Radiotherapy in Developing Countries PROCEEDINGS of a SYMPOSIUM, VIENNA, 1 -5 SEPTEMBER 1986 ORGANIZED by IAEA in CO-OPERATION with WHO

Home , Floppy disk, NR Group

Radiotherapy in Developing Countries PROCEEDINGS OF A SYMPOSIUM, VIENNA, 1 -5 SEPTEMBER 1986 ORGANIZED BY IAEA IN CO-OPERATION WITH WHO

INTERNATIONAL ATOMIC ENERGY AGENCY, VIENNA, 1987 RADIOTHERAPY IN DEVELOPING COUNTRIES

NUCLEAR SAFETY

INFORMATION LIBRARY PLEASE RETURN A2643

PROCEEDINGS SERIES

RADIOTHERAPY IN DEVELOPING COUNTRIES

PROCEEDINGS OF AN INTERNATIONAL SYMPOSIUM ON RADIOTHERAPY IN DEVELOPING COUNTRIES - PRESENT STATUS AND FUTURE TRENDS ORGANIZED BY THE INTERNATIONAL ATOMIC ENERGY AGENCY IN CO-OPERATION WITH THE WORLD HEALTH ORGANIZATION A N D HELD IN VIENNA, 1-5 SEPTEMBER 1986

INTERNATIONAL ATOMIC ENERGY AGENCY VIENNA, 1987 RADIOTHERAPY IN DEVELOPING COUNTRIES IAEA, VIENNA, 1987 STI/PUB/719 ISBN 92-0-010087-2

Permission to reproduce or translate the information contained in this publication may be obtained by writing to the International Atomic Energy Agency, Wagramerstrasse 5, P.O. Box 100, A-1400 Vienna, Austria.

Printed by the IAEA in Austria March 1987 FOREWORD

Cancer is of increasing concern all over the world, but the situation is becoming very serious in developing countries. Timely vaccination, and progress in sanitation and personal hygiene have improved life expectancy in many Third World countries, thus making it more likely that their citizens will eventually contract cancer. The doc tor in the developing world has not much to offer against this disease. Because medi cal care is not readily available and because means of early diagnosis are inadequate, cancer in these countries is often in an advanced stage when a patient is presented for treatment. Surgeons are scarce in the hospitals of the developing world, and chemotherapy is rather expensive. Radiotherapy might be useful either as a single modality of treat ment or in conjunction with surgery and chemotherapy for a significant percentage of patients. According to the World Health Organization (WHO), 40% of the developing countries do not have any radiotherapy services at all, while there is also a severe shortage of drugs used to fight cancer. In most of the countries where radiotherapy is available it functions inadequately because of its isolation from the health care system, the lack of qualified staff and equipment and insufficiency of funds. In countries where there are no radiotherapy facilities or where facilities are used inadequately, a large group of patients are being denied an effective mode of treatment. The Agency is actively involved with this problem because nuclear energy can be used as a means of treatment. WHO, a sister organization, is also vitally interested in the treatment of cancer because by the year 2000 the problem is expected to reach enormous dimensions and something should be done about it now. Many of the Agency activities are performed in close collaboration with WHO, as is this Symposium. By way of concerted co-ordinated research programmes, the Agency is trying to encourage a few of the ideas for improving radiotherapy practice that have direct relevance to the developing countries. One of these programmes is related to the use of adjuvant therapy to enhance the effects of irradiation. When radiotherapy sources are scarce, these potentiating modes of therapy help in conserving resources. Technical assistance is mainly offered in the form of fellowships for training of physicians and physicists from the developing countries in radiotherapy practices. The Agency also undertakes to provide experts wherever needed for on the job training. Unfortunately, the Agency’s resources are limited and support for therapy machines has been possible in a very few cases only. There is also an IAEA/WHO project dealing with the setting up of a network of brachytherapy centres for treatment of early cases of carcinoma of the cervix. This project was carried out in Egypt in 1983 and was highlighted at this Symposium. In addition to the annual training courses designed for this project, the Agency plans two biannual training courses in the region of South-East Asia and the Pacific for radiotherapists and related disciplines. Also directly related to radiotherapy is the IAEA/WHO network of Secondary Standard Dosimetry Laboratories which is concerned with improving the dosimetric accuracy of radiotherapy. This network presently includes some fifty laboratories mainly in the developing world. It is supported by about twelve national primary stan dard laboratories and by the International Office of Weights and Measures. The Agency’s dosimetry laboratory near Vienna functions as a co-ordinating laboratory for the network and organizes dose intercomparison measurements among partici pating laboratories. The aim of the present Symposium was to gather together specialists from deve loped and developing countries to review progress and present their latest findings. It was the first major meeting of its kind organized to take stock of the situation and assess the magnitude of the problem. It is evident that much work needs to be done before the aim of having radiotherapy facilities for treating cancer in each and every central hospital of the developing world is realized. The Symposium might be regarded as a first concrete step towards this goal.

EDITORIAL NOTE

The Proceedings have been edited by the editorial staff of the IAEA to the extent considered neces sary for the reader's assistance. The views expressed remain, however, the responsibility of the named authors or participants. In addition, the views are not necessarily those of the governments of the nominating Member States or o f the nominating organizations. Although great care has been taken to maintain the accuracy of information contained in this pub lication, neither the IAEA nor its Member States assume any responsibility for consequences which may arise from its use. The use of particular designations of countries or territories does not imply any judgement by the publisher, the IAEA, as to the legal status of such countries or territories, o f their authorities and institu tions or of the delimitation of their boundaries. The mention of names of specific companies or products (whether or not indicated as registered) does not imply any intention to infringe proprietary rights, nor should it be construed as an endorsement or recommendation on the part of the IAEA. The authors are responsible for having obtained the necessary permission for the IAEA to reproduce, translate or use material from sources already protected by copyrights. Material prepared by authors who are in contractual relation with governments is copyrighted by the IAEA, as publisher, only to the extent permitted by the appropriate national regulations. CONTENTS

RADIATION THERAPY FOR CARCINOMA OF THE CERVIX

Treatment of carcinoma of the cervix in developing countries (IAEA-SM-290/81) ...... 3 M. Snelling Technical and radiobiological peculiarities in fractionated high dose rate afterloading — A ten year report on the treatment of carcinoma of the cervix (IAEA-SM-290/83) ...... 13 F.H. Glaser High dose rate afterloading in the treatment of cervix carcinoma with external irradiation and brachy therapy (IAEA-SM-290/13) ...... 27 T. Kuipers Preliminary report on the treatment of carcinoma cervix uteri with a combination of conventional irradiation and mitomycin С as radiosensitizer (IAEA-SM-290/29) ...... 33 S. Puribhat, C. Chotigavanich, S. Tangkaratt, P. Puntumchinda, S. Sombooncharoen The role of intra-arterial chemotherapy in the management of uterine tumours with a complex modality treatment (IAEA-SM-290/59) ...... 41 J. Stumpf, G. Vadon, G. Németh Introduction of advanced remote afterloading brachytherapy techniques into developing countries, considering high activity sources, maintenance and care of equipment (IAEA-SM-290/23) ...... 47 U.M.F. Altemark, S.M. Quandt Discussion ...... 55

