porphobilinogen. Porphobilinogen, a monopyrrole, is the basic ring structure of the porphyrins, and like ALA, is highly water-soluble. Via the action of porphobilinogen deaminase and coenzyme uroporphyrinogen III cosynthase, four molecules of porphobilinogen are combined to form uroporphyrinogen III, a cyclic tetrapyrrole. Uroporphyrinogen decarboxylase ultimately transforms uroporphyrinogen III to coproporphyrinogen III, which reenters the mitochondrion to be oxidized to protoporphyrinogen IX by coproporphyrinogen III oxidase. Protoporphyrin oxidase enzymatically converts protoporphyrinogen IX to protoporphyrin IX, which finally, in the presence of ferrochelatase, chelates ferrous ion to complete synthesis of heme.

Feedback Control and Pp IX Accumulation The presence of free heme inhibits the synthesis of ALA [9, 10]. However, this negative feedback mechanism is bypassed by large excess of exogenous ALA application. Due to the fact that the slowest step in the heme biosynthesis happens at the conversion of Pp IX to heme, exogenous ALA administration consequently causes accumulation of the photosensitizing Pp IX [1] (fig. 3).

Endogenous Quantity of ALA It is estimated that approximately 350 mg ALA/day must be synthesized by the human body to support endogenous heme production [11–13]. Liver cells normally make about 15% of the heme that is synthesized in the body, and approximately 50 mg ALA/day are required for this amount of hepatic heme synthesis. The remainder of heme formation takes place in the bone marrow for the synthesis of hemoglobin which requires approximately 300 mg ALA/day.

Wyss

208

Metabolism The water solubility of porphyrins decreases with the reduction in the number of carboxylic acid chains. ALA and porphobilinogen, as well as the polar porphyrins are excreted predominantly in the urine (uroporphyrins), whereas the hydrophobic, decarboxylated porphyrins (coproporphyrins) are excreted predominantly in the stool. Pp IX is so hydrophobic that it is excreted almost exclusively in the bile. Oral application of ALA in a man resulted in unchanged excretion of a large proportion (30–70%) in the urine within 24 h, the majority within 6 h following administration [14, 15]. It seems that tubular resorption of ALA in the kidneys is negligible. Systematic clearance of ALA-induced Pp IX occurs within 24 h, whether the route of ALA administration is systemic or topical [1, 16]. ALA-injected mice were killed at regular intervals following injection and the Pp IX fluorescence in various organs and tissues measured. No tissue showed more than background levels of Pp IX fluorescence at 24 h after injection [1]. Human volunteers given ALA by either topical application to various types of skin lesions or intradermal injection of normal skin developed localized photosensitization which vanished within 24 h [1].

Toxicology ALA, Pp IX and other porphyrin precursors are committed biosynthetic intermediates that have a transient, compartmentalized existence and, under normal conditions, are not likely to have pharmacological or toxicological activity beyond their role in the tightly regulated production of heme. In vivo toxicological effects that may be associated with elevated systemic levels of ALA, Pp IX and other porphyrins are well known, and characterize a group of metabolic diseases known as porphyrias. High systemic concentrations of ALA in acute intermittent porphyria (AIP) may be combined with neurological abnormalities. Photosensitivity, associated with high systemic levels of Pp IX, is a typical finding in erythropoietic protoporphyria. ALA Numerous in vivo studies to clarify the relationship between elevated ALA levels and neurological symptoms have yielded contradictory conclusions. Significant depression in spontaneous activity of rats [17] have been described. The symptoms of porphyria may be due to ALA which competes


209

for synaptic GABA (neurotransmitter) receptor binding [18]. However, the concentration of ALA found to be toxic is well above that measured in serum and cerebrospinal fluid during AIP attacks. Therefore it is not possible to ascribe to ALA an etiological role in the development of the behavioral and neurological manifestation of porphyria [19]. The lack of neurologic effects following ALA administration is due to the failure of ALA to cross the blood-brain barrier in pharmacologically significant amounts [20, 21]. Since limited neurological recovery in patients with AIP may occur without a major reduction in plasma levels of ALA, it is concluded that ALA is not directly neurotoxic [22]. Finally, an alternative explanation for neurological effects associated with AIP has been proposed by other studies. Observing that a majority of investigations have not shown ALA to be toxic in vivo infused in large amounts to animals and humans, that ALA does not appear to cross the blood-brain barrier to a significant extent, and that concentrations of these precursors in the brain or cerebrospinal fluid of porphyric patients are substantially low, these studies [23, 24] hypothesize that depletion of hepatic heme, a hallmark of AIP, leads to elevated tryptophan content and 5-hydroxytryptamine turnover in the brain. Administration of tryptophan or 5-hydroxytryptophan to humans has been reported to result in severe abdominal pain, psychomotor disturbances, nausea and dysuria, all of which are the principal symptoms of porphyria [23]. It should be noted that on this basis, administration of exogenous ALA would not lead to a depletion in hepatic heme and, therefore, would not be expected to elicit the neurological symptoms of AIP. Mutagenicity studies in mice with high amounts of ALA did not reveal toxic effects [25, 26]. Reproductive performance of ALA-treated males were equivalent to control males. Treated females were fertile and the number of implantation sites, resorption and fetuses were normal. Pp IX Erythropoietic protoporphyria, associated with decreased activity of ferrochelatase, the enzyme that converts Pp IX to heme, results in massive accumulations of Pp IX in erythrocytes, plasma and feces. Clinically, porphobilinogen is entirely cutaneous, with none of the hematolytic or neurological problems of some of the other porphyrias. Light exposure leads to erythema followed by edema of the skin [27]. The most common complaint, due to Pp IX photosensitization, is a very unpleasant pricking, itching, or burning sensation under the skin. A consistent observation in animal and human studies [28, 29] is that there is no significant toxicological effect on skin from ALA-induced Pp IX or light alone. It is photosensitization due to Pp IX, rather than metabolic activity, that has toxic potential in skin [28].

Wyss

210

References 1 2 3 4 5

6 7

8 9 10 11 12 13 14 15 16 17

18 19 20 21 22 23 24 25

Kennedy JC, Pottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 1992;14:275–292. Jori G, Spikes JD: Photobiochemistry of porphyrins; in Smith KC (ed): Topics in Photomedicine. New York, Plenum Press, 1984. Poh-Fitzpatrick MB: Molecular and cellular mechanisms of porphyrin photosensitization. Photodermatology 1986;3:148–157. Lim HW: Mechanisms of phototoxicity in porphyria cutanea tarda and erythropoietic protoporphyria. Immun Ser 1989;46:671–685. Menon IA, Persad SD, Haberman HF: A comparison of phototoxicology of protoporphyrin, coproporphyrin and uroporphyrin using a cellular system in vitro. Clin Biochem 1989;22:197– 200. Brun A, Sandberg S: Mechanisms of photosensitivity in porphyric patients with special emphasis on erythropoietic protoporphyria. J Photochem Photobiol B 1991;10:285–302. De la Faille HB, Bijlmer-Iest JC, van Hattum J, Koningsberger J, Rademakers LHPM, van Weelden H: Erythropoietic protoporphyria: Clinical aspects with emphasis on the skin. Curr Probl Dermatol 1991;20:123–134. Burnham BF: The chemistry of the porphyrins. Semin Hematol 1968;4:296–322. Mamet R, Schoenfeld N, Mevasser R, Bomstein Y, Lahav M, Atsmon A: Regeneration of the heme synthesis in the regenerating rat liver. Biochem Med Metab Biol 1990;43:263–270. Stout DL, Becker FF: Heme synthesis in normal mouse liver and mouse liver tumors. Cancer Res 1990;50:2337–2340. Berk PD, Rodkey FL, Blaschke TF, Collison HA, Waggoner JG: Comparison of plasma bilirubin turnover and carbon monoxide production in man. J Lab Clin Med 1974;83:29–37. Kappas A, Sassa S, Galbraith RA, Norman Y: The porphyrias; in Scriver CR (ed): Metabolic Basis of Inherited Disease, ed 6. New York, McGraw-Hill, 1989, pp 1305–1365. Granick S, Sassa S: o-Aminolevulinic acid synthetase and control of heme and chlorophyll synthesis; in Vogel HJ (ed): Metabolic Regulation. New York, Academic Press, 1971, pp 77–141. Berlin NI, Neuberger A, Scott JJ: The metabolism of delta-aminolevulinic acid. Normal pathways, studied with the aid of 15N. Biochem J 1956;64:80–90. Berlin NI, Neuberger A, Scott JJ: The metabolism of delta-aminolevulinic acid. Normal pathways, studies with the aid of 14C. Biochem J 1956;64:90–100. Van Gog H, Schothorst AA: Determination of very small amounts of protoporphyrin in epidermis, plasma and blister fluids. J Invest Dermatol 1973;61:42–45. Moore MR, McGillion FB, Goldberg A: Some pharmacological and behavioural effects of deltaaminolevulinic acid; in Doss M (ed): Porphyrins in Human Diseases. Basel, Karger, 1976, pp 148– 154. Muller WE, Snyder SH: Delta-aminolevulinic acid: Influence on synaptic GABA receptor binding may explain CNS symptoms of porphyria. Ann Neurol 1977;2:340–342. Dichter HN, Taddeini L, Lin S, Ayala GF: Delta-aminolevulinic acid. Effect of a porphyrin precursor on an isolated neuronal preparation. Britain Res 1977;126:189–195. Edwards S, Shanley B, Reynoldson J: Neuropharmacology of delta-aminolevulinic acid. I. Effects of acute administration in rodents. Neuropharmacology 1984;23:477–481. Edwards S, Jackson D, Reynoldson J, Shanley B: Neuropharmacology of delta-aminolevulinic acid. II. Effect of chronic administration in mice. Neurosci Lett 1984;50:169–173. Gorchein A, Webber R: Delta-Aminolevulinic acid in plasma, cerebrospinal fluid, saliva and erythrocytes. Clin Sci 1987;72:103–112. Litman DA, Correira MA: L-Tryptophan: A common denominator of biochemical and neurological events of acute hepatic porphyria? Science 1983;222:1031–1033. Litman DA Correira MA: Elevated brain trypophan and enhanced 5-hydroxytryptamine turnover in acute hepatic heme deficiency: Clinical implications. J Pharmacol Exp Ther 1985;232:337–345. Arnold DW, Kennedy GL Jr, Keplinger ML, Calandra JC: Mutagenicity studies with delta-aminolevulinic acid. Food Cosmet Toxicol 1975;13:63–68.


211

26 27 28

29

Kennedy GL Jr, Arnold DW, Calandra JC: Toxicity studies with delta-aminolevulinic acid. Food Cosmet Toxicol 1976;14:45–47. Magnus IA: The cutaneous porphyrias. Semin Hematol 1968;5:380–408. Divaris DXG, Kennedy JC, Pottier RH: Phototoxic damage to sebaceous glands and hair follicles of mice following systemic administration of 5-aminolevulinic acid correlates with localized protoporphyrin IX fluorescence. Am J Pathol 1990;136:891–897. Goff BA, Bachor R, Kollias N, Hasan T: Effects of photodynamic therapy with topical application of 5-aminolevulinic acid on normal skin of hairless guinea pigs. J Photochem Photobiol B 1992; 15:239–251.

PD Dr. Pius Wyss, Department of Gynecology and Obstetrics, University of Zu¨rich, Frauenklinikstrasse 10, CH–8091 Zu¨rich (Switzerland) Tel. +41 1 255 52 39, Fax +41 1 255 44 33, E-Mail [email protected]

Wyss

212


............................

Photodynamic Endometrial Ablation in Nonhuman Primates D.A. Van Vugt a, b, A. Krzemien a, W. Foster c, S. Lundhal d, S. Marcus e, R.L. Reid a Departments of Obstetrics and Gynecology and b Physiology, Queen’s University, Kingston, Ont., c Health and Welfare Canada, Ottawa, Ont., Canada; d Luminetics, Boston, Mass., and e DUSA Pharmaceuticals Inc., Valhalla, N.Y., USA a

Introduction In order to investigate the feasibility of endometrial photodynamic ablation using 5-aminolevulinic acid (ALA) we conducted a series of experiments in nonhuman primates. We describe in the present report photodynamic endometrial destruction in the monkey. This report documents for the first time acute endometrial ablation in the primate using a photodynamic approach.

Materials and Methods Sixteen rhesus monkeys, 5–20 years of age and weighing 4–8.5 kg, and 1 cynomolgus monkey, 13 years old and weighing 3.7 kg, were used. The monkeys were removed from the cage following an intramuscular injection of ketamine HCl (5 mg/kg; Rogarsetic, rogar/STB, Montreal, Que.) and atropine sulfate (0.4 mg/kg; MTC Pharmaceuticals, Cambridge, Ont.) and intubated. A surgical plane of anesthesia was induced with isoflurane. The uterus was exposed through a 7-cm midline incision. The uterus was stabilized by grasping the round ligament and fallopian tubes adjacent to the uterine fundus. A 2-mm nick was made on the serosal surface midway between the fallopian tubes, and a No. 2 Hegar dilator was inserted into the lumen of the uterus. With application of moderate pressure, the dilator was felt to enter the lumen and to pass without resistance to a depth of approximately 1.5–2 cm. The dilator was removed and an intrauterine catheter (Pharmascience Inc. Montreal, Que.) attached to a syringe and filled with ALA diluted in hyskon (Pharmacia Canada Inc., Baie

D’Urfe, Que.) was inserted. 1 ml of the ALA/hyskon solution was infused over 30 s. The catheter was left in place to tamponade potential bleeding from the site of perforation of the uterine fundus. The incision was loosely closed and the animal remained anesthetized. Four hours after installation of ALA (n>16) or hyskon (n>1), the incision was opened and the catheter removed. A 400-lm diameter quartz fiber (Rare Earth Medical Inc., West Yarmouth, Mass.) with a 1-cm diffusing tip was inserted into the uterine lumen. Light at a wavelength of 635 nm and an intensity of 300 mW was delivered for 60 min using a KTP tunable dye laser (Laserscope, San Jose, Calif.). Light was delivered either in a continuous mode for 60 min or in a fractionated mode (20 min on, 5 min off, 40 min on). Following light treatment, the laparotomy incision was closed in two layers and the animal was allowed to recover. Endometrial Suppression and Progestin Withdrawal The procedure described above was performed in either the postmenopausal state or during the early proliferative phase. The postmenopausal state was induced surgically in 4 monkeys by bilateral oophorectomy 3 weeks prior to ablation. In 4 monkeys, a menopausal state was induced by administration of a GnRH agonist (Zoladex, Zeneca Pharma, Mississauga, Ont.) 4 weeks prior to ablation. Eight experiments were performed in the early proliferative phase of either a natural cycle (n>3) or following progestin withdrawal (n>5). Menstruation was induced by progesterone administration by subcutaneous insertion of silastic capsules containing progesterone. Capsules were implanted subcutaneously in the lumbar region while animals were anesthetized with saffan. Progesterone capsules were removed after 5 days. Menstruation occurred 2–5 days later. Endometrial ablation was performed within 2 days following cessation of menstruation. The extent of acute endometrial destruction was assessed 3–5 days after treatment. Animals were anesthetized, the abdomen was opened as described above, and a subtotal hysterectomy was performed. After removal, the uterus was bissected longitudinally along its lateral aspects, and the anterior and posterior hemisections were placed in 10% buffered formalin. The following day, the hemisections were divided into three equal segments, designated as the fundal, middle and lower uterine segment, and embedded in paraffin. Sections 3 lm thick were cut perpendicular to the longitudinal axis and stained with hematoxylin and eosin. The extent of ablation was microscopically assessed in each of the 3 uterine regions (both anterior and posterior). The depth and width of endometrial ablation was measured using a micrometer. The area of endometrial ablation was estimated and expressed as a percent of the total endometrial area (ablated+residual). An ablation score of 1, 2, 3 or 4 corresponding to 0–25, 25–50, 50–75 or 75–100% endometrial ablation was assigned. A mean (×SEM) ablation score was calculated for each region in each of the 4 treatment groups (proliferative phase/continuous light; proliferative phase/fractionated light; menopausal/continuous light; menopausal/fractionated light). When there was disparity in the degree of ablation between anterior and posterior sections, the higher score was used.

Results Histopathological examination of the uterus 4 days after light exposure revealed no evidence of damage in the absence of ALA (fig. 1a). In contrast,

Van Vugt/Krzemien/Foster/Lundhal/Marcus/Reid

214

Fig. 1. Representative photomicrographs of ALA-induced photodynamic endometrial ablation in the monkey. a–c are uteri from proliferative phase monkeys treated with hyskon+continuous light (a), ALA+fractionated light (b), or ALA+continuous light (c). d Ablation in a menopausal uterus treated with ALA+fractionated light. All panels are oriented with the uterine lumen at the top. No evidence of ablation was observed in the control uterus (a). Partial ablation was observed following ALA+continuous light (c), whereas complete ablation was observed following ALA+fractionated light (b, d).

ALA combined with 300-mW light resulted in significant ablation of the endometrium. The degree of ablation ranged from complete to partial (fig. 1b–d). Evidence of full thickness ablation with partial sparing of the lateral aspects of the endometrium was frequently observed. When the degree of ablation was semiquantified for the various treatment groups, several interesting observations were noted. Endometrial ablation in the middle and lower uterine segments of uteri of menopausal monkeys treated with ALA and continuous light (3.75×0.5 and 3.0×1.2, respectively; n>4) was more pronounced compared to ablation in uteri of monkeys in the proliferative phase (1.8×0.8 and 1.7×0.5, respectively; n>4). The ablation score in uteri of proliferative phase animals was increased to 3.0×1.4 for both the middle and lower uterine segments in the two experiments in which light was

Photodynamic Endometrial Ablation in Nonhuman Primates

215

Table 1. Effect of light fractionation and ovarian status on mean depth of endometrial ablation (mean×SD) Proliferative Cont (n>6) Endometrial thickness 1.55×0.36 Ablation depth 1.14×0.54

Menopausal Fract (n>2)

Cont (n>4)

Fract (n>4)

1.65×0.21 2.15×1.62

1.03×0.24 1.28×0.22

1.45×0.64 1.7×0.96

Cont>Continuous; Fract>fractionated.

delivered in a fractionated mode. Endometrial ablation in the middle and lower uterine segments of menopausal uteri was observed more consistently when light delivery was fractionated (4.0×0 and 3.3×0.9, respectively; note the smaller SD; n>4). In contrast to the middle and lower uterine segments, the ablation scores for the fundal segments were uniformly lower. A micrometer was used to assess the maximal depth of ablation by measuring the thickness of necrosis in the plane perpendicular to the myometrial/ endometrial interface. This measurement was compared to endometrial thickness measured in residual endometrium that was usually present at the lateral aspects (table 1). The average depth of ablation exceeded endometrial thickness for all groups except the proliferative phase treated with continous light.

Discussion The present study demonstrates for the first time that significant and selective endometrial ablation can be achieved in the nonhuman primate using ALA photodynamic therapy (PDT). Full thickness endometrial ablation (from the luminal surface to the basalis layer) was achieved in many animals. No significant destruction of the myometrium was evident. The greatest degree of endometrial ablation was observed most consistently in menopausal uteri. If light penetration is a factor which limits the depth of ablation, it is intuitive that a thin endometrium would constitute a less significant barrier to full thickness ablation. We were initially concerned that endometrial photosensitization might be compromised in the uterus not primed with estrogen since metabolically active tissues are more susceptible to photosensitization [1]. However, the present results confirm our observation made in the rat that ALA-induced protoporphyrin IX (Pp IX) accumulation in


216

endometrium is not compromised by ovariectomy [2]. This finding is reassuring since several medical methods of suppressing the endometrium are available clinically. Ablation was most complete in the middle and lower uterine segments while endometrium was spared in the fundal region. Because the direction of light emission is perpendicular to the fiber axis, incomplete ablation may be explained by inadequate light being directed toward the fundus. In addition, less damage occurred in the endometrium at the lateral aspects of the uterus. If one views the uterine lumen as a triangular potential space defined by the internal cervical os and the two internal fallopian tube ostia, the distance from the light fiber to the basalis layer is greater in the lateral plane compared to the anterior-posterior plane. If we assume that light attenuation is the same in both planes, the amount of light reaching the lateral aspects is less because of the longer distance and explains why ablation of this region was less complete. In vitro measurements of light penetration in the human uterus demonstrated that the penetration depth (defined as the distance from the surface at which the fluence rate drops to 37% of its initial value, is between 2 and 4 mm [3, 4]. Distribution of light to all regions of the uterine cavity may require distention of the uterine cavity or use of multiple light fibers. Both methods have been shown to improve light distribution and penetration within the uterine cavity [3, 4]. Incremental light treatment has been reported to enhance the ablative effects of PDT [1, 5]. Since vasoconstriction occurs during light treatment, it is speculated that light fractionation may cause tissue reperfusion and improved oxygenation and lead to greater generation of singlet oxygen, which is believed to be the primary mechanism of ablation. There was a trend for greater endometrial ablation when the 60-min light treatment was admininstered as a 20 min on/5 min off/40 min on regimen. Because the parameters of light fractionation required to produce optimal ablation may vary in different tissues, additional experiments using different parameters and increased numbers are required in order to demonstrate a significant effect of light fractionation. The anatomy of the monkey cervix necessitated a surgical approach for intraluminal delivery of ALA and light in the present study. A surgical approach introduces the variable of intrauterine bleeding which could have affected light transmission. Light transmission was decreased by more than 40% when shone through a layer of blood that was only 0.15 mm thick [6]. The amount of intrauterine bleeding is difficult to control and assess. This variable may be responsible in part for the variation in the degree of ablation. Intrauterine bleeding should be an insignificant variable in the human since atraumatic instrumentation of the uterus can be accomplished transcervically without anesthesia or sedation.

Photodynamic Endometrial Ablation in Nonhuman Primates

217

ALA-induced PDT has several potential advantages over other methods of endometrial ablation. Unlike surgical or thermal ablation, ALA PDT may not require the use of a general anesthetic. Patients in whom ALA PDT has been used to treat a variety of skin lesions report a level of discomfort during treatment that ranges from no discomfort or a very mild tingling sensation to one of moderate burning and itching [7–9]. Paracervical block combined with nonsteroidal anti-inflamatory agents may be sufficient to prevent discomfort. ALA PDT would be technically easier and therefore would eliminate the many complications that can occur with surgical endometrial ablation, particularly during the early stages of acquiring these skills. Lastly, ALA PDT has the potential of being more effective in producing complete and permanent endometrial ablation.

References 1 2

3

4 5

6 7 8

9

Pogue BW, Hasan T: A theoretical study of light fractionation and dose-rate effects in photodynamic therapy. Radiat Res 1997;147:551–559. Roy BN, Van Vugt DA, Weagle GE, Pottier RH, Reid RL: Effect of 5-aminolevulinic acid dose and estrogen on protoporphyrin IX concentrations in the rat uterus. J Soc Gynecol Invest 1997;4: 40–46. Tromberg BJ, Svaasand LO, Fehr MK, Madsen SJ, Wyss P, Sansone B, Tadir Y: A mathematical model for light dosimetry in photodynamic destruction of human endometrium. Phys Med Biol 1996;41:223–237. Fehr MK, Madsen SJ, Svaasand LO, Tromberg BJ, Eusebio J, Berns MW, Tadir Y: Intrauterine light delivery for photodynamic therapy of the human endometrium. Hum Reprod 1995;10:3067–3072. Messmann H, Mlkvy P, Buonaccorsi G, Davies CL, MacRobert AJ, Bown SG: Enhancement of photodynamic therapy with 5-aminolaevulinic acid-induced porphyrin photosensitisation in normal rat colon by threshold and light fractionation studies. Br J Cancer 1995;72:589–594. Vincent GM, Fox J, Charlton G, Hill JS, McClane R, Spikes JD: Presence of blood significantly decreases transmission of 630 nm laser light. Lasers Surg Med 1991;11:399–403. Kennedy JC, Pottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy (review). J Photochem Photobiol 1994;4:275–292. Szeimies RM, Karrer S, Sauerwald A, Landthaler M: Photodynamic therapy with topical application of 5-aminolevulinic acid in the treatment of actinic keratoses: An initial clinical study. Dermatology 1996;192:246–251. Cairnduff F, Stringer MR, Hudson EJ, Ash DV, Brown SB: Superficial photodynamic therapy with topical 5-aminolaevulinic acid for superficial primary and secondary skin cancer. Br J Cancer 1994; 69:605–608.

Dean Van Vugt, PhD, Department of Obstetrics and Gynaecology, 3022 Etherington Hall, Queen’s University, Kingston, Ont. K7L 3N6 (Canada) Tel. +1 613 533 2899, Fax +1 613 533 6779, E-Mail [email protected]


218


............................

Use of ALA-PDT for Endometrial Ablation in the Treatment of Menorrhagia First Clinical Trials Stanley B. Brown, Michael J. Gannon, J. Andrew Holroyd, Nick Johnson, Mark Stringer, David I. Vernon Centre for Photobiology and Photodynamic Therapy, University of Leeds, Leeds, UK

This article presents a short review of our initial studies and their implications for the future of photodynamic therapy (PDT) as a means of effecting endometrial ablations.

Potential for ALA PDT in the Endometrium: Experiments in the Perfused Uterus Initially there were two major considerations. The first was to determine whether 5-aminolevulinic acid (ALA) introduced into the uterus would lead to the selective production of protoporphyrin IX in the endometrium compared with adjacent tissues such as the myometrium. The second was to determine whether ALA introduced via the uterus could enter the general circulation either directly as ALA or indirectly as protoporphyrin IX formed from exogenous ALA. In order to address both of these questions, a model was developed whereby a human uterus, donated at hysterectomy, could be kept metabolically viable for a period of several hours. Women who were scheduled to undergo hysterectomy for menstrual dysfunction gave informed consent for this work, using the uterus removed at surgery. Immediately after hysterectomy, a uterine artery was cannulated and the organ was perfused with 0.9% saline at 37 ºC. The fallopian tubes were ligated and the uterus was suspended fundus down using a clamp on the cervix. Since it is not practicable to cannulate uterine

veins, the perfusate was collected at appropriate intervals and analysed for ALA and protoporphyrin IX. The experiments were performed under a hood to minimise exposure to light. After 4 h, the uterus was dissected and samples of endometrium and myometrium from the fundus and the lower body of the uterus were taken and frozen in liquid nitrogen for subsequent analysis. The analysis of the perfusate revealed no detectable amounts of protoporphyrin IX and only very small amounts of ALA, less than 3% of the administered dose. Even this very small amount of ALA may have been due to an unavoidable leakage of the instilled ALA from the cervix into the perfusate, rather than passage of ALA through the uterus into the circulation. Analysis of the tissue samples revealed, not only that there was a significant amount of protoporphyrin IX in the endometrial specimens (and that this was of the order of the concentration required for adequate PDT), but also that there was very little protoporphyrin IX in the myometrium; indeed on average the ratio of protoporphyrin IX in the endometrium to the myometrium was 10:1 [1, 2]. The first important conclusion from these experiments was that the uterus did indeed remain metabolically viable for several hours after hysterectomy, as shown by the formation of protoporphyrin IX (this could not occur in cells which were metabolically damaged). Secondly, even though the uterus remained viable, no protoporphyrin IX and very little ALA passed from the uterus into the perfusate. From a safety perspective, these results suggested that if ALA was given to a patient by intrauterine administration, there was unlikely to be any significant transfer of the ALA into the general circulation and therefore no problems with generalised photosensitisation. Finally, and perhaps most significantly from the standpoint of the potential of ALA PDT for endometrial ablation, the very high ratios of ALA in the endometrium to the myometrium suggested that the technique was likely to be highly selective in destroying endometrial tissue (fig. 1).

In vivo Photosensitisation The next phase of the study was to ascertain whether the results obtained in the perfused uterus could be reproduced in vivo. For this study, ALA was mixed in KY gel under asceptic, pharmaceutical conditions. The dose of ALA administered was increased in increments from 25 to 900 mg. Eight women who were scheduled to undergo hysterectomy for menstrual dysfunction, gave informed consent for the injection of ALA into the uterine cavity 2–5 h before hysterectomy. In each case, normal endometrial histology had been confirmed on biopsy and each woman had had a tubal ligation. This was considered

Brown/Gannon/Holroyd/Johnson/Stringer/Vernon

220

Fig. 1. Concentration of protoporphyrin IX (PpIX) in the endometrium and the myometrium as a function of ALA dose in the perfused uterus model.

necessary at this stage of the trial to avoid any possible photosensitisation via the fallopian tubes. Blood samples were taken for analysis of plasma protoporphyrin IX covering a 24-h period and immediately after hysterectomy, the uterus was dissected and samples taken for biochemical analysis, for fluorescence microscopy and for histopathological examination. Plasma protoporphyrin IX levels were found to be very low and indistinguishable after ALA administration from control values taken from the same patient before ALA administration. The tissue concentrations of protoporphyrin IX are shown in figure 2. A number of important points are revealed by these results. First, there are substantial amounts of protoporphyrin IX in the endometrium, approximately double the amounts found in the perfused uterus experiments. This is not surprising since it might be expected that the intact uterus would retain more metabolic activity over the period of exposure to ALA than the perfused uterus. There is no doubt that the levels of protoporphyrin IX found are adequate for PDT, based on comparison with previous data in other tissues. Moreover, the ratio of the amount of protoporphyrin IX found in the endometrium compared to that found in the myometrium, whilst not so high as that observed in the perfused uterus, remains very high and favourable for selective PDT of the endometrium. The data revealed no evidence for any increase in the protoporphyrin IX concentration between 2 and 5 h and also, perhaps surprisingly, no correlation between the total dose of ALA and the amount of protoporphyrin IX formed.

Use of ALA-PDT for Endometrial Ablation in the Treatment of Menorrhagia

221

Fig. 2. Concentration of protoporphyrin IX (PpIX) in the endometrium and the myometrium as a function of ALA dose following ALA administration to patients.

The fluorescence microscopy and histopathological examination of specimens from this study have been described in detail elsewhere [2]. Using a CCD camera, protoporphyrin IX fluorescence was detected in the glandular epithelium of the endometrium and, to a lesser extent, in the endometrial stroma. This occurred following the application of as little as 25 mg of ALA. Again, no relationship between the ALA dose or the duration of treatment and the pattern of protoporphyrin IX fluorescence was detected. It is noteworthy that, although most glands in the ALA treated uteri showed fluorescence due to protoporphyrin IX, it was sometimes possible to find glands which showed no such fluorescence. No protoporphyrin IX fluorescence was observed in the myometrium. No correlation was seen between the date of the menstrual cycle and the endometrial concentration of protoporphyrin IX. As anticipated, the patients in this phase of the study who received ALA showed no indication of any systemic photosensitisation nor indeed any morbidity associated with the ALA administration. At this stage we were therefore satisfied that the treatment was essentially safe, that it had a high potential for photosensitising the endometrium at a level which would produce significant photodamage and that such photosensitisation would be selective, so that little or no adjacent tissue damage would be anticipated following irradiation with light.


222

Clinical Studies Originally, it was planned that the next phase of the study would involve the in vivo photosensitisation of the endometrium, followed by light delivery followed by hysterectomy. It would then have been possible to examine the uterus at hysterectomy to determine the extent of photodamage and whether this would be likely to have led to endometrial ablation. However, it was considered that this approach was ethically undesirable, since patients would have had the full PDT treatment but would not have been permitted to benefit from it. We therefore proceeded directly with a study of PDT of the endometrium, without subsequent hysterectomy, unless it was judged that the PDT treatment had failed. It should be emphasised that this was essentially a phase-I study, the primary aim of which was to assess the patient acceptability of the treatment and any morbidity, rather than being a study of efficacy. Nevertheless, it was considered important to attempt to obtain some measures of efficacy, which would assist in the future design of phase-II trials. This was done in essentially two ways. The first was an objective assessment of photodynamic endometrial ablation by measurement of menstrual blood loss (MBL) before and after PDT. This used a new and simple method for measuring MBL, which has previously been used for a large series of women complaining of menorrhagia [3]. The second method of assessment was simply whether or not hysterectomy was required subsequent to PDT. In the first series of treatments, 5 women with normal endometrial histology who complained of heavy menstrual periods gave informed consent. Each woman had had a tubal ligation. Zoladex (Zeneca pharma, Wilmslow, UK) was given by subcutaneous depot injection 4 weeks before the procedure in order to thin the endometrium. Although there was no previous experience of illumination of the human uterus for PDT, there was considerable experience of light delivery for treatment of the bladder. This was used as a guide to the light dosimetry required in the uterus, though it is necessary to tailor the light delivery in terms of the geometry of the emitter and the duration of treatment specifically to the uterus. Initial work was carried out with a normal uterus which had been removed for dysfunctional uterine bleeding and treatment conditions were simulated with a spherical diffusing fibre introduced through the cervix using a hysteroscope and clamped in a rigid position. A constant flow of 0.25% intralipid was passed into the uterus as a scattering medium under positive pressure. For this experiment, an isotropic fibreoptic probe was inserted through the wall of the uterus from the outside and then withdrawn at a constant rate by attaching the fibre to a chart recorder. This allowed a time-distance calibration which could be processed to record the space irradiance within the uterus as a function of distance


223

from the endometrial surface. Using this approach, a spherical fibretip was developed which appeared to offer adequate illumination at all points within the uterus. For the five women in the initial trial, ALA was administered as described previously [1, 2] and, after an appropriate time interval (1.5–4.5 h) light was delivered from a copper vapour pumped dye laser with power output set to a maximum of 750 mW. The hysteroscopy and light irradiation were performed under general anaesthesia. Women stayed in hospital overnight and kept a diary of symptoms after discharge. Review was at 2 weeks and again at 4 months when an ultrasound scan of the endometrium and MBL measurements were performed. In this first series of patients, the treatment was well tolerated and no side effects of the PDT were reported. In this first clinical series, although there were substantial reductions in MBL for most patients, in no case was this enough to give satisfactory symptomatic improvement and it was concluded that adequate ablation of the endometrium had not been achieved. Three of the 5 patients went on to hysterectomy at 5–7 months, in 1 of the patients the procedure was abandoned and the 5th patient was entered into a second series of five patients and had a repeat ablation. In the light of this first experience of PDT of the human endometrium, a number of amendments were made to the procedures for future treatments. In particular, Zoladex was abandoned as it was thought that the reduction in vascularity which it causes may result in reduction in oxygen levels below the threshold required for PDT. Future treatments were therefore planned without pre-treatment and were scheduled for the early proliferative phase of the menstrual cycle to ensure a thin endometrium. Changes were also made in the ALA formulation since it was shown from diffusion experiments that ALA formulated with Ung Merck was substantially more effective than ALA in KY gel. Improvements were also made in the light delivery system and this has been described in detail elsewhere [2]. A second series of 5 patients were treated using amended procedures as outlined above. In this case, review was at 4 months and 1 year when an ultrasound scan of the endometrium and MBL measurements were performed. Again the treatments were well tolerated and caused little or no morbidity. With the exception of 1 patient who showed an increase in MBL (interestingly this was the patient who was retreated from the first series and who was subsequently shown to have a uterine myoma), these patients showed a reduction in MBL of between 28 and 93% at 1 year. Two patients were satisfied at 1 year and did not elect to receive hysterectomy, but the other 3 patients went on to hysterectomy. We have now completed an initial trial of 30 patients though the later results are not included here.


224

Discussion This study has clearly demonstrated the potential of ALA-based PDT for ablation of the human endometrium. The work has shown that, not only is protoporphyrin IX formed in amounts sufficient for adequate PDT by intrauterine administration of ALA, but also that the selectivity thereby achieved for endometrium versus other tissues is high and certainly enough to give confidence that PDT can be performed in the uterus without risk of damage to normal tissue. The work has also clearly demonstrated the feasibility of carrying out PDT in the endometrium from the standpoint of light delivery. Although some optimisation of the light delivery system may still be required, the balloon catheter system proved easy and reliable. The clinical PDT studies also suggested that effective reductions in MBL could be achieved by PDT, but these were not consistently achieved with all patients studied. Moreover, there was no clear correlation between the measured reduction in MBL and the patient judgement of satisfaction which, ultimately, is the most important endpoint. Whilst the technique therefore shows a high potential, there are clearly a number of problems which will have to be solved and these need to be taken into account when designing a phase-II trial where the primary aim is to measure and improve efficacy. Firstly, there is evidence from our fluorescence microscopy studies that protoporphyrin IX formation may not occur consistently throughout the endometrium. This may be correlatable with those cases where PDT is deemed to have failed, but as yet there is no information available on this point. Ung Merck certainly proved to be a much better vehicle for introduction of ALA than KY gel, but ALA uptake may still be incomplete and further work may be required to develop a better uptake system. Although the measurements of light distribution carried out in our study suggest that adequate light is delivered to the endometrium, this also needs further study and possibly further development of light delivery systems. In our most recent cases, the procedure was performed without general anaesthesia and was well tolerated. We are currently further exploring the possibility of patient administered anaesthesia which, in conjunction with the balloon catheter system for light delivery, would enable outpatient treatment. Finally, it should be emphasised that ALA-PDT in the uterus is a technique which involves a large number of parameters which have to be set for any 1 patient treatment. These include the total dose of ALA, its concentration, the vehicle in which it is administered, the total light dose (in J/cm2), the dose rate (in mW/cm2), the time interval between ALA administration and the light delivery, whether the light is given continuously or fractionated, as well as a number of other parameters associated with the endometrium itself such as


225

the stage of the menstrual cycle and whether any endometrial-thinning drugs are administered. It is quite impossible to carry out statistically significant clinical trials in which there is a systematic variation of each of these parameters. Most of them will have to be set by reference to the initial studies described here in patients, parallel studies in animals and by reference to the general expertise which has been built up by a number of groups in clinical PDT. Even then, it will be difficult for any one centre to recruit the number of patients required to answer the outstanding questions, and there must be a strong case for the several groups around the world who are interested in developing ALA-PDT for endometrial ablation to work to common protocols and to pool data.

Acknowledgement The research described here is generously supported by Yorkshire Cancer Research and we are also grateful to the Medical Research Council for a grant towards the clinical study.

References 1

2 3

Gannon MJ, Johnson N, Roberts DJH, Holroyd JA, Vernon DI, Brown SB, Lilford RJ: Photosensitization of the endometrium with topical 5-aminolevulinic acid. Am J Obstet Gynecol 1995;173: 1826–1828. Gannon MJ, Vernon DI, Holroyd JA, Stringer M, Johnson N, Brown SB: PDT of the endometrium using ALA. SPIE 1997;2972:2–13. Gannon MJ, Day P, Hammadieh N, Johnson N: A new method for measuring menstrual blood loss and its use in screening women before endometrial ablation, Br J Obstet Gynaecol 1996;103: 1029–1033.

Prof. S.B. Brown, School of Biochemistry and Molecular Biology, University of Leeds, Leeds LS2 9JT (UK) Tel. +44 113 233 3166, Fax +44 113 233 3017, E-Mail [email protected]


226


............................

Pharmacokinetics of ALA in Human Uterine Tissue M.K. Fehr a, P. Wyss a, Y. Tadir b, c a

Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Switzerland; b Beckman Laser Institute and Medical Clinic, and c Department of Obstetrics and Gynecology, University of California, Irvine, Calif., USA

As shown in animal models, endogenous production of the photosensitizer protoporphyrin IX (PpIX) can be induced in the endometrium by topical administration of 5-aminolevulinic acid (ALA). Photodynamic therapy (PDT) provides a minimally invasive procedure for endometrial ablation probably not requiring anesthesia and adds a viable alternative to routinely performed surgical treatment modalities. As suggested by the animal studies, selective endometrial destruction using PDT may be feasible. Knowledge of the pharmacokinetic behavior of ALA and PpIX in human uterine tissue is of substantial interest for further therapeutical studies using photodynamic techniques. The purposes of these studies were to determine the distribution of the endogenous photosensitizer PpIX in the human uterus and the time interval leading to maximal endometrial fluorescence following intrauterine instillation of ALA.

Material and Methods Twenty-four premenopausal (mean age ×SD 40 ×7 years) and 11 postmenopausal patients (age 60 ×5 years) scheduled for hysterectomy were included in the study after written informed consent was obtained. 3 premenopausal and 3 postmenopausal women of the patients included served as a control group and no intrauterine drug instillation was performed in order to determine the autofluorescence of uterine tissue on frozen sections. In 2 of 29 patients intended to have drug application, intrauterine instillation could not be performed due to cervical stenosis. In all the 27 remaining patients frozen sections of endometrium could be evaluated by fluorescence microscopy and was confirmed as nonneoplastic by histopathological review.

The study was approved by the Human Subject Review Committee of the University of California, Irvine, and University of Zurich, Switzerland. Indications for hysterectomy were uterine prolaps (n>3), urinary incontinence procedure (n>3), fibroids (n>10), history of endocervical intraepithelial neoplasia III (n>7), early cervical cancer (n>3), early endometrial cancer (n>6) and dysfunctional uterine bleeding (n>3). ALA Application Immediately before instillation, 1 g of crystallized ALA hydrochloride (DUSA Pharmaceuticals, Denville, N.J.) was dissolved in 0.9 ml HyskonÔ (Pharmacia, Piscataway, N.J.) and buffered to pH 6 with 0.8 ml of sterile 8.4% sodium bicarbonate Inj., USP (Abbott Laboratories, North Chicago, Ill.) and 620 mg of sodium bicarbonate powder USP (Spectrum Chemical MFG, Gardena, Calif.) under sterile conditions. The resulting 2.5 ml of a 400-mg ALA/ml solution was pressed through a sterile 0.22-lm filter unit (Millex-GS, Milipore Products Division, Bedford, Mass.) into a sterile syringe. 400 mg/ml was found to be the highest soluble ALA concentration possible in HyskonÔ. Topical application of the ALA solution was performed in the lithotomy position 0.75–24 h prior to the scheduled hysterectomy. ALA instillation was performed immediately after induction of anesthesia for surgery in 14 patients and in 13 subjects instillation was performed on the ward without anesthesia. A standard bivalve speculum was placed. The cervix was cleansed with povidone iodine. A Cook Hystero-Salpingography CatheterÔ (Cook, Billerica, Mass.) with an outer diameter of 2 mm was inserted into the cervical canal. To avoid spillage, the balloon was filled with 1 ml of air and the catheter was slightly retracted to block the os internum. 1 ml of 400 mg/ml ALA/HyskonÔ solution was injected slowly into the uterine cavity in a time span of 30 s at a uniform flow rate via slow manual push. No dilatation of the cervix was attempted. None of the patients complained of pelvic pain either during instillation or during the time span before surgery. One patient experienced exacerbating knee pain and numbness in the thigh after drug instillation. Surgery was postponed and immediate physical examination did not reveal any neurological signs or other pathology. She underwent hysterectomy 21 days after drug instillation. Specimen Retrieval Immediately after hysterectomy the specimen was brought to clinical pathology where 2 sections of the uterine wall from the corpus and fundus were obtained. Specimens were immediately placed in molds containing embedding medium for frozen sections (Tissue Tek, O.C.T. media, Miles, Elkhart, Ind.), snap frozen on dry ice and stored at Ö70 ºC in a lightimpermeable container. All specimens were handled in the dark. Tissues were sectioned in low diffuse light (Cryostat microtome, AO Reichert, Buffalo, N.Y.) to obtain 6-lm-thick slices for fluorescence analysis. Fluorescence Microscopy Low-light-level fluorescence microscopy was performed. On the ALA-induced fluorescence images, fluorescence intensity measurements of different anatomical layers were made for comparative analysis. Mean fluorescence intensities of the superficial endometrial gland (immediately adjacent to the lumen), of the deep endometrial glands (immediately adjacent to the endometrial-myometrial junction) and of the myometrium underlying the endometrium were measured. Since the endometrial stroma is a loosely built tissue and often showed

Fehr/Wyss/Tadir

228

artificial holes on the frozen sections, the endometrial stroma was not included in the analysis. Following fluorescence microscopy the frozen sections were hematoxylin-eosin stained, reviewed for histological diagnosis and endometrial thickness was measured. Urine Analysis for ALA After instillation of the ALA solution into the uterine cavity, urine was collected in 17 consecutive patients for the following 24 h as well as on the 2nd postoperative day for analysis of urine ALA excretion. This was measured by spectrophotometry after condensation of ALA with acetylacetone to form a pyrrole, which then reacts with Ehrlich’s reagent and is quantitated colorimetrically [1]. Data Analysis For statistical analysis of differences in fluorescence intensities at different time points we used analysis of variance. Statistical significance was taken as p=0.05. If a significant overall difference was present, multiple comparisons were performed using Sheffe’s multiple comparison procedure. For comparison of the fluorescence intensities of pre- and postmenopausal endometrium, matched pairs at the same time intervals were formed and the paired, two-tailed t test was used. Data are presented as mean ×1 SD.