DIFFERENT APPROACHES IN RADIATION THERAPY Radioprotection of the immune system during radiation therapy (IAEA-SM-290/84) ...... 61 H. Altmann, H. Tuschl, E. Кип Carcinoma of the oesophagus — Palliation by intracavitary irradiation (IAEA-SM-290/5) ...... 71 C.G. Rowland, K.M. Pagliero Résultats d’irradiations hemicorporelles chez 92 patientes porteuses de cancer du sein polymétastatique traitées à l’Institut Curie (IAEA-SM-290/47) ...... 79 D. Jullien, J.-R. Vilcoq, F. Сатрапа Tratamiento multidisciplinario del tumor de Wilms. Experiencia de trece años (IAEA-SM-290/10) ...... 87 T. Lanché, R. Borrego, G. Martínez, R. Rivera Radioterapia preoperatoria en el tratamiento del cancer de rectosigmoides (IAEA-SM-290/19) ...... 97 C. González-Miranda, L. Badinez, J■ Rajevic The role of radiotherapy in localized non-Hodgkin’s lymphomas (IAEA-SM-290/8) ...... 105 M.B. Patricio, R. Cabral, M. Neves, A. de Ponte, M. Vilhena Neoadjuvant chemotherapy and hypofractionated irradiation in the treatment of head and neck cancers (IAEA-SM-290/33) ...... Ill J.-M. Deneufbourg Technical problems associated with radiotherapy for nasopharyngeal carcinoma. Experience at the Department of Radiotherapy, Kenyatta National Hospital, Nairobi, Kenya (IAEA-SM-290/67) ...... 121 J.N. Onyango, A.M. Babu, N. Tole Intracavitary irradiation for nasopharyngeal carcinoma (IAEA-SM-290/39) ...... 125 R. Uzel, S. Коса, S. Okkan, A. Ertem, S. Turkan Clinical approaches of conventional radiotherapy — Emphasis on Africa (IAEA-SM-290/56) ...... 135 J. Luande Head and neck cancer in Saudi Arabia (IAEA-SM-290/6) ...... 139 B.S. Clubb, C.A. Quick, M.H. Amer, E. Mahboubi, M.A. El-Senoussi, H.P. Schultz, S.M. El-Akkad Accelerated fractionation (IAEA-SM-290/11) ...... 147 V.H.J. Svoboda Fast neutron therapy for the developing world: Is it worthwhile? (IAEA-SM-290/7) ...... 155 A. Aissi, J.M. Feola, B.S. Clubb Treatment of carcinoma of the thyroid gland with iodine-131 (IAEA-SM-290/78) ...... 163 R. Höfer Current prospects for 125I in tumour therapy (IAEA-SM-290/77) ...... 173 L.E. Feinendegen, W. Boecker, J. Booz, E. Pomplun, W. Schaden, F.H.A. Schneeweiss, C.A. Sondhaus, G. Tisljar-Lentulis Discussion ...... 181

HYPERTHERMIA

Current studies of hyperthermia and hypoxic cell radiosensitizers in clinical oncology (IAEA-SM-290/69) ...... 187 N.M. Bleehen, G.C.W. Howard, H.F.V. Newman The role of hyperthermia in radiation therapy in developing countries (IAEA-SM-290/46) ...... 193 T. Sugahara Interaction between radiotherapy and hyperthermia in human solid tumours (IAEA-SM-290/12) ...... 199 T. Vazquez, P. Kasdorf Hyperthermia and radiotherapy of locally recurrent breast cancer (IAEA-SM-290/60) ...... 207 A.O. Badib, S.A. Khalil, M.Y. Gouda, A.M. Salama Combined radiotherapy and hyperthermia: Prognostic variables affecting tumour response (IAEA-SM-290/73) ...... 211 G. Arcangeli, M. Benassi, S. Carpino, F. Mauro Development of thermal radiation therapy of patients with radioresistant tumours (IAEA-SM-290/52) ...... 221 A. Cyb, A. Konoplyannikov Hyperthermia in cancer radiotherapy. A clinical study at the Osaka City University Hospital (IAEA-SM-290/70) ...... 229 Y. Onoyama, T. Nakajima, M. Tsumura, A. Kida, M. Tanaka Discussion ...... 241

CHEMICAL MODIFIERS

Radiotherapeutic potentiation by chemical and physical means in oral squamous cell carcinomas (IAEA-SM-290/43) ...... 249 S. Krishnamurthi, V. Shanta, N. Gopalan, A. Vasantan, N.M.S. Reddy Metronidazole in the treatment of nasopharyngeal carcinoma (IAEA-SM-290/15) ...... 261 A. Hidayatalla, E. Abdel Rahman, H.M.A. Hamad Effect of ornidazole on fractionated irradiation in carcinoma of the cervix and larynx (IAEA-SM-290/40) ...... 271 S. Okkan, R. Uzel, Z. Yazici, A. Akçasu, N. Turan, S. Turkan Experience in the use of hypoxic cell chemical radiosensitizers and the potential use of some newer radiosensitizers (IAEA-SM-290/76) ...... 281 H.K. Awwad, S. El Badawy, M. Ghoneim, M. Barsoum, M. Zaghloul, H. Akoush, M. Abou Zaid, A. Osman, M. El-Naggar Radioresponsiveness of locally advanced breast cancer. Analysis of factors influencing the response to radiation therapy (IAEA-SM-290/4) ...... 291 J. F odor, J. Tôth, G. Gy enes Preclinical and clinical results of the use of Fluosol-DA (20%) in radiotherapy (IAEA-SM-290/54) ...... 301 C.W. Song, I. Lee, T. Hasegawa, J.G. Rhee, S.H. Levitt, N. McIntosh Radiosensitization of hypoxic bacterial cells and animal tumours by membrane active drugs and hyperthermia (IAEA-SM-290/74) ...... 311 K.C. George, M.A. Shenoy, B.B. Singh Discussion ...... 321

DOSIMETRY AND TECHNOLOGY

Caesium manual afterloading intracavitary treatment of carcinoma cervix: A simple method of dose calculation (IAEA-SM-290/53) ...... 325 K.A. El-Ghamrawi, M.M. Mahfouz, R. Mould, O. Zaki Code of Practice for absorbed dose determination in photon and electron beams (IAEA-SM-290/32) ...... 333 H. Svensson, P. Andreo, J. Cunningham, K. Hohlfeld Clinical dosimetry in Czechoslovakia (IAEA-SM-290/61) ...... 345 V. Laginová, J. Novotny Total gamma depth dose distribution from a californium-252 source in a tissue equivalent phantom (IAEA-SM-290/27) ...... 359 F.A. El-Bakkoush, T.S. Akki, R.M. Megahid Simple dose calculation method for brachytherapy with l92Ir (IAEA-SM-290/31) ...... 369 G. Horgas, V. Lokner, B. Pokrajac, S. Spaventi Discussion ...... 377