Results Figure 1 shows the measured mean fluorescence of the superficial endometrial glands, the deep endometrial glands and the myometrium immediately underlying the endometrium as a function of time interval between ALA instillation and uterine removal. Fluorescence above the values of autofluorescence was first detected in the endometrial glands 75 min after drug application. Fluorescence increased substantially after 2 h. Peak fluorescence of accumulated PpIX was measured between 4 and 12 h after ALA instillation. By grouping the patients for statistical analysis, fluorescence of the deep and superficial endometrial glands was highest 4–8 h after ALA instillation (table 1) and showed high selectivity to endometrial tissue (fig. 1). The conclusion that light illumination at this time interval may lead to lasting endometrial ablation is based on two assumptions. First, that peak fluorescence correlates with maximal phototoxic effect. Although studies in malignant and normal cell lines did not show a strict correlation between PpIX cellular content and ALA-induced phototoxicity, there appeared to be a ‘threshold’ effect of cellular PpIX content above which a consistent phototoxic response was noticed [2]. In animal models for photodynamic endometrial ablation maximal tissue destruction seemed to take place at sites of maximal fluorescence [3, 4]. The second assumption is that endometrial regeneration primarily originates by proliferation of stem cells supposed to be located in the gland stumps [5] and

Pharmacokinetics of ALA in Human Uterine Tissue

229

Fig. 1. Mean fluorescence values of superficial endometrial glands, deep endometrial glands and myometrium at the endometrial-myometrial junction of 26 patients operated at different time intervals after intrauterine instillation of a 400-mg ALA/ml HyskonÔ solution. PpIX accumulation in deep and superficial endometrial glands was highest 4–12 h after ALA instillation.

Table 1. Fluorescence signal of superficial endometrial glands, deep endometrial glands and myometrium after grouping for time intervals Controls Time interval, h (n>6) 0.75–1.5 1.5–2 (n>5) (n>5)

2–4 (n>6)

4–8 (n>4)

11–24 (n>6)

Overall difference

Superficial glands (rel. fluorescence)

22×33

145×224 627×434 765×587 2,235×330 664×726

p=0.0001

Deep glands (rel. fluorescence)

24×34

161×264 583×571 553×551 1,948×133 674×852

p>0.0003

Myometrium (rel. fluorescence)

58×51

14×24

61×61

40×54

25×16

2×1

1.5×1.1

1.5×0.9

3.1×2.4

2.8×0.9

Endometrial thickness (mm)

Fehr/Wyss/Tadir

11×8 2×0.9

NS NS

230

that therefore photodynamic destruction of the entire endometrial glands results in lasting endometrial ablation. Mean fluorescence of superficial and deep endometrial glands was 155×75 (61–386) times and 140×73 (48–357) times, respectively, higher than mean myometrial fluorescence (table 1). In the myometrium no significant changes in fluorescence could be detected with increasing time interval. The induced fluorescence of the endometrial glands was highly selective compared to the adjacent myometrium despite instillation of a highly concentrated ALA solution. The myometrial wall may only minimally be involved during photomedical procedures also providing protection of intra-abdominal organs. Although the endometrial stroma also showed fluorescence, quantification was not attempted since this loosely built tissue showed artificial disruption by frozen sectioning. Eleven hours and more after drug instillation glandular fluorescence became inhomogenous with highly fluorescing glands lying next to poorly fluorescing glands. At time intervals exceeding 11 h measurement of glandular fluorescence became problematic due to the patchy pattern of the signal. ALA solution may penetrate certain glands better resulting in an intraluminal drug pool which sustained cellular fluorescence for a longer time period than in other glands where no such intraluminal drug pool was present. To achieve homogenous endometrial destruction, it seems advisable to avoid time ranges longer than 11 h between drug application and intrauterine illumination. In the 1 patient who had hysterectomy postponed by 21 days, no ALA-induced fluorescence could be detected in the uterine tissue. Premenopausal endometrium seemed to convert ALA faster into fluorescing PpIX than postmenopausal endometrium (fig. 2). Forming matched pairs of premenopausal and postmenopausal mean fluorescence of superficial endometrial glands at the same time points (×5 min) a significantly higher fluorescence was found in premenopausal endometrial glands (p>0.01). The high interpatient variability of mean fluorescence may be due to differences in menopausal status and day of menstrual cycle at treatment resulting in differences in endometrial thickness and metabolic activity of the endometrial cells. Since fluorescence is measured in arbitrary units and the threshold value needed for initiation of the phototoxic effect is not known, the critical question is at which time interval photosensitizer accumulation in the gland stumps is maximal and present in all patients. Diffusion of ALA into endometrial cells and its conversion into the fluorescent photosensitizer PpIX seems to require around 75 min in vivo (fig. 1). Since the inactive, atrophic endometrium of postmenopausal women showed lower fluorescence than premenopausal endometrium at the same time points, conversion of ALA into PpIX seems to depend on cellular metabolic activity (fig. 2). This is in good agreement with the enhancement of PpIX accumulation by mitogenic stimulation in cell lines [6].


231

Fig. 2. Mean fluorescence values of superficial endometrial glands of pre- and postmenopausal patients undergoing hysterectomy at different time intervals after intrauterine instillation of a 400-mg ALA/ml HyskonÔ solution. The marks at the time point zero represent 3 pre- and 3 postmenopausal patients of the control group. Modified from Fehr et al. [10].

In 6 of seventeen 24-hour urine samples (35%) collected after drug instillation, ALA excretion was above the reference value of 7 mg/24 h. All values of 24-hour urine samples collected on the 2nd postoperative day were below the reference value. Berlin et al. [7] observed that orally applied ALA in man resulted in unchanged excretion of a large portion (30–70%) in the urine within 24 h, the majority within 6 h following administration. Human volunteers given ALA by intradermal injection exhibited localized photosensitization which vanished within 24 h [8]. Systemic clearance of ALA-induced PpIX occurs within 24 h as well [9]. No phototoxic skin reactions were observed especially not on the skin exposed to intense illumination during operation. To induce PpIX accumulation in the endometrial gland stumps, ALA has to diffuse through the lumen of endometrial glands before cellular uptake for conversion. Interestingly, no significant differences between fluorescence intensities of superficial and deep endometrial glands could be observed (table 1) which can only be explained by rapid endoluminal ALA diffusion. Endoluminal diffusion seems to be facilitated by the low molecular weight and

Fehr/Wyss/Tadir

232

hydrophilic nature of ALA since diffusion velocity is dependent on molecular weight. Due to the low volume of instilled drug solution and slow injection rate we do not think that intracavitary pressure influenced endoluminal ALA propagation. Accordingly, in pathologic endometrial conditions, such as endometrial cancer or polyps, where the communication of endometrial glands to the lumen is disrupted, homogenous photosensitization of the entire endometrial tissue by topical ALA application may be inhibited. It is concluded that selective photosensitizer accumulation in the human endometrial glands is feasible after intrauterine administration of ALA and that maximal fluorescence of the endometrial gland stumps is observed 4–8 h after instillation. At this time point lasting endometrial photodynamic ablation may be feasible if a sufficient light dose can be delivered to the entire endometrium using an appropriate intrauterine light delivery device.

References 1 2 3

4

5

6

7 8 9 10

Fernandez AA, Jacobs SL: Standard Methods of Clinical Chemistry. New York, Academic Press, 1970, p 57. Iinuma S, Farshi SS, Ortel B, Hasan T: A mechanistic study of cellular photodestruction with 5-aminolevulinic acid-induced porphyrin. Br J Cancer 1994;70:21–28. Wyss P, Tromberg BJ, Wyss MTh, Krasieva T, Schell M, Berns MW, Tadir Y: Photodynamic destruction of endometrial tissue using topical 5-aminolevulinic acid (ALA) in rats and rabbits. Am J Obstet Gynecol 1994;171:1176–1183. Steiner RA, Tadir Y, Tromberg BJ, Krasieva T, Ghazarians AT, Wyss P, Berns MW: Photosensitization of the rat endometrium following 5-aminolevulinic acid (ALA) induced photodynamic therapy. Lasers Surg Med 1996;18:301–308. Wyss P, Steiner R, Liaw LH, Wyss MT, Ghazarians A, Berns MW, Tromberg BJ, Tadir Y: Regeneration processes in the rabbit endometrium: A photodynamic therapy (PDT) model. Hum Reprod 1996;11:1992–1997. Rebeiz N, Rebeiz CR, Arkins S, Kelley KW, Rebeiz CA: Photodestruction of tumor cells by induction of endogenous accumulation of protoporphyrin IX: Enhancement by 1,10-phenanthroline. Photochem Photobiol 1992;55:431–435. Berlin NI, Neuberger A, Scott JJ: The metabolism of delta-aminolevulinic acid. Normal pathways, studied with the aid of 15N. Biochem J 1956;64:80–90. Van Gog H, Schothorst AA: Determination of very small amounts of protoporphyrin in epidermis, plasma and blister fluids. J Invest Dermatol 1973:61:42–45. Kennedy JC, Pottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 1992;14:275–292. Fehr MK, Wyss P, Tromberg BJ, Krasieva T, DiSaia PJ, Lin F, Berns MW, Tadir Y: Selective photosensitizer localization in the human endometrium following intrauterine application of 5aminolevulinic acid. Am J Obstet Gynecol 1996;175:1253–1259.

Dr. M. Fehr, Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Frauenklinikstrasse 10, CH–8091 Zu¨rich (Switzerland) Tel. +41 1 255 50 02, Fax +41 1 255 44 33, E-Mail [email protected]


233


............................

Photodynamic Endometrial Ablation: Morphological and Functional Results P. Wyss a, M.K. Fehr a, Y. Tadir b, R. Hornung a, U. Haller a a

b

Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Switzerland; Beckman Laser Institute and Medical Clinic, University of California, Irvine, Calif., USA

The complexity of photodynamic parameters for endometrial ablation required a large number of specific preclinical studies as documented previously. Based on these experimental results, morphological and functional studies were performed on patients suffering menorrhagia.

Patients and Methods The studies were approved by the Ethics Committee of the Department of OB/GYN University Hospital of Zu¨rich. Written informed consent was obtained from the patients. 5-Aminolevulinic acid (ALA)-induced protoporphyrin IX (PpIX) was used as photosensitizer (ASAT, Zug, Switzerland). After intrauterine instillation, ALA is metabolized to fluorescent PpIX in endometrial cells. In P1 (morph.) and P1/P2 (funct.), an ALA/Dextran 70 solution (1.5 ml) was injected in one fraction into the uterine cavity through a sterile 0.22-lm filter unit (Millex-GS, Milipore Products Division, Bedford, Mass, USA) and a hysterosalpingography catheter (Sholkoff Balloon Catheter, Cook, USA) at a concentration of 400 mg ALA/ml. In P2 (morph.) and P3/P4 (funct.), an ALA/water solution was applied intracavitary in two fractions (1.5 and 1.0 ml) at a concentration of 286 mg ALA/ml. In all patients, the acidic ALA solutions were buffered to pH 5.5 with 10 N NaOH to avoid uterine contractions causing reduced endometrial blood flow. No pretreatment with gonadotropinreleasing hormone (GnRH) or Danatrol was performed. Uterine bleeding had to be stopped using 10 mg medroxyprogresteroneacetate daily in both patients. Based on clinical pharmacokinetic studies, illumination of the uterine cavity was performed 6 h after intrauterine ALA instillation using an optical dose of 160 J/cm2 at a wavelength of 635 nm. Laser light was applied during 26–52 min by an intracavitary-positioned

reflecting balloon light diffusor (Medlight SA, Lausanne, Switzerland) following dilation of the cervix to 5.5 mm. Photodynamic endometrial ablation was performed without anesthesia, except in 1 patient using spinal block. Pain score was recorded during PDT using a scale 1–10, where 1 was absolutely no pain and 10 was extreme, unsupportable pain. Antibiotic prophylaxis was performed in all patients using 2.2 g Augmentin i.v. Morphological Study Photodynamic endometrial ablation was performed in a premenopausal patient (age 38 years, P1) suffering from heavy menorrhagia and in a postmenopausal patient (age 70 years, P2) with long-lasting postmenopausal bleeding. Malignancy was ruled out by prior endometrial biopsy in the postmenopausal patient. Hysterectomy was scheduled 3 days (P1) and 5 months (P2), respectively, following photodynamic endometrial damage. Immediately after hysterectomy the specimen was brought to clinical pathology where two transversial sections of the uterine wall each from the isthmus, corpus and fundus were obtained. Sections for histology were stained with hematoxylin and eosin. Functional Study A total of six sessions for photodynamic endometrial ablation have been performed, one in a postmenopausal (P1) and five in premenopausal patients (P2–P4). Malignancy was ruled out by prior diagnostic hysteroscopy and curettage. Daily bleeding patterns were recorded in a scale scoring between 1 (very weak bleeding) and 6 (extreme heavy bleeding) 1–3 months before PDT and during the follow-up. To evaluate the efficacy of photodynamical endometrial destruction, the area under curve (averaged per months) of the bleeding patterns was compared before and after PDT. The thickness of the endometrium was measured by transvaginal sonography pre- and posttreatment.

Clinical Outcome Photodynamic illumination of the endometrium caused some cramping pain probably due to ischemic reactions and prostaglandin release (fig. 1). When light application was started, uterine pain emerged within minutes. The highest pain score recorded at the end of the illumination cycle ranged between 4 and 7, indicating moderate to severe but supportable pain. Once illumination of the endometrium was ceased, the pain score declined quickly to a welltolerable pain level, certainly after the first illumination cycles. It is important to inform patients about possible pain reactions which may entirely vary between different patients. Transvaginal sonographic imaging of a patient, who was treated by PDT for hypermenorrhea causing heavy anemia needing blood transfusion, exhibited a thinning or disappearing of the intracavitary hyperechogenic lining following endometrial PDT. In transversal sections (fig. 2), ultrasound imaging showed absence of endometrium in the isthmus and corpus of the uterus 9 days after PDT. There

Photodynamic Endometrial Ablation

235

Fig. 1. Pain score during photodynamic illumination of the endometrium.

was regenerated or not destroyed endometrium in the fundal region. Similar sonographic documentations were received in the other patients revealing disappearance of endometrial structures in almost all parts of the uterine cavity. The lack of complete endometrial destruction may be due to insufficient light or drug distribution in the endometrium. Histomorphological sections showed intact and destroyed endometrium of the corporeal region in the same patient 3 days following photodynamic endometrial ablation (fig. 3). The dark stained glandular epithelial structures of normal endometrium pervade the whole endometrial thickness and penetrate the superficial layers of the myometrium. The stroma between the glands contains mainly fibroblasts, extracellular matrix and vessels. Photodynamic therapy replaces the thick endometrium by a thin layer of a loose network of fibres (asterisk in figure 3) missing any epithelial structure and results in the damage of vessels as well. Evidently, reduction of uterine bleeding bases on morphological changes of the endometrium. The bleeding patterns of a postmenopausal and premenopausal patient (fig. 4) exhibit severe uterine bleeding before PDT. Once photodynamic endo-

Wyss/Fehr/Tadir/Hornung/Haller

236

Fig. 2. Transvaginal sonographic imaging of the uterus 9 days following PDT.

metrial ablation was performed an increased slough of bleeding occurred followed by a reduction of the uterine bleeding intensity. The premenopausal patient was not content with the result of the first PDT session and inquired about a second photodynamic treatment which resulted in a satisfying and persistent decrease of menstrual blood loss.


237

Fig. 3. Histomorphological sections of the corporeal endometrial region 3 days following photodynamic endometrial ablation.

Discussion The human endometrium is particulary well suited to photomedical application. From a practical standpoint, both drug and light can be easily administered using topical sensitizers and transvaginal optical balloon illumination. Photodynamic endometrial ablation is feasible without general or


238

Fig. 4. Bleeding patterns of a post- and premenopausal patient before and after endometrial PDT.

local anesthesia [1] and may therefore offer a minimal-invasive, nonhysteroscopic method for endometrial ablation indicated for dysfunctional uterine bleeding. Photodynamic tissue destruction is selectively limited to the endometrial layer (i.e. the myometrium being spared) as previously suggested by fluorescence microscopy [2, 3]. Success depends on prevention of endometrial regeneration via photoinactivation of proliferating stem cells at the endometrial-myometrial junction [4]. The regenerative capacity of endometrium is regarded as unique and as one of the most dynamic in humans. It is characterized by cyclic proliferation, differentiation and cell death every menstrual cycle. The duration for menstrual surface re-epithelialization lasts about 48 h and begins on cycle days 2 or 3 [5]. Such potent renewal of tissue is known only in the hematopoietic [6], intestinal [7], and epidermal systems [8]. Several morphologic studies on endometrial regeneration in humans [9–12] and animals [13, 14] demonstrate its regenerative potential. In order to understand the cellular biology on which quickly-renewing tissues are based, several hypotheses have been offered:


239

(1) A small pool of multipotent undifferentiated stem cells located near or within the endometrial-myometrial junction give rise to epithelial, stromal or vascular cells [15, 16]. (2) The regenerating surface epithelium originates from the residual epithelium of gland stumps or undamaged bordering epithelium [5, 9, 17]. (3) Stromal cells transform to endometrial epithelium relining the surface [10, 18]. The genuine mechanism of endometrial regeneration is difficult to determine by microscopic investigations of cellular morphology. Even functional studies may not define the final mechanism [5, 15]. Based on the embryological background, it is more likely that several hypothetic processes may participate in the regeneration of the endometrium. To suppress high menstrual blood loss it is essential to destroy the full thickness of the endometrium including the deep basal glands in the endometrial-myometrial junction. Since the ALAinduced photosensitizer PpIX is metabolized in this deep region of the endometrium, regeneration may be avoided by the selective targeting of photodynamic procedures. Outcome criteria include amenorrhea, reduction of menstrual blood loss, a subjectively desirable outcome (patient’s satisfaction) or even avoidance of hysterectomy. Since significant reduction in menstrual blood loss may represent success in one patient and failure in another [19], the evaluation of patient’s satisfaction provides a valuable indicator for outcome measures. Amenorrhea does not necessarily have to be the endpoint of treatment of dysfunctional uterine bleeding following endometrial ablation [20]. Failures of complete photodynamic endometrial destruction as shown in the sonographic images were due to a technical insufficiency of the balloon light diffusor or inhomogeneous local drug application. Further technical developments of light and photosensitizer application systems are previewed to achieve complete endometrial destruction. Patients considered for endometrial ablation should have hysteroscopy and curettage to rule out a malignancy. Neither hysteroscopic nor nonhysteroscopic endometrial ablation necessarily exclude the development of an endometrial cancer [21–26]. Flow cytometric data [27] revealed a significantly higher ALAinduced accumulation of PpIX in human endometrial adenocarcinoma cells compared to the normal, nonpathological endometrial epithelial cells. Bedwell et al. [28] showed that PpIX fluorescence intensity of rat colonic tumor glands was about 6 times higher than normal glands after systemic administration of ALA. In general, metabolism in malignant cells seems to be highly activated producing increased accumulation of PpIX following ALA application. The increased accumulation of ALA-induced PpIX in neoplasic endometrial cells compared to their normal counterparts is of notable significance for endome-


240

trial diagnostics and therapeutics. Increased PpIX fluorescence in endometrial cancer cells may be visualized by blue light (i.e. at a wavelength of 405 nm) during hysteroscopy, making guided biopsies possible for accurate diagnosis. Additionally, endometrial cancerous tissue with more elevated intracellular PpIX concentrations will be the primary target during photodynamic endometrial ablation. Long-term studies have to prove whether photodynamic endometrial ablation prevents endometrial cancer growth in the follow-up.

References 1 2

3

4

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Wyss P, Fehr M, van den Bergh H, Haller U: Feasibility of photodynamic endometrial ablation without anesthesia. Int J Gynaecol Obstet 1998;60:287–288. Wyss P, Tromberg BJ, Wyss MT, Krasieva T, Schell M, Berns MW, Tadir Y: Photodynamic destruction of endometrial tissue using topical 5-aminolevulinic acid in rats and rabbits. Am J Obstet Gynecol 1994;171:1176–1183. Steiner RA, Tadir Y, Tromberg B, Krasieva T, Ghazarians A, Wyss P, Berns MW: Photosensitization of the rat endometrium following 5-aminolevulinic acid-induced photodynamic therapy. Lasers Surg Med 1996;18:301–308. Wyss P, Steiner R, Liaw LH, Wyss MT, Ghazarians A, Berns MW, Tromberg BJ, Tadir Y: Regeneration processes in the rabbit endometrium: A photodynamic therapy model. Hum Reprod 1996;11: 1992–1997. Ferenezy A: Studies on the cytodynamics of experimental endometrial regeneration in the rabbit. I. Historadioautography and ultrastructure. Am J Obstet Gynecol 1977;128:536–546. Lajtha LJ: Review of leukocytes. Natl Cancer Inst Monogr 1973;38:111–123. Hagemann RF, Lesher S: Intestinal cytodynamics: Adductions from drug radiation studies; in Zimmerman AM, Padilla BM (eds): Drugs and the Cell Cycle. New York, Academic Press, 1973, p 195. Lavker RM, Sun T: Epidermal stem cells. J Invest Derm 1983;81:121–127. McLennan C, Rydell A: Extent of endometrial shedding during normal menstruation. Obstet Gynecol 1965;26:605–621. Baggish MS, Pauerstein CJ, Woodruff JD: Role of stroma in regeneration of endometrial epithelium. Am J Obstet Gynecol 1967;99:459–465. Ferenezy A: Studies on the cytodynamics of human endometrial regeneration. I. Scanning electron microscopy. Am J Obstet Gynecol 1976;124:64–74. Ferenezy A: Studies on the cytodynamics of human endometrial regeneration. II. Transmission electron microscopy and histochemistry. Am J Obstet Gynecol 1976;124:582–595. Schenker JG, Sacks MI, Polshuk WZ: Regeneration of rabbit endometrium following curettage. Am J Obstet Gynecol 1971;111:970–978. David A, Kaplun D, Serr DM, Czernobilsky B: Effect of intrauterine contraceptive device on the regeneration of rabbit endometrium. Am J Obstet Gynecol 1973;117:473–477. Padykula HA: Regeneration in the primate uterus: The role stem cells; in Wynn RM, Jollie WP (eds): Biology of the Uterus. New York, Plenum Medical, 1989, p 279. Prianishnikov VA: On the concept of the stem cell and a model of functional-morphological structure of the endometrium. Contraception 1978;18:213–223. Bartelmez WG: Histological studies on the menstruating mucous membrane of the human uterus. Contrib Embryol. Basel, 1933, vol 34, pp 141–148. Papanicolaou G: Epithelial regeneration in the uterine glands and on the surface of the uterus. Am J Obstet Gynecol 1933;25:30–37. Chimbria TH, Anderson ABM, Turnbull AC: Relationship between measured menstrual blood loss and patient’s subjective assessment of loss, duration of bleeding, numbers of sanitary towels used, uterine weight and endometrial surface area. Br J Obstet Gynecol 1980;100:244–252.


241

20 21 22 23 24 25

26

27

28

Liu DT: Treatment of dysfunctional uterine bleeding. Amenorrhea need not to be an end point. BMJ 1995;310:802. Gimpelson RJ: Not so benign endometrial hyperplasia: Endometrial cancer after endometrial ablation. J Am Assoc Gynecol Laparosc 1997;4:507–511. Margolis MT, Thoen LD, Boike GM, Mercer LJ, Keith LG: Asymptomatic endometrial carcinoma after endometrial ablation. Int J Gynaecol Obstet 1995;51:255–258. Horwitz IR, Copas PR, Aaronoff M, Spann CO, McGuire WP: Endometrial adenocarcinoma following endometrial ablation for postmenopausal bleeding. Gynecol Oncol 1995;56:460–463. Copperman AB, DeCherney AH, Oliv DL: A case of endometrial cancer following endometrial ablation for dysfunctional uterine bleeding. Obstet Gynecol 1993;82(suppl):640–642. Ramey JW, Koonings P, Given FT, Acosta A: The process of carcinogenesis for endometrial adenocarcinoma could be short: Development of malignancy after endometrial ablation. Am J Obstet Gynecol 1994;170:1370–1371. Klein Z, Markovitch O, Altaras M, Beyth Y, Fischman A: Advanced endometrial adenocarcinoma following endometrial ablation: A case report and review of the literature. Int J Gynecol Cancer 1997;7:163–165. Wyss-Desserich MT, Sun CH, Wyss P, Kurlawalla CS, Haller U, Berns MW, Tadir Y: Accumulation of 5-aminolevulinic acid-induced protoporphyrin IX in normal and neoplastic human endometrial epithelial cells. Biochem Biophys Res Commun 1996;224:819–824. Bedwell J, MacRobert AJ, Phillips D, Bown SG: Fluorescence distribution and photodynamic effect of ALA-induced protoporphyrin IX in the DMH rat colonic tumor model. Br J Cancer 1992;65: 818–824.

PD Dr. Pius Wyss, Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Frauenklinikstrasse 10, CH–8091 Zu¨rch (Switzerland) Tel. +41 1 255 52 39, Fax +41 1 255 44 33, E-Mail [email protected]


242


............................

A Light Distributor for Photodynamic Endometrial Ablation R. Bays a, b, A. Woodtli b, L. Mosimann a, P. Wyss c, G. Wagnie`res a, U. Haller c, H. van den Bergh a a

Institute of Environmental Engineering, Swiss Federal Institute of Technology, Lausanne; b Medlight SA, Ecublens, and c Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Switzerland

Photodynamic therapy (PDT) is one of many possible approaches to endometrial ablation in the case of menorrhagia. In recent clinical investigations 5-aminolevulinic acid was applied topically in the uterus to create temporary high concentrations of protoporphyrin IX in the endometrium [1, 2]. The endometrium was then exposed to red light at 635 nm in the hope of its selective destruction. In this procedure it is important to irradiate essentially all of the endometrium with a sufficient and homogeneous dose of light (at the desired dose rate) to destroy the tissue. Hence there is a need for an appropriate fiber optic-based light distributor. In hollow organs like the uterus this problem can be solved in many ways [3, 4] among which the use of a highly reflecting, transmitting balloon exposed to light from the inside. Photons leaving a fiber at the center of such a balloon are reflected many times prior to being transmitted through the balloon wall. This implies that nearly independent of the shape of the balloon, the light is sufficiently ‘scrambled’ prior to being transmitted, that a nearly homogeneous irradiation is attained over the whole balloon surface [5]. The balloon is made of an elastomer loaded with highly scattering particles. The device is introduced by the cervix to the uterine cavity without anesthesia. The balloon, which has the shape of a uterus, is inflated in order to position it and so that its wall is in nearly perfect contact with the endometrial surface as can be seen in figure 1. The balloon is also built so as to give

a

b

c

Fig. 1. a The balloon is introduced into the uterine cavity by the means of an introducing tube. b The introducing tube is pulled out, leaving the deflated balloon in place. c The balloon is then inflated to adjust its shape to the organ, and it supplies optimal illumination to the tissue surface.

Fig. 2. Picture of the complete device.

as homogeneous as possible a pressure on the uterus wall to avoid cutting off the blood circulation and hence the tissue oxygenation important for PDT. Before starting the light exposure, the balloon is also slightly deflated for the same reason. Preliminary clinical investigation has shown promising results. Necrosis analysis has demonstrated that light homogeneity at the balloon surface is good enough to cause complete tissue ablation of the endometrium up to the myometrium. The balloon shown in figure 1 and 2 has been clinically tested by Wyss et al. [1] and was constructed for this purpose by Medlight SA in Lausanne, Switzerland.

Bays/Woodtli/Mosimann/Wyss/Wagnie`res/Haller/van den Bergh

244

References 1 2 3

4 5

Wyss P, Fehr M, van den Bergh H, Haller U: Feasibility of photodynamic endometrial ablation without anesthesia. Int J Gynaecol Obstet 1998;60:287–288. Gannon MJ, Vernon DI, Holroyd JA, Stringer M, Johnson N, Brown SB: PDT of the endometrium using ALA. Proc SPIE 1997;2972:2–13. Van den Bergh H, Mizeret J, Theumann J-F, Woodtli A, Bays R, Robert D, Thielen P, Philippoz JM, Braichotte D, Forrer M, Savary JF, Monnier Ph, Wagnie`res G: Light distributors for photodynamic therapy. Proc SPIE 1996;2631:173–198. Fehr MK, Madsen SJ, Svaasand LO, Tromberg BJ, Eusebio J, Berns MW, Tadir Y: Intrauterine light delivery for photodynamic therapy of the human endometrium. Hum Reprod 1995;10:3067–3072. Beyer W, Baumgartner R, Ell C, Heinze A, Jocham D, Sroca R, Stepp H, Unso¨ld E: Uniform light distribution in hollow organs by means of backscattering layers. Proc SPIE 1990;1201:298–303.

Dr. R. Bays, Institute of Environmental Engineering, Swiss Federal Institute of Technology, CH–1015 Lausanne (Switzerland) Tel. +41 21 697 07 75, Fax +41 21 697 07 79, E-Mail [email protected]

A Light Distributor for Photodynamic Endometrial Ablation

245


............................

Intrauterine Light Probe for Uterine Photodynamic Therapy Y. Tadir a, b, R. Hornung a, c, B.J. Tromberg a a b

c

Beckman Laser Institute and Medical Clinic; Department of Obstetrics and Gynecology, University of California, Irvine, Calif., USA, and Department of Gynecology and Obstetrics, University Hospital of Zu¨rich, Switzerland

It is essential to develop a light applicator that is sufficiently flexible to accommodate a range of uterine optical and physiological properties [1]. The principles of light propagation through the human uterus have been studied extensively [2, 3]. A mathematical model has been derived to describe homogenous light distribution in the uterus [4]. It is possible to apply sufficient light to the entire endometrium with three cylindrical diffusers. A horizontal plane through these three diffusers is defined by three unique axes that are symmetric in all four quadrants. The light at various points in one plane is contributed by one, two, or all three diffusers. The simulations for the mathematical model of our intrauterine light probe were based on worst-case optical properties of the endometrium to ensure that a minimal fluence rate would be achieved everywhere in the uterine endometrium to destroy it irreversibly. Based on these observations, we developed a dedicated intrauterine light probe device designed specifically to illuminate the entire uterine cavity. Using extirpated human uteri, we demonstrated (1) that the intrauterine light probe delivers light along the three unique axes as predicted by the mathematical model; (2) that light distribution is symmetric in all four quadrants, and (3) that the intrauterine light probe is capable of delivering sufficient light for effective photodynamic therapy for endometrial ablation. The intrauterine light probe (US Patent No. 5,478,339) consists of three optical fibers converging to one bundle mimicking the shape of the uterine cavity (fig. 1). Each of the three fibers contains a cylindrical light diffuser at the output end and a universal connector to fit the laser source at the input

Fig. 1. The intrauterine light probe emitting light superimposed on a drawing of a uterine cavity.

end. The length of the cylindrical diffusing tips is 2 cm for the lateral and 4 cm for the central fiber, but may vary according to the size of the uterine cavity. The two lateral fibers are curved to relax parallel to the side walls pointing towards both uterine cornua. The outer diameter of the flexible fibers conducting the light is 400 lm. The diffusers have an outer diameter of 1.2 mm. The intrauterine light probe is introduced into the uterine cavity through a semirigid introducing sleeve (outer diameter 3.5 mm) in a maneuver similar to the insertion of an intrauterine contraceptive device. Once inserted, the sleeve is retracted, and the fibers expand to fit the cavity shape. The feasibility of the intrauterine light probe to deliver the required light dose for photodynamic therapy was demonstrated on three surgically removed human uteri. Laser light at a wavelength of 635 nm, matching the absorption peak of protoporphyrin IX, the photoactive metabolite of the precursor 5aminolevulinic acid, was generated by an argon-pumped dye laser. The laser beam was divided into three by a beam splitter connected to the three fibers of the intrauterine light probe. The intrauterine light probe was positioned in the uterine cavity through the cervical os by inserting the semirigid catheter containing the unspread light diffusers. When the fundus was reached, the sleeve was withdrawn, and the intrauterine light probe was allowed to spread.

Intrauterine Light Probe for Uterine Photodynamic Therapy

247

Fig. 2. Light decay in uterine tissues. Measurements were performed on the anterior walls of three ex vivo uteri. Data are presented as mean values×standard error, normalized to 1. The endomyometrial junction was assumed at 5 mm.

A detecting isotropic fiber tip was placed through the uterus from one outer side of the uterine wall to the opposite outer side of the uterine wall. Starting from the initial position where the spherical tip was at one end of the outer uterine wall, the detecting tip was withdrawn towards the opposite outer uterine wall. This sequence was performed for each of the four uterine orientations. Figure 1 shows the trifurcated intrauterine light probe emitting red light superimposed on a drawing of a uterine cavity. The insertion sleeve is kept at the level of the internal cervical os. In the extirpated uteri, positioning of the intrauterine light probe was feasible without dilation of the cervical canal. In all three uteri, the fibers of the intrauterine light probe spread as expected without additional manipulation. Figure 2 shows the average light decay in the anterior wall of the three uteri. At a distance of 5 mm, where the endomyometrial junction is expected, the light reached a level of 0.31×0.02. At 10 mm, the level of light was only 0.07×0.02. Figure 3 shows the light distribution in the tissue of one removed uterus. The shapes of the curves confirm the validity of our mathematical model. When the detecting fiber passed along

Tadir/Hornung/Tromberg

248

Fig. 3. Light distribution in an ex vivo human uterus. Graphs A–H indicate light decay at various distances from the intrauterine light probe (y axis values normalized to 1; x axis 20 mm). Uterine wall thickness was 21 mm. 1>Dye laser (635 nm); 2>illumination fiber; 3>intrauterine light probe; 4>detecting tip; 5>detecting fiber; 6>analyzer.

Intrauterine Light Probe for Uterine Photodynamic Therapy

249

the coronal plane, partly through the uterine walls and partly inside the uterine cavity, three peaks were measured. The central peak is slightly lower (plots A and E) because the light is emitted through a longer diffusing tip. Light decay was comparable in all planes.

References 1 2 3 4

Tadir Y, Hornung R, Pham TH, Tromberg BJ: Intrauterine light probe for photodynamic ablation therapy. Obstet Gynecol 1999;93:299–303. Fehr MK, Madsen SJ, Svaasand LO, et al: Intrauterine light delivery for photodynamic therapy of the human endometrium. Hum Reprod 1995;10:3067–3072. Wyss P, Svaasand LO, Tadir Y, et al: Photomedicine of the endometrium: Experimental concepts. Hum Reprod 1995;10:221–226. Tromberg BJ, Svaasand LO, Fehr MK, et al: A mathematical model for light dosimetry in photodynamic destruction of human endometrium. Phys Med Biol 1996;41:223–37.

Y. Tadir, MD, Beckman Laser Institute and Medical Clinic, University of California, 1002 Health Sciences Road East, Irvine, CA 92612 (USA) Tel. +1 714 824 4713, Fax +1 714 824 8413, E-Mail [email protected]

Tadir/Hornung/Tromberg

250

Cervic/Vulva/Vagina Wyss P, Tadir Y, Tromberg BJ, Haller U (eds): Photomedicine in Gynecology and Reproduction. Basel, Karger, 2000, pp 251–264

............................

Selective Photosensitization in Vulvar Condyloma and PDT of Vulvar Intraepithelial Neoplasia Mathias K. Fehr a, Pius Wyss a, Diana Dobler a, Curtis F. Chapman b, Tatiana B. Krasieva b, Bruce J. Tromberg b, Michael W. Berns b, Yona Tadir b, c, Viola Schwarz a, Urs Haller a a

Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Switzerland; b Beckman Laser Institute and Medical Clinic, and c Department of Obstetrics and Gynecology, University of California, Irvine, Calif., USA

Introduction Photodynamic therapy (PDT) following topical applicationn of photosensitizers has shown high complete response rates in a variety of skin diseases, including superficial basal cell carcinoma, superficial squamous cell carcinoma, actinic keratosis, and psoriasis [1, 2]. In these diseases it has been shown that topically applied 5-aminolevulinic acid (ALA) readily passes the abnormal layer of keratin and is metabolized to photosensitizing concentrations of porphyrins. ALA is a precursor of protoporphyrin IX (PpIX) in the biosynthetic pathway for heme. Normally, the synthesis of heme regulates the synthesis of ALA via feedback control. The administration of exogenous ALA bypasses this feedback and induces the accumulation of PpIX causing tissue photosensitivity since the step of converting PpIX into heme is a relatively slow process [3]. Because the adjacent normal skin is less permeable, it is not necessary to restrict the topical application of ALA to the lesion itself. Condylomata acuminata are caused by human papillomavirus (HPV) infection, an infection which usually involves the entire lower female genital tract. As noted by the Center for Disease Control, the treatment of condylomata acuminata has not been well studied, and no treatment is completely satisfactory [4]. The management of subclinical HPV infection is even more

controversial than that for overt condyloma. Since ALA-induced photosensitization is highly restricted to epithelial tissues and tumors [1, 3, 5] and a higher conversion rate to PpIX can be expected in proliferative tissues because of higher metabolic activity, we hypothesized that selectivity in photosensitization of condyloma could be achieved using appropriate time intervals and drug concentrations. Selectivity of photosensitivity would consequently imply that multicentric overt condyloma as well as subclinical lesions could be treated simultaneously by PDT. This concept could easily be expanded to multicentric vulvar intraepithelial neoplasia which are associated with HPV infection with certain virus types and similarly show a disturbed epithelial architecture [6, 7]. First we investigated the feasibility of ALA-induced photosensitization of typical vulvar condyloma and its selectivity to the lesion by comparing the fluorescence to the surrounding, macroscopically normal-appearing skin. Different drug concentrations and time intervals were evaluated by in vivo fluorescence imaging and fluorescence microscopy. Results show that selective photosensitization of vulvar condyloma can be achieved within short time intervals. In a second step we assessed the feasibility of PDT of 7 patients with vulvar condyloma and of 5 patients with vulvar intraepithelial neoplasia III using a 10% ALA gel and different optical doses. These patients are part of an ongoing clinical phase-II study evaluating the efficacy of PDT and the needed optical dose for treatment of these diseases.

Material and Methods Selective Photosensitization of Condyloma Twenty-four patients (age 27×1.6 years) referred to the Beckman Laser Institute and Medical Clinic at the University of California, Irvine, for CO2 laser treatment of typical vulvar condylomata acuminata were included in this study. Written informed consent was obtained following approval by the Human Subject Review Committee of the University of California, Irvine. The diagnosis was made by colposcopic inspection of the vulva after application of 4% acetic acid. All aceto-white lesions were mapped on a sketch and photographed. Pregnant or nursing patients were excluded as well as patients with inflammatory changes or colposcopically unclear lesions requiring biopsy. Two ALA formulations were studied: 20% ALA-HCl emollient cream (oil-in-water emulsion) and 2.5% ALA-HCl ointment. We chose a 20% ALA emollient cream because these formulations have been shown to optimize percutaneous ALA penetration in vitro (unpublished data) and induce high PpIX conversion in vivo [1, 2]. With the use of the 2.5% ALA ointment we wanted to test a lower ALA concentration which might induce minimal fluorescence in normal tissues and achieve better adherence of the formulation to the skin. Most clinical studies of skin lesions using topical ALA have been performed with cream formulations and occlusive dressings [1, 2] which can hardly be used on the external genitalia.

Fehr et al.

252

The cream was prepared immediately prior to application by mixing 1 g of crystallized ALA (DUSA Pharmaceuticals, Denville, N.J.) with 4 g of emollient cream (DUSA Pharmaceuticals, Denville, N.J.) for at least 3 min with a spatula. The anhydrous ointment (DUSA Pharmaceuticals, Denville, N.J.) consisted of 2.5% ALA hydrochloride in a preformulated, stable form. Before drug application the vulvar skin was washed with a watery 0.4% chlorhexidine solution to reduce the bacterial flora which could cause autofluorescence or enhance ALAinduced bacterial fluorescence. Before drug application, fluorescence images were acquired to assess autofluorescence. 5 g of the ALA cream or ointment were spread over the entire vulva and perianal skin. The patients were given a nonadhering dressing (ReleaseÔ, Johnson & Johnson Products, Arlington, Tex.) to prevent absorption of the formulation by clothing. Patients who were scheduled for fluorescence evaluation after 24 h were asked not to shower or wipe off the cream until the next morning. In vivo fluorescence was evaluated 1, 3, or 6 h after application of the 2.5% ALA ointment and 1, 3, or 24 h after application of the 20% ALA cream. Patients were surveyed for discomfort at the drug application site and the drug was washed off with wet gauzes. Fluorescence was activated by light from a model B-100 AP ultraviolet lamp (UVP, San Gabriel, Calif.) which was positioned in front of the vulva as a distance of approximately 50 cm. Fluorescence was monitored using an intensified charge-coupled device (CCD) camera (Hamamatsu Photonics, model C2400-86 intensifier head, C2400-77 CCD, Japan) displaying the real-time images on the monitor of a Macintosh Quadra 840AV computer. Images were acquired using 450- and 650-nm bandpass filters (25-nm bandpass, Corion, Holliston, Mass.) and a built-in frame grabber. To assess excitation light distribution, each defined lesion was first imaged using a 450-nm bandpass filter. After quickly exchanging the filter without moving the camera position, the fluorescence image was acquired with the 650-nm filter. After turning off the UV lamp, images were acquired for each filter to determine dark noise signal. Following in vivo fluorescence imaging which took less than 10 min, biopsies were taken after local anesthesia with 1% lidocaine HCl injection. At least one macroscopically identified typical condyloma of the labia minora or vestibule was excised together with normal-appearing adjacent skin using a scalpel. For the sake of simplicity, the stratified squamous epithelium of the hymen, vestibule and labia minora is referred to as vulvar skin without hair in contrast to hair-bearing vulvar skin. Specimens were immediately placed in molds containing embedding medium for frozen specimens (Tissue-Tek II O.C.T. media, Miles, Elkhart, Ind.), frozen on dry ice and stored at Ö70 ºC for 6 days maximum. The blocks were sectioned in low diffuse light (Cryostat microtome, AO Reichert, Buffalo, N.Y.) to obtain 6-lm-thick slices for fluorescence analysis. Biopsies of typical vulvar condyloma of 4 patients without drug application were used as controls. Low-light-level fluorescence microscopy was performed with a slow-scan, thermoelectrically cooled, charge-coupled device camera system (Princeton Instruments, Trenton, N.J.) coupled to a Zeiss Axiovert 10 inverted fluorescence microscope (Carl Zeiss, Oberkochen, Germany). A 10¶ objective (Zeiss Plan-neofluan numerical aperture>0.3) was used to visualize bright-field and fluorescence images of frozen sections. A 100-watt mercury arc lamp filtered through a 405-nm bandpass filter (20-nm band width, Omega Engineering, Stamford, Conn.) provided excitation light. A dichroic filter (Zeiss, FT 420, Carl Zeiss, Oberkochen, Germany) was used to separate excitation from emission signals and a 635nm broad bandpass filter (55-nm band width, Omega Engineering, Stamford, Conn.) was used to isolate the fluorescence emission. Instrument control, image acquisition and pro-

Selective Photosensitization in Vulvar Condyloma and PDT

253

cessing were performed with a Macintosh IIfx computer and IPLab software (Signal Analytics Corp., Vienna, Va). Sample photo degradation was minimized by limiting arc-lamp exposure to 2 s by electronically synchronizing camera and lamp shutters (Uniblitz, model T132, Vincent Associates, Rochester, N.Y.). In order to correct for light distribution, background images were acquired from blank slides under conditions identical to those used for sample measurements. Dark noise levels were determined by acquiring images without source illumination. All fluorescence images were corrected for both nonuniform illumination and dark noise contribution using the following algorithm: Corrected fluorescence image> Fluorescence image – dark noise ¶mean (background image – dark noise), background image – dark noise where mean (background – dark noise) is the mean grayscale value for the dark-noisecorrected background image. Specimens were divided into anatomical layers for comparative analysis. Mean fluorescence intensity of the epithelium without stratum corneum and of the papillary dermis was measured in arbitrary units. Measurements were made on four different sections through the condyloma and on four different sections of the adjacent skin. The mean values for each patient were then calculated. After fluorescence analysis, the sections were stained with hematoxylin eosin and reviewed. Maximal epithelial thickness of the condyloma and the adjacent normal skin was measured with the built-in scale of the ocular. The in vivo fluorescence images were likewise processed with IPLab software and corrected for nonuniform illumination and contributing dark noise using the following algorithm: Corrected fluorescence image> image (650 nm) – dark noise (650 nm) ¶mean image (450 nm) – dark noise (450 nm)). image (450 nm) – dark noise (450 nm) Light distribution without fluorescence signal was determined using the 450-nm bandpass filter and the dark noise was determined for each filter separately. On the corrected fluorescence image the mean fluorescence of the condyloma and of the surrounding, macroscopically appearing normal skin were measured. For statistical analysis, differences in fluorescence intensities and ratios between fluorescence intensity of condyloma and adjacent normal skin were examined at different time points for each drug formulation using the Kruskal-Wallis test. Statistical significance was taken as p=0.05. If a significant overall difference was present, multiple comparisons were performed using the Mann-Whitney test with Bonferroni correction. Data are presented as mean×SE. Photodynamic Therapy of Condyloma and Vulvar Intraepithelial Neoplasia III Seven patients referred to the Department of Obstetrics and Gynecology of the University of Zurich, Switzerland, for treatment of typical vulvar condylomata acuminata, and 5 patients with biopsy-proven vulvar intraepithelial neoplasia III (VIN III) were included in this study (table 1, 2). Written informed consent was obtained following approval by the Human Subject Review Committee of the University of Zurich, Switzerland. The diagnosis of condyloma was made by colposcopic inspection of the vulva after application of 4% acetic acid, the diagnosis of VIN III was made on histologic examination of a 4-mm diameter

Fehr et al.

254

Table 1. Patients treated for vulvar condyloma Age ImmunoLength of Times Time interval Fluence rate Optical dose J/cm2 years suppression disease months pretreated min mW/cm2 22 25 22 45 37 32 18

no no no no AIDS no no

2 2 12 64 132 13 2

1 0 0 4 6 8 0

140 115 160 150 152 114 212

100 100 100 54 64 65 20

125 125 125 105 77 77 25

Table 2. Patients treated for VIN III Age ImmunoLength of Times Time interval Fluence rate Optical dose J/cm2 years suppression disease months pretreated min mW/cm2 45 35 41 27 34

no HIV B3 no no AIDS

6 3 5 30 13

1 0 1 4 3

150 154 120 136 153

71 100 35 100 64

120 125 125 125 115

punch biopsy. All aceto-white lesions were mapped on a sketch and photographed. Pregnant or nursing patients were excluded. Before drug application the vulvar skin was washed with a watery 0.4% chlorhexidine solution to reduce the bacterial flora. 10 g of a 10% ALA gel (Applied Science Technology, Zug, Switzerland) was spread over the entire vulva and perianal skin. The vulva was covered with a nonadhering dressing (TegadermÔ, Johnson & Johnson Products, Arlington, Tex.) to prevent absorption of the formulation by clothing. Patients were asked to lie in bed until light application. Light application was performed using a dye laser (LaserscopeÔ Surgical Systems, 600 series, San Jose, Calif.) pumped by a KTP laser (LaserscopeÔ Surgical Systems, KTP/YAGTM, San Jose, Calif.). The dye laser was tuned to emit 630-nm light and was coupled to a 600lm quartz fiber ending in a front lens for homogenous illumination (Medlight SA, Lausanne, Switzerland). The patient was positioned on the operating table with her legs in stirrups so that the hips were flected and abduced. If necessary, the labia majora was taped with transparent tapes to allow illumination of the entire vulvar vestibulum. The fluence rate at the surface of the vulvar skin was determined using a power meter (Fieldmaster, CoherentÔ, Palo Alto, Calif.). Before, during, and following light illumination the patient was asked to define the experienced pain or discomfort using a visual analog scale ranging from zero (no


255

pain or discomfort) to ten (unbearable pain necessitating immediate action) every 1–2 min. If requested by the patient the vulvar skin was cooled by rinsing with water. Time intervals between drug application and illumination as well as the applied optical doses are shown in tables 1 and 2. Two, 7, 14 and 28 days after treatment patients were asked about the experienced discomfort and, skin reaction was assessed and documented by photographs. Patients with VIN III had a punch biopsy taken of the treated lesions 4 weeks after PDT even if no pathology was suspected by vulvoscopy.

Results Selective Photosensitization of Condyloma Of the 12 women who had the 20% ALA cream applied, 9 reported a mild burning or stinging at the drug application site after specific questioning for this side effect, whereas no discomfort was reported using the 2.5% ALA ointment. No patient felt it necessary to wash off the cream and all experienced a temporal decrease in intensity of discomfort. No subject enrolled in the 24-hour group showered or washed the cream off before the next morning. Autofluorescence was detected by in vivo imaging in only two patients at low fluorescence intensity and was located primarily in the folds surrounding the glans of the clitoris. At longer time intervals, drug-induced fluorescence of the non-hair-bearing vulvar as well as perianal skin was visible, whereas no fluorescence of the hair-bearing skin of the labia majora, mons pubis or perineum could be detected. Weak fluoresence of condyloma could be detected with high selectivity after a 1-hour time interval since fluorescence of surrounding skin including the vestibule was weak at that time point (fig. 1a). Three hours or more after drug application the contrast in fluorescence between condyloma and surrounding skin decreased for nonhair-bearing skin (fig. 1b), whereas condyloma of vulcar skin with hair remained clearly depicted. The decrease in the ratio (condyloma versus surrounding skin) from 1 to 6 h for the 2.5% ALA ointment and from 1 to 3 h for the 20% ALA cream in figure 2 was significant (p>0.021 and p>0.021, respectively). No significant difference between the two drug formulations at 1 or 3 h was evident after Bonferroni correction (p>0.04 and p>0.99, respectively). All typical condyloma and lesions which were mapped as acetowhite lesions showed fluorescence. Moreover, low-intensity fluoresence in the skin folds surrounding the glans of the clitoris could be detected in some patients even after a 1-hour time interval. Fluorescence microscopy revealed that after short time intervals peak fluorescence was localized in the basal epithelial layer (fig. 3a). Three and 6 h

Fehr et al.

256

a

b Fig. 1. In vivo fluorescence imaging with intensified CCD camera and 650-nm bandpass filter. a 1 h after 2.5% ALA ointment application, condyloma on the frenulum of the clitoris, left labium minus and hairless side of the right labium majus are fluorescing (marked with C). On the prepuce of the clitoris there is fluorescence of lower intensity which could not be clearly attributed to typical condyloma or aceto-white epithelium. b 3 h after drug application the whole vestibule and labia minora are fluorescing. Still typical condyloma of the fourchette and right labium minus are depicted (marked with C). Reproduced with permission from Fehr et al. [8].

after drug application the whole epithelium was fluorescing uniformly, whereas after 24 h fluorescence had shifted to the granular and horny layer (fig. 3b). The ratios between the mean fluorescence of the epithelium and the underlying papillary dermis for the 2.5% ALA ointment were 4.0×0.4, 4.0×0.5 and 4.9×0.5 at 1.5, 3 and 6 h, respectively, and for the 20% ALA cream the ratios were 4.6×0.3, 4.5×0.6 and 3.5×0.4 for 1.5, 3 and 24 h, respectively. After 90 min, only the 2.5% ALA ointment showed a further significant temporal increase in overall fluorescence of the condyloma epithelium (fig. 4). 20% ALA cream-induced fluorescence was significantly higher than that induced by 2.5% ALA ointment only at the 3-hour time point (p>0.014). For the 2.5% ALA ointment, the decrease in the ratio between the epithelial fluorescence of the condyloma versus the adjacent skin was significant when comparing 90 min to 6 h and 3 h to 6 h, but not comparing 90 min to 3 h. For the 20% ALA cream, only the ratios at 90 min were significantly different from those at 24 h. In addition, 90-min ratios using the 2.5% ALA ointment were significantly higher (p>0.014) than those using the 20% ALA cream. The maximal epithelial thickness of the condyloma was 0.54×0.04 and 0.28×0.01 mm for the adjacent normal skin.


257

Fig. 2. Ratio of condyloma fluorescence to fluorescence of surrounding normal skin in fluorescence imaging. Points represent means×SE, 4 patients per point. Reproduced with permission from Fehr et al. [8].

Photodynamic Therapy of Condyloma and Vulvar Intraepithelial Neoplasia III In all 12 patients the mean pain score immediately before light application was 0.7. During irradiation discomfort increased rapidly and in all patients some stinging or burning pain was noted. The mean pain score after 1 min of light application was 4.4, after 2 min 5.4, and increased further and peaked at 4 min with a mean pain score of 6.7 (fig. 5, 6). Then the pain score decreased slowly to 4.4 at 20 min although light application was continued. In 4 of 12 patients irradiation had to be interrupted on demand to allow cooling of the skin with cold water. When light application was discontinued the pain score fell rapidly to a mean of 3 at 2 min, 2.1 at 4 min, and 2 at 8 min after the end of irradiation. Although pain experienced during PDT was very individual and numbers were small, it is felt that pain increased more slowly if a lower fluence rate was used and that longer time intervals between drug and light application induced higher pain levels during PDT. All patients with condyloma showed response to therapy with regard to the number of condyloma or size of the lesions. Even the patient who received

Fehr et al.

258

a

b Fig. 3. Fluorescence micrograph showing fluorescence of condylomata acuminata following 20% ALA cream application. a 90 min after drug application the fluorescence is mainly in the basal epithelial layer. b 24 h after drug application fluorescence has shifted to the superficial epidermal layers.

Fig. 4. Fluorescence microscopy results. Mean fluorescence of the epithelium of the condyloma on left Y axis and ratio of epithelial condyloma fluorescence to adjacent skin on right Y axis. 4 patients per point, points represent means×SE. Reproduced with permission from Fehr et al. [8].


259

Fig. 5. Pain score on a visual analog scale ranging from 0 to 10 during and after photodynamic therapy of the 7 patients treated for vulvar condyloma. * Treatment interruptions due to pain.

Fig. 6. Pain score on a visual analog scale ranging from 0 to 10 during and after photodynamic therapy of the 5 patients treated for vulvar intraepithelial neoplasia III. * Treatment interruptions due to pain.

Fehr et al.

260

Table 3. Response of condyloma to PDT Age ImmunoLength of years suppression disease months

Times Time interval Optical dose Response pretreated min J/cm2

22 25 22 45 37 32 18

1 0 0 4 6 8 0

no no no no AIDS no no

2 2 12 64 132 13 2

140 115 160 150 152 114 212

125 125 125 105 77 77 25

CR PR 1/4 PR 1/4 CR CR PR 2/4 NC 6/10

CR>Complete clearing of all lesions; PR>clearing of ?50% of the lesions; NC>lesions became smaller, but =50% cleared completely.

Table 4. Response of VIN III to PDT Age ImmunoLength of years suppression disease months

Times Time interval Optical dose Response biopsy pretreated min J/cm2

45 35 41 27 34

1 0 1 4 3

no HIV B3 no no AIDS

6 3 5 30 13

150 154 120 136 153

120 125 125 125 115

CR CR 2/5 cleared 1/3 cleared CR

an optical dose of 25 J/cm2 showed marked reduction in the size of the condyloma by sloughing of the thickened epidermis. However, the condyloma showed rapid regrowth and new lesions outside the treatment field developed. Three other patients showed clearance of almost all lesions 2 weeks after treatment, but a few lesions survived PDT and showed regrowth at 4 weeks. Three of six patients treated with an optical dose above 70 J/cm2 showed complete clearance of lesions and remained recurrence free for 4 months’ follow-up (table 3). Three of five patients with VIN III showed histologically proven clearance of their precancerous lesions without any signs of scarring (table 4). The 2 patients with recurrent VIN III at 4 weeks had shown sloughing of the lesions 2 and 7 days after treatment with hardly any recognizable disease 2 weeks after PDT. Four weeks after therapy some lesions were again clearly detectable using acetic acid and histology showed persistent VIN II–III. In both patients the number of VIN lesions had been reduced.


261

Comment PDT could provide a simple and safe alternative to more radical surgical procedures for treatment of condylomata acuminata and VIN. Furthermore, if HPV-infected microscopic hyperproliferative intraepithelial cell nests accumulate the photosensitizer selectively, nonvisible hyperproliferations could be destroyed photochemically which might influence the recurrence rate of this disease. At all time intervals ALA-induced fluorescence was more than twofold higher in the epidermis than in the papillary dermis. This feature may limit photodynamic destruction to epithelial tissue and may minimize scarring. This selectivity of ALA-induced fluorescence to epithelial tissues is not limited to topical drug application, since the same selectivity could be observed in animal and human studies using systemic ALA administration [1, 9–11]. Our observation that epidermal fluorescence showed a temporal shift from the basal layer at short time intervals to the superficial layer at 24 h suggests that the metabolically active dividing cells convert ALA faster to PpIX than the more differentiated cells. After 24 h the photosensitizer was cleared from the basal layer. Hence PDT at this time interval may lead to incomplete epidermal destruction. Since this critical shift in photosensitizer localization can only be detected by fluorescence microscopy, the in vivo fluorescence imaging results at long time intervals are of questionable significance for PDT. Ninety minutes after drug application, the 2.5% ALA formulation resulted in higher epithelial selectivity of condyloma fluorescence than the 20% cream. Since the absolute fluorescence did not differ significantly between the two drug formulations, the higher selectivity is due to a lower fluorescence of the surrounding normal skin. Even at 6 h, where absolute fluorescence of the condyloma was maximal, fluorescence of normal skin reduced the ratio of condyloma epithelium to adjacent skin signal. In vivo fluorescence imaging showed that the macroscopically normalappearing epithelium of the labia minora, vestibule and hymen converted ALA readily to fluorescing porphyrins. Towards the hymen, the epidermis on the medial aspects of the labia minora becomes thinner. It is widely believed that the epidermis changes into an epithelium of mucous membrane since the epithelial covering is much less cornified than true skin and the vestibule and hymen are covered with nonkeratinized squamous epithelium [12]. ALA seems to penetrate the nonkeratinized stratified squamous epithelium and induces fluorescence. Therefore, the higher mean fluorescence of condyloma compared to adjacent epithelium at short time points might not be solely explained by impaired ALA penetration into normal skin or enhanced penetration into the condyloma due to disrupted epithelial architecture. It is more likely that

Fehr et al.

262

selectivity in fluorescence results from higher metabolic activity of the condyloma epithelium leading to faster conversion of ALA to PpIX. In this study, the selectivity of condyloma fluorescence was only assessed by comparing the fluorescence of the condyloma to that of the immediately surrounding skin and not by histologic evaluation of all fluorescent skin areas. Hence the specificity of fluorescence imaging for diagnosis of condyloma could not be established. However, the sensitivity of fluorescence detection of condyloma was high since all mapped aceto-white areas showed fluorescence. The clinical significance of the low-intensity fluorescence which was detected in some skin folds after a 1-hour time interval remains unanswered since no biopsies were taken from these areas for histologic classification of the tissue or localization of the photosensitizer. Fluorescence imaging cannot distinguish between ALA-induced porphyrins of bacterial origin on the epithelial surface and epithelial porphyrin production in the basal layer. Furthermore, the correlation between fluorescence signal and actual photodynamic destruction is not yet known. We conclude that vulvar condyloma acuminata can selectively accumulate PpIX photosensitizer using topical ALA formulations at short time intervals. Consequently, these results suggest that the feasibility of selective photodynamic destruction of vulvar condylomata is reasonable and should be examined. First results of an ongoing phase-II study using PDT following nonselective application of a 10% ALA gel show that photodynamic destruction of condyloma and VIN III is possible. Complete clearance from all condyloma was achieved in half of the patients who received an optical dose above 70 J/cm2. Similarly in 3 of 5 patients VIN III was histologically cleared by PDT. All patients showed some response to therapy but in the patients with a partial clearance of condyloma the lesions showed rapid regrowth. These results were obtained in some heavily pretreated patients who were seriously immunocompromised. Data are too scarce to analyze the parameters responsible for success or failure of this treatment modality, and follow-up is limited. The main advantage of PDT for these diseases is that no scarring of the vulvar skin occurred, and the healing time of the diseased skin was shorter compared to other treatment modalities like laser evaporation. One week after PDT all patients were free of discomfort. The main problem of PDT using ALA in this anatomic area is the pain associated with light application. Fortunately pain resolved rapidly following termination of irradiation. Future efforts, like shortening the time interval between drug and light application or fractioning light application will show if this problem can be solved. We conclude that PDT of condyloma and VIN III using ALA is feasible and deserves further clinical investigation.


263

References 1 2 3

4 5 6

7 8

9

10 11 12

Kennedy JC, Pottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 1992;14:275–292. Wolf P, Rieger E, Kerl H: Topical photodynamic therapy with endogenous porphyrins after application of 5-aminolevulinic acid. J Am Acad Dermatol 1993;28:17–21. Divaris DXG, Kennedy JC, Pottier RH: Phototoxic damage to sebaceous glands and hair follicles of mice following systemic administration of 5-aminolevulinic acid correlated with localized protoporphyrin IX fluorescence. Am J Pathol 1990;136:891–897. Centers for Disease Control: 1989 STD Treatment Guidelines. Morb Mortal Wkly Rep 1989;38(S): 18–21. Bedwell J, MacRobert AJ, Phillips D, Bown SG: Fluorescence distribution and photodynamic effect of ALA-induced PpIX in the DMH rat colonic tumour model. Br J Cancer 1992;65:818–824. Gomez Rueda N, di Paola GR, Meiss RP, Vighi S, Llamosas F: Association of human papillomavirus infection and vulvar intraepithelial neoplasia: A morphological and immunohistochemical study of thirty cases. Gynecol Oncol 1987;26:331. Okagaki T: Female genital tumors associated with human papillomavirus infection, and the concept of genital neoplasm-papilloma syndrome. Pathol Annu 1984;19:31. Fehr MK, Chapman CF, Krasieva T, Tromberg BJ, McCullough JL, Berns MW, Tadir Y: Selective photosensitizer distribution in vulvar condyloma acuminatum after topical application of 5-aminolevulinic acid. Am J Obstet Gynecol 1996;174;951–957. Loh CS, Bedwell J, MacRobert AJ, Krasner N, Phillips D, Bown SG: Photodynamic therapy of the normal rat stomach: A comparative study between di-sulphonated aluminum phthalocyanine and 5-aminolevulinic acid. Br J Cancer 1992;66:452–462. Loh CS, MacRobert AJ, Bedwell J, Regula J, Krasner N, Bown SG: Oral versus intravenous administration of 5-aminolevulinic acid for photodynamic therapy. Br J Cancer 1993;68:41–51. Grant WE, Hopper C, Mac Robert AJ, Speight PM, Bown SG: Photodynamic therapy of oral cancer: Photosensitization with systemic aminolevulinic acid. Lancet 1993;342:147–148. Kaufman RH: Anatomy of the vulva and vagina; in Kaufman RH, Friedrich EG, Faro S, Gardner HL (eds): Benign Diseases of the Vulva and Vagina. St Louis, Mosby Year Book, 1994.

Dr. med. Mathias K. Fehr, Department of Obstetrics and Gynecology, University Hospital Zu¨rich, Frauenklinikstrasse 10, CH–8091 Zu¨rich (Switzerland) Tel. +41 1 255 50 02, Fax +41 1 255 44 33, E-Mail [email protected]

Fehr et al.

264


............................

Photodynamic Therapy for Cervical Dysplasia Stephan Schmidt a, Stephan Spaniol b a b

Women’s Hospital University of Marburg, and Institute of Applied Physics, University of Bonn, Germany

Introduction The incidence of virus-induced lesions of the female genital tract has progressively risen during the last decades. Due to radical changes in the living habits human papillomavirus (HPV) infections have become more frequent [1, 2]. The typical findings are pathologic Papanicolaou smear results including the detection of coilocytes [3, 4]. With the occurrence of PAP III DK smear HPV (groups 16/18) are frequently positive in PCR investigations [5]. Photodynamic laser therapy (PDT) has been proposed as a selective treatment modality, and in addition has the potential of virus inactivation [6]. The aim of this pilot study was to evaluate the beneficial effect of PDT in cases of PAP smears indicating cervical dysplasia [7].

Methods Patients The subjects of this open study were 10 outpatients with PAP smears indicating mild dysplasia associated with HPV changes in all cases (PAP III DK). All patients were treated after informed consent. PDT of the cervix was performed as an outpatient procedure without the need of local or general anesthesia. Topical Application of the Photosensitizer The protocol consisted of instillation of a defined solution of aminolevulinic acid (ALA) of 10% buffered at pH of 5.5. After a time interval of 4 h careful rinsing was performed (fig. 1).

Fig. 1. Clinical protocol of ALA therapy. Aminolevulinic acid is applied topically 4 h before laser irradiation at 635 nm. Results are evaluated by means of cytology after 1 and 6 months.

Fig. 2. Schematic drawing of the light diffusor for PDT of the cervix uteri. A spherical silicon component for the surface of the cervix is combined with a cyclindric component for insertion into the cervical canal.

Irradiation Technique Thereafter the cervix was irradiated by a Dye-Laser (Coherent Lambda Plus, USA) providing 2 W red light at 635 nm wave length (50 J/cm2). We used an applicator system designed to provide homogeneous irradiation (PrototypeInstitute of Applied Physics, Bonn, Germany). Light Diffuser The light diffuser for the cervix consists of cylindrical and spherical components (fig. 2). To install the reflected holecylindric spherical form we utilized silicon (RTV 7601, Fa Wacker, Germany) with high-quality scattered medium (TiO2, RTV, Fa Wacker, Germany). The silicon mold has an inner diameter of 7 cm and a thickness of 5 mm. The irradiation focus was 3.25 cm resulting in an irradiation surface of 8.3 cm2 (fig. 3). The light diffuser prototype was tested during in vitro experiments in order to obtain data for dosimetry [8].

Schmidt/Spaniol

266

Fig. 3. Laser fiber with modification to allow intermittent lateral irradiation. Note that notches in the laser fiber provide defined laser power (150 m W/cm2).

Fig. 4. Linear correlation between degree of notchings of the laser fiber and the in vitro detection of laser power (% of total mW/cm2) lateral to the fiber. The cylindric component of the light diffuser was chosen with a small diameter of 3 mm in order to allow insertion into the cervical canal. Homogeneous irradiation was provided due to notching of the laser fiber (Ceram Optec Laser, Germany; fig. 4). Due to the optical properties of the human uterus it seems necessary to adapt the mode of laser irradiation to the anatomical properties [9].

Results Cytologic findings were reevaluated after defined periods of 1–6 months. A normalization of the PAP smear results was found in 9 of 10 cases.

Photodynamic Therapy for Cervical Dysplasia

267

Table 1. Evaluation of the benefit of PDT for dysplasia Patient No.

1 2 3 4 5 6 7 8 9 10

Cytologic classification before laser

after laser

III DK III K III DK III DK III DK III DK III DK III DK II K III DK

I II II I III DK II II II I II

Normalization of the PAP smear results as well as indication of virus devitalization in the majority of cases.

While before treatment coilocytes were a regular finding indicating virus infection, after PDT this cytologic pattern vanished in most cases. The results are demonstrated in table 1. The PCR findings were consistent. The case that did not respond had to undergo conization where mild dysplasia was the histological finding.

Discussion Avoiding conization in young women through ALA-induced PDT holds the promise of providing treatment without influencing the cervical function capacity during pregnancy [7]. As alternative treatment modalities, namely interferon application or medical virustatic therapy, have not yet provided adequate results in patients with HPV infection, the virustatic potential of PDT is a special interest. The effectiveness of PDT during in vitro and in vivo studies have been reported, and these findings could be of great importance when the oncogenic potential of HPV viruses (group 16–18) is taken in account [10, 11]. An additional potential lies in the photodiagnosis due to selective fluorescence of dysplastic cells and carcinoma in situ after ALA application. A pilot study with the Storz D-light system was started and showed promising results (Storz, Tuttlingen, Germany).

Schmidt/Spaniol

268

References 1

2 3 4 5 6

7 8 9

10 11

Go¨ppinger A, Birmelin G, Ikenberg H, Elmenthaler U, Hilgarth M, Hillemanns HG, Wied GL: Human papillomavirus standardization and DNA cytophotometry in cervical intraepithelial neoplasia. J Reprod Med 1987;32:609–613. Beral V: Cancer of the cervix: A sexually transmitted infection? Lancet 1974;i:1037–1043. Ku¨hn W: Diagnostik und Therapie HPV-assoziierter La¨sionen des weiblichen Genitaltraktes. Geburtshilfe Frauenheilkd 1993;53:820. Kurman RJ, Malkasian GD Jr, Sedlis A, Solomon D: From Papanicolaou to Bethesda: The rationale for a new cervical cytologic classification. Obstet Gynecol 1991;77:779–782. Parazzini F, La Vecchia C, Negri E, Fedele L, Franceschi S, Gallota L: Risk factors for cervical intraepithelial neoplasia. Cancer 1992;69:2276–2282. Schmidt S, Schultes B, Wagner U, Oehr P, Decleer W, Lubaschowski H, Biersack HJ, Krebs D: Photodynamic laser therapy of carcinoma: Effects of five different photosensitizers in the colonyforming assay. Arch Gynecol Obstet 1991;249:9–14. Schmidt S, Wagner U, Spaniol S, Krebs D: Photodynamic therapy for dysplasia of the cervix. J Gynecol Pathol 1996;1:6–7. Welch AJ, Yoon G, van Germet MJ: Practical models for light distribution in laser-irradiated tissue. Lasers Surg Med 1987;6:488–493. Madsen S, Svaansand L, Tromberg B, et al: Determination of the optical properties of the human uterus using frequency-domain photon migration and steady-state techniques. Phys Med Biol 1994; 39:1191–1202. ¨ rztebl Zur Hausen H: Papillomavirusinfektionen als Ursache des Geba¨rmutterhalskrebses. Dtsch A 1994;91:1448–1450. Miller AB, Rawis WE: Epidemiology of gynecologic cancer. I. Cervix; in Coppleson (ed): Medical Gynecologic Oncology. Livingstone, Edinburgh, 1981, pp 9–18.

Prof. Dr. S. Schmidt, Women’s Hospital University of Marburg, Pilgrimstein 3, D–35037 Marburg (Germany) Tel. +49 6421 28662 13, Fax +49 6421 28664 13, E-Mail [email protected]

Photodynamic Therapy for Cervical Dysplasia

269


............................

Application and Characteristics of Photodynamic Therapy for Cervical Cancer Tetsuya Muroya, Keiko Kawasaki, Takaya Kunugi, Tsukasa Akiya, Hiroshi Iwabuchi, Hotaka Sakunaga, Masaru Sakamoto, Tadashi Sugishita, Yoshio Tenjin Department of Gynecology, Sasaki Institute, Kyoundo Hospital, Tokyo, Japan

Introduction The advantages of photodynamic therapy (PDT) over other methods for the treatment of cervical cancer have been proved and recognized. To meet the demands of the increasing number of younger patients wanting to preserve fertility and to accommodate high-risk patients, elderly patients, and those who refuse surgery, PDT has been used as a breakthrough method. PDT can be performed without anesthesia since there is no pain. Further, there is no bleeding with PDT, fertility can be preserved while leaving the cervix relatively intact, and there are no negative effects on pregnancy and delivery.

Materials and Methods We have been employing PDT at our hospital to accommodate patient demands to preserve fertility, mainly among those patients with CIS and dysplasia [1–6]. Two types of low-pulse lasers are used, excimer dye laser (EDL; fig. 1) or YAG-OPO laser both of which have a considerably higher degree of tissue penetration, in conjunction with Porfimer sodium (PHE), a tumor affinity photosensitizer that displays localizing effects of tumor tissues. PDT is performed 48 h after intravenous injection of 1.5–2mg/kg PHE, the time when the PHE concentration on the malignant cells becomes thickest compared to normal cells. Use of a colposcope with an optical path for laser enables observing and checking lesions

Fig. 1. Excimer Dye Laser (EDL). The EDL is a low energy pulse laser, and its 630mm wavelength dye laser is generated when rhodmine 640 pigment solution is irradiated by 308-mm ultraviolet rays generated by an XeCl Excimer laser. The laser wavelength is 630–635 mm, width of pulse is 10×5 ns, and pulse radiation energy is 4–5 mJ/pulse maximum. Pulse repeated frequency is 40 Hz in normal cases (interchangeable to 40, 60 and 80 Hz).

directly while photoirradiation is being performed (fig. 2). Also, for irradiation of the cervical canal, the cervical probe can deliver 30% light in a forward direction in the cervical canal and 70% circumferentially (fig. 3). Formerly the operator needed to mark the position where the cervical probe had originally been inserted in order to withdraw it in 1-mm increments. Presently, a cervical probe manipulator is installed in the colposcope which can move 0.5 mm in half revolutions and 1 mm in one complete revolution (fig. 4). This enables the operator to withdraw the cervical probe in exact 1-mm increments.

Application and Characteristics of PDT for Cervical Cancer

271

Fig. 2. The Olympus Optical Co. and the Department of Gynecology, Kyoundo Hospital, jointly developed this colposcope equipped with a light path for introduction of laser irradiation. The colposcope includes an excimer dye laser pathway. At the focal point about 23 cm from the colposcope, the laser spot is 10 mm in diameter. Compared with cut fiber, this method provides a more uniform and accurate laser spot. 100 J/cm2 photoirradiation was administered to the ectocervix and cervical canal.

Results In our practice, among all patients who received PDT, 85% wanted to preserve fertility. Only 1 patient had complications. Seventeen patients refused to undergo conventional surgery and 2 were elderly. In 1 case a partial response (PR) became a complete response (CR) after a second PDT. At the present time, studies using the PDT method have been conducted on 131 patients: 95 CIS; 31 dysplasia, and recently 3 squamous cell carcinomas with microinvasion (fig. 5) and 1 CIS+endocervical adenocarcinoma, microinvasion and vulval dysplasia (fig. 6). Of 131 cases, 126 had a CR (96.18%). In addition, all microinvasive cases had CRs. In only 5 cases, remnants were found in a very limited area but most of the lesions had disappeared (table 1). The no change (NC) case would have been a PR if the present standard had been applied then. Lesions evenly spread on the vaginal wall were highly

Muroya/Kawasaki/Kunugi/Akiya/Iwabuchi/Sakunaga/Sakamoto/Sugishita/Tenjin

272

Fig. 3. Cervical probe. This probe was developed to administer PDT in the cervical canal, i.e. endocervix. A special sapphire chip or ceramic chip is mounted on the tip of the cut fiber.

Fig. 4. Cervical probe manipulator.


273

Fig. 5. Case 1 (SCC Ia). Upper row shows squamous cell carcinoma, microinvasion. Lower left picture shows the condition 11 weeks after PDT. Lower right picture indicates condition 5 months after PDT.

suspicious of invasive cancer. The patient became pregnant after receiving PDT, and delivered a normal child. Subsequently she received a semiradical hysterectomy because remnants were found in the deep cervical canal. Histopathological findings on the extracted uterus showed that the lesions originally found on the vaginal wall and in the cervical canal had almost disappeared and the only remnant found was in the deep internal cervical canal, and it was suspected that the first PDT did not reach it. We found remnants only in the cervical canal in 1 PR case who had CIS. The remnants disappeared giving a CR after a second PDT. Three and a half years have passed since the second PDT, and so far no recurrence has been reported. One severe dysplasia and two CIS cases had PRs. Some remnants within a very limited area were seen after PDT and are under observation. One of the CIS patients is scheduled for a second PDT. The first PDT was administered about 9 years ago. After PDT treatment 22 patients became pregnant, and of these patients, 12 had normal transvaginal deliveries and 4 had cesarean sections (3 are currently pregnant and 3 had abortions). In 3 cases even a second child was born. They developed no


274

Fig. 6. Case 2 (vulval dysplasia). Upper left picture before PDT. Upper right picture 1 day after PDT. Lower left picture 6 weeks after PDT. Lower right picture 6 months after PDT. The original lesions have completely disappeared.

Table 1. Response Number of cases Dysplasia CIS SCC Ia EC-Ad-ca Ia VIN Total

CR

PR

NC

31 95+1 3 1 1

30 (96.8%) 92+1 (96.88%) 3 (100%) 1 (100%) 1 (100%)

1 (3.2%) 2 (2.08%) 0 (0%) 0 (0%) 0 (0%)

0 (0%) 1 (1.04%) 0 (0%) 0 (0%) 0 (0%)

131+1

127+1 (96.97%)

3 (2.27%)

1 (0.75%)

VIN>Vulvar intraepithelial neoplasia.

adhesion or constriction of the cervical canal, and no difficulties were reported connected with childbirth after PDT. Also, even in a case with a stump remnant, for which total hysterectomy is usually required after cold knife conization, the patient had a CR after PDT and a normal pregnancy and delivery (patient had previously given birth by cesarean section).


275

No side effects or abnormal findings have been seen in blood and urine tests. However, it is necessary that the patients strictly avoid the sun for some time in order to prevent side effects from sunburn, but as long as this is adhered to, almost no problem is reported.

Discussion After PDT, all 4 microinvasive cancer cases and a few cases with suspected microinvasive cancer on cytological findings had CRs. Therefore, we presume that PDT is fully effective up to and including stage Ia1. One invasive cancer case which was found to have some remnants after PDT became pregnant and had a normal delivery. This would not be possible using other methods. After delivery she had a semiradical hysterectomy. This case shows the effectiveness of PDT in advanced stages. The patient was believed to be at the advanced stage IIa, but after PDT her condition was downgraded to stage Ib. One case of post-irradiation dysplasia, who had had a radical hysterectomy and irradiation, had a CR. Three other cases with stump remnants after cold knife conization (in such cases usually a total hysterectomy is needed) the patients had CRs after PDT. One of the patients became pregnant and had a normal delivery. These cases prove that PDT is effective and can be applied to treat various cases. Also, 1 of 2 cases in which colposcopic findings suggested type-III UCF (as a rule, type-III cases were excluded from PDT because irradiation would have been blind) had a CR. It was possible to expand the use of PDT to this case as well as to elderly patients and patients with complications which make surgery impossible. In 3 other cases who had PRs, it was determined that irradiation was insufficient. The number of younger patients who wish to preserve fertility has been increasing, and in order to cope with these demands, PDT has been receiving special attention due to its use as a breakthrough method. Clinical trials of lasers with variable laser wavelength (YAG-OPO, manufactured by IHI) have commenced, and a new type of smaller semiconductor laser is also in the preparatory stages for clinical trials. These latest developments should be of benefit and promise the unfolding of even more technological advancements in PDT. We expect that in the near future more clinics will adopt PDT due to ongoing progress and development of new types of photosensitizers, such as BPD-MA, a second-generation photosensitizer developed by QLT PhotoTherapeutics Inc., ATX-S 10, and NPe 6, all of which have a shorter retention


276

time and do not require hospitalization. First-generation compounds such as Photofrin, necessitate hospitalization of the patient for a period ranging from 4 weeks to 2 months and the strict avoidance of sun.

References 1

2 3

4 5

6

Muroya T, Sakunaga H, Sakamoto M, Sugishita T, Tenjin Y: Clinical tests for phase III study of PDT for the early cervical cancer and precervical cancer lesions by PHE (Porfimer sodium) and Eximer Dye Laser (PDT EDL-1). Oncol Chemother 1992;8:302–307. Muroya T, Sakunaga M, Sugishita T, Tenjin Y: Fertility preservation treatment for early cervical cancer and dysplasia by PDT (photodynamic therapy). Oncol Chemother 1993;9:21–32. Muroya T, Sakunaga H, Sakamoto M, Sugishita T, Tenjin Y: Photodynamic therapy (PDT) clinical trials. From the viewpoint of colposcopic, cytological, hysteroscopical changes. J Jpn Soc Laser Med 1994;15:41–52. Muroya T, Suehiro Y, Umayahara K, Akiya T, Iwabuchi H, Sakunaga H, Sakamoto M, Sugishita T, Tenjin Y: Photodynamic therapy for early cervical cancer. Cancer Chemother 1996;23:47–56. Muroya T, Suehiro Y, Umayahara K, Akiya T, Iwabuchi H, Sakunaga H, Sakamoto M, Sugishita T, Tenjin Y: Uterus preservation operation for CIN. PDT’s clinical trial results J Jpn Soc Obstet Gynecol Surg 1996;7:27–38. Muroya T: New strategies in cancer treatment – Photodynamic therapy. J Tokyo Med Coll 1997; 55:408. Dr. Tetsuya Muroya, Department of Gynecology, Sasaki Institute, Kyoundo Hospital, 1-8 Kanda, Surugadai, Chiyoda-ku, Tokyo 101-0062 (Japan) Tel. +81 3 3292 2051, Fax + 81 3 3292 3376, E-Mail [email protected]


277


............................

Photodynamic Therapy in Recurring Gynecologic Cancer Luigi Corti, Caterina Boso Department of Radiotherapy, University Hospital, Padova, Italy

Introduction Recurrent cancer at any site has a gloomy prognosis. Cancer of the cervix, corpus uteri, vulva, vagina or rectum that recurs after radical surgery or curative radiation therapy is a perplexing problem confronting both gynecologic and radiation oncologists. In many centers, oncologists are not enthusiastic about retreating these patients, seeing that retreatment is considered practically futile. The results of retreatments at different institutions vary in technique, area and foci [1]. Our experience in the radiotherapy of gynecological recurrences gives a more than 5-year survival rate of 33.3%, but with a high cost for a patient already plagued by a number of complications. Other authors show that curative radiotherapy in gynecological cancer has a 50% complication rate, with 12% being severe. Late radiotherapeutic negative side effects occur, especially in the rectum (40%) and bladder (20%); other sites are not significant. The critical dose given in these locations is 60 Gy [2]. Our clinical experience with photodynamic therapy (PDT) began in 1982 with a xenon lamp, and we started gynecological treatment in 1984. Previous to this we seached for the concentration of endogenous porphyrin needed in gynecological tumors [3]. PDT is a new technique utilizing the properties of a photosensitizer (hematoporphyrin, Hp) and Hp derivates (HpDs) to selectively localize the neoplastic tissue and, when activated by red light (630 nm), to produce free radicals lethal to the neoplastic cell. Another potential clinical application of this method is the diagnosis of malignant neoplasms: when illuminated by a blueviolet light (410 nm), the

photosensitizer, located in tumors, emits a fluorescent light in the red band and hence will also reveal those neoplastic areas which are not macroscopically visible [4]. This technique has been developed all over the world, especially in the USA, Japan and Australia where the most important experiences in this field have been reported [5, 6]. In some clinical studies PDT was used in gynecological neoplastic pathology, usually in patients with vulvar or vaginal vault relapses who had already been submitted to conventional treatment (surgery, radiotherapy) with poor results. In Japan this method is mainly used for diagnostic purposes, and it precedes PDT in order to localize neoplastic areas which are not visible [5]. The first observations on the possibility of detecting neoplastic areas by porphyrins date back to Policard [7] in 1924. He observed that a typical red fluorescence was emitted by different types of tumors when illuminated by a Wood lamp, and he assumed that the presence of high concentrations of endogenous porphyrins in the tumors was due to the presence of hemolytic bacteria. Afterwards, many studies were carried out in animals with different tumors, and injected with Hp, and such studies showed how this substance tends to accumulate in malignant tissue in larger amounts than in most normal tissue. Worthy of note is an extensive study by Gregorie et al. [8] utilizing HpD as a diagnostic tool: 226 patients, with suspected squamous carcinoma or adenocarcinoma were examinated for HpD fluorescence following intravenous administration of HpD. The positive correlation between fluorescence and histology was found to be equal to 77% for squamous carcinoma and 84% for adenocarcinoma [8]. Ryan [personal commun.] found that the extinction coefficient at 405 nm and the relative yield (compared with quinine sulfate) for the Hp fluorescence in serum are 7.4¶10 cm–1 mol–1 and 0.02, respectively. Theoretically, for very small concentrations 0.126 lc of the incident excitation light is absorbed (when c is the Hp concentration in lg/ml and l is the depth in cm). Using these values, the theoretically detectable mass of Hp in a ‘spot’ 3 mm in diameter and 70 mm thick should be approximately 0.1 pg. Materials and Methods According to these studies, in our hospital tissue endogenous porphyrins in gynecological tumors have been studied to show their possible clinical application in the detection and, consequently the prevention and staging of malignant neoplasms of the female genital system. Immediately following surgical removal of the organs, 92 applications of tissue

Photodynamic Therapy in Recurring Gynecologic Cancer

279

endogenous porphyrins were used on specimens from 40 patients divided in 2 groups. In group 1, 21 patients had primitive cervical cancer, based on a cytological or histological presurgical diagnosis. Of these, 2 patients were no longer considered because histological examination of the organ revealed endometrial cancer. Neoplastic relapses and tumors irradiated before surgery were excluded from this group. In group 2, 19 patients underwent a hysterectomy with or without adnexectomy for various reasons but, for the most part, for uterine fibromas. These patients formed the control group. In patients with cervical cancer, 2 samples were taken from tumoral tissue, and 1 or 2 samples from tissue believe to be free of neoplastic infiltration. In the control group, 1 or 2 samples were taken from the cervix in which colposcopic and cytological examination did not reveal the presence of preneoplastic or neoplastic lesions. Tissue endogenous porphyrins were quantified according to a modified procedure of Lemberg and Legge [9]. The data thus obtained were statistically elaborated by comparing the values of the porphyrins present in neoplastic tissue of patients affected by cervical carcinoma with those values of porphyrins present in peritumoral tissue (of the same woman) using the Wilcoxon test. The data relative to the peritumoral tissue porphyrins were subsequently compared to the data of the porphyrins present in the healthy cervical tissue of the control group; similarly, the values of tumor tissue porphyrins were compared with the values of the porphyrins present in the cervical tissue of the control group. These last two comparisons were carried out using the Mann-Whitney test. This study shows that there is no statistically significant difference between the concentration of porphyrins present in tumoral tissue and the concentration present in peritumoral tissue (p>0.42), and also that the difference in concentration between the peritumoral and healthy tissue of the control group is not significant (p>0.10). On the other hand, there is a significant difference between the concentration of porphyrins in neoplastic and healthy tissue (p>0.005). Confirmation of the above was obtained by evaluation of the percent distribution of the data as shown in table 1. On first sight, it can be noted that the neoplastic tissue does not present a large difference in endogenous porphyrin concentration, most likely due to the existence of a ‘neoplastic condition’ (cellular infiltration of lymphoplasmacellular type, alteration of tissue vascularization, vasal neoformation). This may be true in an advanced stage neoplasm, but the observation that the endogenous porphyrin concentration is significantly lower in non-neoplastic tissue may lead one to think that in the initial stages this tendency is respected and this could give credence to PDT as a future treatment in such cases. The second step was to introduce PDT in the treatment of recurrences after traditional oncological therapy. Twenty-six patients were treated for vaginal and/or pelvic recurrences. The primary sites were gynecological tumors in 24 cases and rectal tumors in 2 cases. The median age was 67.5 years. All patients underwent previous treatment for the primary tumors: surgery in 25 cases; radiotherapy in 23; brachytherapy in 17, and chemotherapy in 10. The recurrence site was the vaginal vault in the majority of cases (23/26). The histology was: 17 epidermoid cancers; 5 endometrial adenocarcinomas; 2 adenocarcinoma of the rectum; 1 carcinoma in situ, and 1 severe dysplasia. The histological grading was: 3 G1; 17 G2, and 4 G3 (table 1).