ORGANIZATION OF RADIATION THERAPY IN DEVELOPING COUNTRIES (Panel)

Radiotherapy in developing countries — Constraints and possible solutions (IAEA-SM-290/79) ...... 381 N.T. Racoveanu Earlier detection of carcinoma of the cervix uteri in slowly developing countries (IAEA-SM-290/80) ...... 403 M.M. Mahfouz, S. El-Haddad, K.A. El-Ghamrawi, O. Zaki, F. Haggag The establishment of radiotherapy in a developing country. Experience from a project in Nairobi, Kenya (IAEA-SM-290/82) ...... 413 R. Walstam, J. Einhorn Training of radiation therapy technologists (IAEA-SM-290/85) ...... 419 R.J. Morton Estado actual de la radioterapia en Venezuela (IAEA-SM-290/65) ...... 425 N. JJrdaneta, R. Millan, A. Rodríguez, A. Colina, E. Alvarez, A. Vera, M. Bittar, J. Mattout, M. Aguilera, L. Rúan, R. Perdomo, C. Nieves, R. Vera Panel Discussion ...... 439 Chairmen of Sessions and Secretariat of the Symposium...... 443 List of Participants ...... 445 Author Index ...... 453 Index of Papers by Number...... 455

RADIATION THERAPY FOR CARCINOMA OF THE CERVIX

IAEA-SM-290/81

Invited Paper TREATMENT OF CARCINOMA OF THE CERVIX IN DEVELOPING COUNTRIES

M. SNELLING* London, United Kingdom

Abstract

TREATMENT OF CARCINOMA OF THE CERVIX IN DEVELOPING COUNTRIES. Squamous cell carcinoma of the cervix is the most frequent female cancer in developing countries. While the cure rate in stage I may be 85%, this falls rapidly to 50%, 30% and lower in more advanced stages of the disease. Few centres see more than 10% stage I cases. Earlier diagnosis and immediate treatment are essential. Similar problems exist in other common, accessible squamous cell carcinomas, in the oral cavity, skin, and penis. While carcinoma of the breast is a more complex problem, the squamous tumours are easy to detect on inspection, to prove on biopsy or smear, to assess at a cancer centre with radiology and pathology, and to treat at an early stage by simple surgery and radiotherapy (brachytherapy) or later by brachytherapy and teletherapy. Modern management necessarily involves publicity with education of the public, the medical profession and ancillaries on the causes, symptoms, natural history and treatment of cancer. This requires the help of public health services, obstetric and birth control colleagues, village nurses, primary health services and others. Regular screening projects are necessary including cervical smears and improvements in urban and rural health with the elimination of known causes of cancer, improvement in hygiene and water supply and, where cancer of the cervix is concerned, some changes in social mores, improvement in birth control (obstructive methods) and control of virus infection. Cervical smears on a national scale have been proved effective in Canada and elsewhere. Such a programme requires careful planning with the institution of laboratories and the training of cytopathologists and technicians. The IAEA/WHO Project in Egypt is planning a national scheme for earlier diagnosis by means of cervical smears and of early treatment near the patients’ homes in district hospitals by teams trained in Cairo using manual afterloading machines with caesium-137 sources. The International Working Party for the Treatment of Cancer of the Uterus in the Developing World (founded in 1972 and sponsored by the IAEA and WHO) is an informal group of experienced radiotherapists from many countries. It meets every two years in a developing country; the next meeting is to be held in Calcutta in January 1987. Discussions include all aspects of the disease and its treatment, including advanced disease. The paper includes some details concerning the use and advantages of manual and remote afterloading techniques in both intracavitary and interstitial therapy including the use of high dose rate remote afterloading devices in developing countries.

* Professor Emeritus of the Meyerstein Institute of Radiotherapy and Oncology, The Middlesex Hospital, London, United Kingdom.

3 4 SNELLING

1. IN T R O D U C T IO N

Carcinoma of the cervix is the most frequent cancer in females in many developing countries, where it may account for one-third of all cancers recorded [1 ]. Although a well understood, common and accessible tumour with many cures following treatment with surgery, radiotherapy or combinations of these treatments, thousands of women die each year with undiagnosed or untreatable advanced disease because diagnosis has not been made in a premalignant or early invasive stage. Cure rates of more than 85% are found in stage I cases in industrialized and developing countries, but with more extensive tumours this drops to 50— 60% in stage II, and to less than 25% in stage III cases. Patients delay seeking advice and may have to travel long and expensive journeys before reaching a cancer centre and, when they arrive there, the disease is often very advanced and the prognosis after treatment bad or hopeless. This is shown in patients in two decades reported at the Chittaranjan Hospital in Calcutta [2]. Between 1954 and 1963, 4647 cases were seen of whom 4.8% were stage I, 19.2% stage II and 55.3% stage III. Between 1969 and 1978 out of 5567 cases there were 5.1% stage I, 22.5% stage II and 53.3% stage III. Only a fraction of these could receive treatment. Between 1954 and 1963 there were 2061 cases treated of whom 8.8% were stage I, 16.2% stage II and 73.5% stage III. Between 1969 and 1978 there were 2172 cases treated of whom 8.8% were stage I, 25.7% stage II and 63.6% stage III. The prognosis of the treated cases can be deduced from their stages. Similar reports have been published from centres all over the developing world. More than 90% of these tumours are squamous cell carcinoma varying in malignancy and of moderate radiosensitivity. Prognosis is clearly related to the extent of the disease, thus emphasizing the need for early diagnosis with increased facilities for diagnosis and immediate treatment.

2. GROUPS OF PATIENTS INVOLVED

The late diagnosis and consequent bad prognosis are related to the circum stances of the patients; cancer of the cervix is most common in the lowest socio economic groups with patients who tend to suffer from malnutrition, and poor housing without easy access to water, deficient hygiene and open to infection (herpes infection is important). Cancer of the cervix is a venereal disease, related to sexual intercourse, to the age at marriage and the beginning of sexual activity, to the frequency of intercourse, the number of partners and the number of IAEA-SM-290/81 5 children. In this group education is defective and illiteracy common. There is little knowledge of the behaviour and cause of cancer or of its symptoms and signs. It is generally believed to be incurable and there is ignorance of the importance of seeking advice and of the means of getting diagnosis or treatment. In this social group early marriage and extreme multiparity have been the rule for many years. Early marriage is now being gradually restricted legally and birth control is being encouraged. The use of the condom as an obstructive method of birth control has a real advantage in reducing the premalignant changes in the cervical epithelium. A change in social mores and acceptable behaviour in the industrializéd countries which has resulted in the appearance of very malignant cancers of the cervix in young sexually active women has not yet been reported in developing areas. The education of both the urban and the rural population in the nature and cause of cancer with information concerning the more common diseases must be planned on a national scale and carried out together by education and medical services and spread in schools and colleges and by the media such as newspapers, radio and television. Teaching concerning cancer of the cervix is reinforced by family doctors and by the staff at birth control centres, district and village nurses, cancer societies and, increasingly, in primary health services visiting remote areas. As regards the financial situation the women concerned are often very poor and cancer is an expensive disease often involving the mother of a large, young family. Diagnosis and treatment may mean a long journey, while care at home may be impossible without special help. More interest is needed from the administrators in assisting with these expenses and in the organization of diagnostic and therapeutic facilities closer to the patients’ homes than those existing at present.

3. NATIONAL PLAN - THE IAEA/WHO EGYPT PROJECT

Planning can only be done on a national scale since elimination of such a disease must be the right of every woman and to achieve this there must be collaboration between a network of hospitals — university, regional and district — with the smaller units and village services involved. The IAEA and WH O are working together in a project designed to establish earlier diagnosis and treatment of cancer of the cervix in district hospitals all over the country by means of a transfer of information and technology from a university hospital in the capital to the staff of the district hospitals. Such a project, generously financed for four years by the Italian Government, is being developed in Egypt. It is hoped that, when completed, such a scheme can be reproduced in other developing countries in every continent. 6 SNELLING

4. EARLY DIAGNOSIS OF CARCINOMA OF THE CERVIX

We know from experience that because of fear, ignorance and inconvenience a patient rarely asks for medical advice until the symptoms — usually bleeding — have been present for some months. The patients reported at the Chittaranjan Hospital [2] gave histories of 270 and 208 days before arriving at the cancer centre and there were further delays of 40 and 31 days before treatment commenced. The importance of education has already been emphasized and an increase in the screening clinics, visits to factories and villages, etc., are also necessary if premalignant changes and early stage invasive carcinoma are to be found in this alerted population. This again involves detailed planning on a nation-wide scale including the provision of new laboratories and the training of cytopathologists and their staff of technicians. It also involves the appointment of an efficient and foolproof system of communication with dedicated and efficient administrators and staff.

5. CYTOLOGY - FEMALE POPULATION COVERED

National schemes are being planned in many countries and reports on their efficacy and cost are eagerly awaited. While we would all wish every woman to be covered from the age when she becomes sexually active, such an ambition demands the impossible until education and social habits change, in addition to offering unacceptably low yields in positive findings at too high a cost. The women examined should ttterefore in the first place be those most liable to the disease through early marriage, multiparity, etc., as well as those with early smears showing premalignant changes. Until we hear the results of pilot surveys in different populations we cannot be certain of the useful age at which to start this examina tion nor of the frequency needed in later repetitions. It must also as usual become a decision on the best use of the money and manpower available. In the IAEA/WHO Egypt Project, where a cytological survey is being started on a nation-wide basis to find earlier diagnosed cases, cervical smears combined with gynaecological examination are carried out on all women between 25 and 60 attending the gynaecological departments of the hospitals collaborating in the study. The survey there is centred on the Professor of the Reference Cytology Laboratory at a Cairo University Hospital where more than 60 000 smears each year are expected. This Reference Cytology Laboratory will also receive copies of all slides and reports from five peripheral cytology laboratories set up in three university and two general hospitals covering the country These also receive smears from nineteen collecting centres in smaller hospitals. Equipment was necessary for these new laboratories and there was an urgent need for the training of more technicians and their establishment in adequately paid posts in the health service. Administrators of high calibre, personnel and transport were also necessary in the central and peripheral registries. IAEA-SM-290/81 7

A computerized central registry collects clinical and treatment data of the patients treated in the project as well as information on the cytological survey. As mentioned earlier, this work is funded by a generous gift from the Italian Government. It is thought that the cost of this cytological survey carried out in Egyptian hospitals will amount to about US $60 000 for equipment and US $16 000 for training. The running costs might amount to US $35 000 for salaries and US $16 000 for disposable stores. Similar work is developing elsewhere and at a Seminar on Transfer of Technology, held in New Delhi in March 1986, there were discussions on similar activities with a network of collaborating hospitals centred in the Cytology Research Centre at the Indian Council of Medical Research. At the meeting of the International Working Party in Thailand in 1985 we were told of the collection of smears in the villages for examination at large regional hospitals.

6. TREATMENT OF CARCINOMA OF THE CERVIX

Treatment results have demonstrated that early diagnosis followed by treat ment without delay should be available to all women. Protocols for the management by surgery or radiotherapy or by a combination of both should be agreed on - and adhered to - by members of all the disciplines involved in such treatment in the hospitals collaborating in the treatment. Agreement, collabora tion and constant communication are all essential. In developing countries gynaecological surgery is available at university, regional and often also large district hospitals. A fully equipped and staffed radiotherapy department is found in university and, occasionally, in large regional hospitals. A vast number of patients, however, live and need diagnosis in remote areas served by district hospitals where they encounter the difficulties already described. While surgery is necessary in the treatment of premalignant conditions and is used in early invasive disease and for some radioresistant tumours the main line of treatment for the great mass of patients with carcinoma of the cervix must be radiotherapy.

7. TREATMENT OF PATIENTS IN REMOTE AREAS

Among the methods suggested for making treatment more accessible to rural patients are the peripheral clinics, new small treatment centres and the system of transfer of information and technology from a central university to distant district hospitals as described in the IAEA/WHO Egypt Project. The peripheral clinic is much used in industrialized countries where the distance between the regional centres and district hospitals is not very great but inconvenient and expensive for the patients. Staff members of the regional cancer centres visit the hospitals weekly, consult with the staff and see new and 10 SNELLING and administrators of all kinds with experience in the treatment of these tumours are invited. The meetings are open to all with interest and experience in the subject. The International Working Party has been sponsored by the IAEA and W H O for many years and discussions in Istanbul and Mexico City led to their joining in the project taking shape in Egypt. Many subjects have been discussed at the meetings of this group such as the hazards of radium-226 and its replace ment by cobalt-60, caesium-137 and iridium-192, the design of applicators and the use of manual and remote afterloading with modern microsources. We have discussed all aspects of conventional and high dose rate therapy, dose, time fractionation and dosimetry. We have considered the aetiology and incidence of the disease, organization and socio-economic aspects of its treatment, the provision of earlier diagnosis, public health aspects and the transfer of technology and knowledge. Discussions of the next meeting at Calcutta will be centred on the problem of carcinoma of the cervix in village women in South-East Asia.