Corti/Boso

280

Table 1. Characteristics of patients Patients, n Median age (range), year 1st tumor site Gynecological tract Rectum Previous treatment Surgery External radiation therapy Brachytherapy Chemotherapy Recurrence site Vaginal vault Histology Epidermoid cancer Endometrial andenocarcinoma Adenocarcinoma of the rectum Carcinoma in situ Severe dysplasia Histological grading G1 G2 G3

26 67.5 years (55–82) 24 2 25 23 17 10 3 17 5 2 1 1 4 17 4

In many cases, the light source was an Argon dye laser (Meditec). The most recent model is able to deliver 3 different wavelengths: green (514 nm) with a maximun of efficiency of 3 W; blue green (488 nm) 9 W, and red (630 nm) 1.5 W. The dye laser has two aligned tubes that pass through a mother optical fiber which has an SMA connection for coupling with cylindrical flat-cut operative fibers with a terminal microlens. A new fiber with a bundle and microlens for irradiation of the vaginal vault was recently introduced in our practice (Ceramoptec, Germany). In some cases we utilized an applicator made of plastic material that has a concave wall covered with reflecting material in order to correctly irradiate the lateral vaginal wall. The fiber runs along the middle of the concavity. In cases of combined treatment with more wavelengths, a bundle with a regulating focalizing lens is used, in this way it is possible to choose the most adequate spot. The photosensitizing drug used was Hp and the injectable solution was made by Monico Farmaceutici, Venice (Italy) by mixing 1 part Hp with 550 parts per volume of NaOH 0.1 M; the solution was fixed at 7.2–7.4 by adding an adequate dose of NaCl. Finally the solution was diluted with 0.9% NaCl giving a final volume of 200 parts. The solution obtained was sterilized by filtration using Millipore and tested for sterility and apirogenity. The final product is normally 5 mg/Kg/body weight. In a few cases we also used HpD as photosensitizer.


281

The selection criteria of patients were: age; general conditions; previous therapy, and lesion dimensions. For staging we considered the macroscopic description of the lesion, the histological diagnosis of tumor recurrence in the gynecological sites, and the study of the locoregional involvement, with cystoscopy, rectoscopy and a pelvic scan; when the patient is admitted for therapy blood tests are taken. The curative or palliative aims of PDT are determined by locoregional or widespread diffusion of the cancer: palliation is for patients with a wide vaginal recurrence, pelvic recurrence that invades the vaginal vault or distant metastases; curative aims and for those patients with esophytic or endophytic lesions, with dimensions of =2 cm [10, 11]. The symptomatic group consists of patients with actual bleeding; only in 1 case was bleeding associated with pain. Injection of the photosensitizing drug is carried out after informed consent. Patients should be fasted before the injection, since we noted a moderate dyspepsia when the injection was performed after eating. The dose was 5 mg/kg body weight for HpD and 2 mg/kg for Photofrin II. The injection of the photosensitizer (Hp) lasts for about 1 h since sometimes a quick injection caused nausea and dyspepsia. The patients must avoid direct sunlight exposure for 30 days to prevent possible skin effects. Irradiation is usually performed 48 h after drug injection. According to the dimensions of the lesion and the therapeutic aim, it is possible to use a combination of more than one laser. For curative treatments an Argon dye laser (red light 630 nm) was used. In palliative cases, limited to the vagina, we used an Argon dye laser, and Argon (514 nm) if the bleeding was great, and an Argon dye and CO2 laser when bleeding was moderate. To allow light penetration through a bleeding field we used alternate treatments. The light doses ranged from 60 to 500 J/cm2. 18 patients received only one treatment, 5 had double treatments, 24 and 48 h after injection of the photosensitizer, and 3 patients were treated up to five times. The patients were prepared for treatment with a vaginal wash if bleeding was present, and then placed on a gynecological bed. Usually anesthesia was not necessary, in some patients mild sedation was helpful. We used a transparent plastic speculum to make it possible to irradiate every site of the vaginal vault and wall. The site that had to be treated was cleaned using gauze to avoid any reflection produced by physiological or pathological fluids. If the use of more than one laser was thought to be necessary, we first preferred to use the Argon laser, green light, with a photocoagulative aim for some seconds, followed by PDT with an Argon dye laser. We preferred the CO2 laser when hyperthermic effects were needed, since it can have wide defocalization at low doses (maximum 3.5 W/cm).

Results Forty-five days after treatment patients were evaluated using a gynecological examination and after 75–90 days using a vaginal smear. In the curative group a biopsy was necessary; in the palliative group a pelvic scan and/or a CT was performed. The results were evaluated as either objective or symptomatic. The survival rate and the progression-free survival were also evaluated.

Corti/Boso

282

The cytological and/or histological absence of the lesion at least 75 days after therapy was considered an objective complete response (CR); a reduction of ?50% of the lesion a partial response (PR), and all other cases were considered as no change (NC). The symptomatic response was evaluated in the palliative group. In this case a CR was complete absence of symptoms for at least 60 days. In the palliative cases there were 6 CR (66.66%) and 3 PR (33.33%). In the curative group there were 12 CR (70.58%), 1 PR (5.88%) and 4 NC (23.53%). According to the histology, CR was seen in 2 of 7 patients with adenocarcinoma (28%), 7 of 17 with epidermoid cancer (41.2%); 1 of 1 with carcinoma in situ (100%), and 1 case of severe dysplasia (100%). Grading was evaluated in 18 cases. CR was seen in 1 of 2 (50%) for G1, 6 of 12 (50%) for G2, and 1 of 4 (25%) for G3. Esophytic lesions had a 100% CR compared to endophytic lesions. In the curative group, patients with PR were treated later with colectomy and are still alive without evidence of disease. Of 12 CR 5 patients had a new recurrence: 3 were local recurrences (2 were treated with radiotherapy and 1 underwent surgery), and 2 were intra-abdominal recurrences and no curative treatment was possible. 2 patients with CR died at 72 and 86 months of other causes. The survival rate of patients treated only by PDT ranged from 3 to 92 (mean 50.7) months. In the palliative group, the survival rate ranged between 5 and 29 (mean 13.3) months. Acute and side effects were also evaluated. The treatment-related mortality was nil. No vaginal perforations or stenosis were recorded. The main drawback of PDT was skin photosenitization which forbade sunlight exposure for up to 4 weeks.

References 1 2

3

4 5 6

Nori D, Hilaris BS, Kim HS, Clark DG, Kim WS, Jones WB, Lewis JL: Interstitial irradiation in recurrent gynaecological cancer. Int J Radiat Oncol Biol Phys 1981;7:1513–1517. Fenton J, Decroix Y: Radiothe´rapie exclusive des e´pithe´liomes du col ute´rin de stade II distal et III: re´sultats et complications the´rapeutiques de 393 cas traite´s a` l’Institut Curie. Bull Cancer Paris 1979;66:542–548. Corti L, Maggino T, Maluta S, Calzavara F, Scavazza Fiore R, Marchetti M, Marchesoni D: Photodynamic therapy (PDT) in gynaecological neoplasms: Endogenous porphyrins; in Onnis A, Maggino T (eds): Proc Int Meet Gynaecological Oncology, Venice, 1985, pp 521–524. Kinsey JH, Cortese DA, Sanderson DR: Detection of hematoporphyrin fluorescence during fiberoptic broncoscopy to localize early bronchogenic carcinoma. Mayo Clin Proc 1978;53:594. Soma H, Akiya K, Nitahara S: Treatment of vaginal carcinoma with laser photoirradiation following administration of haematoporphyrin derivate. Ann Chir Gynecol 1982;71:133. Bruce GW, Forbes IJ, Cowled PA: The treatment of vaginal recurrence of gynecological malignancy with phototherapy following hematoporphyrin derivative pretreatment. Am J Obstet Gynecol 1982; 142:356.


283

7 8 9 10 11

Policard A: Etudes sur les aspects offerts par des tumeur experimentales examineé a` la lumie`re de wood. CR Soc Biol 1924;91:1423. Gregoire HB, Horger EO, Ward JL: Hematoporphyrin derivative fluorescence in malignant neoplasm. Ann Surg 1968;167:829. Lemberg R, Legge JW: Hematin Compounds and Bile Pigments. New York, Interscience, 1949, pp 85–87. Corti L, Tomio L, Maluta S: Recurring gynaecologic cancer treated with photodynamic therapy. Photochem Photobiol 1987;46:949–952. Corti L, Maluta S, Tomio L: Photodynamic therapy in gynaecological cancer. Laser Med Sci 1989; 4:155.

Luigi Corti, MD, Department of Radiotherapy, University Hospital Padova, Via Giustiniani 1, I–35128 Padova (Italy) Tel. +39 49 8212940, Fax +39 49 8212958, E-Mail [email protected]

Corti/Boso

284

Ovaries and Peritoneal Cavity Wyss P, Tadir Y, Tromberg BJ, Haller U (eds): Photomedicine in Gynecology and Reproduction. Basel, Karger, 2000, pp 285–295

............................

Intraperitoneal PDT for the Treatment of Recurrent Ovarian Cancer Stephen M. Hahn, William F. Sindelar, Thomas F. Delaney, Stephen C. Rubin, Douglas Fraker, R. Alex Hsi, Eli Glatstein Department of Radiation Oncology, University of Pennsylvania, Philadelphia, Pa., USA

Introduction Photodynamic therapy (PDT) is a promising new cancer treatment modality which requires the administration of a photosensitizing agent and light of a wavelength specific to the absorption characteristics of the photosensitizer. The appeal of PDT in oncology is that it has the potential for selective destruction of cancerous tissue compared to normal tissues. The basis for selectivity is threefold. First, there is evidence that some photosensitizers, including the first-generation photosensitizer, hematoporphyrin derivative (HpD), are retained in tumors to a greater extent than in some normal tissues [1, 2]. Secondly, selectivity can be obtained by restricting the application of light to the region of the malignancy, thereby avoiding some normal tissues and minimizing damage. Finally, the restricted depth of penetration of visible light in tissues limits the damage to deeper structures. The potential for minimal normal tissue toxicity and thus tumor selectivity has prompted increasing research in evaluating PDT as a cancer treatment. PDT has been a proposed treatment for a variety of malignancies and premalignant conditions. PDT in the treatment of malignancies has included head and neck cancers [3–5], lung cancer [6, 7], mesothelioma [8, 9], esophageal cancer [10], Barrett’s esophagus [11, 12], brain tumors [13], breast cancer [14, 15], and bladder cancer [16, 17]. There has been tremendous interest in PDT recently in the United States since the US Food and Drug Administration has approved this therapy for esophageal and lung cancer. However, it should be noted that PDT is not likely to be an approprite locoregional treatment

for all cancers. Nonetheless, PDT may play a role together with other modalities such as surgery and chemotherapy in the treatment of selected malignancies. The critical issue for clinical researchers is to define the situations where PDT has the greatest chance of having a positive impact upon the natural history of cancer. There are many important questions about PDT that remain unanswered. External illumination of light has a limited depth of penetration in tissues. While this may have a positive impact upon reducing damage to normal tissues, limited light penetration restricts the biological impact of PDT to superficial structures. The treatment of deep-seated tumors with PDT, therefore, requires either surgical exploration, tumor resection, and exposure for light administration or the interstitial delivery of light. Another unresolved, but related issue is the treatment of regional lymph nodes, which in some cancers, are important sites of microscopic disease that may required treatment. The accurate determination of light dose, photosensitizer concentration and oxygen concentration in tissue also requires further study. Several researchers have made significant strides in light and photosensitizer dosimetry [18–21]. Developing a rational approach to photosensitizer and light dosimetry and correlating these to the biological effect of PDT in vivo will be important to the future design of PDT treatments. Despite these questions surrounding the use of PDT clinically, some predictions can be made about potential clinical targets. The effective limit of light penetration in tissue is likely to lend this therapy to the treament of surface malignancies, including peritoneal carcinomatosis. This is especially true after surgical debulking where the residual tumor may be microscopic or =5 mm in depth. Theoretically, PDT might be ideal for malignancies, such as ovarian cancer which have the propensity to spread to peritoneal surfaces. For many disseminated intraperitoneal (IP) malignancies including ovarian cancer and gastrointestinal cancers, involvement of regional lymph nodes and micrometastatic disease are clinical concerns. Therefore, a locoregional treatment such as IP PDT is likely to be successful only as a part of a multimodality treatment regimen including surgery and/or chemotherapy. Ovarian cancer is a leading cause of cancer-related morbidity and mortality in women. It is estimated that there were 25,400 cases of ovarian cancer diagnosed in the US in 1998 and 14,500 deaths [22]. Many patients will present with advanced disease and for most of them a cure remains elusive. It is not our expectation that IP PDT will be successful as the sole modality of treatment for ovarian cancer. We believe that IP PDT should be integrated with surgery and combination chemotherapy. As a very early step in this process of defining the role of IP PDT, we have initiated a phase-II trial of PDT using the photosensitizer, Photofrin, and visible light for disseminated IP malignancies

Hahn/Sindelar/Delaney/Rubin/Fraker/Hsi/Glatstein

286

including patients with refractory ovarian cancer, gastrointestinal cancers, and sarcomas. The preclinical rationale for this therapy and our clinical experience to date with this therapy will be presented.

Preclinical Studies The scientific basis for the human clinical experience with IP PDT in the United States derives from a number of preclinical studies evaluating the efficacy and toxicity of IP PDT in animal models [23–30]. These studies demonstrated that IP PDT could be administered to animals with some degree of efficacy and reasonable safety. Douglass et al. [30] were the first to study intra-abdominal PDT in an animal model. Using HpD, 5 mg/kg, and 631 nm light, necrosis of tumors implanted in the serosa of the bowel, liver, pancreas, and bladder was reported. High total doses of light and fluence rates were used. Tochner et al. [23] evaluated PDT with HpD (50 mg/kg) and 514 nm light in a murine ovarian embryonal carcinoma ascitic model. The tumor was injected intraperitoneally and by 9 days after injection a tumor burden of 2–4 g was apparent. The untreated tumor was uniformly lethal 20–23 days after injection. Both HpD and light were delivered intraperitoneally and the light dose was 9.6 J delivered over 16 min. One group of 16 mice received only one treatment of light and HpD administered on day 9. A prolongation in survival was observed in this group and 1 mouse survived long term (?50 days) and was apparently cured. A second group of 16 mice received two treatments of HpD and light on days 9 and 15. Cures were seen in 6 of 16 mice [23]. It should be noted that in this tumor model, eradication of the tumor is difficult to achieve with the administration of IP chemotherapy. A 70% cure rate is observed with IP doxorubicin but only if the agent is administered 2 days after tumor inoculation, when the tumor burden is low. If doxorubicin is administered on the same day as the PDT was given (day 9), a cure rate of =20% is observed, presumably because of a higher tumor burden [24, 31]. Tochner et al. [24] extended these observations using a murine ascitic malignant teratoma model. The mice in this study were treated with a total of 4 PDT treatments using HpD and 514 nm light. The light was delivered to eight separate portions (octants) of the abdomen using a flat cut fiber. Each octant was treated for 2 min with a laser output of 10 mW for a total treatment time of 18 min. A 100% complete response rate and an 85% cure rate was observed [24]. The results demonstrate that IP PDT produces a cure of murine ascitic tumors. These data also suggest that multiple sequential treatments might be necessary in order to achieve a high percentage of cures. The two

Intraperitoneal PDT for the Treatment of Recurrent Ovarian Cancer

287

preclinical studies published by Tochner et al. [23, 24] provided the impetus for the development of the first phase-I clinical trial of IP PDT at the National Cancer Institute (NCI) [32–34]. Since the initiation of the phase-I trial at the NCI, other researchers have investigated the efficacy of IP PDT. Molpus et al. [29] studied the photosensitizer, benzoporphyrin derivative mono-acid and 690 nm light in a xenograft murine model of the human ovarian cancer, NIH:OVCAR-5. The light was administered intraperitoneally using a dose of 20 J. Several multi-dose regimens were studied and all led to a reduction in tumor burden at necropsy and a median survival benefit. Veenhuizen et al. [27] have studied IP PDT in rats bearing the CC531 colon carcinoma cell line. Porfirmer sodium and 628 nm light were delivered. IP PDT led to a significant growth delay for a single tumor implanted in the peritoneum and repeated treatments improved the treatment results. The clinical development of IP PDT has been aided by preclinical studies which have evaluaed the potential toxicities of this treatment. Toxicity to the bowel, especially bowel perforation, has always been a major concern. The potential mechanisms of bowel toxicity are transmural penetration of light and interference with blood flow in the bowel leading to ischemia. There has been a suggestion that IP PDT with HpD interferes with jejunal blood flow [35]. However, others have found no significant damage to major blood vessels after IP treatment [36]. Veenhuizen et al. [26] found that the intestines of Wag/ RijA rats were the most sensitive organs in a study evaluating IP PDT with either porfimer sodium or meso-tetrahydroxyphenylchlorin (mTHPC). A steeper toxicity-dose response curve was reported for mTHPC compared to porfimer sodium but a similar spectrum of toxicities was observed. One bowel perforation in the mTHPC group was reported. Mild reversible damage to the kidneys was seen on histological analysis without functional impairment. Elevations in liver transaminases were also reported in animals treated with porfimer sodium. Acute lethality caused by IP PDT was reported to be the result of toxic shock and rhabdomyolysis leading to circulatory failure. The normal tissue toxicity of PDT in the peritoneum has also been studied using a dog model [25]. HpD (1.2 mg/kg) was administered intravenously and intraperitoneally. 630 nm light was delivered to the entire peritoneal surface 48 hours after intravenous injection and 2 h after IP injection of HpD. The doses of light ranged from 0.57 to 0.74 J/cm2. Four dogs received 3 treatments. No significant side effects were observed in this study. A reversible decrease in lymphocyte counts and a modest elevation in liver function tests were observed. Mild peritonitis was seen in biopsy specimens of the treated peritoneum. Given the extensive preclinical evidence that bowel toxicity might be the dose-limiting toxicity of IP PDT, concerns were raised regarding the tolerance


288

of bowel anastomoses [37]. Since the initial thoughts were to integrate IP PDT with surgical debulking, it was likely that patients would require resection of small bowel as part of the surgical procedure. Small bowel anastomoses were created in New Zealand white rabbits followed by IP PDT. HpD was administered in doses of 1.5–2.5 mg/kg 24 h prior to surgery and light doses of 0–20 J/cm2 were evaluated. No adverse effects on the anastomoses were observed at these doses. Higher doses of HpD (10 mg/kg) administered with light doses of 20 J/cm2 resulted in a high rate of anastomotic breakdown.

Human Clinical Trials A phase-I study of surgery and PDT with laser light and porfimer sodium was conducted by the Surgery and Radiation Oncology Branches of the NCI for disseminated intraperitoneal malignancies [32, 33]. Seventy patients were enrolled in the study. Twenty-five of these patients had refractory ovarian cancer. Patients received porfimer sodium by intravenous injection prior to laparotomy. At surgery, an attempt was made to debulk tumor deposits to =5 mm in thickness because it was felt that the effective tissue penetration of 630 nm light was only about 5 mm. If possible, all gross disease was resected. Forty-six patients who were adequately debulked underwent light delivery to all peritoneal surfaces. After resection, the peritoneal cavity was lavaged with saline to eliminate residual blood. Light doses were recorded in real time by means of photodiodes that were sewn into the peritoneal cavity and connected to a computerized online dosimetry system. These flat diodes which measured only incident light were placed in the right upper quadrant, left upper quadrant, right and left peritoneal gutters, and pelvis. A mobile diode was also used. Illumination began with treatment of the mesentery followed by the small and large bowel. These areas were treated with external illumination from a flat cut optical fiber. Dilute intralipid (0.02–0.05%) was then placed within the abdomen to scatter the light and improve the homogeneity of light distribution to all areas of the peritoneum. Light to the peritoneal cavity was delivered with an optical fiber enclosed in a modified endotracheal tube. This lightdiffusing wand was moved over anatomic regions that were isolated to ensure uniform delivery of light. In this phase-I study, the PDT dose was sequentially escalated by increasing the sensitizer dose from 1.5 to 2.5 mg/kg, by shortening the interval between the injection and the surgery, and by increasing the light dose. Initially 630 nm red light alone was used but later a combination of 514 nm green light and 630 nm light was used. The reason for this change in wavelengths was that bowel toxicity was initially observed and because of the greater (and presum-


289

ably transmural) penetration by red light, 514 nm green light was substituted for illumination of the bowel and mesentery. Patients also received boost doses (10–15 J/cm2) with 630 nm of red light or 5–7.5 J/cm2 with 514 nm of green light to areas of gross disease on the diaphragms, gutters and/or pelvis. Treatment of the entire peritoneum caused intra-abdominal fluid sequestration and small bowel edema. Four patients developed intestinal fistulae. Three patients developed a bowel perforation which was the dose-limiting toxicity. One patient who suffered a colonic perforation, died after multiple procedures and multi-organ failure. All patients who developed a bowel perforation received either 630 nm light to the bowel or a dose of 514 nm light to the bowel of 3.8 J/cm2 or greater. Seven patients who received light dose of 10 J/cm2 to the diaphragms developed pleural effusions that caused respiratory compromise and required thoracentesis. Other major (but not dose-limiting) complications included postoperative hemorrhage, necrotizing pancreatitis, splenic rupture, and ureteral leak and urinoma. Sun sensitivity, thrombocytopenia, and asymptomatic liver function test abnormalities were also observed. The distinction between surgical and PDT-related complications was difficult to establish in this study. The patients had advanced, refractory disease and often required extensive resections. Some of the complications described above are not atypical of debulking surgery in this patient population. Based upon the toxicities observed in the phase-I study, the maximally tolerated dose of PDT was determined to be porfimer sodium 2.5 mg/kg administered intravenously 48 h prior to debulking surgery. The maximally tolerated light doses were determined to be 2.5 J/cm2 green light to the mesentery, small and large intestine, 5 J/cm2 red light to the stomach, 7.5 J/cm2 red light to the liver, spleen, omental bursa, and diaphragm, and 10 J/cm2 red light to the retroperitoneal gutters and pelvis. Boosts of 630 nm light up to a total dose of 15 J/cm2 to limited areas of gross disease in the pelvis, gutters, or diaphragms were also considered tolerable. The follow-up of these patients was recently updated [34]. Peritoneal cytology was obtained in 17 patients. Thirteen of seventeen patients with malignant peritoneal cytology were found to have negative follow-up cytologic analysis for an overall peritoneal cytologic response rate of 76%. This included 7 of 11 ovarian cancer patients. The median survival of the ovarian cancer patients was 28 (4–98+) months. More importantly, long-term disease-free survivors were reported in the subgroup of patients with ovarian cancer. Three of twenty-five ovarian cancer patients were disease-free more than 24 months after treatment. One patient died of lymphoma 28 months after treatment and 1 patient died of metastatic colon cancer 95 months after treatment, both free of ovarian cancer recurrence at the time of their deaths. It should be


290

emphasized that these patients had no treatment other than surgical debulking and a single exposure to PDT. Based upon the results of the phase-I clinical trial, we have initiated a phase-II clinical trial of intraperitoneal PDT for disseminated intraperitoneal malignancies. Patients are stratified according to cancer type (ovarian cancer, gastrointestinal malignancies, and sarcoma). This will include patients with ovarian cancer who have had only a partial response to standard combination chemotherapy or who develop recurrent disease after combination chemotherapy. Patients will be given doses of porfimer sodium and light at the maximally tolerated dose defined in the NCI trial. 532 nm light will be used instead of 514 nm light. The primary objective of this trial is to define the response rate of IP PDT in these 3 groups of patients. Secondary objectives will be to report the toxicities of this treatment in each patient population and to assess FDGPET scanning as a measure of PDT response. A pathologic restaging of all patients will be requested 6 months after treatment with PDT. Incident light measurements will be made intraoperatively with the same system used in the NCI trial [33]. This light system measures incident light only but will allow us to deliver similar doses of light to critical structures. We plan to compare the light doses measured using the NCI system with a newer light dosimetry system developed by Star [19] that measures total light dose, including both incident and scattered light [9, 38]. The isotropic probes from the newer dosimetry system will be placed in the abdominal cavity in proximity to the flat photodiodes. It is expected that this will allow us to measure the contribution of scattered light to the total fluence received in tissues. We also expect to evaluate the heterogeneity of scattered light dose in individual patients and regions within the peritoneum. In a parallel laboratory study, biopsies of cancer and normal tissue will be analyzed for porfimer sodium concentration. We expect to evaluate if differential retention of the photosensitizer is present in cancer tissue compared to normal tissues of interest, specifically, intestine, peritoneum, and skin. Comparisons between the differential tissue uptake of photosensitizer and clinical outcome will be made. At the time of the preparation of this article, 13 patients have been treated in the phase-II clinical trial of IP PDT. Response data are not available; however, limited toxicity data are emerging. In our expeience, IP PD is associated with significant toxicity which is reversible and tolerable. The toxicities are most pronounced in the patients who have the greatest tumor burden and who have undergone the most extensive debulking procedures. Postoperative fluid sequestration is a common problem observed in nearly all patients. Massive fluid resuscitation is mandatory and on occasion, pressor support is required. One patient died after suffering what was likely a perioper-


291

ative myocardial infarcation and a low cardiac output state. Pleural effusions have been noted on chest radiographs but these have not been clinically significant.

Future Directions The phase-I clinical trial at the NCI defined the spectrum of toxicities that are to be expected from IP PDT. Not surprisingly, bowel perforation was the dose-limiting toxicity. To our surprise, however, long-term survivors have been observed after this treatment. The number of patients who received the highest doses of light and porfimer sodium are small and therefore a full assessment of efficacy or dose response cannot be made. Nonetheless, we feel these results are sufficiently encouraging to continue with the phase-II clinical trial. The choice of porfimer sodium for the phase-I clinical trial was made because it was the only photosensitizer in clinical use in the United States at that time. We have chosen to perform the phase-II clinical trial with porfimer sodium because of the suggestion of efficacy from the phase-I trial. There is, however, significant interest for exploring the use of newer second-generation photosensitizers for IP PDT. Some of the second-generation photosensitizers have the advantage of being synthetic compounds and are activated by wavelengths of light that have the potential for greater tissue penetration. Greater tissue penetration may have great advantages for treating more bulky disease in the peritoneum or microscopic disease located in regional lymph nodes. However, this may also lead to greater bowel toxicity or clinically significant liver and renal toxicity. The evaluation of these second-generation photosensitizers should proceed, in our opinion, first with a preclinical assessment of toxicity in a large animal model. After establishing the safety of these agents, a phase-I clinical trial in humans should be performed. The preclinical studies of IP PDT suggest that multiple treatments are necessary to achieve a cure. These models did not include surgical debulking and therefore, it is possible that extensive surgical resection could obviate the need for repeated treatments. Nonetheless, repeated treatments may be necessary in some patients. There are significant clinical obstacles to repeated IP PDT treatments. At present, we require that the patients undergo a full laparotomy in order to adequately debulk the cancer and for full exposure of the peritoneum during light delivery. It is probably not likely that patients would tolerate (or agree to) repeated treatments delivered with a full laparotomy. Delivery of light under laparoscopic guidance would solve this problem. However, significant technical barriers to light delivery and optical


292

measurement exist. Additional preclinical work will be necessary to determine if this is feasible.

Conclusions Intraperitoneal PDT is an experimental adjunctive therapy to surgical debulking in patients with advanced recurrent ovarian cancer and other disseminated intraperitoneal malignancies. There are preclinical data which suggest that tumor responses can be induced with IP PDT. The early clinical data and the preclinical toxicity studies suggest that the bowel is the dose-limiting organ. Future studies from our group will focus on integrating light and photosensitizer dosimetry with clinical outcome. Evaluation of second-generation photosensitizers and repeated IP PDT treatments given through a laparoscope will be the focus of future preclinical studies. The patients who are being enrolled on the phase-II clinical trial have malignancies that cannot be cured with any known therapy. Conventional treatments such as whole abdomen radiation therapy, combination chemotherapy, and surgical debulking alone, while associated with a modest response rate will not lead to long-term disease-free survival. In this clinical context, it is reasonable to aggessively investigate IP PDT as a new cancer treatment modality. References 1 2

3

4 5

6 7

8

Gomer CJ, Dougherty TJ: Determination of [3H]- and [14C]-hematoporphyrin derivative distribution in malignant and normal tissue. Cancer Res 1979;39:146–151. Young SW, Woodburn KW, Wright M, Mody TD, Fan Q, Sessler JL, Dow WC, Miller RA: Lutetium texaphyrin (PCI-0123): A near-infared, water-soluble photosensitizer. Photochem Photobiol 1996; 63:892–897. Wenig BL, Kurtzman DM, Grossweiner LI, Maree MF, Harris DM, Lobraico RV, Prycz RA, Applebaum EL: Photodynamic therapy in the treatment of squamous cell carcinoma of the head and neck. Arch Otolaryngol Head Neck Surg 1990;116:1267–1270. Biel MA: Photodynamic therapy and the treatment of neoplastic diseases of the larynx. Laryngoscope 1994;104:399–403. Gluckman JL, Waner M, Shumrick K, Peerless S: Photodynamic therapy. A viable alternative to conventional therapy for early lesions of the upper aerodigestive tract? Arch Otolaryngol Head Neck Surg 1986;112:949–952. Cortese DA, Edell ES, Kinsey JH: Photodynamic therapy for early stage squamous cell carcinoma of the lung. Mayo Clin Proc 1997;72:688–690. Furuse K, Fukuoka M, Kato H, Horai T, Kubota K, Kodama N, Kusunoki Y, Takifuji N, Okunaka T, Konaka C, Wada H, Hayata Y: A prospective phase II study on photodynamic therapy with Photofrin II for centrally located early-stage lung cancer. J Clin Oncol 1993;11:1852–1857. Pass HI, DeLaney TF, Tochner Z, Smith PE, Temeck BK, Pogrebniak HW, Kranda KC, Russo A, Friauf WS, Cole JW, Mitchell JB, Thomas G: Intrapleural photodynamic therapy: Results of a phase I trial. Ann Surg Oncol 1994;1:28–37.


293

9

10

11 12

13 14

15 16 17 18 19 20

21

22 23

24 25

26

27

28 29

Baas P, Murrer L, Zoetmulder FAN, Stewart FA, Ris HB, Zandwijk N, Peterse JL, Rutgers EJT: Photodynamic therapy as adjuvant therapy in surgically treated pleural malignancies. Br J Cancer 1997;76:819–826. Lightdale CJ, Heier SK, Marcon NE, McCaughan JSJ, Gerdes H, Overholt BF, Sivak MVJ, Stiegmann GV, Nava HR: Photodynamic therapy with porfirmer sodium versus thermal ablation therapy with Nd:YAG laser for palliation of esophageal cancer: A multicenter randomized trial. Gastrointest Endosc 1995;42:507–512. Overholt BF, Panjehpour M: Photodynamic therapy for Barrett’s esophagus. Gastrointest Endosc Clin North Am 1997;7:207–220. Barr H, Shepherd NA, Dix A, Roberts DJ, Tan WC, Krasner N: Eradication of high-grade dsyplasia in columnar-lined (Barrett’s) oesophagus by photodynamic therapy wih endogenously generated protoporphyrin IX. Lancet 1996;348:561–562. Muller PJ, Wilson BC: Photodynamic therapy of malignant primary brain tumours: Clinical effects, post-operative ICP, and light penetration of the brain. Photochem Photobiol 1987;46:929–935. Sperduto PW, DeLaney TF, Thomas G, Smith P, Dachowski LJ, Russo A, Bonner R, Glatstein E: Photodynamic therapy for chest wall recurrence in breast cancer. Int J Radiat Oncol Biol Phys 1991;21:441–446. Schuh M, Nseyo UO, Potter WR: Photodynamic therapy for palliation of locally recurrent breast carcinoma. J Clin Oncol 1987;5:1776–1772. Benson RC: Treatment of diffuse transitional cell carcinoma in situ by whole bladder hematoporphyrin derivative photodynamic therapy. J Urol 1985;134:675–678. Harty JI, Amin M, Wieman TJ, Tseng MT, Ackerman D, Broghamer W: Complications of whole bladder dihematoporphyrin ether photodynamic therapy. J Urol 1989;141:1341–1346. Marijnissen JPA, Star WM: Quantitative light dosimetry in vitro and in vivo. Lasers Med Sci 1987; 2:235–242. Star WM: In vivo action spectra, absorption fluorescence excitation spectra of photosensitizers for photodynamic therapy. J Photochem Photobiol B 1995;28:101–102. Patterson MS, Wilson BC, Graff G: In vivo tests ofthe concept of photodynamic threshold dose in normal liver photosensitized by aluminum chlorosulphonated phthalocyanine. Photochem Photobiol 1990;51:343–349. Chen Q, Wilson BC, Shetty SD, Patterson MS, Cerny JC, Hetzel FW: Changes in in vivo optical properties and light distributions in normal canine prostate during photodynamic therapy. Radiat Res 1997;147:86–91. Landis SH, Murray T, Bolden S, Wingo PA: Cancer Statistics, 1998. CA 1998;48:6–29. Tochner Z, Mitchell JB, Harrington FS, Smith P, Russo DT, Russo A: Treatment of murine intraperitoneal ovarian ascitic tumor with hematoporphyrin derivative and laser light. Cancer Res 1985;45: 2983–2987. Tochner Z, Mitchell JB, Smith P, Harrington F, Glatstein E, Russo D, Russo A: Photodynamic therapy of ascites tumours within the peritoneal cavity. Br J Cancer 1986;53:733–736. Tochner Z, Mitchell JB, Hoekstra HJ, Smith P, DeLuca AM, Barnes M, Harrington F, Manyak M, Russo D, Russo A: Photodynamic therapy of the canine peritoneum: Normal tissue response to intraperitoneal and intravenous photofrin followed by 630 nm light Lasers Surg Med 1991;11: 158–164. Veenhuizen RB, Ruevekamp-Helmers MC, Helmerhorst TJM, Kenemans P, Mooi WJ, Marijnissen JPA, Stewart FA: Intraperitoneal photodynamic therapy in the rat: Comparison of toxicity profiles for photofrin and mTHPC. Int J Cancer 1994;59:830–836. Veenhuizen RB, Marijnissen JPA, Kenemans P, Ruevekamp-Helmers MC, Mannetje LWC, Helmerhorst TJM, Stewart FA: Intraperitoneal photodynamic therapy of the rat CC431 adenocarcinoma. Br J Cancer 1996;73:1387–1329. Goff BA, Blake J, Bamberg MP, Hasan T: Treatment of ovarian cancer with photodynamic therapy and immunoconjugates in a murine ovarian cancer model. Br J Cancer 1996;74:1194–1198. Molpus KL, Kato D, Hamblin MR, Lilge L, Bamberg M, Hasan T: Intraperitoneal photodynamic therapy of human epithelial ovarian carcinomatosis in a xenograft murine model. Cancer Res 1996; 56:1075–1082.


294

30

31

32

33

34

35

36 37

38

Douglass HO, Nava HR, Weishaupt KR, Boyle D, Sugerman MG, Halpern E, Dougherty TJ: Intra-abdominal applications of hematoporphyrin photoradiation therapy. Adv Exp Biol Med 1981; 160:15–21. Ozols RF, Locker GY, Doroshow JH, Grotzinger KR, Myers CE, Fisher RI, Young RC: Chemotherapy for murine ovarian cancer: A rationale for ip therapy with adriamycin. Cancer Treat Rep 1979; 63:269–273. Sindelar WF, DeLaney TF, Tochner Z, Thomas GF, Dachowski LJ, Smith PD, Friauf WS, Cole JW, Glatstein E: Technique of photodynamic therapy for disseminated intraperitoneal malignant neoplasms. Arch Surg 1991;126:318–324. DeLaney TF, Sindelar WF, Tochner Z, Smith PD, Friauf WS, Thoms G, Dachowski L, Cole JW, Steinberg SM, Glatstein E: Phase I study of debulking surgery and photodynamic therapy for disseminated intraperitoneal tumors. Int J Radiol Biol Phys 1993;25:445–457. Sindelar WF, Sullivan FJ, Abraham E, DeLaney TF, Smith PD, Friauf GF, Smith TJ, Okunieff PO: Intraperitoneal photodynamic therapy shows efficacy in phase I trial (abstract 1550). Proc Am Soc Clin Onc 1995;14:477. Selman SH, Kreimer-Birnbaum M, Goldblatt PJ, Anderson TS, Keck RW, Britton SL: Jejunal blood flow after exposure to light in rats injected with hematoporphyrin derivative. Cancer Res 1985;45:6425–6427. Suzuki S, Nakamura S, Sakaguchi S: Experimental study of intra-abdominal photodynamic therapy. Lasers Med Sci 1987;2:195–203. DeLaney TF, Sindelar WF, Thomas GF, DeLuca AM, Taubenberger JK: Tolerance of small bowel anastomoses in rabbits to photodynamic therapy with dihematoporphyrin ethers and 630 nm red light. Lasers Surg Med 1993;13:664–671. Van Staveren HJ, Marijnissen JPD, Aalders MCG, Star WM: Construction, quality control and calibration of spherical isotropic fibre-optic light diffusers. Lasers Med Sci 1995;10:137–147.

Stephen M. Hahn, MD, Department of Radiation Oncology, University of Pennsylvania, 3400 Spruce Street, 2 Donner, Philadelphia, PA 19104-4283 (USA) Tel. +1 215 662 7296, Fax +1 215 349 5445, E-Mail [email protected]


295


............................

PDT for Cytoreduction in Cases of Ovarian Cancer S. Schmidt a, U. Wagner b a b

Women’s Hospital University of Marburg, and Women’s Hospital University of Bonn, Germany

Introduction Ovarian cancer treatment is one of the obstacles of gynecologic therapy. In spite of extraordinary efforts to improve the results of screening, diagnosis, and treatment of ovarian cancer, the annual incidence of ovarian cancer in the United States has increased from 18,200 in 1983 to 22,000 in 1993 with increasing death rate [1, 2]. Because ovarian cancer is often asymtomatic in its earliest stages, over 70% of women present with widespread disease at the time of diagnosis (FIGO III and IV) [2]. In spite of the potential of chemotherapy with platin compounds, the most important single effective factor influencing the survival time is the degree of initial debulking of the tumor site during the curative attempt of primary radical surgery [1, 2]. Disseminated malignant disease within the peritoneal cavity is a difficult clinical problem that can cause a dissection problem in the area of the diaphragma, the liver surface, the mesenterium and the adjacent tissue of the great vessels [3]. The 5-year survival time is reported to range from 20 to 40%. Photodynamic therapy (PDT) has been suggested as an additional intraperitoneal treatment of ovarian cancer [4]. A goal of cancer treatment is the selective destruction of malignant cells with the preservation of normal tissues and function. The current study investigates the utilization of an innovative modification of PDT for disseminated ovarian cancer.

Fig. 1. Chemical structure of the photosensitizer. Phthalocyanin is coupled with an antibody of high selectivity for ovarian tumor cells (B43/13, Biomira).

We have developed a new technique of antibody targeted photodynamic laser therapy using phthalocyanine as a photosensitizer in order to increase the selectivity of PDT [5, 6].

Material and Methods Patient Selection A controlled trial analyzing the course in 72 patients was performed using a clinical protocol including preoperative staging by CT scan, endoscopy and tumor marker analysis. In 31 patients a preoperative application of antibody-bound phthalocyanin (2.0 mg/ B4313, Biomira, Canada) was performed by instillation into the peritoneal cavity (fig. 1) [7]. All patients were classified as FIGO III. As a control group 31 patients with same pathological staging were treated without PDT and evaluated by means of the survival pattern. Operative Procedure All patients underwent surgical exploration through a midline abdominal incision to determine the extent of disease [3]. The tumor volume was semiquantitatively defined at this

PDT for Cytoreduction in Cases of Ovarian Cancer

297

Fig. 2. Absorption spectrum of phthalocyanin with a peak at 675 nm for PDT of deep penetration and low absorption at the visible light range avoiding phototoxic side effects. >phthalocyanin. – – – –>Hematoporphyrin derivate.

stage of surgery to be compared with the volume after radical surgery. A T0 surgical result was the goal of the intervention including tumor extirpation, adnectomy, hysterectomy, omentectomy and lymph nodectomy. In a minority of cases an additional resection of the ilium (n>5) or colon (n>6) was performed in order to achieve adequate tumor reduction. Techniques of Peritoneal Irradiation After patients had undergone tumor debulking, the abdomen was explored with attention directed to complete hemostasis. Saline lavage of the abdominal cavity was performed to eliminate residual blood that could attenuate light at the time of phototherapy (fig. 2). Tunable coherent dye and titanium lasers were utilized at a wavelength of 675 nm at 2-watt output. The biological tissue was irradiated with an energy density of 5 J/cm2 (fig. 3). Light delivery was prescribed by calculating the exposure time to deliver the target energy density. Usually the laser light was directed by a hand piece into the region of micrometastasis during the treatment of different anatomic areas.

Results An improvement in survival time and a number of complications that are potentially light induced were the most striking findings. The clinicals results of the controlled trial are summarized in figure 4.

Schmidt/Wagner

298

Fig. 3. Clinical protocol for PDT in ovarian cancer using antibody-bound phthalocyanin. 2 mg of antibody-coupled sensitizer is instilled with 500 ml Ringer’s lactate solution 72 h before laser irradiation into the peritoneal cavity. Postirradiation evaluation is performed by electron microscopy (after 6 h) and patient monitoring after 1, 30 and 60 months.

Fig. 4. Kaplan evaluation of survival of ovarian cancer patients after PDT and the control group. After 30 weeks the percentage of living patients is significantly improved in the PDT group.


299

Fig. 5. Complications after phothodynamic therapy. A high percentage of patients (23%) showed subileus, whereas leakage at anastomosis sites was the most critical finding.

Due to the Kaplan evaluation the survival in the PDT group was 90% after 30 months and 40% after 60 months. In the control group a significant difference was stated after 30-month (78%) survival. After 60 months the survival in this group was 37%. The same was seen for the tumor-free survival time (56% in the PDT group and 40% in the control group at 30 months). The clinical cause of the PDT group was complicated by intestinal irritations: 23% showed subileus and 6.8% anastomosis leakage (fig. 5). While all incidences of subileus were managed conservatively leakages required relaparotomy. One women died from generalized sepsis.

Discussion While cytoreduction during primary surgery has been reported to be the most important single factor influencing the survival of patients with ovarian cancer stage FIGO III, radical dissection is limited due to a tumor adjacent to large veins and the diaphragma as well as the radix mesenteri [8]. In addition multiple micrometastases are difficult to identify on the large peritoneal surface. PDT has been shown to provide selective tumor destruction without major functional side effects and can be applied to the entire peritoneal cavity [8, 9]. The potential to reduce the tumor load implicates an improved effectivity of chemotherapy in these patients [9]. Intra-abdominal instillation of antibodies avoid the problem of accumulation of photosensitizer in the liver were phototoxic reactions due to laser irradiation could be of deleterious outcome [4, 10]. Elevated liver enzymes have been reported in this context.

Schmidt/Wagner

300

On the other hand the bowel is still endangered during radical surgery with operating room lamps. We calculated the irradiation to exceed 5 J/cm2 in the exposed areas of the bowel in spite of the blue filters of the lamps. Subileus and anastomosis leakage might be less dominant with NaCl lamps that do not activate phthalocyanin [11]. Furthermore micrometastasis might be destroyable by laser light irradiation with less deeply penetrating light [9], such as green light from an argon laser avoiding destruction of the muscularis. In this way side effects could be diminished [11]. In conclusion the pilot study implicates a potential clinical benefit that should be ratified by means of a randomized study.

References 1 2

3 4

5

6 7 8

9 10

11

Lele SB, Piver MS, Mang TS, Dougherty TJ, Tomczak MJ: Photodynamic therapy in gynecologic malignancies. Gynecol Oncol 1989;34:350–352. Baker TR, Piver MS, Hempling RE: Long-term survival by cytoreductive surgery to =1 cm, induction weekly cisplatin and monthly cisplatin, adriamycin, and cyclophosphamide in advanced ovarian adenocarcinoma. Cancer 1994;74:656–663. Ka¨ser O: Die Entwicklung radicaler Operationstechniken in der Gyna¨kologie am Beispiel des Ovarial- und Zervixkarzinoms. Geburtshilfe Frauenheilkd 1989;49:1025–1030. Sindelar WF, Delaney TF, Tochner Z, Thomas GF, Dachoswki LJ, Schmith PD, Friauf WS, Cole JW, Glatstein E: Technique of photodynamic therapy for disseminated intraperitoneal malignant neoplasms. Phase I study. Arch Surg 1991;126:318–324. Schmidt S, Decleer W, Lubaschowski H, Kindermann D, Krebs D: Photodynamic laser therapy targeted by antibody linked dyes against gynecologic tumors. J Cancer Res Clin Oncol 1990;7: 116–324. Schmidt S, Wagner U, Schnurbein UV, Ko¨hler S, Spaniol S, Krebs D: Photodynamic laser therapy: 110 applications in gynecologic oncology. Laser Technol 1995;5:123–126. Eckhauser M: Biodistribution of Fe (II) phthalocyanin tetrasulphonate. Lasers Med Sci 1990;5: 21–24. Delaney TF, Sindelar WF, Tochner Z, Schmith PD, Friauf WS, Thomas G, Dachowski L, Cole JW, Steinberg SM, Glatstein E: Phase I study of debulking surgery and photodynamic therapy for disseminated intraperitoneal tumors. Int J Radiat Oncol Biol Phys 1993;25:445–457. Corti L, Maluta S, Tomio L: Photodynamic therapy of gynecologic cancer. Lasers Med Sci 1989; 4:155–158. Schmidt S, Wagner U, Oehr P, Krebs D: Klinischer Einsatz der photodynamischen Therapie bei gyna¨kologischen Tumorpatienten: Antiko¨rpervermittelte photodynamische Lasertherapie als neues onkologisches Behandlungsverfahren. Zentralb Gyna¨kol 1992;114:45. Delaney TF, Sindelar WF, Thomas GF, Deluca AM, Taubenberger JK: Tolerance of small bowel anastomoses in rabbits to photodynamic therapy with dihematoporphyrin ethers and 630 nm red light. Laser Surg Med 1993;13:664–671.

Prof. Dr. S. Schmidt, Women’s Hospital University of Marburg, Pilgrimstein 3, D–35037 Marburg (Germany) Tel. +49 6421 28662 13, Fax +49 6421 28664 13, E-Mail [email protected]


301


............................

Potential Application of PDT for Treatment of Endometriosis and Ectopic Pregnancy: Animal Models R.L. Reid a, J.Z. Yang a, A. Krzemien a, M.F. Melchior a, J.P. Van Dijk a, P.M. Hahn a, D.A. Van Vugt a, b, P.A. Greer c, E. Dickson d, R.H. Pottier d Departments of a Obstetrics and Gynecology, b Physiology and c Pathology and Biochemistry, Queen’s University, and d Royal Military College, Kingston, Ont., Canada

Ectopic Pregnancy The number of ectopic pregnancies (EPs) is on the rise [1]. More sensitive diagnostic techniques have now changed the clinical challenge from management of life-threatening rupture of the fallopian tubes to the optimization of future reproduction by eliminating the EP from the tube in a way that minimizes iatrogenic tubal damage and allows rapid recovery. When the diagnosis of EP has been established unequivocally, the characteristics of the EP meet predefined criteria [2], and follow-up can be assured, systemic treatment with methotrexate (50 mg/m2) appears to be the treatment of choice. However, often the diagnosis cannot be confirmed without laparoscopy or patient compliance with follow-up after methotrexate can not be assured. In these instances laparoscopy is used to confirm the diagnosis and surgical excision is typically performed. For preservation of reproductive potential a means to selectively destroy a tubal EP without the need to surgically incise the tube would be advantageous. To test the potential of 5-aminolevulinic acid (ALA) photodynamic therapy (PDT) to destroy an early implanted pregnancy, we first conducted experiments on pregnancies in the rat uterine horn as a model for human tubal EP [3]. Pregnant Sprague-Dawley rats (n>36) were anesthetized on day 10 of gestation. ALA dissolved in 2 ml of saline was infused over 5 min via a jugular venous cannula at doses of 20 (n>6) or 200 (n>24) mg/kg. Three hours later

Table 1. Light exposure for 6 experimental groups Groups of rats (n>6)

Intravenous ALA

Uterine horn left (time)

right

1 2 3 4 5 6a

0 20 200 200 200 200

Y Y Y Y Y Y

N Y (30 min) N Y (15 min) Y (5 min) N

a

(30 min) (30 min) (30 min) (15 min) (5 min) (15 min)

Rebred 8–12 weeks later.

laparotomy was performed. The number of fetuses was counted in each uterine horn and light treatments were selectively administered according to the protocol in table 1. Light treatments for 0, 5, 15, and 30 min were achieved with a Kodak projector equipped with a 500 CBA halogen lamp and a Hoya R60 red filter to deliver light at a wavelength of 630 nm and a distance of 18 cm with a tissue surface power of 66.2 mW/cm2. Light dosages at 5, 15, and 30 min were 20, 60, and 120 J/cm2. One week later the number of surviving fetuses was determined. Six animals were rebred 8–12 weeks later to determine the integrity of the luminal epithelium after PDT of the ‘ectopic’ pregnancy. Complete regression of fetuses was typically observed in the uterine horns of ALAtreated rats after light exposure. Survival was 90.8×2.8% in saline-treated controls, 16.0×4.9% at 20 mg/kg ALA and 0% with 200 mg/kg. This photodynamic effect required between 5 and 15 min of light. Survival at 15 min of light exposure was 0.9×1.3% while at 5 min it was 8.9×6.2%. Following rebreeding 4 of 6 animals (66.2%) showed fetal development, albeit 28% fewer fetuses than before, in the uterine horns previously treated with PDT. These observations indicate that relatively brief exposure of an intact pregnancy to transmural light 3 h following systemic ALA could lead to pregression of pregnancy. The rebreeding experiment was reassuring since we [4, 5] and others [6] have shown the potential of PDT following intrauterine ALA to permanently ablate the rat endometrium. It was apparent that the mechanism for regression of early pregnancy induced by PDT might have involved transient disruption to the vascularity of the endometrium. Other investigations involving PDT of tumors have confirmed that blood flow decreased markedly shortly after photosensitized tissues were exposed to activating light [7, 8]. However, prior work has shown that animal pregnancies

Potential Application of PDT for Treatment of Endometriosis and EP

303

concentrate hematoporphyrin in placental tissues after intravenous administration [9]. If the PDT effect primarily involved the placenta, the EP could be destroyed with minimal impact on the fallopian tube mucosa. In a subsequent series of experiments [10], mouse embryos flushed from the oviducts of mated, pregnant mare’s serum gonadotropin-superovulated, female CD1 mice were incubated in 0, 0.1, 0.5, 1.0 and 5.0 mM ALA with and without subsequent exposure to light (source as above). Progression of embryo growth to the blastocyst stage was 100% in the absence of ALA and light, exceeded 80% even at the highest ALA concentration in the absence of light, exceeded 90% with light and no ALA, but was dramatically reduced to near 0% at ALA concentrations above 0.1 nM with both 5 and 15 min of light. These observations confirm the deleterious effect of PDT using ALA upon preimplantation mouse embryos and indicate that such embryos are capable of converting ALA to protoporphyrin IX (PpIX). Implanted embryos removed from the pregnant rat uterus show characteristic PpIX fluorescence concentrated in the placenta when exposed to black light (unpubl. observ.) suggesting that production of PpIX can be expected in implanted gestations as well. These findings suggest that brief exposure of an EP (tubal) to transmural light of appropriate power and spectral characteristics following systemic ALA treatment may selectively ablate the unwanted pregnancy at a light fluence that has minimal impact on the subsequent health of the tubal mucosa. We are presently working with a prototype nonlaser light source capable of delivering up to 1.5–2 W of power between 600–700 nm at the distal end of a 2-mm fiber as a light delivery device for these studies.

Endometriosis Endometriosis, the presence of endometrial glands and cytogenic stroma in extrauterine locations, is known to be associated with dysmenorrhea, chronic or cyclic pelvic pain, and infertility. Recent evidence has confirmed the appropriateness of resecting or destroying visible endometriotic implants, even small ones, detected at laparoscopy [11]. Standard methods for resecting endometriosis require specialized training and equipment, and carry significant risk of damage to adjacent pelvic organs, blood vessels, and nerves. In addition, resection is often incomplete since the heterogeneous appearances of endometriosis make recognition of early lesions difficult [12–14]. The development of a technique which would ablate all active endometriosis, even lesions that may be indistinct visually, with mere exposure of the pelvis to light would afford significant advantages over current methods.

Reid et al.

304

We first sought to determine whether experimental endometriotic lesions in the rat model would take up and covert 5-ALA to PpIX [14]. Endometriosis was surgically induced in 28 rats using full-thickness uterine explants sutured to the peritoneal walls, ovaries, and mesentery of bowel 12–16 weeks prior to the experiments. Topical (intralesional) injection of ALA (50–100 ll of 4%) solution was followed in 3 h by examination of peritoneal cavity fluorescence under a long wave ultraviolet lamp (black light). Lesions resulting from explants with outward-facing endometrium (but not those with inward-facing endometrium) showed pink fluorescence characteristic of PpIX when exposed to black light. In other experiments oral and intravenous ALA (200 mg/kg) were administered 4, 8 and 12 h prior to resection and fluorescent microscopic assessment of lesions. Both oral and intravenous ALA, but not control solutions, resulted in fluorescence of the lining (active endometrial tissue) within endometrial implants that peaked between 3 and 4 h after ALA and was nonexistent at 8 h. Spectrophotofluorometric measurements confirmed that PpIX was the source of this fluorescence. The relative PpIX fluorescence intensity of other pelvic organs was contrasted to that of endometriotic implants. No fluorescence was evident in skeletal muscle, peritoneum, mesentry of bowel, or eye (retina) after systemic ALA. In contrast, skin and bladder showed fluorescence that was dose-dependent and short-lived (not present at 8 h) after both intravenous and oral ALA [15]. Others have demonstrated the ability to destroy ‘endometriotic’ implants in rabbits using lasers directed at specific lesions following treatment with dihematoporphyrin ether (DHE) [16] and hematoporphyrin derivative (HpD) [17]. These approaches offer little advantage over standard laser ablation of endometriosis since individual lesions must be identified and targeted. In addition, both DHE and HpD when given systemically result in photosensitization that is nonspecific and long-lasting in contrast to ALA [18]. To evaluate the ability of ALA PDT to ablate endometriotic lesions we performed an additional set of experiments 9–11 weeks following surgical creation of the endometriosis model (endometrial explants) in 18 rats. All animals received ALA (400 mg/kg) via the jugular vein 3 h prior to laparotomy. Lesion characteristics were recorded and photographed before light treatment (nonlaser prototype, 5, 10 and 15 min at 600–700 nm, 2-mm fiber 2 cm from lesion) was directed to two of the four peritoneal implants in each animal employing a conical shield to limit light to each specific site. Initial evaluation revealed 53 assessable endometrial implants of which 33 appeared flat and 20 cystic. Histologic examination of control implants (treated with systemic ALA but no light) revealed cysts that were lined with endometrial epithelium and contained serous fluid. Walls of the cysts contained


305

a well-defined layer of circular myometrium and a thin layer of endometrial stroma which did not appear to contain endometrial glands except in two instances. Ten minutes of light exposure resulted in damage which was seen in all photosensitized implants when these were harvested 3–4 days following PDT (n>11). Flat-appearing lesions became necrotic and in some instances hemorrhagic. Cystic lesions showed destruction of the endometrial epithelium, loss of cellularity and necrosis of the circular myometrial layer usually more pronounced in the anterior wall of the cyst. In some instances PDT resulted in damage to the underlying skeletal muscle. Fifteen minutes of light treatment of photosensitized implants (n>5) resulted in similar damage except the necrosis of the underlying skeletal muscle was more pronounced and was seen in all treated implants. Examination of two implants exposed to 5 min of light treatment did not reveal any conclusive damage. No damage was evident on histological examination of implants receiving vehicle treatment only followed by exposure to 15 min of light (n>2). To observe the permanency of the photodynamic damage to endometriotic cysts, 2 animals (six implants) were exposed to high estrogen levels employing estrogen-containing silastic capsules following PDT. The lesions were examined and harvested 3 weeks following the original laparotomy and treatment (ALA+light n>3; ALA+no light n>3). At the time of tissue harvest two of the ALA- and light-exposed implants appeared cystic while one seemed to regress to form a flat-appearing lesion. All of the three control implants (ALA, no light) remained cystic in appearance. Histology of the control cysts showed the lumen to be lined with high columnar epithelium (consistent with estrogen stimulation) overlying a lush endometrial stroma with a few glands surrounded by a well-defined myometrial layer. The implants exposed to light showed definite damage, specifically, complete lack of an epithelial lining of the cyst (n>3), and a cyst wall consisting of only an extremely thin layer of myometrium (n>2). The flat-appearing implant was actually a collapsed cyst. These experiments confirm the ability of PDT employing systemic ALA to result in lasting destruction of endometrial implants in a rat model of endometriosis. Early human endometriosis may be more ammenable to PDT since lesions are often very small and superficial with little obstruction to light penetration. Lesions too small to be seen would be ablated and only active endometriosis would be destroyed avoiding the clinical challenge of having to determine if a scarred or ‘burnt out’ area of endometriosis needs to be surgically resected. Selectivity of PDT for the endometriotic tissue would avoid the potential for current therapies to ‘overshoot’, the mark and damage adjacent pelvic structures.

Reid et al.

306

References 1 2 3

4

5

6

7

8 9

10 11 12 13 14 15

16

17

18

Pansky M, Golan A, Bukovsky I, Caspi E: Nonsurgical management of tubal pregnancy. Am J Obstet Gynecol 1992;164:888–895. Stovall TG, Ling FW, Carson SA, Buster JE: Nonsurgical diagnosis and treatment of tubal pregnancy. Fertil Steril 1990;54:537–538. Yang JZ, Van Vugt DA, Melchior MF, Hahn PM, Reid RL: Photodynamic ablation of early pregnancy in the rat with 5-aminolevulinic acid: A potential new therapy for tubal ectopic pregnancy in the human. Fertil Steril 1994;62:1060–1065. Yang JZ, Van Vugt DA, Kennedy JC, Reid RL: Intrauterine 5-aminolevulinic acid induces selective fluorescence and photodynamic ablation of the rat endometrium. Photochem Photobiol 1993;57: 803–807. Yang JZ, Van Vugt DA, Kennedy JC, Reid RL: Evidence of lasting functional destruction of the rat endometrium after 5-aminolevulinic acid-induced photodynamic ablation: Prevention of implantation. Am J Obstet Gynecol 1993;168:995–1001. Wyss P, Tromberg BJ, Wyss MT, Krasieva T, Schell M, Berns MW, Tadir Y: Photodynamic destruction of endometrial tissue with topical 5-aminolevulinic acid in rats and rabbits. Am J Obstet Gynecol 1994;171:1176–1183. Henderson BW, Waldow SM, Mang TS, Potter WR, Malone PB, Dougherty TJ: Tumor destruction and kinetics of tumor cell death in two experimental mouse tumors following photodynamic therapy. Cancer Res 1985;45:572–576. Fingar VH, Siegal KA, Wieman TJ, Weber Doak K: The effects of thromboxane inhibitors on the microvascular and tumor response to photodynamic therapy. Photochem Photobiol 1993;58:393–399. Figge FHJ, Weiland GS, Manganiello LOJ: Cancer detection and therapy. Affinity of neoplastic, embryonic, and traumatized tissues for porphyrins and metalloporphyrins. Proc Soc Exp Biol Med 1948;68:640–641. Yang JZ, Greer PA, Van Vugt DA, Reid RL: Treatment with 5-aminolevulinic acid and photoactivating light causes destruction of preimplantation mouse embryos. Fertil Steril 1995;63:1088–1093. Marcoux S, Maheux R, Berube S: Laparoscopic surgery in infertile women with minimal or mild endometriosis. Canadian Collaborative Group on Endometriosis. N Engl J Med 1997;337:217–222. Redwine DB: The visual appearance of endometriosis and its impact on our concepts of disease; in: Current Concepts in Endometriosis. New York, Liss, 1990, pp 393–412. Murphy AA, Green WR, Bobbie D, dela Cruz ZC, Rock JA: Unsuspected endometriosis documented by scanning electron microscopy in visually normal peritoneum. Fertil Steril 1986;46:522–524. Strepling MC, Martin DC, Chatman DL, Vander Zwaag R, Poston WM: Subtle appearance of pelvic endometriosis. Fertil Steril 1988;49:427–431. Yang JZ, Van Dijk-Smith JP, Van Vugt DA, Kennedy JC, Reid RL: Fluorescence and photosensitization of experimental endometriosis in the rat after 5-aminolevulinic acid administration: A potential new approach to the diagnosis and treatment of endometriosis. Am J Obstet Gynecol 1996; 174:154–160. Manyak MJ, Russo A, Nelson LM, Thomas GF, Solomon D, Stillman RJ: Photodynamic therapy of rabbit endometrial transplants: A model for treatment of endometriosis. Fertil Steril 1989;52: 140–145. Petrucco OM, Sathananden M, Petrucco MF, Knowles S, McKenzie L, Forbes IJ, Cowled PA, Keye WE: Ablation of endometriotic implants in rabbits by hematoporphyrin derivative photoradiation therapy using the gold vapor laser. Lasers Surg Med 1990;10:344–348. Loh CS, MacRobert AJ, Bedwell J, Regula J, Krasner N, Bown SG: Oral versus intravenous administration of 5-aminolevulinic acid for photodynamic therapy. Br J Cancer 1993;68:41–51.

Robert L. Reid, MD, Department of Obstetrics and Gynaecology, Victory 4, Kingston General Hospital, Kingston, Ont. K7L 2V7 (Canada) Tel. +1 613 542 9473, Fax +1 613 533 6779, E-Mail [email protected]


307


............................

Photodynamic Diagnosis of Endometriosis in Patients Peter Hillemanns, Matthias Korell Department of Obstetrics and Gynecology, Klinikum Grosshadern, Ludwig Maximilian University, Munich, Germany

Endometriosis is the presence of endometrial glands and cytogenic stroma in extrauterine locations. It is a frequent clinical problem for women in the reproductive age and can influence markedly both reproductive prognosis and quality of life. Typically, this disorder causes dysmenorrhea, chronic or cyclic pelvic pain, and infertility, resulting in prolonged medical treatment and repeated hospitalization for surgery. Estimates of the prevalence of endometriosis range from 18 to 30% [1]. The main diagnostic technique is laparoscopy which should be performed carefully with a good manipulator by an operator who is skilled in diagnosing the disease. However, endometriosis may present in a large variety of forms and color manifestations such as nodular implants, black or red peritoneal lesions. As a result, these lesions can be missed easily and may lead to recurrent endometriosis with all its clinical consequences. Light-induced fluorescence is a relatively new technique with unique properties which make it attractive for the diagnosis and local treatment of surface disorders. It uses photoreactive drugs which concentrate preferentially in tumors and other hyperproliferating tissues. It involves a series of interactions within the tissue between molecules of a photosensitizer, oxygen, and photons of light. The topical application of 5aminolevulinic acid (ALA) has been used clinically for the endoscopic detection of neoplastic lesions of the bladder and early stage lung cancer by means of induced fluorescence imaging [2, 3]. Photodynamic therapy (PDT) using topical ALA has already been shown to be effective in the treatment of various neoplastic epithelial lesions such as basal cell carcinoma of the skin and premalignant lesions of the oral cavity [4, 5], but not in high-grade cervical intraepithelial neoplasia [6]. Vulvar dystrophy could be treated successfully with minimal side effects [7]. In experimentally induced endometriosis lesions,

Fig. 1. Spectral measurements 4 h after oral administration of 5-aminolevulinic acid (10 mg/kg body weight) in fimbrial mucosa, active endometriosis, nodular endometriosis, and peritoneum (from upper to lower curve).

the intensity of protoporphyrin IX fluorescence after intravenous and oral delivery of ALA was significantly higher than in adjacent normal peritoneum [8]. The aim of this study was to assess the diagnostic potential of protoporphyrin fluorescence after oral administration of ALA in patients referred for endometriosis (fig. 1). We evaluated the protoporphyrin IX fluorescence in women referred for diagnostic and/or operative laparoscopy because of suspected endometriosis. A dose of 10 mg/kg body weight ALA was administered orally. After a time interval of 120–280 min, fluorescence spectral analysis and video inspection was performed. Fluorescence was excited using a filtered short-arc xenon lamp at 380–440 nm and a power output of 200 mW (D-Light, Karl Storz GmbH, Tuttlingen, Germany). A modified video-laparoscope attached to a sensitive color endo-camera (Telecam SL PAL; Karl Storz GmbH, Tuttlingen, Germany) with additional features proved to be applicable to endoscopic fluorescence imaging of ALA-induced protoporphyrin IX in the abdominal cavity. During regular laparoscopy, switching of camera and light modes between blue light for fluorescence detection and normal white light mode was easy to perform and did not hinder the endoscopic procedure. Rejection of excita-

Photodynamic Diagnosis of Endometriosis in Patients

309

tion light was filtered by using a yellow glass filter on the eyepiece of the camera head to select the emitted wavelength range of interest. Furthermore, spectral measurements performed by imaging the area of interest through the endoscope optics via a beam splitter onto a quartz fiber connected to a spectrometer (intensified OMA, SI, Gilching, Germany) were user-friendly and gave reproducible results. Exposure times of less than 3 h after oral application of ALA resulted in very low protoporphyrin IX fluorescence which could only be detected in the mucosa of the tubal fimbriae. Increased intra-abdominal porphyrin fluorescence intensities were found between 4 and 5 h after administration. We observed the highest value of ALA-induced porphyrin fluorescence in the fimbrial mucosa. Also, the tubal mucosa presented with pronounced fluorescence which was visible through the tubal serosa and muscularis. Upon laparoscopic endoscopy, the uterus and ovaries were fluorescence negative. A moderate fluorescence intensity was noted at the edge of the liver. The cul de sac fluid showed a weak to moderate porphyrin fluorescence. Endometriotic lesions imposing as red peritoneal areas under white light were associated with a specific ALA-induced fluorescence of moderate intensity under blue light. However, the fluorescence distribution in these lesions appeared to be heterogeneous. On the contrary, no fluorescence was seen in black peritoneal lesions and nodular implants of endometriosis. Regular peritoneum was negative for blue light-induced fluorescence. The oral administration of ALA using a concentration of 10 mg/kg body weight was well tolerated and no systemic effects were recorded. None of our patients reported symptoms of cutaneous photosensitization. The hypothesis of photodynamic diagnosis being a diagnostic tool for endometriosis represents a very attractive approach. Furthermore, the potential of selective accumulation of a photosensitizing drug may be useful for the treatment of endometriosis when exposed to photoactivating light. Yang et al. [8] evaluated the ALA-induced fluorescence and photosensitization of experimental endometriosis in rats after systemic administration. Fluorescence intensity of ALA-induced protoporphyrin IX was significantly higher in endometriotic lesions than in adjacent normal peritoneum between 2 and 4 h after oral and intravenous delivery of ALA. In our study, high values of ALA-induced porphyrin fluorescence in the abdominal cavity were observed between 3 and 5 h after oral ALA administration. This is consistent with findings in the endometrium and endometriosis implants of various studies in animal models [8–10]. Active peritoneal endometriotic lesions showed a significantly higher level of fluorescence than normal peritoneum. Nodular-type and pigmented peritoneal endometriosis were fluorescence negative and could not be detected by this photodynamic approach. However, these endometriotic lesions are being detected more easily

Hillemanns/Korell

310

by conventional laparoscopy. Therefore, porphyrin fluorescence after oral administration of ALA may prove to be beneficial in the diagnosis of active peritoneal endometriosis. Surprisingly, the fimbrial mucosa showed a very intense photosensitization with oral ALA which may bear the potential of phototoxic damage. Young women of reproductive age in whom preservation of the tubal and fimbrial mucosa is crucial for fertility may have to be excluded from ALAinduced porphyrin fluorescence detection for endometriosis.

Acknowledgment This investigation was supported by the BIOMED project of the European Community (No. Pl 96-2260). P.H. was supported in part by a grant from the Wissenschaftliches Herausgeberkolloquium der Mu¨nchner Medizinischen Wochenschrift.

References 1 2

3

4

5 6

7 8

9

10

Goldmann DC, Cramer DW: The epidemiology of endometriosis. Prog Clin Biol Res 1990;323:15–31. Baumgartner R, Huber RM, Schulz H, Stepp H, Rick K, Gamarra F, Leberig A, Roth C: Inhalation of 5-aminolevulinic acid: A new technique for fluorescence detection of early stage lung cancer. J Photochem Photobiol B 1996;36:169–174. Kriegmair M, Baumgartner R, Knuchel R, Stepp H, Hoftstadter F, Hofstetter A: Detection of early bladder cancer by 5-aminolevulinic acid induced porphyrin flurescence. J Urol 1996;155: 105–109. Fan KFM, Hopper C, Speight PM, Buonaccorsi G, MacRobert AJ, Bown SG: Photodynamic therapy using 5-aminolevulinic acid for premalignant and malignant lesions of the oral cavity. Cancer 1996;78:1374–1383. Kennedy JC, Pottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol 1992;14:275–292. Hillemanns P, Korell M, Schmitt-Sody M, Baumgartner R, Beyer W, Untch M, Hepp H: Photodynamic therapy in women with cervical intraepithelial neoplasia using topically applied 5-aminolevulinic acid. Int J Cancer 1999;81:34–38. Hillemanns P, Untch M, Pro¨ve F, Baumgartner R, Hillemanns M, Korell M: Photodynamic therapy of vulvar lichen sclerosus with 5-aminolevulinic acid. Obstet Gynecol 1999;93:71–74. Yang JZ, Van Dijk-Smith JP, Van Vugt DA, Kennedy JC, Reid RL: Fluorescence and photosensitization of experimental endometriosis in the rat after systemic 5-aminolevulinic acid administration: A potential new approach to the diagnosis and treatment of endometriosis. Am J Obstet Gynecol 1996;174:154–160. Wyss P, Tromberg BJ, Wyss MT, Krasieva T, Schell M, Berns MW, Tadir Y: Photodynamic destruction of endometrial tissue with topical 5- aminolevulinic acid in rats and rabbits. Am J Obstet Gynecol 1994;171:1176–1183. Gannon MJ, Johnson N, Roberts DJ, Holroyd JA, Vernon DI, Brown SB, Lilford RJ: Photosensitization of the endometrium with topical 5-aminolevulinic acid. Am J Obstet Gynecol 1995;173: 1826–1828. Dr. Peter Hillemanns, Department of Obstetrics and Gynecology, Klinikum Grosshadern, Ludwig-Maximilians-Universita¨t, D–81377 Munich (Germany) Tel. +49 89 70950, Fax +49 89 7095 5844, E-Mail [email protected]

Photodynamic Diagnosis of Endometriosis in Patients

311

Breast Cancer Wyss P, Tadir Y, Tromberg BJ, Haller U (eds): Photomedicine in Gynecology and Reproduction. Basel, Karger, 2000, pp 312–315

............................

Introduction P. Wyss a, S. Schmidt b a

b

Department of Obstetrics and Gynecology, University Hospital of Zu¨rich, Switzerland, and Women’s Hospital University of Marburg, Germany

Breast cancer is the most common oncological disease in woman. It was diagnosed in more than 175,000 people in the USA in 1992 and was lethal for more than 44,000 patients [1]. Approximately 5–19% of breast cancer patients suffer chest wall recurrences following classical primary treatment such as lumpectomy, mastectomy, radiation therapy, chemotherapy, or hormone manipulation therapy [2–6]. Often, two or more methods are used in combination to manage the local and systemic disease. Breast cancer, which is metastatic to skin of the chest wall, is usually seen in advanced cases and these patients have normally failed other therapies [7, 8]. Breast cancer recurrences have a significant physical and psychological impact on well-being [9], with pain, odor and secretion constantly reminding the patients of the presence of a progressing disease [10]. Surgical removal and site-specific radiotherapy are common treatment procedures of chest wall metastasis. In cases of extensive spread of recurrences, especially in radiotherapeutically pretreated regions, resection often requires placement of a myocutaneous flap or mesh skin graft. Since 1979 photodynamic therapy (PDT) has provided an emerging alternative to major surgical procedures [11, 12]. The aim of this treatment technique is to achieve a palliative effect such as pain relief, reduction of tumor size, reduction of secretion from the tumor site and improvement of quality of life (table 1). Most clinical experience in photodynamic therapy of recurrent breast cancer is reported using hematoporphyrin derivative and photophrin as photosensitizers [8, 11–21], which are accumulated in tumor tissue and show a selective and prolonged retention. These studies revealed complete response in 7–20% of patients [11, 14, 18, 19, 22]. A recent study using

Table 1. Chronology of photodynamic therapy of recurrent breast cancer Author/reference

Photosensitizer

Mode (i.v./locally)

Drug dose mg/kg

Wavelength nm

Light dose J/cm2

Dougherty et al. [8], 1979

HpD

i.v.

?

?

?

Dougherty [13], 1981

HpD

i.v.

2.5–5.0

?

?

CR %

Patients n

35

Bandieramonte et al. [14], 1984

HpD

i.v.

3.0

630

60–120

2

Carruth et al. [15], 1985

HpD

i.v.

3.0

630

25

5

Waldow et al. [16], 1987

PF HpD PF

i.v. i.v. i.v.

1.5 3.0 2.0

630 630 630

25 8–11 20–60

1 1 3

Schuh et al. [11], 1987

PF

i.v.

1.0–2.0

630

36–288

McCaughan et al. [12], 1989

HpD PF

i.v. i.v.

3.0 2.0

630 630

20–30

12

Buchanan et al. [17], 1989

HpD PF

i.v. i.v.

3.0–4.0 1.5–2.0

630 630

25–200

7 8

Sperduto et al. [18], 1991

PF

i.v.

1.5

630

20–359

Ris et al. [26], 1992

mTHPC

i.v.

0.3

650

10–30

7

20

Lowdell et al. [27], 1993

PHP

i.v.

1.5–2.0

630

5–1,500 cm

Schmidt [28], 1993

AB-PhC

i.v.

1 mg

670

50

PF

i.v.

0.57–2.5

630

30–244

Cairnduff et al. [24], 1994

ALA (20%)

locally

–

630

150

Koren et al. [29], 1994

HpD (metal-free)

i.v.

2.0

632 white light

200

Baas et al. [20], 1996

PF (+mitomycin C)

i.v.

0.75 0.75+MMC

630

125–200 75 or 87.5

Lapes et al. [25], 1996

TPPS4

locally

0.15–0.3 mg

630

150

Taber et al. [21], 1998

PF

i.v.

1.8–2.0

630

25–100

9 0

1

13.5

37 5

9

22 4 9

91

Photofrin and a light dose of 100 J/cm2 reported complete response in 91% (10/11) of patients [21]. The only major clinical side effect of these systemically applied photosensitizers is skin phototoxicity which may last up to 1–2 months [23]. Patients have to avoid exposure to direct sunlight and bright indoor light following treatment. Less success was stated by local application of drugs such as aminolevulinic acid and TPPS4 [24, 25], probably due to insufficient penetration and less homogeneous distribution of the photosensitizers in the tumor. The following studies present the first clinical results using second-generation photosensitizers such as purpurin and chlorin for PDT of breast cancer recurrences.

Breast Cancer: Introduction

20 1

–3

Khan et al. [19], 1993

14

313

7

References 1 2

3

4

5

6 7 8 9 10 11 12 13 14

15

16 17

18

19 20

21

De Vita V, Hellmann S, Rosenberg S: Important Advances in Oncology 1990. Philadelphia, Lippincott, 1990. Overgaard M, Hansen PS, Overgaard J, Rose C, Andersson M, Bach F, Kjaer M, Gadeberg C, Mouridsen HT, Jensen MB, Zedeler K: Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial (see comments). N Engl J Med 1997;337:949–955. Casolo P, Mosca D, Amorotti C, Raspadori A, Drei B, Di Blasio P, Colli G, De Maria R, De Luca G, Ganz E, Amuso D: Our experience in the surgical treatment of early breast cancer. Results of a prospective study of 204 cases. Ann Ital Chir 1997;68:195–205. Elkhuizen PH, van de Vijver MJ, Hermans J, Zonderland HM, van de Velde CJ, Leer JW: Local recurrence after breast-conserving therapy for invasive breast cancer: High incidence in young patients and association with poor survival. Int J Radiat Oncol Biol Phys 1998;40:859–867. Gage I, Recht A, Gelman R, Nixon AJ, Silver B, Bornstein BA, Harris JR: Long-term outcome following breast-conserving surgery and radiation therapy. Int J Radiat Oncol Biol Phys. 1995;33: 245–251. Wyss P, Rageth JC, Kohler K, Unger C, Hochuli E: Prognosis of local-regional recurrence in breast carcinoma. Schweiz Rundsch Med Prax 1991;80:556–559. Berger H, Bu¨chler M, Reisfeld R, Schulz G: Cancer Therapy. New Developments in Surgical Oncology and Chemo- and Hormonal Therapy. Heidelberg, Springer, 1989. Dougherty T, Lawrence G, Kaufman JH, Boyle D, Weishaupt KR, Goldfarb A: Photoradiation in the treatment of recurrent breast carcinoma. J Natl Cancer Inst 1979;62:231–237. Stevenson JM, Bochenek P, Jamrozik K, Parsons RW, Byrne MJ: Breast cancer in Western Australia in 1989. V. Outcome at 5 years after dignosis. Aust NZ J Surg 1997;67:250–255. McDonald AE: Skin ulceration; in Groenwald SL, Frogge MH, Goodman M, Yarbo CH (eds): Cancer Symptom Management. Boston, Jones Bartlett, 1997, pp 364–376. Schuh M, Nseyo UO, Potter WR, Dao TL, Dougherty TJ: Photodynamic therapy for palliation of locally recurrent breast carcinoma. J Clin Oncol 1987;5:1766–1770. McCaughan JS Jr, Guy JT, Hicks W, Laufman L, Nims TA, Walker J: Photodynamic therapy for cutaneous and subcutaneous malignant neoplasms. Arch Surg 1989;124:211–216. Dougherty TJ: Photoradiation therapy for cutaneous and subcutaneous malignancies. J Invest Dermatol 1981;77:122–124. Bandieramonte G, Marchesini R, Melloni E, Andreoli C, di Pietro S, Spinelli P, Fava G, Zunino F, Emanuelli H: Laser phototherapy following HpD administration in superficial neoplastic lesions. Tumori 1984;70:327–334. Carruth JA, McKenzie AL: Pilot study of photodynamic therapy for the treatment of superficial tumours of the skin and head and neck; in Jori G, Perria C (eds): Photodynamic Therapy of Tumors and Other Diseases. Padova, Progetto, 1985, pp 281–287. Waldow SM, Lobraico RV, Kohler IK, Wallk S, Fritts HT: Photodynamic therapy of treatment of malignant cutaneous lesions. Lasers Surg Med 1987;7:451–456. Buchanan RB, Carruth JA, McKenzie AL, Williams SR: Photodynamic therapy in the treatment of malignant tumours of the skin and head and neck. Eur J Surg Oncol 1989;15:400– 406. Sperduto PW, DeLaney TF, Thomas G, Smith P, Dachowski LJ, Russo A, Bonner R, Glatstein E: Photodynamic therapy for chest wall recurrence in breast cancer. Int J Radiat Oncol Biol Phys 1991;21:441–446. Khan SA, Dougherty TJ, Mang TS: An evaluation of photodynamic therapy in the management of cutaneous metastases of breast cancer. Eur J Cancer 1993;29A:1686–1690. Baas P, van Geel IP, Oppelaar H, Meyer M, Beynen JH, van Zandwijk N, Stewart FA: Enhancement of photodynamic therapy by mitomycin C: A preclinical and clinical study. Br J Cancer 1996;73: 945–951. Taber SW, Fingar VH, Wieman TJ: Photodynamic therapy for palliation of chest wall recurrence in patients with breast cancer. J Surg Oncol 1998;68:209–214.

Wyss/Schmidt

314

22 23 24

25

26

27

28 29

Dougherty TJ: Hematoporphyrin derivative for detection and treatment of cancer. J Surg Oncol 1980;15:209–210. Dougherty TJ, Kaufman JE, Goldfarb A, Weishaupt KR, Boyle D, Mittleman A: Photoradiation therapy for the treatment of malignant tumors. Cancer Res 1978;38:2628–2635. Cairnduff F, Stringer MR, Hudson EJ, Ash DV, Brown SB: Superficial photodynamic therapy with topical 5-aminolaevulinic acid for superficial primary and secondary skin cancer. Br J Cancer 1994; 69:605–608. Lapes M, Petera J, Jirsa M: Photodynamic therapy of cutaneous metastases of breast cancer after local application of meso-tetra-(para-sulphophenyl)-porphin (TPPS4). J Photochem Photobiol B 1996;36:205–207. Ris HB, Altermatt HJ, Nachbur B, Stewart JCM, Wang Q, Lim CK, Bonnett R, Althaus U: Clinical evaluation of photodynamic therapy with mTHPC for chest malignancies; in Spinelli P, Dal Fante M, Marchesini R (eds): Photodynamic Therapy and Biomedical Lasers. Amsterdam, Elsevier Science Publishers, 1992, pp 421–425. Lowdell CP, Ash DV, Driver I, Brown SB: Interstitial photodynamic therapy: Clinical experience with diffusing fibres in the treatment of cutaneous and subcutaneous tumours. Br J Cancer 1993; 67:1398–1403. Schmidt S: Antibody-targeted photodynamic therapy. Hybridoma 1993;12:539–541. Koren H, Alth G, Schenk GM, Jindra RH: Photodynamic therapy – An alternative pathway in the treatment of recurrent breast cancer. Int J Radiat Oncol Biol Phys 1994;28:463–466.


Breast Cancer: Introduction

315


............................

Photodynamic Therapy in Breast Cancer Patients: Application of SnET2 for Skin Metastases Stephan Schmidt Women’s Hospital, University of Marburg, Germany

The first clinical results of a multicenter study aiming at the evaluation of purpurin in patients with breast cancer and local recurrence are presented in this article.

Methods The presented study, Clinical Trial Number PDTTP-920101-CA013, is entitled, ‘A phase-II multinational, open label study of single-dose tin ethyl etiopurpurin (SnET2) photodynamic therapy (PDT) in patients with advanced breast cancer for the management of cutaneous metastasis of breast carcinoma’. Photosensitizer SnET2 is a chemically pure purpurin with maximal absorbance at a wavelength of 664 nm (fig. 1). Although the terminal half life is several days long, pharmacokinetic analysis of the drug has shown that the value rapidly decreases and does not cause severe prolonged phototoxicity of the skin (fig. 2). Study Design The primary objectives of the open study are to determine the tumor response/remission rate of cutaneous metastases of breast carcinoma and to determine the systemic and local toxicity/morbidity/safety profile of SnET2 PDT. Secondary objectives are the evaluation of the patient’s disease burden during the 52 weeks following SnET2 PDT to determine the patient’s duration of response during the 52 weeks following SnET2 PDT, and to determine lesion treatment outcomes by patient assessments before SnET2 PDT and at 4, 12, 24, 32, 40 and 52 weeks after treatment relative to cosmetic outcome and quality of life improvement.

Fig. 1. Chemical structure of tin ethyl etiopurpurin (SnET2). This purpurin photosensitizer has a defined chemical structure. The maximum absorbance is a wavelength of 664 nm.

Fig. 2. Pharmacokinetics of SnET2. Purpurin has a rapid clearance with a short period of skin toxicity.

During the last year it was our pleasure to coordinate the multicenter study with purpurin (Upjohn Miravant). In Germany the study was performed in 9 study sites including 8 women’s hospitals and 1 department of radiology. Patient Selection 45 patients were included in this study. Inclusion Criteria. Preconditions were discrete lesions =6 cm, ?1 cm and/or lesion areas with a total surface area of p200 cm2. Furthermore a biopsy-confirmed cutaneous

Photodynamic Therapy in Breast Cancer Patients

317

Fig. 3. Skin area with cutaneous metastasis after PDT utilizing SnET2 as photosensitizer. Due to diode laser irradiation at 664 nm, selective swelling and erythema are observed at the tumor site.

metastasis from breast cancer was a precondition. The patient was to be aged q18 years, the Karnofsky status has to exceed q70%. Signed consent was essential. Exclusion Criteria. Local therapy including radiotherapy, local or systemic chemotherapy within the past 4 weeks was a reason to exclude the patient from the trial. Also hematopoietic, coagulation, hepatic, renal, cardiovascular, pulmonary dysfunction as well as sepsis or a history of hypersensitivity reactions to light and porphyria were contraindications. Treatment Modality A single-dose intravenous infusion was performed 24 h before laser irradiation. Here a 1.2-mg/kg dose was provided. Defined laser irradiation was performed with a diode-laser (PDT Systems, Santa Barbara, Calif., USA). A power density of 50–150 W/cm2 and energy close to 200 J/cm2 were chosen.

Schmidt

318

Fig. 4. Skin area 4 weeks after PDT with purpurin. Due to repair and healing, normalization of the treated area can be seen.

Identification of the location became possible after photo documentation and numeric registration of tumor sites.

Results Due to active participation of the study sites, an adequate number of patients were recruited during a short period of only 9 months. At this time the observation period of single patients is insufficient for a general evaluation. It can be stated though that no systemic toxicity or phototoxic skin reactions were reported. All patients showed an acute selective reaction after irradiation (fig. 3).


319

Table 1. Results of a pilot study of PDT using hematophorphyrin (Photosan) as a photosensitizer in patients with cutaneous metastases of breast cancer [4] Photosensitizer HPD Photosan

Laser Dye laser

CR 0

PR 18

NR 2

In the large majority an improved quality of life was achieved through partial remission (PR), while 2 of 20 patients had no improvement (NR).