10. PRACTICE OF INTRACAVITARY BRACHYTHERAPY IN DEVELOPING AREAS

Radium-226 sources should no longer be used: they should be collected and stored safely, and should be replaced by caesium-137 or, where suitable arrangements can be made for the short half-life, by iridium-192 sources. Because of its high energy, cobalt-60 is used now for high dose rate therapy with appropriate radioprotection. Microsources should be used with afterloading techniques, either manual or remote. When a new centre starts using these techniques it is wise for the staff to study them in an experienced centre with many patients and to commence with manually loaded sources of conventional (or only slightly increased) dose rates. The use of microsources permits most easily the arrange ment of the sources as required by the dosimetry prescribed. The technique chosen depends upon the work load and the expertise available and on the radioprotection required for the staff, visitors, and other patients, etc. Where there are few cases irradiation is limited to the wards and manual afterloading techniques may be used (up to 10 cases/week). Conventional dose rate remote afterloading reduces this irradiation but occupies beds and rooms for some hours or days. High dose rate therapy is increasingly used. The shorter treatment times, reduced to some minutes only, increase the patient’s comfort considerably and are convenient for medical staff and administration of the department. When used with great care their effect on tumour and normal tissues is similar to that of radium and an equivalence has been worked out in such parameters as time, dose rate and fractionation. There are obvious hazards associated with these short, high dose rate treatments and they are only suitable for use in departments already experienced in low dose rate treatments and where there is reliable expertise. Medium dose rate treatment is increasingly IAEA-SM-290/81 ]] used but without the great convenience of high dose rate therapy, which also offers an accuracy and repeatability of individualized treatments not offered by other means. Medium dose rate therapy offers a certain saving of time but the equivalence of its effects with the parameters established by experience with conventional dose rates does not appear to be as yet agreed on and awaits reports on pilot trials in progress. There is a need for agreement in centres on protocols for the staging of cases and their management and on the exact details of treatment to be given at each stage. These should be exactly followed with records made of clinical findings, treatment and follow-up, which after three and five years can be analysed to allow an assessment of the management under examination. The present existence of many slightly differing protocols and individual variations with alterations often after only a short period render impossible the advance of these important techniques.

11. PRACTICE OF TELETHERAPY: TELECOBALT OR LINEAR ACCELERATOR

As previously stated, all but stage I squamous cell carcinoma of the cervix require complementary therapy with external irradiation either before or after brachytherapy. Treatment with this equipment is also necessary for well over 50% of the cases of cancer at other sites referred to the centre. Treatment of cancer of the cervix can be carried out accurately with a modern cobalt unit or with a small linear accelerator. The use of a larger linear accelerator is needed for electron therapy, especially in the treatment of carcinoma of the head and neck. There is no indication that its use in carcinoma of the cervix is desirable. There is a real need for the provision of fully equipped departments of radiotherapy in large hospitals throughout the country with appropriate linking and technology transfer with the district hospitals in their regions.

12. PALLIATIVE TREATMENT OF ADVANCED CANCER

This paper has been concerned with the eventual elimination of carcinoma of the cervix by identifying and treating premalignant and early invasive disease. While this process becomes implemented our hospitals will remain filled with advanced and virtually untreatable cases. Our planning must therefore be extended to them and in the first place we must assess this clinical material and the palliation and assistance that we can offer. SNELLING

REFERENCES

TUNGSUBUTRA, K., TUNGVORAPONG CHAI, V., PESI, M., “ Cancer epidemiology in Thailand” , Diagnosis and Treatment of Carcinoma of the Cervix in Developing Areas (Proc. Int. Working Party Meeting Thailand, 1985), Adam Hilger Ltd, Bristol and Boston (1985) 187. GOSH, S., BOSE, C.K., “ Her plight to misery: Problems of carcinoma of the cervix studied in Calcutta, West Bengal” , Diagnosis and Treatment of Carcinoma of the Cervix in Developing Areas (Proc. Int. Working Party Meeting Thailand, 1985), Adam Hilger Ltd, Bristol and Boston (1985) 89. UNHANAND, SAI-SANGUAN, “ A scheme for the decentralization of cancer radiotherapy services to provincial areas in Thailand”, Diagnosis and Treatment of Carcinoma of the Cervix in Developing Areas (Proc. Int. Working Party Meeting Thailand, 1985), Adam Hilger Ltd, Bristol and Boston (1985) 231. IAEA-SM-290/83

TECHNICAL AND RADIOBIOLOGICAL PECULIARITIES IN FRACTIONATED HIGH DOSE RATE AFTERLOADING - A TEN YEAR REPORT ON THE TREATMENT OF CARCINOMA OF THE CERVIX

F.H. GLASER Clinic and Polyclinic of Radiology, Medical Academy Erfurt, Erfurt, German Democratic Republic

Abstract

TECHNICAL AND RADIOBIOLOGICAL PECULIARITIES IN FRACTIONATED HIGH DOSE RATE AFTERLOADING - A TEN YEAR REPORT ON THE TREATMENT OF CARCINOMA OF THE CERVIX. High dose rate afterloading (HDR-AL) eliminates the risk of radiation hazard to the staff, facilitates optimization of dose distribution and makes treatment easier for patients and the hospital. HDR-AL breaks with the classic principle of protraction in conventional low dose rate brachytherapy (LDR-BT), so a higher fractionation is necessary to get the advantage of the time factor for the recovery and repair of normal healthy tissue. Since 1974, more than 2500 patients have been treated with HDR-AL using the afterloading equipment DECATRON. This remotely controlled device works with 192 Ir sources and the isodose shaping is carried out by a step by step movement and different times of stay at the single points on the source’s path. A total of 1131 women with cervical carcinoma were treated with HDR-AL alone or in combination with external beam therapy. In 612 of them the irradiation was given as a primary treatment intracavitarily, and in 5 19 cases HDR-AL was performed as a post-operative treatment intravaginally. All patients were followed up for at least 12 months up to more than 5 years. The results of 5 year survival in primary and post operative HDR-AL were compared with the international statistics from the Annual Report, Vol. 18. The results, related to the stages, are at least equivalent to those in conventional LDR-BT, and in several groups are better. Stages II and III show a statistically significant improvement of treatment results in comparison with LDR-BT. The incidence of early and late side effects on bladder and rectum showed a statistically highly significant decrease in HDR-AL and was dependent on single and total exposure in a highly significant manner (p = 0.001). The therapeutic effectiveness is increased and the risk of side effects on bladder and rectum is decreased by the better radiobiological (constant dose rates) and dosimetric adaptation of HDR-AL and external beam therapy.

Remotely controlled short term afterloading with high dose rates (HDR-AL) eliminates the risk of staff exposure in brachytherapy and offers advantages for optimization of the dose distribution, both for the patients and the clinic. On the other hand, brachytherapy with high dose rates raises many radiobiological problems requiring modification of the treatment concept with regard to the choice of single and total doses, fractionation, interval and overall treatment time.

13 14 GLASER

FIG. 1. TheDECATRON

HDR-AL breaks with the classic principle of protraction in brachytherapy so that the advantage of the time factor for the recovery capacity of healthy tissue through a higher fractionation must be sought. Clinical observations, animal experiments and hypothetical model concepts show that with high dose rate irradiation of 50 cGy/min and more, compared with low dose rates of about 1-10 cGy/min, an increase in radiobiological effectiveness can be achieved [ 1 — 7]. In HDR-AL at a nearly linear dose-time relationship an increase in the relative radiobiological effectiveness up to about 60% develops at almost the same dosage as in low dose rate brachytherapy (LDR-BT) [7]. The total doses necessary in HDR-AL can therefore be reduced to a range of 30— 50 Gy compared with pro tracted LDR-BT with a dosage of 60—80 Gy [8 — 12]. In 1974 HDR-AL was started with the DECATRON, a remotely controlled afterloading device, which was developed and produced in the German Democratic Republic (Fig. 1 ). The DECATRON works with 192 Ir sources, its source activity lying between 185 and 740 GBq (5—20 Ci). The isodose forming is achieved by an electromechanically controlled step by step switch. The size of a step is 5 to 10 mm. The metal applicator is externally fixed at the treatment couch, which guarantees exact and geometrically determined constant dose distribution in a higher fractionated treatment regimen. The position of the applicator is controlled by orthogonal X-ray examinations and the position can be changed in accordance with the treatment plan. IAEA-SM-290/83 15

From 1974 to December 1983 more than 2500 patients with gynaecological tumours were treated, of whom 1131 had cervical carcinoma. The staging of all cases of cervical carcinoma is shown in Table I. A total of 612 women were irradiated primarily intracavitarily, and 519 post-operatively intravaginally. The grouping by age and stage is a m atter of negative selection, as the referring clinics prefer to orientate themselves by surgery. The average age of the primarily treated patients was 61.6 years, that of the post-operatively irradiated patients was 46.6 years.

TABLE I. STAGING OF CERVICAL CARCINOMA TREATED WITH HDR-AL FROM 1974-1983

Stage Post-operative treatment Primary treatment Total (No.) (%) (No.) (%) (No.) (%)

I 138 51.9 128 48.1 266 23.5 II 312 63.2 182 36.8 494 43.7 III 65 19.4 270 80.6 335 29.6 IV 4 11.1 32 88.9 36 3.2

Total 519 45.9 612 54.1 1131 100

Average age 46.6 61.6 54.7 (years) (21-74) (30-89) (21-89)

TABLE II. HISTOLOGICAL CLASSIFICATION OF CERVICAL CARCINOMA TREATED WITH HDR-AL FROM 1974-1983

Histology Primary treatment Post-operative treatment Total (%) (No.) (%) (No.) (%) (No.)

Keratinizing squamous cell 10.6 120 9.1 103 19.7 223 carcinoma Non-keratinizing squamous cell 40.6 459 32.5 368 73.1 827 carcinoma Adenocarcinoma 2.7 30 3.7 42 6.4 72 Other kinds 0.3 3 0.5 6 0.8 9

Total 54.2 612 45.8 519 100.0 1131 16 GLASER

TABLE III. TREATMENT REGIMEN IN PRIMARY IRRADIATION OF CERVICAL CARCINOMA Comparison between LDR-BT and HDR-AL

Conventional technique Afterloading technique (low activities) (high activities)

Activity 10-17 GBq 137Cs 185-740 GBq 192 Ir Dose rate at —point A up to 5 cGy/min up to 3 Gy/min —point В up to 1 cGy/min up to 0.6 Gy/min Fractions 3 5 -6 Single dose at -p o in t A 15-20 Gy 6 -7 Gy —point В 3 -4 Gy 1.2-1.5 Gy Treatment time 8 -1 2 h 3—30 min Interval 1 -2 weeks 1 week Overall time 2 -4 weeks 4 -5 weeks Total dose at -p o in t A 4 5 -6 0 Gy 3 5 -4 2 Gy -p o in t В 9 -1 2 Gy 6 -9 Gy

Additional external beam therapy 20-25 X 1.3 Gy = 26-32.5 Gy at point A X 2.0 Gy = 4 0 -5 0 Gy at point В

Complete dose at — point A 70 -8 6 Gy 6 1 -6 8 Gy -p o in t В 50 -6 0 Gy 50-55 Gy

In Table II the histological classification is presented. More than 90% of all cases of cervical carcinoma were variously differentiated squamous cell carcin o m a. The treatm ent regimen in primarily irradiated cervical carcinoma is shown in Table III and is compared with the conventional methodology of low dose rate brachytherapy (LDR-BT). In 5 to 6 fractions with single doses of 7 Gy, a total dose of 35 to 42 Gy at Point A was applied. This was followed by additional external beam therapy. The total doses lay between 61 and 68 Gy at Point A and between 50 and 55 Gy at the Point B. Table IV shows the results of the primary treatm ent of cervical carcinoma in 612 patients, stages I—IV. IAEA-SM-290/83 17

TABLE IV. RESULTS OF PRIMARY IRRADIATION OF CERVICAL CARCINOMA Absolute survival rate in all 612 patients

Stage I = 128 II = 182 III = 270 IV = 32 I-IV = 612

Relapse free % No. % No. % No. % No. % No. survival

After years 1 94.5 (121/128) 90.1 (164/182) 69.3 (187/270) 0 (0/32) 77.1 (472/612) 2 78.4 (80/102) 74.7 (109/146) 51.8 (132/225) 67.9 (321/473) 3 74.7 (65/87) 63.2 (7 2 /1 14) 51.8 (86/166) 60.8 (223/367) 4 60.7 (37/61) 59.7 (46/77) 49.2 (59/120) 55.0 (142/258) 5 59.6 (28/47) 57.1 (32/56) 44.9 (35/78) 52.5 (95/181)

Control biopsy (271 /349) = 77.7% histologically negative Autopsy (64/106) = 60.4% histologically negative Average age: 61.6 years

As mentioned above this is a negative selection regarding age and stage. This becomes especially apparent in the comparatively low survival rates of only about 60% in stages I and II during which many of the patients died intercurrently and free of tumour. The effectiveness of a combination of after loading and external beam therapy shows in stage III with a survival rate of about 45%. In stage IV only a palliative or symptomatic treatm ent was carried out; all of those patients died within the first year of control. In 271 of 349 clinically tumour-free patients a control biopsy or cytological control after 6 to 8 months did not reveal any tum our tissue. This fact corresponds to a local freedom from tumours of 77.7%, which was proved histologically. In 64 of 106 (60%) dead patients autopsy demonstrated local freedom from tumours, too. A total of 204 patients died intercurrently or from distant métastasés; eight of them from secondary tumours of the lungs, bladder, pancreas and kidneys. In 19 cases (3.1%) a local relapse occurred. In 10 of these 19 patients a therapeutic effect was achieved by secondary irradiation with HDR-AL. A total of 119 patients in this group could only be treated palliatively or symptomatically. For curative aims 493 patients could be analysed. The results of the curative treatment of these 493 stage I—III patients are recorded in Table V. The relapse free 5 year stage related survival rate was 62.2% in stage I, 59.3% in stage II, 57.4% in stage III and 59.4% for all cases in stages I—III. The treatm ent regimen in post-operative irradiation in cervical carcinoma is shown in Table VI. In 4 fractions with single doses between 7.5 and 10 Gy, 18 GLASER