After periods of 2, 4 and 8 weeks all patients showed remission and healing of the treated skin area (fig. 4). During the purpurin study in Germany a total of 45 patients were included over a time period of 10 months. The evaluated tolerability and safety profile were satisfactory. Most patients reported an improvement in quality of life. Because the evaluation is in process, showing modes of tissue reaction and repair, no final statement is possible.

Discussion Ideally photosensitizer agents must have an active absorbance wavelength in the near red range of the light spectrum. In this range of the ‘optical window’, which regular tissue does not absorb, the second-generation photosensitizers are of great interest as they overcome the problem of phototoxicity of the skin areas [1]. A number of pilot studies and clinical trials have indicated a potential benefit for the desperate group of patients with advanced breast cancer and local recurrence. Aiming at registration for routine clinical use, multicenter protocols have been evaluated for different photosensitizers [2–4] (table 1). While some agents, such as phthalocyanin, have less phototoxic side effects due to absorbance spectra, purpurin avoids the daylight sensitivity due to its pharmacokinetic pattern with early clearance [3]. As a sufficient number of patients have already been included in the German purpurin trial and no toxic side effects have been reported we should be able to evaluate the clinical benefit in the near future. Thereafter the technique could be performed as a routine procedure and an additional palliative tool in oncogynecology [4].

Schmidt

320

References 1 2 3

4

Tromberg BJ, Orenstein A, Kimel S: In vivo tumor oxygen tension measurements for the evaluation of the efficiency of photodynamic therapy. Photochem Photobiol 1990;52:375. Wolford SR: A two-week intravenous toxicity study of CL 318,952/Benzoporphyrin derivate monoacid ring A, a photodynamic therapeutic agent, in dogs. Pearl River Study No 93004. Selman SH, Fitkin DL, Keck RW, Morgan AR, Doiron DR: Treatment of the transplantable FANFT-induced bladder tumors with the purpurin SnET2 and red light emitted by a pulsed frequency doubled Nd:YAG laser. J Laser Appl 1996;3:44–47. Schmidt S: Photodynamische Laser-Therapie bei fortgeschrittenen Mammakarzinomen. Gebfra 1996;56:153–156.

Prof. Dr. S. Schmidt, Women’s Hospital, University of Marburg, Pilgrimstein, 3, D–35037 Marburg (Germany) Tel. +49 6421 28662 13, Fax +49 6421 28664 13, E-Mail [email protected]


321


............................

Photodynamic Therapy of Recurrent Breast Cancer P. Wyss, M.K. Fehr, R. Hornung, V. Schwarz, U. Haller Department of Obstetrics and Gynecology, University Hospital Zu¨rich, Switzerland

Quality of life and patient’s self-image are substantially impaired by the visible lesions of breast cancer recurrences [1]. Several studies sustained photodynamic technique as an alternative treatment modality for superficial metastasis as seen in table 1 of the introduction (p. 313). In this chapter, clinical experiences with a very potent second-generation photosensitizer for photodynamic palliation of breast cancer recurrences are documented. mTHPC has been suggested to be superior to HPD and Photofrin for the following reasons: (1) It is a single component of 98% purity and not a mixture of multiple active components; (2) it is much more phototoxic for a given light dose (5–10 vs. 20–359 J/cm2) [2]; (3) it requires less drug doses (0.1–0.15 vs. 1.0–5.0 mg/kg), and (4) duration of skin phototoxicity requiring strict avoidance of direct sunlight is much shorter (1–2 vs. 8–10 weeks). PDT of recurrent breast cancer using mTHPC as photosensitizer exhibited promising treatment results.

Patients and Methods PDT was performed as compassionate use in 4 patients aged 44–58 (mean 50) years. All patients had previously been treated for both primary breast cancer and recurrent disease. Three of 4 patients had undergone radiotherapy of the chest wall. Informed consent was signed by all patients, and approval was granted by the Ethics Committee of the Department of OB/GYN University Hospital of Zu¨rich. A total of 45 chest wall nodules were illuminated. As photosensitizer, meta-tetra(hydroxyphenyl)chlorine (mTHPC) (Scotia Pharmaceuticals. Guildford, UK) was intravenously applied. When a drug dose of 0.10 mg/kg body weight was used (P1–P3), illumination was performed 48 h after drug application at a light dose of 5 J/cm2. When a dose of 0.15 mg/kg body weight was used (P4), metastases were illuminated 96 h after drug application at a light dose of 10 J/cm2 (table 1). mTHPC was activated at a

a

b

c

d Fig. 1. Follow-up after photodynamic illumination of breast cancer recurrences. a Clinical aspect before PDT. b One week after PDT. c One month after PDT. d Two months after PDT.

Table 1. Dependence of light dose and time interval on mTHPC concentration mTHPC concentration

Interval drug/light application

Light dose

0.10 mg/kg 0.15 mg/kg

48 h 96 h

5 J/cm2 10 J/cm2

wavelength of 652 nm using a diode laser (Applied Optronics, USA). Skin metastases were illuminated by a front lense light diffuser (EPFL, Lausanne, Switzerland). The surrounding healthy skin was covered by a hole-containing plaster corresponding to the size of the illumination field. Following mTHPC injection, patients were advised to avoid direct sunlight and bright indoor light, and to protect their skin by clothes or sun blockers (Excipial pigment cream, Spirig AG, Egerkingen, Switzerland) during 1 week.

Photodynamic Therapy of Recurrent Breast Cancer

323

Fig. 2. Wound healing time following photodynamic illumination of skin metastases.

Performance and Clinical Outcome Analgesia was not necessary in all patients during photodynamic therapy. Depending on the chosen fluence rate, irradiation lasted between 4 and 6 min for each treated region. The illuminated field could contain one or more skin metastases. Therefore, the duration of a PDT session was about 40–60 min. At the time of laser illumination, recurrences exhibited slight dark coloring surrounded by minor ischemic reactions. Seven days following PDT (fig. 1b), the site of irradiated regions showed blistered metastatic lesions and erythema. Until 1 month (fig. 1c), superficial ulceration was converted to dark and dry necrosis, the erythema had almost disappeared. At 2 months (fig. 1d), the eschar was usually sloughed off and most of the photodynamically treated lesions were healed with minimal scar formation. Healing time ranged mainly between 8 and 10 weeks (fig. 2). One patient (P2) with large irradiated areas causing necrosis up to a diameter of 15 cm exhibited prolonged healing time. Additionally, this patient exhibited longlasting corticoid therapy for rheumatism. Treated skin areas up to a diameter of 2–3 cm healed within 2–3 months (P1, P3). The healing procedure and time of patient 4, who was illuminated 4 days after application of higher dosed mTHPC, did not differ from the other patients. Intensity and duration of pain following PDT varied between different patients (fig. 3). One patient (P1) felt only little skin irritation whereas another patient (P2) had extreme pain after PDT. This phenomenon could be explained by the extension of the illuminated field. The larger the size of an illuminated region, the longer the lesion needed to heal and the higher the pain intensity. The threshold of the illumination size to avoid these disadvantages is approximately 3¶3 cm. Therefore we intend to keep the area of illumination smaller than 3¶3 cm. It seems preferable to treat multiple small lesions instead of a few large fields. However, photodynamic therapy using mTHPC resulted in

Wyss/Fehr/Hornung/Schwarz/Haller

324

Fig. 3. Intensity and duration of pain following PDT.

complete response in all patients and offers a minimal-invasive, low-cost treatment modality for recurrent breast cancer with few side effects, high patient satisfaction and with possible repetitive application since tissue does not evolve resistance against photodynamic therapy.

References 1 2

Stevenson JM, Bochenek P, Jamrozik K, Parsons RW, Byrne MJ: Breast cancer in Western Australia in 1989. V. Outcome at years after diagnosis. Aust NZ J Surg 1997;67:250–255. Savary JF, Monnier P, Fontolliet C, Mizeret J, Wagnie`res GA, Braichotte DR, van den Bergh HE: Photodynamic therapy of early squamous cell carcinomas of the esophagus: A review of 31 cases. Endoscopy 1998;30:258–265.


Photodynamic Therapy of Recurrent Breast Cancer

325

Photon-Assisted Reproduction Wyss P, Tadir Y, Tromberg BJ, Haller U (eds): Photomedicine in Gynecology and Reproduction. Basel, Karger, 2000, pp 326–339

............................

Ten Years of Laser-Assisted Gametes and Embryo Manipulations Yona Tadir Beckman Laser Institute and Medical Clinic, University of California, Irvine, Calif., and Department of Obstetrics and Gynecology, University of California Irvine Medical Center, Orange, Calif., USA

Introduction During the summer of 1988, a team of scientists and clinicians gathered to evaluate the potential of laser microbeams in the fast moving area of assisted reproductive technologies (ARTs). Conventional micromanipulation was at its early stage, and the availability of a large variety of laser equipment delivered through microscopes to submicron spot sizes enabled testing its potential on gametes and embryos [1, 2]. The main applications tested during the past decade were: (1) sperm manipulations with optical tweezers to improve fertilization in vitro and to study basic sperm physiology, and (2) drilling of oocytes and embryos to improve fertilization, assist hatching, and asses zona pellucida (ZP) properties. The purpose of this article is to review the progress and evaluate the current status of laser microbeams vis-a`-vis technologies available in the ART laboratory.

Lasers and Delivery Systems Available for Micromanipulations Lasers (light amplification by stimulated emission of radiation) are electromagnetic waves with unique properties. The beam is collimated, monochromatic and coherent. Lasers differ from each other by the wavelengths (WLs) which are in the visible (red, green or blue), invisible ultraviolet (UV) or infrared (IR). Effects on gametes may also vary as a result of different parameters and application modes. Some heat may be generated in the micromanipulated object if exposure time is long enough. Conversely, heat formation may be minimized by short exposure in the order of micro- or nanoseconds.

Fig. 1. Absorption of light in water (a), DNA (b), and protein (bovine serum albumin, c) depicted on the scale of the electromagnetic spectrum in the ultraviolet (UV), visible, and infrared (IR) range. The region of wavelengths (WLs) in which noncontact laser manipulation is feasible (d). Types of lasers at different WLs tested for gamete and embryo manipulations are listed as referenced in the text (in italics). Absorption of the better known COH2 laser (at 10,600 nm) in water is mentioned as a reference.

Laser beams for gamete manipulation are typically reduced to a spot size of 1–5 lm [3]. In principle, lasers can be delivered to the target as a free beam or via flexible quartz fibers. This is dependent of the WL and the absorption by the nurturing liquid medium which is relatively low between 200 and 2,000 nm [4]. As such, WLs that are shorter or longer than this range require fiber delivery as will be discussed later. Light absorption in proteins and DNA is also WL-dependent, and this should be considered as an important factor when selecting the optimal laser gamete manipulations (fig. 1). The advantage of using light as an accurate cutting tool is the ability to eliminate the need for disposable or reusable tools. As such, the noncontact free beam delivered through the microscope objective is a preferred approach. Moreover, the availability of solid-state compact diode lasers inserted into the body of the microscopes makes this combination very practical (fig. 2).

Sperm Manipulations The ability to trap and immobilize cells with optical tweezers initially described by Ashkin et al. [5] has opened new possibilities for manipulating

Laser-Assisted Manipulations

327

Fig. 2. Inverted microscope designed for gamete and embryo manipulations. A compact diode laser (1,480 nm) is inserted into the body of the microscope. The laser beam reflects on mirrors to pass through the objective and the medium-containing dish. More than one beam can be delivered through the optical system to perform combined laser procedures such as cutting and trapping, or combined with conventional tools used in gamete manipulations.

sperm [6, 7]. The principles of cell trapping are based on mechanical force which is exerted on a microscopic particle by light. A single beam gradient force trap consists of a laser beam with a Gaussian intensity profile, focused to a spot smaller than the particle being trapped. This trap confines the particle to a location just below the focal point of the laser beam in the axial direction and centered in the beam in the transverse direction. The magnitude and direction of the net force on the particle is determined by the scattering of the laser light through the object. The force generated by the light is greater

Tadir

328

than all other forces acting on the particle and, as such, creates a trapping effect. In a set of studies, we demonstrated how single motile sperm can be trapped and subsequently released by reducing the trapping power [8–10]. Single sperm can be guided from one location to another without any mechanical tool. Although sperm can be optically trapped and guided through a hole in the ZP to fertilize an egg [11, 12], it is probably not the preferred approach to improve the fertilization potential in modern assisted reproduction since other technologies, such as the intra-cytoplasmic sperm injection (ICSI), has already been proven superior [13]. However, it is too early to predict if a combined approach of sperm trapping, tail cutting (with another laser beam) and ICSI will be valuable. Using these principles, a laser-generated optical trap was applied to manipulate sperm in two [6] and three dimensions [7]. Initially, the continuous wave neodymium:yttrium-aluminum-garnet (Nd:YAG) laser operating at 1,064 nm was used to determine relative force generated by single sperm [8]. The results demonstrated that zigzag motile sperm swam with more force when compared to straight-swimming sperm. Other experiments revealed that similar effects could be achieved with tunable continuos wave titanium sapphire laser (700– 800 nm WL) [9, 10]. Several studies were performed to explore relative and absolute sperm force under various physiologic conditions. This demonstrated a significant increase in swimming force following interaction with a cumulus mass [14] and a significant force increase following exposure to pentoxifylline, a motility-enhancing agent [15]. The relative force of human sperm before and after cryopreservation demonstrated that there was no significant difference when a yolk buffer-freezing media was used as cryoprotectant [10]. In another study, the relative escape force of human epididymal sperm (aspirated microsurgically for in vitro fertilization, IVF) was tested and compared to normal sperm. Data suggested that the relative swimming force of the epidydimal sperm was significantly lower (60%) than ejaculated sperm [9]. ATP-driven motility forces were calculated from calibrated trapping forces generated during the interaction of an 800-nm laser beam with single sperm cells [16]. Sperm heads were obtained microsurgically removing the flagellum with a pulsed laser beam (‘laser scissors’). A trapping efficiency of 0.12×0.02 and a mean intrinsic motility force of 44×20 pN were determined for motile spermatozoa from healthy donors. In a recent study, we demonstrated that UV exposure to the optical trap =400 nm) causes cell paralysis within 35 s and cell death within 65 (XX) it was concluded that near IR light should be used for further physiological studies.


329

Oocyte Manipulations Laser Zona Drilling Noncontact Mode Laser-zona interaction and the fate of oocytes exposed to light beams delivered through the microscope objective have been studied by several groups. Initially, a tunable dye laser at various WLs (266–532 nm) was tested on mouse, hamster and discarded human oocytes [2, 3]. The beam was delivered through the microscope objective, and the depth of incision was observed on a television monitor and adjusted by a joystick-activated motorized stage. This method is simple and accurate when compared to conventional micromanipulations. Subsequently, other groups used a krypton fluoride laser (operating at 248 mm) [17], nitrogen laser (337 nm) [11, 18], and a nitrogen-pumped dye laser (440 nm) [19] to test simplicity, accuracy and local effects. The investigators concluded that from the technical point of view these lasers can be used for zona incision. In order to further elucidate the laser the effect on the ZP, we designed a set of experiments with various laser parameters. Oocytes were exposed to two different XeCL excimer laser systems (both operating at 308 nm) that offer a large variety of parameters such as pulse duration or pulse repetition rate. High-quality images were video recorded and analyzed by computerized image processing, and the oocytes were further processed from scanning electron microscopy [20, 21]. Ablation holes smaller than 1 lm were obtained in a reproducible fashion without causing any apparent damage to neighboring cells. Pulse energy and the beam focal plane position were shown to be the most critical parameters in defining the ablated spot diameters. It was concluded that excimer lasers of 308 nm operating in a short pulse duration (15–250 ns) are effective microsurgical tools for achieving ‘clean’ ZP removal in a noncontact mode. At this particular WL, the optical absorption is strong enough to cause selective interactions with the ZP, yet weak enough to induce generate heat or explosive ablation. In addition, the 308-nm radiation can be delivered through glass slides, microscope objectives, and liquid medium or oil. It can facilitate easy, accurate and highly reproducible material removal without the need for handling and maintaining a contact delivery system. However, the known potential damage of UV irradiation still raise some concerns and the sensitivity of gamete’s genetic material deserve extra caution. Ng et al. [22] studied the potential use of the nitrogen laser (337 nm) delivered through an inverted microscope to provide a spot of =1 lm in the noncontact mode. The laser was used at 2.5 lJ/pulse with a repetition rate of 10 pulses/sec. A 10-lm opening was made in each ZP of mouse oocytes. The drilled oocytes were then inseminated in micro-droplets with murine sperm

Tadir

330

at 2¶105 sperm/ml. There was a significant improvement in fertilization and blastocyst formation at day 5 following laser zona drilling (LZD): 89 of 158 (65.2%) compared to 46 of 127 (36.2%; p=0.001). Lasers used in these studies were in the UV or the visible range. A new system that can selectively disrupt the ZP and be delivered as a ‘free beam’ is the compact diode laser operating in the near IR range (1,480 nm). This option will be discussed in the laser-assisted hatching section where it is more clinically relevant. Contact Mode A different approach to zona drilling that uses a glass pipette or laser fibers in a contact mode has been suggested by several investigators. In these studies, the argon fluoride excimer laser (ArFl) at 193 nm [23], Nd:YAG laser (1,640 nm) [24], holmium:YAG laser (2,100 nm) [25] and erbium:YAG laser (2,940 nm) [26, 27] were applied to oocytes. The 193-nm short WL was delivered to mouse oocyte ZP [23] through a series of mirrors and a long focal length lens connected to an alumina silicate pipette. The glass pipette was pulled from capillaries with a 1-mm outer diameter to a tip of about 3–5 lm and filled with positive air pressure. Insemination at low sperm densities led to fertilization and further development to the blastocyst stage [28]. Successful fertilization and pregnancy in humans following Er:YAG-LZD indicated the feasibility of the technique [26, 27]. Laufer et al. [28] examined the safety and efficacy of the 193-nm laser by drilling the ZP of mouse oocytes to improve fertilization rate. The LZD significantly enhanced the fertilization rate over controls, and the rate of hatching was also enhanced. Normal litters were born following the transfer of the embryos into the uteri of pseudopregnant recipients. An improved fertilization rate following ICSI and the ability to use sperm of very poor quality in IVF suggest that zona drilling for this application is obsolete.

Laser-Assisted Hatching Assisted hatching (AH) was introduced into clinical practice in 1990 by Cohen et al. [29, 30] to improve the implantation rate in patients with thick ZP (?15 lm) or in patients over the age of 38. Ten years later, the place of AH following IVF is still controversial. It is being offered to selective groups of patients as a standard procedure in some institutions. A literature search revealed 22 articles discussing AH. Ten articles provided statistical data suggesting that AH improved pregnancy rates for selective groups of randomized patients [29, 31–38], and three other reports revealed nonsignificant improvement or no


331

improvement at all [38–41]. One study confirmed early implantation of embryos when compared to controls [42]. Micromanipulation techniques used in these studies were: mechanical slitting, microinjection of acid Tyrode solution, chemical zona thinning and laser drilling. More information is needed to define the common denominators for patients and embryos that may benefit from AH. However, from the technical point of view, it appears that the laser is the most accurate technique. The procedure can be easily performed, controlled, monitored, and repeated in multiple locations in the same embryo or in multiple embryos. Moreover, a computer-controlled IVF workstation can predefine the size and location of the crater in the ZP, document various parameters, and automatically transfer the data to a spreadsheet or to the patient’s records. WLs tested for laser AH (LAH) range from the UV 308 nm to the IR 2,940 nm and in the contact or free beam modes.

308-nm XeCl Laser Several investigators [21, 42] studied the topical effects of this UV laser on mouse blastomeres using a commercially available system coupled to an inverted microscope. Effects were determined by microinjection of a vital fluorescence dye (fluorescein isothiocynate dextran) into the cell immediately adjacent to the site of zona photoablation. This dye is only passed onto daughter blastomeres and therefore permits study of specific cell lines. Embryonic growth was assessed following cell separation at the morula and blastocyst stage. Four cell stage embryos treated with this laser had significantly fewer cells 12 h after zona photoablation than control embryos. This information suggests that the 308-nm UV excimer laser has some detrimental effects on precompacted mouse embryos. However, a different set of laser parameters at the same WL may eliminate this problem.

337-nm Nitrogen Laser Antinori et al. [37] in a randomized trial evaluated the pregnancy and implantation rate in 3 groups of women with repeated IVF failures. One hundred and seven patients received mixed embryos (with or without LAH), 72 patients received only laser-treated embryos, and a control group of 98 patients were treated by regular IVF. The resulting clinical pregnancies were 39 (36.4%) in the mixed embryos group, 32 (44.4%) in the LAH group and 19 (19.3%) in the IVF controls. The implantation rates per embryo were 9.3, 16, and 5.1% in the three groups, respectively. In total, 17 normal babies were

Tadir

332

born (10 in the mixed group and 7 in the pure LAH group). These results demonstrate that LAH increased the pregnancy and implantation rates. The increase was slight but significant in the mixed embryos group (p=0.01 and p=0.02); it was even higher in the LAH group (p=0.05). The laser parameters used did not cause any visible damage to the embryos as assessed immediately after birth.

2,100-nm Ho:YAG and Ho:YSGG Lasers Light absorption by fluid at this infrared WL (Ho:YAG or holmium: yttrium-scandium-gallium-garnet, Ho:YSGG) is significant, and several factors may affect the amount of energy deposited in the ZP: (a) quality and thickness of the Petri dish; (b) protein content of the culture medium, or (c) distance that the beam has to pass in fluid before it hits the ZP. For this reason, some investigators used this WL in the contact mode and some used it as a free beam. It is not the intent of this review to discuss technical details, such as pulse duration, energy, or pulse repetition rate; however, it is important to realize that such details will further determine the laser effects [44]. 2,100 nm in the Contact Mode Reshef et al. [25] applied the Ho:YAG laser delivered via fibers on the ZP of two-to eight-cell stage mouse embryos to assist hatching. The rate of development to the blastocyst stage and the rate of hatching between the laser-treated and control embryos were compared. Further development was assessed 72 h after lasing. Thirty-three of 49 laser-drilled embryos (67%) progressed to hatching blastocysts as compared to 36 of 82 (44%) untreated controls (p=0.01). 2,100 nm in Noncontact Mode Our group [44, 45] used the Ho:YSGG (2,100 nm) as a free beam delivered through an inverted microscope and quartz glass dish to perform LAH in two-cell mouse embryos. Control embryos were treated with human tubal fluid (HTF) culture with or without serum (HTF-s, HTF-o) or with late serum supplementation (HTF-o/s). Fewer (p=0.05) embryos developed to the blastocyst stage in the HTF-s group (81%) in contrast to the LAH (90%), HTF-o (94%) and HTF-o/s (92%) treatments. The level of hatching was significantly increased (p=0.01) in the LAH treatment (57%) compared to HTF-o/s (32%), HTF-s (18%) or HTF-o (5%). Implantation rates were not impaired following the LAH treatment (21%). These data suggested that LAH using the Ho: YSGG laser is accurate and effective; however, in view of technical limitations


333

of light delivery through fluids near the 2,000-nm range, there might be better WLs to perform LAH in the noncontact mode.

2,940-nm Er:YAG Laser This infrared WL has a high absorption peak in water and can be delivered to the ZP only via fibers in a contact mode. Embryos must be kept stable with a holding pipette during the procedure [4]. Strohmer and Feichtinger [46] tested it in mouse embryos and, subsequently, in human embryos. Groups of 10–15 mouse embryos were placed under oil on two slides. A control slide was maintained on a warming stage while embryos on the other slide were subjected to the laser to produce 20- to 30-lm holes in the ZP. Subsequently, embryos were assessed up to the blastocyst stage. There was no difference between the laser-treated mouse embryos and the untreated controls on days 1 and 2 of culture. On day 3, however, complete hatching was significantly enhanced in the laser-treated group (44/55 [80%] vs. 17/58 [29.3%] for controls, p>0.0001). The same laser was used in a multicenter study for human AH [32]. Embryos obtained from 129 patients who previously experienced repeated IVF failures were exposed to similar laser effects for AH. Ablation was performed by applying some pressure with the laser fiber to deposit approximately 10 lJ in the ZP. Five to 8 pulses were employed to penetrate the ZP creating a 20- to 30-lm opening. A pregnancy rate of 36% (30/84) and 29% (13/45) in the two centers was achieved which was encouraging considering the selective groups of patients studied. Preliminary results of prospective randomized study in patients undergoing a first IVF attempt exhibited a 50% pregnancy rate (10/20 patients) in the LAH group compared to 44% (10/23) in patients without AH. Differences in the implantation rate per embryo in this preliminary study was also not significant (23.8 vs. 21%, LAH vs. control). An improved pregnancy rate following LAH using the 2,940-mm laser in patients with repeated IVF failures and first IVF attempt as compared to controls was reported by Antinori et al. [36]. The embryo implantation rate was 7.3–12.2% in patients with previous repeated failures.

1480-nm Diode Laser Near IR solid-state lasers are small and can emit light at power levels sufficient to cause selective damage to the ZP. The laser module that contains the diode, the electronic board, and the collimated lens are all small (about

Tadir

334

the size of a cigarette box) and can be inserted into the inverted microscope. The 1,480-nm WL is ideal since it is not absorbed by water but is highly absorbed by ZP glycoproteins. This system may be an optimal cutting tool for the IVF laboratory as suggested by Rink et al. [47]. In these studies, the beam was delivered through a 45¶objective of an inverted microscope (2–4 lm spot diameter) to produce zona dissection in mouse and human oocytes and zygotes. One laser exposure was sufficient to drill openings in the ZP ranging from 5 to 20 lm depending on laser power and exposure time. The same group [48] demonstrated that the energy needed to drill a hole of a given diameter is greater for mouse and human zygotes than for oocytes. This confirmed a previous observation with the XeCl laser that ethanol-induced zona hardening can be verified and quantified with a noncontact laser [49]. The safety by microdrilling the ZP of mouse oocytes with a 1.48-lm diode laser has been investigated by determining the ability of mouse oocytes to develop in vivo [50]. Mice born after transfer of control and ZP-microdrilled embryos into foster mothers were submitted to anatomical and immunohistochemical investigations, and their aptitude to breed was assessed in two subsequent generations. Decoronization of the oocytes with hyaluronidase induced a reduction in the fertilization and implantation rates which was attributed to a zona hardening phenomenon. After laser ZP microdrilling, these rates were restored to those obtained with embryos derived from untreated oocytecumulus complexes. Pups derived from ZP microdrilled embryos were comparable with those obtained from control embryos, confirming the lack of deleterious effects of the laser treatment. Another group of mouse zygotes were microdrilled by exposing their ZP to a short pulse of the same 1.4-lm diode laser and allowed to develop in vitro [51]. Various sharp-edged holes could be generated and sizes varied by changing irradiation time (3–100 ms) or laser power (22–55 mW). Drilled zygotes presented no signs of thermal damage under light and scanning electron microscopy (fig. 3). Embryos allowed to develop in vitro showed no sign of abnormality. Preliminary human studies using the same diode laser demonstrated an improved pregnancy rate following AH of cryopreserved embryos [52]. Multicenter AH trials are currently in progress to assess the safety and efficacy of the 1.4-lm diode laser in selective groups of patients and embryos.

Conclusions Competitive technologies should be tested by nonprejudicial investigators in clinical studies to avoid bias regarding the role and outcome of any new


335

Fig. 3. Scanning electron micrograph of a bovine embryo superficially drilled with a 1,480-nm diode laser in a noncontact mode. No thermal damage is visible in the area adjacent to the crater. Multiple superficial drills in the zona pellucida are visible in the smaller magnification (courtesy of Dr. J. Conia, Cell Robotics, Inc.).

technique. This is especially true when expensive tools are introduced into clinical practice. Advances in gamete manipulations has changed indications for IVF and opened new avenues for research and clinical applications. Although lasers may not be beneficial to the process of fertilization, they may play a role in implantation by AH. The combined dissecting and the diode laser microscope may be the ultimate approach for gamete manipulations since no disposable tools are needed. The computerized workstation offers significant benefits since no extra handling is needed to assess various parameters, and data can automatically be stored and loaded in the patient’s records. Other delicate procedures, such as blastomere or polar body biopsy, can be assisted by the noncontact effects laser method [53]. Systems that combine mechanical tools for gamete manipulations with diode lasers, or even systems that provide more than one laser beam delivered through the same optical

Tadir

336

path, may be useful for cutting and trapping. This may be used in sperm tail cutting prior to ICSI or for targeting subcellular organelles that should be removed or inactivated during the IVF.

References 1

2

3 4 5 6 7

8

9 10

11 12 13

14

15 16 17 18 19

Berns MW, Asist J, Edwards J, Strahs K, Girton J, McNeill P, Rattner JB, Kitzes M, HammerWilson M, Liaw LH, Siemens A, Koonce M, Peterson S, Brenner S, Burt J, Walter R, Bryant PJ, Van Dyk D, Coulombe J, Cahill T, Berns GS: Laser microsurgery in cell and developmental biology. Science 1981;213:305–513. Tadir Y, Wright WH, Berns MW: Cell micromanipulation with laser beams; in Capitanio GL, Asch RH, De Cecco L, Croce S (eds): GIFT: From Basics to Clinics. New York, Raven Press, 1989, pp 359–368. Tadir Y, Wright WH, Vafa O, Liaw LH, Asch R, Berns MW: Micromanipulation of gametes using laser microbeams. Hum Reprod 1991;6:1011–1016. Tadir Y, Neev J, Ho P, Berns MW: Lasers for gamete micromanipulations: Basic concepts. J Assist Reprod Genet 1993;10:121–125. Ashkin A, Dziedzic JM, Bjorkholm JE, Chu S: Observations of a single beam gradient force optical trap for dielectric particles. Optics Lett 1986;11:288–290. Tadir Y, Wright WH, Vafa O, Ord T, Asch R, Berns MW: Micromanipulation of sperm by a laser generated optical trap. Fertil Steril 1989;52:870–873. Colon JM, Sarosi P, McGovern PG, Ashkin A, Dziedzic JM, Skurnick J, Weiss G, Bonder EM: Controlled micromanipulation of human spermatozoa in three dimensions with infrared laser optical trap: Effect on sperm velocity. Fertil Steril 1992;57:695–698. Tadir Y, Wright WH, Vafa O, Ord T, Asch R, Berns MW: Force generated by human sperm correlated to velocity and determined using a laser generated optical trap. Fertil Steril 1990;53: 944–946. Araujo E, Tadir Y, Patrizio P, Ord T, Silber S, Berns MW, Asch R: Relative force of human epididymal sperm correlated to the fertilization capacity in vitro. Fertil Steril 1994;62:585–590. Zoentania ND, Araujo E, Tadir Y, Berns MW, Schell MW, Stone SC: Effect of freezing on the relative escape force of sperm as measured by laser optical trap. Fertil Steril 1995;63:185– 188. Schutze K, Clemeny-Sengewald A, Berg FD: Laser zona drilling and sperm transfer into the perivitelline space. Hum Reprod 1993;8:390. Schutze K, Clement-Sengewald A, Ashkin A: Zona drilling and sperm insertion with combined laser microbeam and optical tweezers. Fertil Steril 1994;61:783–786. Van Steirteghem AC, Liu J, Joris H, Nagy Z, Janssenswillen C, Tournaye H, et al: Higher success rate by intracytoplasmic sperm injection than by subzonal insemination. Report of a second series of 300 consecutive treatment cycles. Hum Reprod 1993;8:1055–1060. Westphal L, El-Danasouri IE, Shimizu S, Tadir Y, Berns MW: Exposure of human sperm to the cumulus oophorus results in increased relative force as measured by a 760 nm laser optical tram. Human Reprod 1993;8:1083–1086. Patrizio P, Liu Y, Sonek G, Berns MW, Tadir Y: Effect of pentoxfyllin on the intrinsic force of human sperm. American Academy of Andrology, Minneapolis, April 1996. Konig K, Tadir Y, Patrizio P, Berns M, Tromberg B: Effects of ultraviolet exposure and near infrared laser tweezers on human spermatozoa. Hum Reprod 1996;11:2161–2164. Blanchett BB, Russel JB, Finshcer CR, Portman M: Laser micromanipulation in the mouse embryo: A novel approach to zona drilling. Fertil Steril 1992;57:1337–1347. Schutze K, Clement-Sengewald A: Catch and move – Cut or fuse. Nature 1994;368:667–669. Godke RA, Beetem DD, Burleigh DW: A method of zona pellucida drilling using a compact nitrogen laser (abstract 258). 7th World Congr on Human Reproduction, Helsinki, 1990.


337

20 21 22

23 24

25

26

27

28 29

30 31 32 33 34

35

36 37

38 39

40

41

Neev J, Tadir Y, Ho P, Asch RH, Ord T, Berns MW: Microscope-delivered UV laser zona dissection: Principles and practices. J Assist Reprod Genet 1992;9:513–523. Li L, Munne S, Licciardi F, Neev Y, Tadir Y, Berns MW, et al: Microinjection of FITC-dextran into mouse blasotomeres to assess topical effects of zona penetration. Zygote 1993;1:43–48. Ng SC, Liow SL, Schutze K, Vasuthevan S, Bongso A, Ratnam SS: The use of ultra violet microbeam laser zona dissection in the mouse (abstract 273). 8th World Congr of In Vitro Fertilization and Assisted Reproductive Technologies, Japan, 1993. J Assist Reprod Prog Suppl 1993. Palanker D, Ohad S, Lewis A, Simon A, Shenkar J, Penchas S, et al: Technique for cellular microsurgery using the 193 nm excimer laser. Lasers Surg Med 1991;11:580–586. Coddington CC, Veeck LL, Swanson RJ, Kaufman RA, Lin J, Simonetti S, Bocca S: The YAG laser used in micromanipulation to transect the zona pellucida of hamster oocytes. J Assist Reprod Genet 1992;9:557–563. Reshef E, Haaksma CJ, Bettinger TL, Haas GG, Schafer SA, Zavy MT: Gamete and embryo micromanipulation using the holminum:YAG laser. 49th Am Fertil Soc, Montreal, 1993. Fertil Steril 1993;88(suppl):16. Feitchinger W, Strohmer H, Fuhberg P, Radivojevic K, Antoniori S, Pepe G, Versaci C: Photoablation of oocyte zona pelludcia by erbium:YAG laser for in-vitro fertilization in severe male infertility. Lancet 1992;339:811. Antinori S, Versaci C, Fuhrberg P, Panci C, Caffa B, Hossein Gholami G: Seventeen births after the use of an erbium:YAG laser in the treatment of male factor infertility. Hum Reprod 1994;9: 1891–1896. Laufer N, Palanker D, Shufaro Y, Safran A, Simon A, Lewis A: The efficacy and safety of zona pellucida drilling by a 193-nm excimer laser. Fertil Steril 1993;59:889–895. Cohen J, Elsner C, Kort H, Malter H, Massey J, Mayer MP, et al: Impairment of the hatching process following IVF in the human and improvement of implantation by assisting hatching using micromanipulation. Hum Reprod 1990;5:7–13. Cohen J: Assisted hatching of human embryos. J In Vitro Fertil Embryo Transfer 1991;8:179–190. Cohen J, Alikani M, Trowbridge J, Rosenwaks Z: Implantation enhancement by selective assisted hatching using zona drilling of human embryos with poor prognosis. Hum Reprod 1992;5:685–691. Obruca A, Stohmer H, Sakkas D, Menezo Y, Kogosowski A, Barak Y, et al: Use of lasers in assisted fertilization and hatching. Hum Reprod 1994;9:1723–1726. Schoolcraft WB, Schlenker T, Jones GS, Jones HW Jr: In vitro fertilization in women age 40 and older: The impact of assisted hatching. J Assist Reprod Genet 1995;9:581–584. Takahashi K, Takenaka M, Ishizuka B: The effect of assisted hatching on patients repeatedly failed to conceive by in vitro fertilization (in Japanese). Nippon Sanka Fujinka Gakkai Zasshi 1994;46: 1009–1012. Stein A, Rufas O, Amit S, Avrech O, Pinkas H, Ovadia J, Fisch B: Assisted hatching by partial zona dissection of human pre-embryos in patients with recurrent implantation failure after in vitro fertilization. Fertil Steril 1995;4:838–841. Antinori S, Panci C, Selman HA, Caffa B, Dani G, Versaci C: Zona thinning with the use of laser: A new approach to assisted hatching in humans. Hum Reprod 1996;11:590–594. Antinori S, Selman HA, Caffa B, Panci C, Dani GL, Versaci C: Zona opening of human embryos using a non-contact UV laser for assisted hatching in patients with poor prognosis of pregnancy. Hum Reprod 1996;11:2488–2492. Parikh FR, Kamat SA, Nadkarni S, Arawandekar D, Parikh RM: Assisted hatching in an in vitro fertilization programme. J Reprod Fertil Suppl 1996;50:121–125. Hellebaut S, De Sutter P, Dozortsev D, Onghena A, Qian C, Dhont M: Does assisted hatching improve implantation rates after in vitro fertilization or intracytoplasmic sperm injection in all patients? A prospective randomized study. J Assist Reprod Genet 1996;13:19–22. Tucker MJ, Cohen J, Massey JB, Mayer MP, Wicker SR, Wright G: Partial dissection of the zona pellucida of frozen-thawed human embryos may enhance blastocyst hatching, implantation, and pregnancy rates. Am J Obstet Gynecol 1991;165:341–344. Tucker MJ, Luecke NM, Wicker SR, Wright G: Chemical removal of the zona pellucida of day 3 human embryo has no impact on implantation rate. J Assist Reprod Genet 1993:10;187–191.

Tadir

338

42 43

44 45

46 47 48

49

50

51

52

53

Liu HC, Cohen J, Alikani M, Noyes N, Rosenwarks Z: Assisted hatching facilitates earlier implantation. Fertil Steril 1993;5:871–875. Neev J, Gonzales A, Licciardi F, Alikani M, Tadir Y, Berns MW, et al: A contact-free microscope delivered laser ablation system for assisted hatching of the mouse embryo without the use of a micromanipulator. Hum Reprod 1993;8:939–944. Neev Y, Schiewe MC, Sung WV, Kang D, Berns MW, Tadir Y: Assisted hatching in mouse embryos using a non-contact Ho:YSSG laser system. J Assist Reprod Genet 1995;12:228–293. Schiewe MC, Neev J, Hazeleger NL, Balmaceda JP, Berns MW, Tadir Y: Developmental competence of mouse embryos following zona drilling using a noncontact holmium:yttrium scandian gallium garnet (Ho:YSGG) laser system. Hum Reprod 1995;10:1821–1824. Strohmer H, Feichtinger W: Successful clinical application of laser for micromanipulation in an in vitro fertilization program. Fertil Steril 1992;58:212–214. Rink K, Delacretaz G, Salathe RP, Senn A, Nocera D, Germond M, et al: 1.5 lm diode laser microdissection of the zona pellucida of mouse oocytes (abstarct 53). Biomed Optics 1994;2134. Germond M, Nocera D, Senn A, Rink K, Delacretaz G, Fakan S: Microdissection of mouse and human zona pellucida using a 1.48-microns diode laser beam: Efficacy and safety of the procedure. Fertil Steril 1995;64:604–611. Tadir Y, Neev Y, Schiewe M, Balmaceda JP, Ord T, Asch RH, Berns MW: Spontaneous and induced zona pellucida hardness: Measurements using enzyme assay and a non-contact laser micromanipulation (abstract). Pacific Coast Fertility Society, Palm Springs, 1993. Fertil Steril Prog Suppl 1993; O-21. Germond M, Nocera D, Senn A, Rink K, Delacretaz G, Pedrazzini T, et al: Improved fertilization and implantation rates after non-touch zona pellucida microdrilling of mouse oocytes with a 1.48 micron diode laser beam. Hum Reprod 1996;11:1043–1048. Rink K, Delacretaz G, Salathe RP, Senn A, Nocera D, Germond M, et al: Non-contact microdrilling of mouse zona pellucida with an objective-delivered 1.48-microns diode laser. Lasers Surg Med 1996:18:52–62. Germond M, Senn A, Rink K, Delacretaz G, De Grandi P: Is assisted hatching of frozen-thawed embryos enhancing pregnancy outcome in patients who have several previous nidation failures? (abstract). J Fertil Reprod 1995;3:41. Licciardi F, Gonzalez A, Tang YX, Grifo J, Cohen J, Neev Y: Laser ablation of the mouse zona pellucida for blastomere biopsy. J Assist Reprod Genet 1995;12:462–466.

Yona Tadir, MD, Beckman Laser Institute and Medical Clinic, University of California Irvine, 1002 Health Sciences Road East, Irvine, CA 92612 (USA) Tel. +1 714 824 4713, Fax +1 714 824 8413, E-Mail [email protected].


339


............................

PALMÔ Robot-MicroBeam for Laser-Assisted Fertilization, Embryo Hatching and Single-Cell Prenatal Diagnosis An Overview A. Clement-Sengewald a, K. Schu¨tze b, S. Sandow c, C. Nevinny c, H. Po¨sl b

a b

c

I. Frauenklinik der Universita¨t, Mu¨nchen; Applikatives Laserzentrum der I. Medizinischen Abteilung, Krankenhaus Harlaching, Mu¨nchen, und Immunologische Laboratorien, Mu¨nchen, Deutschland

The PALMÔ Robot-MicroBeam as well as the Robot-MicroTweezers have proven to be well suited for noncontact gamete micromanipulation without harming the specimen [1]. The cutting ability of the microbeam was used to induce fusion of blastomeres, to stop sperm motility or to cut holes in the zona pellucida of oocytes in order to facilitate fertilization and/or to assist embryo hatching. Microtweezers could catch and move even vigorously moving sperm and introducing them into the perivitelline space of oocytes in order to induce fertilization [2]. A new application of the Robot-MicroBeam is the noncontact procurement of single cells from histological tissue slices, cell smears or from cytocentrifuged samples for cell and chromosome specific genetic analysis.

Laser-MicroBeam Technology The high photon density that arises in the narrow focal spot of an extremely focused, pulsed nitrogen laser (i.e. Robot-MicroBeam; fig. 1) causes a photochemical reaction that decomposes organic material without a heat effect, i.e. cold ablation [3]. This way it is possible to precisely microdissect

Fig. 1. PALMÔ Robot-MicroBeam system.

small molecules or to capture single cells or subcellular particles without any mechanical contact. The high-intensity gradient within the focal point of a continuous-wave, strongly focused red or infrared laser beam (i.e. optical tweezers) causes microscopically small objects to be assembled within the laser focus. They are trapped there by means of radiation pressure forces [4]. Both laser systems were combined to the PALMÔ CombiSystem (P.A.L.M. GmbH, Wolfratshausen, Germany). All lasers are coupled in an inverted research microscope and focused through high-numerical aperture objectives to yield focal spot sizes of about 0.5–1 lm at the object plane. A special laser microscope interface enables independent laser focus adjustment and beam positioning of each laser. The lasers can be used simultaneously and also together with fluorescence illumination. The laser systems are equipped with a motorized, computer-controlled microscope stage, which was developed for precise sample positioning and convenient working at a computer monitor. Mouse movements are translated into smooth stage displacements with nanometer precision. The video-generated microscope image is overlaid with the user interface and displayed on the computer monitor. This allows realtime control of the laser position(s) on the screen. The advantage of optical micromanipulation is the lack of any mechanical contact. Thus, contamination and evaporation can be avoided.

PALMÔ Robot-MicroBeam

341

Table 1. Laser-induced fusion of blastomeres in mice Laser-treated two-cell embryos pulsed n 109

fused n 34 (31%)

cleaved n 9 (26%)

Control two-cell embryos M+B n 3 (33%)

cultured n 25

cleaved n 25 (100%)

M+B 25 (100%)

M+B>Morulae and blastocysts.