TABLE V. RESULTS OF PRIMARY TREATMENT OF CERVICAL CARCINOMA Relative survival rate with curative aim in 493 cases

Stage I = 125 II = 180 III = 188 I—III = 493

Relapse free % No. % No. % No. % No. survival

After years 1 96.8 (121/125) 91.1 (164/180) 99.5 (187/188) 95.7 (472/493) 2 80.0 (80/100) 75.7 (109/144) 83.5 (132/158) 79.9 (321/402) 3 76.5 (65/85) 64.3 (72/112) 70.5 (86/ 122) 69.9 (223/319) 4 62.7 (37/59) 61.3 (46/75) 67.0 (59/88) 64.0 (142/222) 5 62.2 (28/45) 59.3 (32/54) 57.4 (35/61) 59.4 (95/180)

TABLE VI. TREATMENT REGIMEN IN POST-OPERATIVE IRRADIATION OF CERVICAL CARCINOMA Comparison between LDR-BT and HDR-AL

Conventional technique Afterloading technique (low activities) (high activities)

Activity 7-10 GBq 137Cs 185-740 GBq 192Ir Dose rate in up to 4 cGy/min up to 2.5 Gy/min 1 cm tissue depth Fractions 3 4 Single dose 20 Gy 7.5-10 Gy Treatment time 8 -1 2 h 2—40 min Interval 1 —2 weeks 1 week Overall time 2 -4 weeks 3 weeks Total dose 60 Gy 3 0 -4 0 Gy

Additional external beam therapy 25 X 2 Gy = 50 Gy iliacal in stage II + III IAEA-SM-290/83 19

TABLE VII. RESULTS OF POST-OPERATIVE IRRADIATION IN CERVICAL CARCINOMA Absolute survival rate in all 519 cases, stages I—IV

Stage 1 = 138 II = 312 III = 65 IV = 4 I—IV = 519

Relapse free % No. % No. % No. % No. % No. survival

After years 1 98.6 (136/138) 94.2 (294/312) 75.3 (49/65) 0 (0/4) 92.3 (479/519) 2 94.8 (127/134) 87.3 (240/275) 50.9 (26/51) 85.4 (393/480) 3 91.5 (1 18/129) 81.2 (186/229) 36.6 (15/41) 79.9 (319/399) 4 87.6 (106/121 76.4 (146/191) 40.7 (1 1/27) 77.6 (263/339) 5 82.2 (88/107) 66.7 (84/126) 42.1 (8/19) 71.4 (180/252)

Average age: 46.6 years

total doses between 30 and 40 Gy were applied in 10 mm tissue depth above the vaginal stump. In stages II and III percutaneous external beam therapy was additionally applied. The total doses to the vaginal stump lay between 60 and 65 Gy and in the area of the pelvis wall at 55 Gy. The results of post-operative irradiation of cervical carcinoma in 519 patients in all stages are presented in Table VII. As can be seen the average age of this group lies below that of the primarily treated patients and the results in the various stages are as follows: 82.2% in stage 1, 66.7% in stage II, 42.1% in stage III and 71.4% in all stages I—IV. All 4 patients in stage IV died within the first year of control. A total of 31 patients with post-operative treatment could only be irradiated palliatively or symptomatically. A total of 488 patients could be analysed for curative aims. The results of the curative treatm ent of these 488 patients are presented in Table VIII. The relapse free 5 year survival rate was 83.8% in stage I, 69.4% in stage II, 53.3% in stage III, and for all the curatively treated cervical carcinoma in stages I—III was 74.7%. So far 112 women have died; 18 of them intercurrently and clinically free of tumours. In 14 patients (2.7%) métastasés developed in the vaginal stump. The results of primary and post-operative treatment of all 1131 patients with cervical carcinoma in relation to the stage are shown in Table IX. In stage I 75.3% survived for 5 years free of tum our, in stage II 63.7%, in stage III 44.3% and for all cases a total 5 year survival rate was analysed at 63.5%. 2 0 GLASER

TABLE VIII. RESULTS OF POST-OPERATIVE IRRADIATION IN CERVICAL CARCINOMA Relative survival rate in 488 cases with curative treatment

Stage I = 136 II = 303 III = 49 I-III = 488

Relapse free % No. % No. % No. % No. survival

After years 1 100 (136/136) 97.0 (294/303) 100 (49/49) 98.2 (479/488) 2 96.2(127/132) 90.2 (240/266) 76.5 (26/34) 91.0 (393/432) 3 92.9(1 18/127) 84.5 (186/220) 60.0 (15/25) 85.8 (3 19/372) 4 89.1(106/119) 79.3 (146/184) 57.9 (11/19) 81.7 (263/322) 5 83.8 (88/105) 69.4 (84/121) 53.3 (8/15) 74.7 (180/241)

TABLE IX. RESULTS OF PRIMARY AND POST-OPERATIVE IRRADIATION OF CERVICAL CARCINOMA Absolute survival rate in all 1131 patients

Stage I = 266 II = 494 III = 335 IV = 36 I—IV = 1131

Relapse free % No. % No. % No. % No. % No. survival

After years 1 96.6 (257/266) 92.7 (458/494) 70.4 (236/335) 0 (0/36) 84.1 (951/1131 2 87.7 (207/236) 82.9 (349/421) 57.2 (158/276) 76.5 (714/933) 3 84.7 (183/216) 75.2 (258/343) 48.8 (101/207) 70.8 (542/766) 4 78.6 (143/182) 71.6 (192/268) 47.6 (70/147) 67.8 (405/597) 5 75.3 (116/154) 63.7 (1 16/182) 44.3 (43/97) 63.5 (275/433)

Average age: 54.7 years

As mentioned above, of these 1131 patients, 150 women could be treated only palliatively or symptomatically and 981 patients were irradiated curatively. The results of these curatively treated women are presented in Table X. The 5 year stage related survival rate was 77.3% in stage I, 66.3% in stage II, 56.6% in stage III, and 68.6% for all cases in stages I—III. IAEA-SM-290/83 21