Laser Micromanipulation to Study Gametes and Embryo Development The Robot-Microbeam was successfully used to fuse blastomeres in mouse embryos in order to study the development of tetraploid embryos. Poking small holes into the contact area of two adjacent cells can selectively induce their fusion [5]. 31% of laser-treated 2-cell embryos fused within 1 h after the laser pulses and 26% of the fused embryos cleaved at least once while in in vitro culture (table 1). We could demonstrate the feasibility of laser induced blastomere fusion. As compared to electrofusion, the laser-based fusion technique is more selective but further studies have to be done to evaluate its efficiency.

Stopping Sperm Motility With a few laser pulses, placed close to the waving tail, it is possible to reversibly stop sperm motility [2]. The time of recovery was dependent on the laser energy applied. After stopping, the laser could carefully be focused onto the tail, and with one single shot the tail was cut at any desired position (fig. 2). This feature may facilitate the needle-based intracytoplasmic sperm injection (ICSI), since it is possible to incubate oocytes and sperm within the same microdroplet. This also avoids the application of special and harmful agents like PVP, which are used to slow down sperm motion. During ICSI the sperm tail is routinely hit with the injection pipette to increase the rate of fertilization. Stopping sperm motion and subsequently cutting the sperm tail with the laser MicroBeam speeds up the ICSI procedure, where it is sometimes difficult to immobilize, perforate and catch sperm with the ICSI pipette.

Clement-Sengewald/Schu¨tze/Sandow/Nevinny/Po¨sl

342

Fig. 2. Sperm tail cut by a single UV laser shot. Optical tweezers can trap the separated sperm head in order to move it around.

Laser Zona Drilling with the Robot-MicroBeam Various assisted reproduction technologies have been applied to improve fertilization. These include needle-based zona drilling and subzonal or intracytoplasmic sperm injection. Several methods have been tested in order to open the zona pellucida of oocytes or embryos, to facilitate fertilization, assist blastocyst hatching or perform embryo biopsy. Besides chemical dissolution of the zona, mechanical opening with microtools has been performed. As these techniques are not easy to control and/or require the time-consuming and strenuous use of microtools like glass micropipettes, the development of a noncontact and precise micromanipulation technique was desirable. Therefore, the application of laser technology for zona drilling has been suggested. There are two principles of laser zona drilling. If an Erb:YAG laser is used [6], the laser has to be guided through a glass capillary and the oocyte has to be held in position with a holding pipette. On the contrary, using a focused 308-nm excimer, the 337-nm nitrogen laser is directed tangentially to the oocyte and a precise hole of any desired diameter beyond 1 lm and at any selected site can be drilled without any mechanical contact or holding pipette (fig. 3). With a 1.48-diode laser, the minimum drilling size is about 3 lm in diameter. The Robot-MicroBeam, equipped with a 337-nm nitrogen laser was successfully used for micromanipulation of a varitety of biological specimens without impairing their viability. As shown by the comet assay [7] the 337-nm wavelength does not introduce strand breaks into DNA. This indicates that


343

Fig. 3. Laser zona drilling performed by the UV laser.

Table 2. Effect of laser zona drilling on embryo development in cattle [12] Experiment

Oocytes n

Cleavage (day 2)

LZD-pol+IVF LZD-op+IVF Control (Öcc)+IVF Control (+cc)+IVF

52 57 64 91

34 38 28 66

(65.3%)a (66.6%)b (43.7%)c (72.5%)

Morula/blastocyst (days 6–8)

4 10 12 29

(7.6%)a (17.5%) (18.7%)c (31.8%)

Expanded/hatched blastocyst (days 9–15) 0b 6 (10.5%)c 10 (15.6%)c 21 (23%)

+cc/Öcc>With/without cumulus cells; LZD-pol/LZD-op>laser zona drilling close to/ opposite of the polar body. a, c 2 v analysis: p=0.05; results of five trials. b, c 2 v analysis: p=0.01; results of five trials.

the wavelength of the 337-nm nitrogen laser is sufficiently removed from the absorption maximum of DNA, which is at 260 nm. In mice, laser zona drilling with the 337-nm nitrogen laser resulted in increased fertilization rates, enhanced embryo hatching, and normal offspring [8, 9], whereas embryo development after laser zona drilling with a 308-nm excimer was impaired [10, 11]. We examined the dependency of fertilization with respect to the drilling site in cattle oocytes (table 2). In this experiment, the zona pellucida of the


344

Fig. 4. Laser zona drilled four-cell embryo.

oocytes was drilled twice either in close vicinity of the polar body or at the opposite side. Subsequent fertilization yielded equal fertilization and expanded/hatched blastocyst rates as in control embryos (nondrilled), if laser pulses were applied at the opposite side of the polar body [12].

Laser-Assisted Embryo Hatching Several healthy babies were born after the use of the above-described 268-nm Erb:YAG laser [4]. The noncontact 337-nm Robot-MicroBeam was successfully used for laser zona drilling of human embryos (fig. 4) resulting in many healthy babies [13]. It was also shown that the 1.48-nm diode laser could successfully assist embryo hatching. Assisted hatching of frozen-thawed embryos with a 1.48lm diode laser enhances pregnancy outcome in patients who had several previous nidation failures.

Sperm Trapping with Optical Tweezers Optical tweezers can be used to catch and move motile sperm and to measure the force of their movement [14–16]. Furthermore, with optical tweezers sperm were successfully inserted into the perivitelline space of mouse, cattle, and human oocytes through a previously laser-drilled hole [17].


345

a

b

c

d Fig. 5. Subzonal sperm insertion in a cattle oocyte. a Intact zona pellucida. b Channel drilled into the zona pellucida. c Trapped sperm before insertion under zona pellucida. d Inserted sperm attached to the oocyte membrane.

Solely Laser-Induced in vitro Fertilization of Cattle Oocytes We could demonstrate, that fertilization of cattle oocytes was possible by exclusively using the force of focused laser light (fig. 5) [18]. For laser zona drilling the laser energy was set to about 3–5 lJ/pulse with a frequency of about 15 pulses/s and a pulse duration of 3 ns. For sperm trapping, the laser power at the object plane was between 30 and 100 mW, depending on sperm motility and the speed of transportation. As a control, in vitro matured cattle oocytes from a slaughterhouse, either with the surrounding cumulus cells or after removal of cumulus cells, were fertilized using a high and low sperm concentration, respectively (table 3). One experimental group of cumulus-freed oocytes was perforated with a single, laser-drilled channel of 10–15 lm in diameter and incubated with highly diluted sperm. The second


346

Table 3. Fertilization of bovine oocytes after sperm insertion with combined laser scalpel and laser tweezers [18] Experiment

Oocytes n

Fertilized oocytes n

Fertilization rate %

+cc,+sperm(hc) Öcc,+sperm(hc) Öcc,+sperm(lc)+LZD Öcc,+sperm(lc)+LZD+laserSUZI

163 160 96 79

88 80 0 3

53.9 50 0 3.7

+/Öcc>With/without cumulus cells; sperm(hc)/(lc)>sperm concentrations 106/ml or 5,000/ml, respectively; LZD>laser zona drilling; laserSUZI>laser-mediated subzonal sperm insertion.

experimental group of oocytes was laser drilled and incubated with the same low sperm concentration but, in addition, 3–5 sperm were transferred into the perivitelline space by optical tweezers. After 20–22 h of in vitro culture, the oocytes were fixed and stained with acetoorcein and checked for the appearance of two pronuclei and a sperm tail in the cytoplasm, which proves fertilization. Three of 79 laser-fertilized oocytes showed pronuclei, whereas no fertilization occurred in the solely zona-drilled group. The fertilization rate in the 2 control groups was about 50%. Pronuclear formation provided first evidence that laser manipulations were not harmful to bovine gametes. The low fertilization rate might be increased by improved specimen handling during laser treatment. As no fertilization took place in the first experimental group, we could prove that fertilization of bovine oocytes was induced solely by laser light-mediated manipulation of gametes.

Single-Cell Prenatal Diagnosis Prenatal diagnosis enables identification of genetic defects in the fetus underlying acquired or inherited disorders. Until now, the diagnosis was performed after amniocentesis or chorionic villus sampling, but these still bare a low risk to the ongoing pregnancy. The detection of fetal cells in maternal blood and in maternal transcervical mucus offers a noninvasive diagnosis of genetic defects of the fetus. As molecular technologies are extremely sensitive, there is a high demand for clean sample preparation. Numerous efforts have been made in order to selectively procure single cells from heterogeneous tissue using needle-based microdissection or laser welding of selected tissue onto a


347

Fig. 6. Schematic drawing of laser microbeam microdissection with subsequent laser pressure catapulting.

transfer membrane [19–23]. However, the major problem of these techniques are inefficient sample recovery and contamination with unselected material. One source of this problem seems to be the mechanical contact during microdissection or specimen transfer. With histological tissue sections it was shown that the Robot-MicroBeam can efficiently be used to micodissect selected cells with laser microbeam microdissection (LMM), i.e. to isolate them from their surroundings, and that the isolated specimen can be catapulted directly into the cap of a microfuge tube using laser pressure catapulting (LPC; fig. 6, 7).


348

Fig. 7. Isolation of a single cell with combined microbeam microdissection and laser pressure catapulting.

Laser treatment does not impair DNA, RNA or protein recovery and the captured specimens are well suitable for further PCR-based DNA amplification or RT-PCR [24]. Nucleated red blood cells (NRBCs) were isolated from maternal blood by Ficoll density gradient centrifugation followed by CD 71 magnetic-activated cell sorting. Subsequently, the isolated cells were subjected to a slide by cytocentrifugation and stained (May-Gru¨nwald-Giemsa). NRBCs were microscopically identified and single cells were isolated from neighboring cells by LMM. With one single laser shot the laser-isolated cells were catapulted directly into the sample tube. In several cases, LPC could be performed directly, i.e. without previous LMM isolation. The single cells captured by the laser were amplified using an oligonucleotide-degenerated primer PCR. Followed by a second PCR step, the origin of the cell could be characterized [25]. However, in most cases the isolated NRBCs were of maternal origin. Therefore, we used maternal cervical mucus obtained from transcervical swabs in a noninvasive manner to isolate and capture fetal cells. Successful single-cell PCR for determining the


349

Rhesus blood group, in the case of a Rhesus-negative mother, and for individual genotyping with microsatellites (short tandem repeats) showed that fetal cell isolation from maternal mucus is indeed a promising method for noninvasive prenatal diagnosis [26].

Conclusion Robot-MicroBeam technology is a valuable tool for a variety of applications in cell biology, embryo development and in vitro fertilization. A singlecell preparation in combination with molecular medicine has the potential to develop into a key technology for prenatal and preimplantation diagnosis and can simultaneously be used to assist fertilization procedures. LMM and LPC are valuable methods to specifically capture cells from formalin-fixed or frozen tissue sections in a convenient and rapid manner. In addition, samples derived from cell smears or after cytocentrifugation can be procured. It is also possible to capture subcellular structures like nuclei [Cremer et al., in prep.], chromosomes or chromosomal parts [Thalhammer et al., in prep.]. Noncontact laser microinjection of genetic material into living cells is a further promising application.

References 1 2 3 4 5 6 7 8 9 10 11 12

Schu¨tze K, Clement-Sengewald A: Catch and move – Cut and fuse. Nature 1994;368:667–669. Schu¨tze K, Clement-Sengewald A, Ashkin A: Zona drilling and sperm insertion with combined laser microbeam and optical tweezers. Fertil Steril 1994;61:783–786. Srinivasan R: Ablation of polymers and biological tissue by ultraviolet lasers. Science 1986;234: 559–565. Ashkin A: Applications of laser radiation pressure. Science 1980;210:1081–1088. Clement-Sengewald A, Schu¨tze K, Heinze A, Palma GA, Po¨sl H, Brem G: Laser assisted cell fusion and cytoplast transfer in early mammalian embryos. SPIE 1993;1876:187–194. Feichtinger W, Strohmer H, Fuhrberg P, et al: Photoablation of oocyte zona pellucida by erbiumyag laser for in-vitro fertilisation in severe male infertility. Lancet 1992;339:811. deWitt A, Greulich G: Wavelength dependence of laser-incduced DNA damage in lymphocytes observed by single-cell gel electrophoresis. J Photochem Photobiol Biol 1995;30:71–76. Liow SL, Ng SC, Bongso A, Ratnam SS: Micromanipulation of gametes using laser microbeams. Hum Reprod 1995;6:1011–1016. Ng SC, Liow SL, Bongso A, Kumar J, Ratnam SS: The use of non-contact microbeam laser for assisted reproduction. J Assist Reprod Genet 1995;12(suppl):7. Neev J, Gonzalez A, Liccardi F, et al: Opening of the mouse zona pellucida by laser without a micromanipulator. Hum Reprod 1993;8:939–944. Tadir Y, Neev J, Berns MW: Lasers in micromanipulation of preimplantation embryos and gametes. Semin Reproduct Endocrin 1994;12:169–176. Clement-Sengewald A, Schu¨tze K, Berg FD: Effect of laser zona drilling on embryo development in cattle. Hum Reprod 1994;9:77.


350

13

14 15 16

17 18

19

20 21 22

23 24 25 26

Antinori S, Selman HA, Caffa B, Panci C, Dani GL, Versaci C: Zona opening of human embryos using a non-contact UV laser for assisted hatching in patients with poor prognosis of pregnancy. Hum Reprod 1996;11:2488–2492. Ashkin A, Dziedzic JM, Yamane T: Optical trapping and manipulation of single cells using infrared laser beams. Nature 1987;330:769–771. Tadir Y, Wright WH, Vafa O, Ord T, Asch R, Berns MW: Micromanipulation of sperm by a laser genereated optical trap. Fertil Steril 1989;52:870–873. Schu¨tze K, Clement-Sengewald A, Berg FD: ‘Non-contact’ laser micromanipulation of gametes and embryos. I. The UV-laser microbeam. II: IR-optical tweezers trap. Hum Reprod Update 1995; 1:1. Enginsu ME, Schu¨tze K, Bellanca S, et al: Micromanipulation of mouse gametes with laser microbeam and optical tweezers. Hum Reprod 1995;10:1761–1764. Clement-Sengewald A, Schu¨tze K, Ashkin A, Palma GA, Kerlen G, Brem G: Fertilization of bovine oocytes induced solely with combined laser microbeam and optical tweezers. J Assist Reprod Genet 1996;13:259–265. Zhuang Z, Bertheau P, Emmert-Buck MR, Liotta LA, Gnarra J, Linehan WM, Lubensky IA: A microdissection technique for archival DNA analysis of specific cell populations in lesions =1 mm in size. Am J Pathol 1995;146:620–625. Deng G, Lu Y, Zlotnikov G, Thor AD, Smith HS: Loss of heterozygosity in normal tissue adjacent to breast carcinomas. Science 1996;274:2057–2059. Meier-Ruge W, Bielser W, Remy E, Hillenkamp F, Nitsche R, Unso¨ld R: The laser in the Lowry technique for microdissection of freeze-dried tissue slices. Histochem J 1976;8:387–401. Kubo Y, Klimek F, Kikuchi Y, Bannasch P, Hino O: Early detection of Knudson’s two-hits in preneoplastic renal cells of the Eker rat model by the laser microdissection procedure. Cancer Res 1995;55:989–990. Emmert-Buck MR, Bonner RF, Smith PD, Chuaqui RF, Zhuang Z, Goldstein SR, Weiss RA, Liotta LA: Laser capture microdissection. Science 1996;274:998–1001. Schu¨tze K, Lahr G: Identification of expressed genes by laser-mediated manipulation of single cells. Nature Biotechnol 1998;16:737–742. Sandow S, Meenzen C, Schu¨tze K, et al: Isolation and characterization of fetal cells from maternal blood. 9th Annu Meet Gesellschaft fu¨r Humangenetik, Innsbruck, 1997. Burgemeister R, Hinderer S, Gloning K: Single fetal cells separated by the laser MicroBeam technique. Bioforum 1999; in press.

Dr. Annette Clement-Sengewald, I. Frauenklinik der Universita¨t, Maistrasse 11, D–80337 Munich (Germany) Tel. +49 89 5160 4308, Fax +49 89 5160 4149, E-Mail [email protected]


351


............................

Diode Laser for Assisted Hatching Marc Germond, Marie-Pierre Primi, Alfred Senn, Klaus Rink, Laurent Descloux, Guy Delacre´taz Fertility Unit, Department of Obstetrics and Gynecology, CHUV, Lausanne, Switzerland

Introduction At their initial stage of development, mammalian embryos are surrounded by a 10- to 20-lm thick glycoprotein coating, the zona pellucida (ZP), which has to be breached prior to implantation in utero. Hatching of the embryo is the process whereby the expanded blastocyst undergoes several expansion/ contraction cycles thereby inducing a progressive plastic deformation of the ZP matrix. Once maximum deformation of this structure is reached, a local disruption of the ZP matrix occurs and the blastocyst hatches. Any functional or structural disorder of the ZP may lead to an absent or incomplete opening resulting in hatching failure. Only 10–20% of human embryos replaced in the uterus after in vitro fertilization (IVF) complete the initial stages of implantation, a process which presents today’s greatest challenge in assisted reproduction techniques. The low implantation rate has been attributed to several factors, such as aneuploidy leading to a developmental arrest [1, 2], suboptimal in vitro culture media [3, 4], a hatching failure of an otherwise healthy blastocyst [5], or a lack of endometrial receptivity [6–8]. The possible causes of hatching failure have been classified in two main categories: (1) embryos presenting thick ZP (?15 lm) may have a reduced ability to hatch and implant [9] and the ZP thickness has been shown to be correlated with age of the patients, basal FSH level and preovulatory estradiol [10], and (2) a secondary hardening of the ZP is suspected to occur during in vitro culture [11], after hyaluronidase treatment for decoronization [12] or after cryopreservation [13]. In order to help the embryos escape from their protective envelope, and therefore facilitate their implantation [5, 14], two different micromanipulation

procedures, mechanical [15–17] and chemical [18, 19], have been applied to produce holes in the ZP or to thin it superficially. The mechanical methods, such as partial zona dissection, are easy to perform but they produce holes or slits of variable sizes which, when too small, are not favorable for the integrity of the embryo and can interfere with the hatching [20]. The application of acidic Tyrode’s solution to open the ZP, even if it results in more standardized holes, requires greater technical expertise as it has been reported to be toxic to human oocytes and embryo development, especially in embryos with thin zonae [9]. This method was also proposed as a means for quantifying zona hardening in human embryos by measuring the duration of the drilling procedure [21]. More recently, a third method for drilling the ZP using laser systems has been proposed [22] and since then several types of laser sources have been investigated [23–26]. This article will focus on the use of the 1.48-lm diode laser that was developed by the Institut d’Optique Appliqueé (Ecole Polytechnique Fe´de´rale de Lausanne, Switzerland) in association with the Reproductive Medicine Unit (DGO, CHUV) and commercialized as a functional unit (FertilaseÔ; Medical Technologies Montreux SA, Clarens, Switzerland). After a description of the physical characteristics of the diode laser system and of the drilling process, the biological and clinical applications of this laser in assisted hatching (AH) will be presented.

The 1.48-lm Diode Laser: Technical Aspects Lasers represent ideal tools for microsurgical procedures due to the fact that their radiation is easily focused on the targeted area. Although many different laser wavelengths ranging from the very energetic 193-nm radiation through the visible have been initially proposed, infrared (IR) radiation is the most appropriate especially when taking into account mutagenic risks. Additionally, several water absorption bands are available in the IR, allowing the use of water as the main absorber and thus ensuring an efficient laser ablation process. Among the potentially interesting wavelengths in IR radiation, we chose 1.48 lm in the near IR, which exhibits a penetration depth of several hundred microns in water as well as in culture media [26] and simultaneously a sufficient absorption to induce significant tissular effects when properly focused. The 1.48-lm radiation is emitted by a compact InGaAsP laser diode. The fact that this diode has a high brightness (very high density of emitted light per unit of surface and angle) as compared to typical laser diodes is essential. It warrants an efficient refocusing of the laser beam in a spot of some microns

Diode Laser for Assisted Hatching

353

Fig. 1. Scheme of the 1.48-lm diode zona pellucida drilling arrangement. The fluorescence port of the inverted microscope is used to couple both the surgical and the aiming lasers. The microscope objective is used to precisely focus the laser radiation onto the egg ZP.

in diameter by means of a microscope objective. In cw-operation, a laser power of typically 100 mW is developed. The reduced dimensions of the diode laser (9¶5¶5 mm3) allow the design of a compact system easily adaptable to any inverted microscope. We mounted the laser diode in a robust embodiment together with the necessary collimating optics to ensure an optimal injection of the laser beam. A typical schematic of the 1.48-lm laser system as well as the laser beam injection path is shown in figure 1. The invisible 1.48-lm diode laser beam is matched to a 1-mW visible 670-nm aiming diode laser. The collinear 1.48-lm and 670-nm laser beams are then injected into the inverted microscope through the fluorescence port. The two collinear laser beams are directed along the microscope’s optical axis by an additional mirror having a high reflectivity at 1.48 lm and a high transmission in the visible. The two beams are focused on the object plane by a 40¶ microscope objective. The spot diameter is 8–16 lm. ZP opening is performed according to the following procedure. The culture dish is placed on the displacement stage of the microscope (fig. 2). Due to the specific wavelength used, laser drilling can be performed directly in the culture dish while keeping the eggs in their original culture medium. With the

Germond/Primi/Senn/Rink/Descloux/Delacre´taz

354

Fig. 2. Zona pellucida drilling strategies. The focused laser beam is either directed tangentially to the ZP to produce a trench (a) or in a more inward location to produce a hole (b).

help of the displacement stage, a region of the ZP where the perivitelline space is widest is positioned at the location of the aiming spot. Opening is performed by exposing the ZP to laser light, typically 12–30 ms. Several drilling strategies have been considered. If a single irradiation procedure is preferred, the hole size can be chosen by varying the irradiation time, typically a hole having a diameter of 20 lm is produced during a 12to 30-ms irradiation time. Larger hole diameters are obtained by increasing the irradiation time. If the egg is placed tangentially to the diode laser beam a trench is induced in the ZP. By precisely positioning the laser focalization point with respect to the ZP width a complete opening or only a local thinning of the ZP can be generated at will. If the egg is placed in such a way that the laser beam intersects the ZP in an area closer to the polar axis, while ensuring that the laser beam trajectory still remains in the perivitelline space, a cylindrical hole is formed. Direct interaction of the laser beam with the cytoplasm is avoided. A multiple irradiation procedure might be preferred by some operators in order to generate an opening with a specific shape. As an example, in figure 3 the drilling has been performed by two consecutive irradiations.


355

Fig. 3. Zygote drilled at the 2-cell stage on day 2. The laser drilled trench completely opens the ZP; it has been obtained by two consecutive 35-ms irradiations with the FERTILASEÔ laser system.

Because of the short irradiation time needed to open the ZP, no micromanipulators are needed to stabilize the eggs during the drilling procedure in the standard conditions. For the operator, ZP drilling is virtually instantaneous. Even if a multiple irradiation procedure has been chosen, no micromanipulators are needed. The drilling procedure is monitored and systematically recorded with a CCD video camera for retrospective studies. An example of the striking quality and precision achieved by the 1.48-lm laser irradiation can be seen on an electron micrograph of a drilled mouse zygote shown in figure 4. The hole appears clearly delimited in an area circumscribing the laser impact point.

Animal Experimentation Safety of ZP drilling with the diode laser was tested in mice. Zygotes were obtained from B6D2F1 mice, stimulated by intraperitoneal injections of FSH (10 IU/0.2 ml) on the first day and of hCG (10 IU/0.2 ml) 2 days later just prior to mating in order to induce superovulation, and sacrified 20 h after mating (day 1). The zygotes were suspended in HTF/HEPES medium (groups of 10–15 zygotes in a 4-well multidish) and exposed to laser light. In each experiment, 4 groups (control, 1, 2, 3) of zygotes were treated on day 1. The control group was left with an intact ZP while groups 1, 2 and 3 were exposed to a single laser exposure of 5, 15 and 30 ms, respectively. The mean ZP


356

Fig. 4. Two electron micrographs showing a hole and a trench drilled with the 1.48lm diode laser system. Note the sharpness of the walls of both openings.

thickness of zygotes measured was 7.42×0.07 lm (×SD, n>40). The diameter of the drilled holes in ZP varied from 2–3.5 lm in group 1, 5–7.5 lm in group 2 and 8–10 lm in group 3. After treatment the zygotes were transferred to a standard incubation medium (HTF/HCO3 containing 0.5% bovine serum albumin and equilibrated with 5% O2, 5% CO2 and 90 N2) and stored in a


357

Table 1. Effect of the drilled hole diameter on mouse blastocyt formation (day 5) and hatching (day 7) and on ZP thinning (day 7) Group

Hole diameter lm

Zygotes on day 1

Blastocysts on day 5

Hatching blastocysts on day 7

ZP thickness on day 7 lm×SD

Control 1 2 3

– 2.0–3.5 5.0–7.5 8.0–10.0

182 180 183 180

98 109 100 102

27 27 80 57

3.41×0.07 3.75×0.08 6.45×0.06* 6.60×0.06*

(54%) (61%)NS (55%)NS (57%)NS

(15%) (15%)NS (44%)* (32%)*

NS>Not significantly different from control. * p=0.001 compared to control.

gassed incubator. The control group was treated identical to the other groups except for the drilling procedure. On day 5, the embryo development was evaluated by counting the number of full blastocysts. They were then transferred to a different medium (B2, INRA Menezo) and replaced in the gassed incubator. On days 6 and 7, the hatching process was observed and taped on video for further measurements of ZP thickness and countings of fully hatched embryos and embryos in the middle of the hatching process (at hatching). The results are summarized in table 1. First it appears that the embryos developed identically to the blastocyst stage (day 5) whether the ZP was drilled or not, indicating that zona drilling did not affect the early development. The numbers of blastocysts reported in table 1 are slightly inferior to the total number of embryos reaching that stage as a few of them could not yet be classified as full blastocysts on day 5. The in vitro culture of drilled zygotes did not show signs of damage due to the laser irradiation, as cleavage and blastocyst formation occurred at comparable rates in control and treated eggs. Hatching was initiated 1 day earlier in groups 2 and 3. Furthermore, the total number of embryos at hatching on day 7 was higher in groups 2 and 3 compared to the control and group 1. Finally, in the latter groups, the thickness of ZP had decreased by half in size by day 7 compared to day 1, while it was found to be only slightly thinner in groups 2 and 3, showing that the natural thinning of ZP resulting from successive expansions and constrictions of the blastocyst did not occur in these 2 groups. The ZP of fully hatched blastocysts appeared thicker and its external diameter smaller in the treated group as a consequence of the lower mechanical forces exerted on the ZP during the blastocyst expansion. Drilled embryos hatched through the drilled hole in a figure ‘8’. Transfers


358

of control and drilled mouse zygotes into foster mothers showed no differences in implantation rates [27]. This supports the idea that the intermediary figure8-shaped blastocysts give rise to viable fetuses. In terms of hatching ability, group 1 behaved similar to the control. It thus appeared that the hole diameter had to be superior or equal to the ZP thickness for hatching to be assisted efficiently. The safety of microdrilling the ZP with the 1.48-lm diode laser has been investigated further by determining the ability of mouse oocytes to fertilize in vitro and develop in vivo [28]. Decoronization of the oocytes with hyaluronidase induced a reduction in the fertilization and implantation rates, which could be attributed to a zona-hardening phenomenon. After laser ZP microdrilling, these rates were restored to those obtained with embryos derived from untreated oocyte-cumulus complexes. Pups derived from ZP microdrilled embryos were comparable to those obtained from control embryos, confirming the lack of deleterious effects of the laser treatment. The mice born after transfer in foster mothers of control and ZP-microdrilled embryos were submitted to anatomical and immunohistochemical investigations, and their aptitude to breed assessed in two subsequent generations (F3–F4). In conclusion, the 1.48-lm diode laser allows a safe ZP microdrilling of mouse oocytes after decoronization with hyaluronidase. Based on the health of the F2, F3 and F4 generations, and the lack of neuroanatomical and neurochemical differences, we extended this technology to the human.

Clinical Studies Clinical studies in IVF patients published so far have come to various conclusions regarding the effect of AH: beneficial [9, 29–35], none [9, 36–38] or detrimental [9]. However, a number of parameters, such as the patient’s age, the ZP thickness, the number of previous IVF failures, need to be taken into consideration. A preliminary study was undertaken at the CHUV in Lausanne during 1994–1997 on patients who had failed to conceive in at least two transfer cycles of frozen-thawed embryos and who came for a third transfer cycle. Sixty-eight patients were enrolled in the study and matched retrospectively with 65 control patients to whom undrilled cryopreserved embryos were transferred. The treatment period of the 2 groups overlapped partly, but the laboratory and transfer techniques remained identical. Drilling of the embryos was performed before the embryo transfer (ET) with the 1.48-lm diode laser by exposing the ZP to up to three 20- to 50-ms laser pulses. The cumulated embryo score (CES), which takes into account the number of transferred


359

Table 2. Clinical results after transfer of control and laser drilled (+AH) frozen-thawed embryos according to the cumulated embryo score (CES) CES

=20

>20

Number of

Control

+AH

Patients Transfers Clinical pregnancies % Deliveries Babies born Transfers Clinical pregnancies % Deliveries Babies born

65 28 1 3.6 1 1 38 4 10.5 3 3

68 54 3 5.6 2 2 54 16 29.6 14 20

p

NS

0.0287

All patients had experienced at least two previous unsuccessful transfers of cryopreserved embryos before their inclusion in the study.

embryos and their grade, was calculated at the time of ET (CES>R(number of blastomeres¶grade)). An immunosuppressive and antibiotic treatment was prescribed for 7 days, from 2 days prior to ET to the 5th day following ET. Clinical pregnancies were recorded when a gestational sac with heart activity was visible by ultrasound 28–35 days after ET. Results are summarized in table 2. For transfers with low CES (=20), AH did not significantly improve the clinical pregnancy or implantation rates. When the CES was P20, both rates were increased significantly by AH, implantation rate from 3.6 to 14.7% and clinical pregnancy from 10.5 to 29.6%. Sixteen deliveries were obtained with the birth of 22 healthy children in the AH group, compared to 4 deliveries and 4 children in the control group.

Discussion The efficacy of the described set-up was first studied on mouse zygotes. Compared to other described laser systems, the 1.48-lm diode laser presents many advantages. Oocytes or embryos can be maintained in their usual culture dish and medium [25] without requiring special slides and optics as for the UV lasers [39, 40], or micromanipulators and microfibers as for the Er:YAG laser [24]. The drilling procedure is fast; depending on the ZP width and


360

embryo diameter, an opening can be achieved with one single shot (up to 50 ms) [25, 41]. Holding of the egg with a suction pipette is unnecessary for firing in safe conditions, as the drilling event is instantaneous compared to the sometimes observed slow rotating movement of the eggs. The other UV or IR lasers necessitate firing with train pulses which may last 100 times longer (more than 5 s). The diode laser energy deposition does not produce any visible mechanical side effects as microscope observation shows no displacement or rotation of the eggs following the laser irradiation. Transmission electron microscopy does not detect ultrastructural alterations in the ZP matrix or of the oocyte cytoplasm in the vicinity of the hole [27]. Just after firing, tiny fragments are often seen moving out of the drilled crater confirming that the hole results from a thermolytic process. The drilled hole is stable over time and persists in culture up to the hatching stage. When comparing the conditions in which holes of reproducible sizes could be generated by the diode laser in human versus mouse eggs, the only difference was that the irradiation time needed was about twice as long [27]. This is explained by the greater thickness of the human ZP (10–25 versus 6–10 lm) and by the larger diameter of the human egg (120–140 versus 60–80 lm). Nervertheless, the energy level needed for human eggs is still in a range where no thermal side effects have been observed in the mouse. The work on mice has also shown that higher pulse durations are required for drilling holes of a given diameter in zygotes compared to oocytes [26]. Similar observations were made on the human [27]. This fact demonstrates that the biochemical and structural transformations induced by the cortical reaction in the ZP can be quantified by the laser drilling procedure. Acid microdrilling has already been proposed as a means for quantifying zona hardening in human embryos [21]. Contrary to the subjective appreciation of zona resistancy to acid dissolution, the laser energy needed to drill a hole of a given size constitutes a more precise way of quantifying the zona hardeness. By extension this parameter could help reveal secondary ZP hardening as it is suspected to occur in older patients or following in vitro culture (personal observation, unpublished data). Hatching of the blastocyst out of the ZP is an essential step which has to occur prior to uterine implantation. The poor implantation rate following tranfer of apparently normal looking embryos is one of the unsolved problems encountered during IVF. Beside intrinsic abnormalities of the embryo or defective uterine receptivity, a hatching failure could partly explain the low implantation rate. Blastocyst hatching might be impaired in some patients when the ZP is too thick or hardened by in vitro culture [42], hyaluronidase treatment for decoronization [11], freezing [12] or simply physiological aging [10]. It has been hypothesized that AH may enhance embryo implantation


361

Fig. 5. The 1.48-lm diode laser-assisted hatching unit. The compact FERTILASEÔ system, with its control unit on the right, is attached to the fluorescence port at the back of the inverted microscope.

not only by mechanically facilitating the hatching process but also by permitting earlier embryo-endometrium contact, improving the synchronization between embryo development and uterine receptivity which is mandatory for implantation efficiency. Partial zona dissection or acid drilling have been used for several years in clinical settings to improve pregnancy rates after embryo transfer. Cohen et al. [9] showed that the implantation rates of human embryos correlated to the ZP thickness. These authors also reported that 15% of IVF embryos presented a ZP wider than 15 lm thickness [5]. It has been shown that zona thinning alone is not sufficient to promote implantation, which suggests that the inner layer of the human ZP has to be fully breached [43]. Apart from a single study [31], all published reports stated that implantation is not improved by AH in nonselected IVF patient populations [9, 36–38]. However, in some subgroups of poor prognosis patients, i.e. with elevated basal FSH, repeated IVF failures, thick ZP or frozen-thawed embryo transfers, AH was repeatedly associated with higher implantation and pregnancy rates [9, 30, 32, 33, 36, 38, 44]. These last conclusions were also reached when AH was performed with Er:YAG [34] or PALM [35] systems. The presented 1.48-lm diode laser has an important advantage compared to other tools, because the drilling can be


362

done at early stages of embryo development (from day 1 on), which allows shorter in vitro culture. In the present study on frozen-thawed embryo, the drilling was performed on day 2 at a stage where the embryo is sensitive to chemical drilling. An enhanced pregnancy rate can be achieved in the group with laser AH, provided the transferred embryos have a CES of P20. First of all, this study indicates that the laser microdrilling at day 2 is safe even when performed at early stages of embryo development. Furthermore, the AH-induced enhancement of implantation suggests that the freeze/thaw procedure may indeed have a direct effect on the ZP hardness, at least for patients who have already experienced several unsuccessful transfers of cryopreserved embryos. As further studies are needed to get a comprehensive overview of the cases who might benefit from AH, a multicentric study, involving 4 different European IVF centers equipped with the Fertilase system (fig. 5), is currently underway. In conclusion, compared to other laser systems described, the 1.48-lm diode laser presented is the most suited for ZP microdissection and offers many advantages in clinical situations where the ZP needs to be opened. The benefits will be revealed by the various clinical studies that are presently in progress.

References 1 2 3 4

5

6 7 8

9

10

Plachot M: Viability of preimplantation embryos, Baillie`res Clin Obst Gynaecol 1992;6:327–338. Bongso A, Ng SC, Fong CY, Mok H, Ng PL, Ratnam SS: Cocultures in human assisted reproduction. Support of embryos in vitro and their specificity. Ann NY Acad Sci 1991;626:438–444. Bongso A, Ng SC, Fong CY, Ratnam S: Cocultures: A new lead in embryo quality improvement for assisted reproduction. Fertil Steril 1991;56:179–191. Wiemer KE, Cohen J, Amborski GF, Wright G, Wiker S, Munyakazi L, Godke RA: In-vitro development and implantation of human embryos following culture on fetal bovine uterine fibroblast cells. Hum Reprod 1989;4:595–600. Cohen J, Elsner C, Kort H, Malter H, Massey J, Mayer MP, Wiemer K: Impairment of the hatching process following IVF in the human and improvement of implantation by assisting hatching using micromanipulation. Hum Reprod 1990;5:7–13. Yaron Y, Botchan A, Amit A, Peyser MR, David MP, Lessing JB: Endometrial receptivity in the light of modern assisted reproductive technologies. Fertil Steril 1994;62:225–232. Turnbull LW, Lesny P, Killick SR: Assessment of uterine receptivity prior to embryo transfer: A review of currently available imaging modalities. Hum Reprod Update 1995;1:505–514. Schwartz LB, Chiu AS, Courtney M, Krey L, Schmidt-Sarosi C: The embryo versus endometrium controversy revisited as it relates to predicting pregnancy outcome in in-vitro fertilization-embryo transfer cycles. Hum Reprod 1997;12:45–50. Cohen J, Alikani M, Trowbridge J, Rosenwaks Z: Implantation enhancement by selective assisted hatching using zona drilling of human embryos with poor prognosis. Hum Reprod 1992;7:685– 691. Loret De Mola JR, Garside WT, Bucci J, Tureck RW, Heyner S: Analysis of the human zona pellucida during culture: Correlation with diagnosis and the preovulatory hormonal environment. J Assist Reprod Genet 1997;14:332–336.


363

11 12

13

14 15

16 17 18 19 20 21 22 23 24 25 26

27

28

29 30 31 32 33 34

De Felici M, Siracusa G: Spontaneous hardening of the zona pellucida of mouse oocytes during in vitro culture. Gamete Res 1982;6:107–113. Drobnis EZ, Andrew JB, Katz DF: Biophysical properties of the zona pellucida measured by capillary suction: Is zona hardening a mechanical phenomenon? J Exp Zool 1988;245:206– 219. Carroll J, Depypere H, Matthews CD: Freeze-thaw induced changes of the zona pellucida explains decreased rates of fertilization in frozen-thawed mouse oocytes. J Reprod Fertil 1990;90:547– 553. Gordon JW, Dapunt U: Restoration of normal implantation rates in mouse embryos with a hatching impairment by use of a new method of assisted hatching. Fertil Steril 1993;59:1302–1307. Depypere HT, McLaughlin KJ, Seamark RF, Warnes GM, Matthews CD: Comparison of zona cutting and zona drilling as techniques for assisted fertilization in the mouse. J Reprod Fertil 1988; 84:205–211. Odawara Y, Lopata A: A zona opening procedure for improving in vitro fertilization at low sperm concentrations: A mouse model. Fertil Steril 1989;51:699–704. Malter HE, Cohen J: Partial zona dissection of the human oocyte: A nontraumatic method using micromanipulation to assist zona pellucida penetration. Fertil Steril 1989;51:139–148. Gordon JW, Talansky BE: Assisted fertilization by zona drilling: A mouse model for correction of oligospermia. J Exp Zool 1986;239:347–354. Garrisi GJ, Talansky BE, Grunfeld L, Sapira V, Navot D, Gordon JW: Clinical evaluation of three approaches to micromanipulation-assisted fertilization. Fertil Steril 1990;54:671–677. Cohen J, Feldberg D: Effects of the size and number of zona pellucida openings on hatching and trophoblast outgrowth in the mouse embryo. Mol Reprod Dev 1991;30:70–78. Cohen J, Alikani M, Reing AM, Ferrara TA, Trowbridge J, Tucker M: Selective assisted hatching of human embryos. Ann Acad Med Singapore 1992;21:565–570. Tadir Y, Wright WH, Vafa O, Liaw LH, Asch R, Berns MW: Micromanipulation of gametes using laser microbeams. Hum Reprod 1991;6:1011–1016. Palanker D, Ohad S, Lewis A, Simon A, Schenker JG, Penchas S, Laufer N: Technique for cellular microsurgery using the 193-nm excimer laser. Lasers Surg Med 1991;11:580–586. Strohmer H, Feichtinger W: Successful clinical application of laser for micromanipulation in an in vitro fertilization program. Fertil Steril 1992;58:212–214. Rink K, Delacretaz G, Salathe RP, Senn A, Nocera D, Germond M: 1.48 mm diode laser microdissection of the zona pellucida of mouse zygotes. SPIE 1994;2134A:412–422. Rink K, Delacretaz G, Salathe RP, Senn A, Nocera D, Germond M, De Grandi P, Fakan S: Noncontact microdrilling of mouse zona pellucida with an objective-delivered 1.48-microns diode laser. Lasers Surg Med 1996;18:52–62. Germond M, Nocera D, Senn A, Rink K, Delacretaz G, Fakan S: Microdissection of mouse and human zona pellucida using a 1.48-microns diode laser beam: Efficacy and safety of the procedure. Fertil Steril 1995;64:604–611. Germond M, Nocera D, Senn A, Rink K, Delacretaz G, Pedrazzini T, Hornung JP: Improved fertilization and implantation rates after non-touch zona pellucida microdrilling of mouse oocytes with a 1.48 microm diode laser beam. Hum Reprod 1996;11:1043–1048. Schoolcraft WB, Schlenker T, Jones GS, Jones HW Jr: In vitro fertilization in women age 40 and older: The impact of assisted hatching. J Assist Reprod Genet 1995;12:581–584. Schoolcraft WB, Schlenker T, Gee M, Jones GS, Jones HW Jr: Assisted hatching in the treatment of poor prognosis in vitro fertilization candidates. Fertil Steril 1994;62:551–554. Hu Y, Hoffman DI, Maxson WS, Ory SJ: Clinical application of nonselective assisted hatching of human embryos. Fertil Steril 1996;66:991–994. Check JH, Hoover L, Nazari A, O’Shaughnessy A, Summers D: The effect of assisted hatching on pregnancy rates after frozen embryo transfer. Fertil Steril 1996;65:254–257. Tao J, Tamis R: Application of assisted hatching for 2-day-old, frozen-thawed embryo transfer in a poor-prognosis population. J Assist Reprod Genet 1997;14:128–130. Obruca A, Strohmer H, Sakkas D, Menezo Y, Kogosowski A, Barak Y, Feichtinger W: Use of lasers in assisted fertilization and hatching. Hum Reprod 1994;9:1723–1726.


364

35

36

37

38

39 40 41 42 43

44

Antinori S, Selman HA, Caffa B, Panci C, Dani GL, Versaci C: Zona opening of human embryos using a non-contact UV laser for assisted hatching in patients with poor prognosis of pregnancy. Hum Reprod 1996;11:2488–2492. Stein A, Rufas O, Amit S, Avrech O, Pinkas H, Ovadia J, Fisch B: Assisted hatching by partial zona dissection of human pre-embryos in patients with recurrent implantation failure after in vitro fertilization. Fertil Steril 1995;63:838–841. Hellebaut S, De Sutter P, Dozortsev D, Onghena A, Qian C, Dhont M: Does assisted hatching improve implantation rates after in vitro fertilization or intracytoplasmic sperm injection in all patients? A prospective randomized study. J Assist Reprod Genet 1996;13:19–22. Tucker MJ, Morton PC, Wright G, Ingargiola PE, Sweitzer CL, Elsner CW, Mitchell-Leef DE, Massey JB: Enhancement of outcome from intracytoplasmic sperm injection: Does co-culture or assisted hatching improve implantation rates? Hum Reprod 1996;11:2434–2437. Laufer N, Palanker D, Shufaro Y, Safran A, Simon A, Lewis A: The efficacy and safety of zona pellucida drilling by a 193-nm excimer laser. Fertil Steril 1993;59:889–895. Schutze K, Clement-Sengewald A, Ashkin A: Zona drilling and sperm insertion with combined laser microbeam and optical tweezers. Fertil Steril 1994;61:783–786. Rink K, Delacretaz G, Salathe RP, Senn A, Nocera D, Germond M: Laser surgery at the micrometer scale: Possibilities and limits. SPIE 1994;2323:262–272. Wassarman PM, Liu C, Litscher ES: Constructing the mammalian egg zona pellucida: Some new pieces of an old puzzle. J Cell Sci 1996;109:2001–2004. Tucker MJ, Luecke NM, Wiker SR, Wright G: Chemical removal of the outside of the zona pellucida of day 3 human embryos has no impact on implantation rate. J Assist Reprod Genet 1993;10: 187–191. Chao KH, Chen SU, Chen HF, Wu MY, Yang YS, Ho HN: Assisted hatching increases the implantation and pregnancy rate of in vitro fertilization (IVF)-embryo transfer (ET), but not that of IVF-tubal ET in patients with repeated IVF failures. Fertil Steril 1997;67:904–908.

Marc Germond, MD, PD, MER, Reproductive Medicine Unit, Department of Obstetrics and Gynaecology, CHUV, CH–1011 Lausanne (Switzerland) Tel. +41 21 314 32 76, Fax +41 21 314 32 74, E-Mail [email protected]


365


............................

Er:YAG (2,940 lm) Laser for Assisted Hatching Andreas Obruca, Wilfried Feichtinger Institute for Sterility Treatment, Vienna, Austria

Introduction A variety of laser systems have been considered for microsurgical procedures on human gametes. Thus, attention has focused on the possibility of using infrared laser optical traps to capture or transport spermatozoa [1, 2] or to measure their velocity [3]. Moreover, numerous recent experiments have attempted to use laser and light delivery systems instead of the chemical or mechanical methods currently in use to achieve partial zona dissection by localized thinning or opening the zona pellucida for purposes of microsurgical fertilization [4] and as means of preimplantation-assisted hatching following in vitro fertilization (IVF) [5]. The majority of previous experimental studies, most of which have been performed on animal cells, have used ultraviolet laser systems at 193- to 532-nm wavelengths [6–9] or, less frequently, infrared lasers. Taking into account the prerequisites for the application of lasers in human embryology [12] the following 4 parameters are recommended: (1) absolute avoidance of thermic effects; (2) prevention of genetic damage by a wavelength sufficiently distant from the absorption maximum of DNA; (3) low ablation threshold to ensure precision and to minimize mechanical vibration, and (4) easy handling using existing micromanipulation equipment as short-time manipulations are always the more efficient. Special laser equipment fulfilling all these criteria has been developed in our laboratory [13, 14]. The system consisted of an erbium-yttrium-aluminiumgarnet (ER:YAG) laser (Lisa Laser Products, Katlenburg, Germany) and a laser fiber of 20 lm tip diameter. The application of low energy resulted in photoablation – a nonthermic effect of the laser on biological tissue. A wavelength of 2.9 lm was selected to avoid damaging the genetic structure of the embryo [13], yet offering a high absorption in water (fig. 1).

Fig. 1. Energy absorption in water related to different wavelength.

At the 2.9-lm wavelength, the penetration depth in biological tissue is about 3 lm, creating an ablation of the zona pellucida in layers by laser pulses of approximately 10 lJ energy. Additionally the specially developed glass fiber allowed a procedure similar to conventional mechanical partial zona dissection. The hatching procedure was carried out on an inverted microscope (Diaphot, Nikon, Tokyo, Japan) equipped with a heated stage and two hydraulic micromanipulators (Narishige, Tokyo, Japan). All embryos to be treated were placed simultaneously on a microscope slide and covered with paraffin oil. During the procedure the embryo was held by negative pressure using a holding pipette of 130 lm outer and 20 lm inner diameter. The glassfiber, fitted to the manipulator by pipette holder, was brought into direct contact with the zona, but the tip was not forced further into it, to prevent any contact of the fiber with the surface of the blastomere once it penetrated the zona (fig. 2). Five to eight pulses were necessary to penetrate the zona, creating a 20- to 30-lm opening. We can easily control and visualize the efficiency of the system by inserting the fiber tip into the hole. In a mouse experiment we showed that laser-assisted hatching (AHA) shortens the hatching time by comparison with a nonhatched control group. At day 4.5 after hCG, 80% of the embryos has already started the hatching process in the laser-treated group vs. 29.3% in the control group [15]. In another study we tried to evaluate the ultrastructural effects of this laser on the zona pellucida and the underlying cell membrane of unfertilized

Er:YAG (2,940 lm) Laser for Assisted Hatching

367

Fig. 2. The embryo is held by the holding pipette while the laser fiber is brought into direct contact with the zona.

human oocytes and pathologically fertilized preimplantation embryos using light and electron microscopy. The Er:YAG laser produces an almost circular zona opening in the shape of a truncated cone tapering off towards the inside, with a mean diameter of 18 lm. The exact diameter of the drilled site depends on the diameter of the fiber tip and the total number of pulses applied. After laser interaction, the zona matrix and the surface of the underlying ooplasm membrane showed no degenerative alterations. We conclude that the Er:YAG laser is an effective microsurgical tool for achieving reproducible, precise zona openings particularly suitable for purposes of AHA because of their characteristic shape [16].

Results Assisted Hatching in Repeated IVF Failures We present the overall results of AHA by Er:YAG laser between 1992 and 1997. AHA was performed on 331 patients with repeated IVF failures. The results were compared to a control group (n>1,978) consisting of IVF patients without any micromanipulation procedure treated during the same period. There were no differences for either group in the mean number of transferred embryos, but the mean age was significantly higher in the lasertreated group (34.3×4.9 vs. 32.9×4.8 years; table 1). Laser AHA increased the clinical pregnancy rate significantly (laser AHA 37.2 vs. without AHA 31.5%; table 2).

Obruca/Feichtinger

368

Table 1. Age distribution IVF patients

Age, years

Without AHA Er:YAG laser AHA

32.9 34.3

AHA vs. non-AHA: p=0.0001.

Table 2. Results of assisted hatching IVF patients

Clinical pregnancy rate %

Without AHA 31.5 (622/1,978) Er:YAG laser AHA 37.2 (123/331) p=0.05.

Discussion Our results show that in human IVF, laser AHA increased the pregnancy rate per embryo transfer. Nevertheless the discussion of the value of AHA is still controversial. On one hand ‘assisted hatching is considered a valuable addition to the clinical armamentarium of IVF clinics. Because of suboptimal culture conditions, embryo cryopreservation, the increasing presence of older patients seeking our care, assisted hatching is vital to maximize the prognosis of implantation of human embryos’ [17]. On the other hand some authors question the potential benefit of AHA. Tucker [18] claims that with optimization of oocytes and embryo environments any real need for AHA will disappear. However, at the moment AHA is a useful tool in optimizing results, especially in a selected group of patients with a poorer prognosis. However, there is not only discussion on the overall benefit of AHA but also on the use of the various techniques. Mechanical partial zona dissection or performed chemically by acidic Tyrode solution are widely used compared to scarce data on laser AHA in humans. In a study by Cohen and Feldberg [19] it was demonstrated that complete hatching and normal trophoblast outgrowth depend on the size, shape, and number of zona pellucida openings. Thus, mechanical partial zona dissection, producing narrow incisions p5 lm in diameter similar to those obtained by lasers delivered tangentially, led to complete hatching in only 16% of embryos,


369

whereas the remainder were trapped in a typical figure-eight shape. In contrast, zona drilling by acidified Tyrode solution, creating larger, round holes 11–25 lm in diameter comparable with the openings produced by our Er:YAG laser, allowed normal hatching in 92% of embryos [19]. A subsequent study [20] performing AHA in mouse embryos zona-drilled with acidified Tyrode solution or lased by means of a 308-nm noncontact excimer laser showed that significantly more embryos were hatching in the chemically drilled group (50%) than in the laser-treated group (24%). Although direct comparisons between contact and noncontact laser systems are lack ing, it would appear from these results that large, round holes produced by our contact laser are more effective for promoting complete embryo hatching. The advantages of applying lasers in the noncontact mode are the following: no need for disposable fibers and for sterilization possibility of intracellular application and more simple procedure. However, when using wavelengths in the ultraviolet range potential hazards to the genetic structure are possible. On the other hand the question of heat deposition using systems in the low infrared range has yet to be determined, especially by creating large gaps in the zona because they require longer exposure and higher laser energies. In conclusion Er:YAG laser AHA offers an alternative for repeated failures of embryo transfer. It can be considered as safe and efficient as shown by the children born after using this method [15]. However, a randomized comparison between a contract and noncontact laser system is missing to confirm the effectiveness of both systems.

References 1 2 3

4 5

6 7 8

Schu¨tze K, Clement-Sengewald A, Ashkin A: Zona drilling and sperm insertion with combined laser microbeam and optical tweezers. Feril Steril 1994;61:783–786. Tadir Y, Wright WR, Vafa O, Ord T, Asch RH, Berns MW: Micromanipulation of sperm by a laser generated optical trap. Fertil Steril 1989;52:870–873. Colon JM, Sarosi, P, McGovern PG, Askin A, Dziedzic JM, Skurnick J, Weiss G, Bonder EM: Controlled micromanipulation of human sperm in three dimensions with an infrared laser optical trap: Effect on sperm velocity. Fertil Steril 1992;57:695–698. Cohen J, Malter H, Fehilly C, Wright G, Elsner C, Kort H, Massey J: Implantation of embryos after partial opening of oocyte zona pellucida to facilitate sperm penetration. Lancet 1988;ii:162. Cohen J, Elsner C, Kort H, Malter H, Massey J, Mayer MP, Wiemer K: Impairment of the hatching process following ICF in the human and improvement of implantation by assisted hatching using micromanipulation. Hum Reprod 1990;5:7–13. Blanchet GB, Russell JB, Fincher CR, Portmann M: Laser micromanipulation in the mouse embryo: A novel approach to zona drilling. Fertil Steril 1992;57:1337–1341. Laufer N, Palanker D, Shufaro Y, Safran A, Simon A, and Lewis A: The efficacy and safety of zona pellucida drilling by a 193-nm excimer laser. Fertil Steril 1993;59:889–895. Simon A, Palanker D, Harpaz-Eisenberg V, Lewis A, Laufer N: Interaction between human sperm cells and hamster oocytes after argon fluoride excimer laser drilling on the zona pellucida. Fertil Steril 1993;60:159–164.

Obruca/Feichtinger

370

9 10

11

12 13 14

15 16

17 18 19 20

Tadir Y, Wright WH, Vafa O, Liaw LH, Asch A, Berns MW: Micromanipulation of gametes using laser microbeams. Hum Reprod 1991;6:1011–1016. Coddington CC, Veeck LL, Swanson RJ, Kaufman RA, Lin J, Simonetti S, Bocca S: The YAG laser used in micromanipulation to transect the zona pellucida of hamster oocytes. J Assist Reprod Genet 1992;9:557–563. Germond M, Nocera D, Senn A, Rink K, Delacre´taz G, Fakan S: Microdissection of mouse and human zona pellucida using a 1.48-lm diode laser beam: Efficacy and safety of the procedure. Fertil Steril 1995;64:604–611. Strohmer H, Feichtinger W: Application of laser for micromanipulation: Relevance of biophysical criteria. Fertil Steril 1992;48:540. Strohmer H, Feichtinger W: Successful clinical application of laser for micromanipulation in an in vitro fertilization program. Fertil Steril 1992;58:212–214. Feichtinger W, Strohmer H, Fuhrberg P, Radivojevic K, Antinori S, Pepe G, Versaci C: Photoablation of oocyte zona pellucida by erbium-yag laser for in-vitro fertilization in severe male infertility. Lancet 1992;339:811. Obruca A, Strohmer H, Sakkas D, Menezo Y, Kogosowski A, Barak Y, Feichtinger W: Use of lasers in assisted fertilization and hatching. Hum Reprod 1994;9:1723–1726. Obruca A, Strohmer H, Blaschitz A, Scho¨nickle E, Dohr G, Feichtinger W: Ultrastructural observations in human oocytes and preimplantation embryos after zona opening using an erbium-yttriumaluminium-garnet (Er:YAG) laser. Hum Reprod 1997;10:2242–2245. Schoolcraft WB: Assisted hatching improves the implantation rate. J Assist Reprod 10th World Congr on IVF-AR, Vancouver, 1997 Tucker M: Assisted hatching: A therapy of questionable value. J Assist Reprod 10th World Congr on ICF-AR, Vancouver, 1997. Cohen J, Feldberg, D: Effects of the size and number of zona pellucida openings on hatching and trophoblast outgrowth in the mouse embryo. Mol Reprod Dev 1991;30:70–78. Neev J, Gonzalez A, Licciardi F, Alikani M, Tadir Y, Berns M, Cohen J: Opening of the mouse zona pellucida by laser without a micromanipulator. Hum Reprod 1993;8:939–944.

Andreas Obruca, MD, Department of Obstetrics and Gynecology, University of Vienna Medical School, Wa¨hringer Gu¨rtel 18–20, A–1090 Vienna (Austria) Tel. +43 1 40400 2816, Fax +43 1 40400 2817, E-Mail [email protected]


371

............................

Author Index

Akgu¨n, N. 39 Akiya, T. 270 Ansell, J.K. 147 Bays, R. 243 Beck, G. 39 Bergh, H. van den 243 Berns, M.W. 183, 251 Boso, C. 278 Brown, S.B. 219 Butler, J. 116 Carre´, J. 169 Chapman, C.F. 251 Clement-Sengewald, A. 340 Coquoz, O. 116 Corti, L. 278 Delacre´taz, G. 352 Delaney, T.F. 285 Descloux, L. 352 Dickson, E. 302 Dobler, D. 251 Eleouet, S. 169 Fehr, M.K. 96, 227, 234, 251, 322 Feichtinger, W. 366 Fishkin, J.B. 116 Foster, W. 213 Fraker, D. 285

Gannon, M.J. 176, 219 Germond, M. 352 Glatstein, E. 285 Grahn, M.F. 147 Greer, P.A. 302 Hahn, P.M. 302 Hahn, S.M. 285 Haller, U. 1, 183, 234, 243, 251 Heckelsmiller, K. 39 Hillemanns, P. 308 Holroyd, J.A. 219 Hornung, R. 234, 246, 322 Hsi, R.A. 285 Iwabuchi, H. 270 Jode, M.L. de 147 Johnson, N. 219 Kawasaki, K. 270 Kimel, S. 14 Ko¨nig, K. 86 Korell, M. 308 Krasieva, T.B. 133, 183, 251 Krzemien, A. 213, 302 Kunugi, T. 270 Kunzi-Rapp, K. 39 Lajat, Y. 169

Lavie, G. 14 Lundhal, S. 213 Major, A.L. 96 Marcus, S. 213 Melchior, M.F. 302 Mosimann, L. 243 Muroya, T. 270 Nevinny, C. 340 Obruca, A. 366 Orenstein, A. 14 Patrice, T. 169 Po¨sl, H. 340 Pottier, R.H. 302 Primi, M.-P. 352 Reid, R.L. 176, 213, 302 Richter, C. 157 Rink, K. 352 Rousset, N. 169 Rubin, S.C. 285 Ru¨ck, A. 39 Sakamoto, M. 270 Sakunaga, H. 270 Sandow, S. 340 Schu¨tze, K. 340 Schmidt, S. 265, 296, 312, 316

372

Schwarz, V. 251, 322 Senn, A. 352 Sindelar, W.F. 285 Spaniol, S. 265 Steiner, R.A. 176, 183 Stringer, M. 219 Sugishita, T. 270 Sun, C.H. 76 Suppan, P. 53 Svaasand, L.O. 76, 96, 116

Author Index

Tadir, Y. 96, 176, 183, 227, 234, 246, 251, 326 Tenjin, Y. 270 Tromberg, B.J. 116, 183, 246, 251 Van Dijk, J.P. 302 Van Vugt, D.A. 213, 302 Vernon, D.I. 219 Vonarx, V. 169

Wagner, U. 296 Wagnie`res, G. 243 Woodtli, A. 243 Wyss, P. 4, 12, 76, 96, 176, 183, 206, 227, 234, 243, 251, 312, 322 Wyss-Desserich, M.-T. 76, 183 Yang, J.Z. 302

373

............................

Subject Index

Acridine dye, paramecium toxicity 5, 6 Al(III)phthalocyanine tetrasulfonate accumulation in vascular and tumor cell systems 43–45 aggregation 26–28 photokinetics, subcellular 45, 46 spectroscopic studies 26 structure-activity relationships 25, 26, 28 5-Aminolevulinic acid active compound, see Protoporphyrin IX cervical dysplasia photodynamic therapy irradiation technique 266 light diffuser 266, 267 outcome 267, 268 topical application 265 ectopic pregnancy photodynamic therapy studies mouse 304 rat 302–304 endogenous levels 208 endometrial ablation menorrhagia treatment advantages and disadvantages 225 bleeding outcomes 236, 237, 240 clinical studies 223, 224, 234–237 light applicators 243, 244, 246–248, 250 morphological analysis 235, 236 overview 176, 177, 179, 180, 238, 239 parameters for clinical use 225, 226 patient selection 240, 241 perfused uterus 219, 220

photosensitization in vivo 220–222 regeneration of tissue 239, 240 ultrasound findings 235, 236 monkey studies endometrial suppression and progestin withdrawal 214 histopathology 214–216 segments of ablation 216, 217 technique 213, 214, 217 pharmacokinetics of 5-aminolevulinic acid and protoporphyrin IX in uterus endoluminal diffusion 232, 233 fluorescence microscopy 228, 229, 231 specimen retrieval 228 study design 227, 228 topical drug application 228 urine analysis 229, 232 rat and rabbit studies heat generation 198, 204 histology 185, 192–194, 201 laser delivery system 186, 187 macroscopic changes 191 pharmacokinetics 184, 185, 187, 189, 199, 200 reproductive performance assessment 186, 195 skin photosensitivity 186 temperature measurements 187 endometriosis diagnostics advantages 310, 311

374

laparoscopy and fluorescence detection 309, 310 rationale 308, 309 photodynamic therapy studies rabbit 305 rat 305, 306 fluorescence studies of induced protoporphyrin IX in gynecological cell lines cell culture 76, 77 flow cytometry measurements 77 incubation of 5-aminolevulinic acid 77 intensity of fluorescence 78, 82 laser treatment 77 photobleaching 79, 80, 82, 83 photodynamic effect 83, 84 phototoxicity towards cells and mitochondria 80, 81 viability assays 77, 78 metabolism biosynthesis 208 excretion 209 selectivity of tissues 206 tumors 169 toxicology 209, 210 vulvar condyloma photodynamic therapy fluorescence imaging 262, 263 outcomes 258, 261 principle 251 rationale 251, 252 selective photosensitization 252–254, 256, 257, 263 technique 254–256 vulvar intraepithelial neoplasia photodynamic therapy fluorescence imaging 262, 263 outcomes 261 principle 251 technique 254–256 Apoptosis definition 157, 158 mediators 159 mitochondria targeting for induction Bcl-2 role 159, 163, 164 calcium release role homeostasis 161 overview 157–159

Subject Index

pathways for release 161 peroxynitrite stimulation 162, 163 prooxidant-induced, NAD+-linked release 162 membrane potential in cancer cells 164, 165 nitric oxide role biology 160 cytochrome oxidase regulation 160, 163 overview 157–159, 164 synthesis 160, 161 tetra(m-hydroxyphenyl)chlorin damage 158, 164, 165 Arrhenius-Eyring model, chemical reactions 53 Assisted reproductive technologies, see Photon-assisted reproduction Bcl-2, apoptosis role 159, 163, 164 Beer’s law 102, 120 Benzoporphyrin derivative mono acid, endometrial ablation studies in rat and rabbit heat generation 198, 204 histology 185, 192–194, 201 laser delivery system 186, 187 macroscopic changes 191 pharmacokinetics 184, 185, 189 reproductive performance assessment 186, 195, 202, 203 skin photosensitivity 186 temperature measurements 187 Breast cancer epidemiology 312 mTHPC photodynamic therapy for recurrent disease advantages 322 outcome 324, 325 study design 322–324 optical characterization with frequency-domain photon migration 124, 125, 127–130 photodynamic therapy, history and overview 312, 313

375

Breast Cancer (continued) SnET2 photodynamic therapy for skin metastases outcomes 319, 320 patient selection 317, 318 study design 316, 317 treatment modality 318, 319 treatment overview 312

uptake in vivo 22 vascular injury 22, 23 ZnPc accumulation 43–45 CIN, see Cervical dysplasia Condylomata acuminata, see Vulvar condyloma Confocal microscopy, subcellular localization of photosensitizers 41, 42

Calcium, apoptosis role via mitochondria homeostasis 161 overview 157–159 pathways for mitochondrial release 161 peroxynitrite stimulation 162, 163 prooxidant-induced, NAD+-linked release 162 CD44V, photodynamic therapy effects on expression 175 Cell adhesion, photodynamic therapy effects 174, 175 Cercosporin, photosensitizer features 32, 33 Cervical carcinoma in situ Porfimer sodium photodynamic therapy advantages 276 outcomes 272, 274–276 prospects 276, 277 technique 270, 271 recurrent cancer treatment with photodynamic therapy 278–283 Cervical dysplasia 5-aminolevulinic acid photodynamic therapy irradiation technique 266 light diffuser 266, 267 outcome 267, 268 topical application 265 clinical findings 265 Porfimer sodium photodynamic therapy advantages 276 outcomes 272, 274–276 prospects 276, 277 technique 270, 271 treatment options 268 Chorioallantoic membrane phthalocyanine accumulation 43–45 porphycenes cell culture 21, 22

Diode laser, 1480-nm, laser-assisted hatching advantages and overview 334, 335, 356, 360, 361, 363 drilling technique 354–356 human studies 335, 359, 360 instrumentation 353, 354 mouse studies 335, 356–359, 361 DNA lesions, photodynamic therapy induction 171, 172 Dosimetry, see Optical dosimetry

Subject Index

Ectopic pregnancy 5-aminolevulinic acid photodynamic therapy studies mouse 304 rat 302–304 diagnosis and conventional treatment 302 Endometrial ablation menorrhagia treatment with 5-aminolevulinic acid advantages and disadvantages 225 bleeding outcomes 236, 237, 240 clinical studies 223, 224, 234–237 light applicators 243, 244, 246–248, 250 morphological analysis 235, 236 overview 176, 177, 179, 180, 238, 239 parameters for clinical use 225, 226 patient selection 240, 241 perfused uterus 219, 220 photosensitization in vivo 220–222 regeneration of tissue 239, 240 ultrasound findings 235, 236 monkey studies with 5-aminolevulinic acid endometrial suppression and progestin withdrawal 214 histopathology 214–216 segments of ablation 216, 217 technique 213, 214, 217

376

pharmacokinetics of 5-aminolevulinic acid and protoporphyrin IX in uterus endoluminal diffusion 232, 233 fluorescence microscopy 228, 229, 231 specimen retrieval 228 study design 227, 228 topical drug application 228 urine analysis 229, 232 photodynamic therapy, overview advantages 177 history of study 177, 179, 180 photosensitizers 177 photosensitizer studies in rat and rabbit endometrium 5-aminolevulinic acid 183, 187, 189, 195, 199–202 benzoporphyrin derivative mono acid 183, 184, 189, 195, 202, 203 heat generation 198, 204 histology 185, 192–194, 201 laser delivery system 186, 187 macroscopic changes 191 pharmacokinetics 184, 185, 187, 189, 190, 199–201 Photofrin 184, 189, 195, 200 Photofrin+azone 190, 196, 198, 200, 201 reproductive performance assessment 186, 194–196, 202, 203 skin photosensitivity 186 temperature measurements 187 thermal techniques 176, 177 Endometrial cancer Protoporphyrin IX accumulation 240, 241 Endometriosis 5-aminolevulinic acid diagnostics advantages 310, 311 laparoscopy and fluorescence detection 309, 310 rationale 308, 309 5-aminolevulinic acid photodynamic therapy studies rabbit 305 rat 305, 306 clinical features 308

Subject Index

surgical resection 304 Endometrium, see endometrial ablation Er:YAG laser, 2,940-nm, laser-assisted hatching criteria for success 369, 370 human studies 334, 368–370 instrumentation 366, 367 mouse studies 334, 367 technique 367 Fluorescence imaging, see also Confocal microscopy algorithms 135–138 artifact sources 136 dark noise image 137, 138 history 134 image formation 136 instrumentation 134, 135 light distribution image 137,138 photodynamic therapy applications 140–142, 144, 145 saturation 139, 140 target tissue-to-normal tissue ratio calculation 140, 142, 144 Frequency-domain photon migration applications 117, 130 breast tumor characterization 124, 125, 127–130 instrumentation 122–124 principle 121, 122 Gamete micromanipulation, see Photon-assisted reproduction Graft vs host disease, hypericin therapy 31, 32 Hatching failure, see Photon-assisted reproduction Hematoporphyrin history of use 6–8 photoproduct formation 86, 87, 91–95 photoreaction 86 recurrent cancer treatment with photodynamic therapy 278–283 Hematoporphyrin derivative cancer treatment 8, 9, 14 development 8, 9, 14

377

Hematoporphyrin derivative (continued) intraperitoneal photodynamic therapy for recurrent cancer treatment, preclinical studies 287–289 limitations 15 photoproduct formation 86, 87, 91–95 photoreaction 86 History, photomedicine clinical applications 5–9 Egypt 4, 5 India 4 Ho:YAG laser, laser-assisted hatching 333 Ho:YSGG laser, laser-assisted hatching 333, 334 Human papillomavirus, see Cervical dysplasia, Vulvar condyloma Hypericin absorption properties 29 applications and mechanisms of action 30–32 binding proteins 32 charge 29 lipophilicity 29 phototoxicity 29, 30 protein kinase C inhibition 31 sources 28, 29 Hypocrellins, photosensitizer features 32, 33 Immunosuppression, light irradiation 174 Intraperitoneal photodynamic therapy ovarian cancer cytoreduction with phthalocyanin-tagged antibody advantages 300, 301 irradiation technique 298 operative procedure 297, 298 outcomes 298, 300 patient selection 297 rationale 296, 297 recurrent ovarian cancer limitations 286 Porfimer sodium trials 289–292 preclinical studies 287–289, 293 prospects 292, 293 rationale 285 In vitro fertilization, see Photon-assisted reproduction

Subject Index

Laser microbeam microdissection, single-cell prenatal diagnosis 347–350 Laser pressure catapulting, single-cell prenatal diagnosis 348–350 Laser-assisted hatching diode laser, 1480-nm advantages and overview 334, 335, 356, 360, 361, 363 drilling technique 354–356 human studies 335, 359, 360 instrumentation 353, 354 mouse studies 335, 356–359, 361 Er:YAG laser, 2,940-nm criteria for success 369, 370 human studies 334, 368–370 instrumentation 366, 367 mouse studies 334, 367 technique 367 hatching failure causes 352, 361, 362 lasers Ho:YAG 333 Ho:YSGG 333, 334 nitrogen laser 332, 333 XeCl 332 outcomes 331, 332 PALM Robot-Microbeam system 345 patient selection and indications 331, 335, 336 Light acceleration 96, 97 applicators for endometrial ablation via photodynamic therapy 243, 244, 246–248, 250 distribution in body cavities 107–109, 114, 115 electromagnetic fields 96 frequency 96 medical applications 97, 116 optical energy flux 121 photon energies 97 pressure 98, 99 propagation matter 99, 100 tissue 100–105, 107, 117–121 radiant energy fluence rate 121 scattering 99–105, 117–121 solar radiation features 98

378

Membrane, photodynamic therapy targets 172, 173 Menorrhagia, treatment with 5-aminolevulinic acid photodynamic therapy advantages and disadvantages 225 bleeding outcomes 236, 237, 240 clinical studies 223, 224, 234–237 light applicators 243, 244, 246–248, 250 morphological analysis 235, 236 overview 176, 177, 179, 180, 238, 239 parameters for clinical use 225, 226 patient selection 240, 241 perfused uterus 219–220 photosensitization in vivo 220–222 regeneration of tissue 239, 240 ultrasound findings 235, 236 8-Methoxypsoralen, history of use 4 Methylene blue cytotoxicity mechanisms 47 nonlinear dynamics during photodynamic therapy 48–50 pH effects in photodynamic therapy 49 redox potentials 46, 47 Micromanipulation, see Photon-assisted reproduction Mitochondria membrane targeting in photodynamic therapy 172 phototherapy targeting for apoptosis induction Bcl-2 role 159, 163, 164 calcium release role homeostasis 161 overview 157–159 pathways for release 161 peroxynitrite stimulation 162, 163 prooxidant-induced, NAD+-linked release 162 membrane potential in cancer cells 164, 165 nitric oxide role biology 160 cytochrome oxidase regulation 160, 163 overview 157–159, 164 synthesis 160, 161

Subject Index

tetra(m-hydroxyphenyl)chlorin damage 158, 164, 165 Nitric oxide role, apoptosis role via mitochondria biology 160 cytochrome oxidase regulation 160, 163 overview 157–159, 164 peroxynitrite stimulation of calcium release 162, 163 synthesis 160, 161 Nitrogen laser, laser-assisted hatching 332, 333 Optical dosimetry temperature considerations 113 tissue vs cell suspensions 113 wavelength dependence of singlet oxygen generation 113, 114 Optical properties, see Frequency-domain photon migration Optical tweezers, see Photon-assisted reproduction Ovarian cancer cytoreduction with phthalocyanintargeted antibody advantages 300, 301 irradiation technique 298 operative procedure 297, 298 outcomes 298, 300 patient selection 297 rationale 296, 297 epidemiology 286, 296 intraperitoneal photodynamic therapy for recurrent cancer limitations 286 Porfimer sodium trials 289–292 preclinical studies 287–289, 293 prospects 292, 293 rationale 285 PALM Robot-Microbeam system blastomere fusion in mice 342 laser pressure catapulting 348–350 laser zona drilling cattle 344–347 mice 344

379

PALM Robot-Microbeam system, laser zona drilling (continued) laser-assisted hatching 345 principles 343 microbeam microdissection for single-cell prenatal diagnosis 347–350 overview 340, 341 sperm immobilization by tail-cutting 342 trapping with optical tweezers 345 Photobleaching 79, 80, 82, 83 Photochemical reaction bimolecular reactions overview 72, 73 photo-oxidation 74, 75 photoreduction 74 electron transfer 63, 64 excited states of molecules diatomic molecules 58 linear conjugated molecules 59, 60 light induction 54, 55 monophotonic vs multiphotonic process 64, 65 orbital localization 55–57 pathway 53, 54 primary vs secondary processes 65 principle of exclusion 57 quantum yield 66 quenching, see Quenching, excited states rate constants 66, 67 steps 65, 110, 112, 133, 134 unimolecular reactions acid-base equilibria 71, 72 dissociation reactions 68 intermediates 70, 71 rearrangements 68 stereoisomerizations 69, 70 valence isomerizations 70 yield 66 Photodynamic therapy cancer treatment, see specific cancers history, see History, photomedicine intraperitoneal, see Intraperitoneal photodynamic therapy overview 1–3

Subject Index

photosensitizers, see specific compounds primary vs secondary phototoxicity 39 principle 12, 13, 110, 133 reaction principles, see Photochemical reaction research centers in gynecology 2 Photofrin cervical cancer photodynamic therapy advantages 276 outcomes 272, 274–276 prospects 276, 277 technique 270, 271 endometrial ablation studies in rat and rabbit heat generation 198, 204 histology 185, 192–194 laser delivery system 186, 187 macroscopic changes 191 pharmacokinetics 184, 185, 189, 200 Photofrin+azone 190, 196, 198, 200, 201 reproductive performance assessment 186, 195 skin photosensitivity 186 temperature measurements 187 intraperitoneal photodynamic therapy for recurrent cancers 289–292 Photon-assisted reproduction laser-assisted hatching diode laser, 1480-nm advantages and overview 334, 335, 356, 360, 361, 363 drilling technique 354–356 human studies 335, 359, 360 instrumentation 353, 354 mouse studies 335, 356–359, 361 Er:YAG laser, 2,940-nm criteria for success 369, 370 human studies 334, 368–370 instrumentation 366, 367 mouse studies 334, 367 technique 367 hatching failure causes 352, 361, 362 lasers Ho:YAG 333 Ho:YSGG 333, 334

380

nitrogen laser 332, 333 XeCl 332 outcomes 331, 332 patient selection and indications 331, 335, 336 laser zona drilling contact mode 331 noncontact mode 330, 331 micromanipulation of gametes with lasers laser types 327 overview 326, 327 sperm 327–329 PALM Robot-Microbeam system blastomere fusion in mice 342 laser pressure catapulting 348–350 laser zona drilling cattle 344–347 mice 344 principles 343 laser-assisted hatching 345 microbeam microdissection for single-cell prenatal diagnosis 347–350 overview 340, 341 sperm immobilization by tail-cutting 342 sperm trapping with optical tweezers 345 Photoproducts 86, 87, 91–95 Photosensitizer, see also specific compounds absorption spectra features 16, 17 bleaching sensitivity 112 charge and tissue uptake 19 delivery systems 18 ideal criteria 15 lipophilicity and cellular internalization 18 stereochemistry 19 uptake and localization mechanisms 39–41 Phthalocyanines, see also Al(III)phthalocyanine tetrasulfonate, Zinc-phthalocyanine advantages 25 aggregation 26–28 ovarian cancer cytoreduction with phthalocyanin-targeted antibody

Subject Index

advantages 300, 301 irradiation technique 298 operative procedure 297, 298 outcomes 298, 300 patient selection 297 rationale 296, 297 spectroscopic studies 26 structure-activity relationships 25, 26, 28 Porfimer sodium, see Photofrin Porphycenes intraliposomal location 20, 21 photoinjury cell culture 21, 22 vascular injury 22, 23 structure 19, 20 structure-activity relationships 24 uptake in vivo 22 Protein kinase C, hypericin inhibition 31 Protoporphyrin IX cervical dysplasia photodynamic therapy irradiation technique 266 light diffuser 266, 267 outcome 267, 268 topical application 265 ectopic pregnancy photodynamic therapy studies mouse 304 rat 302–304 endometrial ablation menorrhagia treatment advantages and disadvantages 225 bleeding outcomes 236, 237, 240 clinical studies 223, 224, 234–237 light applicators 243, 244, 246–248, 250 morphological analysis 235, 236 overview 176, 177, 179, 180, 238, 239 parameters for clinical use 225, 226 patient selection 240, 241 perfused uterus 219–220 photosensitization in vivo 220–222 regeneration of tissue 239, 240 ultrasound findings 235, 236 monkey studies endometrial suppression and progestin withdrawal 214 histopathology 214–216

381

Protoporphyrin IX, monkey studies (continued) segments of ablation 216, 217 technique 213, 214, 217 pharmacokinetics in uterus endoluminal diffusion 232, 233 fluorescence microscopy 228, 229, 231 specimen retrieval 228 study design 227, 228 topical drug application 228 urinalysis 229, 232 rat and rabbit studies heat generation 198, 204 histology 185, 192–194, 201 laser delivery system 186, 187 macroscopic changes 191 pharmacokinetics 184, 185, 187, 189, 199, 200 reproductive performance assessment 186, 195 skin photosensitivity 186 temperature measurements 187 endometriosis diagnostics advantages 310, 311 laparoscopy and fluorescence detection 309, 310 rationale 308, 309 photodynamic therapy studies rabbit 305 rat 305, 306 gynecological cell lines, 5-aminolevulinic acid-induced fluorescence studies cell culture 76, 77 flow cytometry measurements 77 incubation of 5-aminolevulinic acid 77 intensity of fluorescence 78, 82 laser treatment 77 photobleaching 79, 80, 82, 83 photodynamic effect 83, 84 phototoxicity towards cells and mitochondria 80, 81 viability assays 77, 78 heme synthesis 206 metabolism biosynthesis 208

Subject Index

excretion 209 feedback control of accumulation 208 tumors 169 photoproduct formation 86, 87, 91–95 photoreaction 86 precursor, see 5-Aminolevulinic acid toxicology 209, 210 vulvar condyloma photodynamic therapy fluorescence imaging 262, 263 outcomes 258, 261 principle 251 rationale 251, 252 selective photosensitization 252–254, 256, 257, 263 technique 254–256 vulvar intraepithelial neoplasia photodynamic therapy fluorescence imaging 262, 263 outcomes 261 principle 251 technique 254–256 Quantum yield, photochemical reactions 66 Quenching, excited states definition 60, 61 electron transfer 63 energy transfer Dexter mechanism 61, 62 Fo¨rster mechanism 62 heavy atom effect 62 paramagnetic quenching 62 Rate constants, photochemical reactions 66, 67 Reactive oxygen species, tumor damage in photodynamic therapy 14, 40, 41, 110 Recurrent gynecologic cancer, treatment with photodynamic therapy cervix and endometrium 278–283 mTHPC for recurrent breast cancer advantages 322 outcome 324, 325 study design 322–324 ovarian cancer 285–293 Reproduction, see Photon-assisted reproduction

382

Robot-Microbeam system, see PALM Robot-Microbeam system Skin metastasis, photodynamic therapy for breast cancer metastases outcomes 319, 320 patient selection 317, 318 study design 316, 317 treatment modality 318, 319 SnET2 advantages in phototherapy 320 pharmacokinetics 316 photodynamic therapy for skin metastases outcomes 319, 320 patient selection 317, 318 study design 316, 317 treatment modality 318, 319 Sperm immobilization by tail-cutting with lasers 342 micromanipulation with lasers 327–329, 345 Target tissue-to-normal tissue ratio, calculation 140, 142, 144 Tetra(m-hydroxyphenyl)chlorin breast cancer, photodynamic therapy for recurrent disease advantages 322 outcome 324, 325 study design 322–324 mitochondrial damage 158, 164, 165 nuclear affinity 171 tissue surface fluorescence correlation with photodynamic effect acute photodynamic effects 148, 150 drug-light interval photodynamic therapy activity 151–153, 155 response 150, 151, 155 irradiation 148 mouse model 147 photosensitizer content in tissue 151 remittance fluorescence spectrophotometry 151 2,7,12,17-Tetrakis(methoxyethyl)porphycene, see Porphycenes

Subject Index

Thermal reactions, photochemical vs dark 54 Tissue surface fluorescence, correlation with photodynamic effect acute photodynamic effects 148, 150 drug-light interval photodynamic therapy activity 151–153, 155 response 150, 151, 155 irradiation 148 mouse model 147 photosensitizer content in tissue 151 remittance fluorescence spectrophotometry 151 Uterus, see Endometrial ablation, Endometrial cancer VIN, see Vulvar intraepithelial neoplasia Vulvar condyloma, 5-aminolevulinic acid photodynamic therapy fluorescence imaging 262, 263 outcomes 258, 261 principle 251 rationale 251, 252 selective photosensitization 252–254, 256, 257, 263 technique 254–256 Vulvar intraepithelial neoplasia, 5-aminolevulinic acid photodynamic therapy fluorescence imaging 262, 263 outcomes 261 principle 251 technique 254–256 Wavelength dependence of singlet oxygen generation by photosensitizers 113, 114 light propagation in tissues 117–119 XeCl laser, laser-assisted hatching 332 Zinc-phthalocyanine accumulation in vascular and tumor cell systems 43–45

383

Zinc-phthalocyanine (continued) fluorescence emission vs triplet state formation 140 photobleaching 45 photokinetics, subcellular 45, 46 Zona pellucida drilling laser zona drilling contact mode 331 diode laser, 1480-nm advantages and overview 334, 335, 356, 360, 361, 363 drilling technique 354–356 human studies 335, 359, 360 instrumentation 353, 354 mouse studies 335, 356–359, 361 Er:YAG laser, 2,940-nm criteria for success 369, 370 human studies 334, 368–370 instrumentation 366, 367

Subject Index

mouse studies 334, 367 technique 367 hatching failure causes 352, 361, 362 lasers Ho:YAG 333 Ho:YSGG 333, 334 nitrogen laser 332, 333 XeCl 332 noncontact mode 330, 331 outcomes 331, 332 PALM Robot-Microbeam system cattle 344–347 mice 344 principles 343 patient selection and indications 331, 335, 336 mechanical and chemical drilling 352, 353, 362

384

Photomedicine in Gynecology & Reproduction

Berek & Novak's Gynecology

Deja review: obstetrics & gynecology

Case Files: Obstetrics & Gynecology

Feminism & Bioethics: Beyond Reproduction

Berek & Novak's Gynecology (Berek and Novak's Gynecology)

Deja Review Obstetrics & Gynecology

CURRENT Diagnosis & Treatment Obstetrics & Gynecology

Berek & Novak's Gynecology, 14th Edition (Novak's Textbook Gynecology)

Infectious Diseases in Obstetrics and Gynecology, fifth Edition (Infectious Diseases in Obstetrics & Gynecology)

Comprehensive Gynecology

Blueprints Obstetrics & Gynecology, 5th Edition

Ultrasonography in Obstetrics and Gynecology

Recent Advances in Obstetrics & Gynecology Volume 21

Reproductive Endocrinology and Infertility: The Requisites in Obstetrics & Gynecology

Clinical Gynecology

Obstetrics & Gynecology: PreTest Self-Assessment & Review, 11th Edition

PreTest Self-Assessment & Review: Obstetrics & Gynecology

Reproduction in Context: Social and Environmental Influences on Reproduction

Current Opinion in Obstetrics & Gynecology JUNE 2010

Reproduction in Cattle 3rd Edition

Reproduction in Domestic Ruminants VII

Clinical Gynecology

Comprehensive Gynecology

Cultural Reproduction

Pediatric and Adolescent Gynecology

Doppler Ultrasound in Obstetrics and Gynecology

TeLinde's Operative Gynecology

Diagnostic Imaging: Gynecology

Ginecologia Novak's Gynecology

Hematological Complications in Obstetrics, Pregnancy, and Gynecology

Photomedicine in Gynecology & Reproduction

Berek & Novak's Gynecology

Deja review: obstetrics & gynecology

Case Files: Obstetrics & Gynecology

Feminism & Bioethics: Beyond Reproduction

Berek & Novak's Gynecology (Berek and Novak's Gynecology)

Deja Review Obstetrics & Gynecology

CURRENT Diagnosis & Treatment Obstetrics & Gynecology

Berek & Novak's Gynecology, 14th Edition (Novak's Textbook Gynecology)

Infectious Diseases in Obstetrics and Gynecology, fifth Edition (Infectious Diseases in Obstetrics & Gynecology)

Comprehensive Gynecology

Blueprints Obstetrics & Gynecology, 5th Edition

Ultrasonography in Obstetrics and Gynecology

Recent Advances in Obstetrics & Gynecology Volume 21

Reproductive Endocrinology and Infertility: The Requisites in Obstetrics & Gynecology

Clinical Gynecology

Obstetrics & Gynecology: PreTest Self-Assessment & Review, 11th Edition

PreTest Self-Assessment & Review: Obstetrics & Gynecology

Reproduction in Context: Social and Environmental Influences on Reproduction

Current Opinion in Obstetrics & Gynecology JUNE 2010

Reproduction in Cattle 3rd Edition

Reproduction in Domestic Ruminants VII

Clinical Gynecology

Comprehensive Gynecology

Cultural Reproduction

Pediatric and Adolescent Gynecology

Doppler Ultrasound in Obstetrics and Gynecology

TeLinde's Operative Gynecology

Diagnostic Imaging: Gynecology

Ginecologia Novak's Gynecology

Hematological Complications in Obstetrics, Pregnancy, and Gynecology

Recommend Documents