The above mentioned results are compared with those of 623 patients who were treated from 1970-1976 with conventional LDR-BT. This comparison is shown in Table XI. The results in stage I are similar, with 77.3% in HDR-AL and 79.5% in LDR-BT. The improvement of the 5 year results in stage II with 66.3 to 58.9% and in stage III with 56.6 to 21.6% as well as of all cases treated with HDR-AL is statistically highly significant with p = 0.001. In Tables XII and XIII the early and late side effects on bladder and rectum observed with HDR-AL are described and compared with those observed with conventional LDR-BT. The early and late side effects on the urinary bladder are shown in Table XII. Early reactions were observed with HDR-AL only in 2.2% as compared with 14.1% with LDR-BT. The frequency of late reactions could be reduced from 15.4 to 0.8%. Only one vesicovaginal fistula was observed. The early and late side effects on the rectum and bowel are shown in Table XIII. Early reactions could be decreased from 31.0 to 2.3% and the proportion of late reactions was diminished from 22.9 to 2.0%. Only two rectovaginal fistulae occurred. This decrease in early and late side effects with HDR-AL in comparison with LDR-BT is statistically significant (p = 0.01). Table XIV shows frequencies of early and late side effects on bladder and rectum in dependence on the single and total exposures. A statistically significant dose dependence for the increase in early and late side effects at single exposures of more than 10 Gy (p = 0.01) and at total exposures of more than 40 Gy (p = 0.001) was observed. In Table XV our 5 year results are compared with those of the international statistics in the Annual Report, Vol. 18. As regards cervical carcinoma the

TABLE X. RESULTS OF PRIMARY AND POST-OPERATIVE IRRADIATION OF CERVICAL CARCINOMA Relative survival rate with curative aim in 981 cases

Stage I = 261 11 = 483 III = 237 1-111 = 981

Relapse free % No. % No. % No. % No. survival

After years 1 98.5 (257/261) 94.8 (458/483) 99.6 (236/237) 96.9 (951/981) 2 89.2 (207/232) 85.1 (349/410) 82.3 (158/192) 85.6 (714/834) 3 86.3 (183/212) 77.7 (258/332) 68.7 (101/147) 78.4 (524/691) 4 80.3 (143/178) 74.1 (192/259) 65.4 (70/107) 74.4 (405/544) 5 77.3 (116/150) 66.3 (1 16/175) 56.6 (43/76) 68.6 (275/401) 2 2 GLASER

TABLE XI. COMPARISON OF THE RELAPSE FREE 5 YEAR SURVIVAL RATES BETWEEN LDR-BT AND HDR-AL IN THE TREATMENT OF CERVICAL CARCINOMA

Low dose rate brachytherapy High dose rate afterloading (60Co or 137Cs) 1970-1976 ( 192Ir) DECATRON 1974-1983

No. = 623 No. = 981 288 primary 335 post-op. 493 primary 488 post-op.

Relapse free survival 32.6 64.5 59.4 74.7 after 5 years (%)

Among them stage I 79.5 77.3 11 58.9 66.3 III 21.6 56.6

I—III 49.8 68.6

TABLE XII. COMPARISON OF EARLY AND LATE SIDE EFFECTS ON THE BLADDER BETWEEN LDR-BT AND HDR-AL IN CERVICAL CARCINOMA

Early Late reactions reactions Cystitis Shrinking Ureter Fistulae bladder stenosis

Frequency % No. % No. % No. % No. % No.

Low dose rate brachy 14.1 (88/623) 4.5 (28/623) 2.8 (17/623) 3.8 (24/623) 4.3 (27/623) therapy

High dose rate afterloading 2.2 (25/1131) 0.3 (3/1131) 0.2 (2/1131) 0.2 (2/1131) 0.1 (1/1131) by DECATRON IAEA-SM-290/83 23

TABLE XIII. COMPARISON OF EARLY AND LATE SIDE EFFECTS ON RECTUM AND BOWEL WITH LDR-BT AND HDR-AL IN CERVICAL CARCINOMA

Early Late reactions reactions Proctitis Ulcus; Stenosis Fistulae bleeding

Frequency % No. % No. % No. % No. % No.

Low dose rate brachy 31.0 (187/623) 10.0 (62/623) 1.9 (13/623) 4.7 (29/623) 6.3 (38/623) therapy

High dose rate afterloading 2.3 (26/1131) 1.2 (14/1131) 0.2 (2/1131) 0.4 (5/1131) 0.2 (2/1131 by DECATRON

TABLE XIV. FREQUENCIES OF EARLY AND LATE SIDE EFFECTS ON BLADDER AND RECTUM IN DEPENDENCE ON SINGLE AND TOTAL RADIATION EXPOSURE

Dose Bladder Rectum (Gy) side effects (%) side effects (%) Early Late Early Late

Single exposure -7 .5 1.0 - 0.3 - - 1 0 3.9 - 10.3 3.1 > 1 0 22.9 5.4 36.2 28.9

Total exposure - 2 0 - - - - -3 0 7.4 - 13.2 5.7 -4 0 16.5 3.1 23.1 10.8 >40 30.0 13.0 86.7 73.3 2 4 GLASER

TABLE XV. COMPARISON OF THE 5 YEAR RESULTS WITH INTERNATIONAL STATISTICS FROM THE ANNUAL REPORT, VOL. 18

Stage All stages Institution I II III IV I-IV No. (%) (%) (%) (No.) (%) (No.)

UFK Innsbruck 323 83.2 69.2 42.5 (0/4) 61.9 (200/323) UFK Helsinki 266 77.3 53.2 35.5 (1/3) 57.9 (154/266) UFK Jena 243 89.1 43.6 25.0 (0 / 1) 67.9 (165/243) FK Karl-Marx-St. 243 85.1 66.7 32.0 (0/8 ) 65.0 (158/243) UFK Leipzig 662 79.8 45.6 21.7 (1/4) 61.9 (410/662) UFK Rostock 362 85.3 53.7 31.9 (0/5) 59.4 (215/362) UFK Gießen 442 80.9 52.6 30.6 (1/9) 64.2 (271/422) UFK Göttingen 255 83.3 57.5 22.4 (1/15) 55.7 (142/255) UFK Würzburg 298 76.4 55.8 35.3 (0/4) 55.4 (165/298)

Halle 1974-1979 442 75.3 69.2 44.3 (0/4) 64.4 (285/442)

TABLE XVI. ADVANTAGES AND RADIOBIOLOGICAL PECULIARITIES OF HDR-AL COMPARED WITH LDR-BT

Advantages Problems - Radiation protection — Refusal of protraction - Short treatment time — Absolutely changed dose-time relationships - External applicator fixation — Decreased therapeutic scope compared - Constant reproducible irradiation geometry with low dose rate - Psychological and physical relief — Higher radiobiological effectiveness of - Abolition of primary treatment mortality higher single doses - Irradiation planning more exact and variable — Higher fractionation necessary - Out-patient treatment (c. 40% of all patients) — Adapted overall treatment time — Equivalent dosage to achieve isoeffects Aims - Proved efficiency of treatment - Rate of healing - Good compatibility - Acceptable proportion of side effects IAEA-SM-290/83 25 results achieved at the Martin Luther University, Halle, in stage I with 75.3% are in keeping with the international average; with 69.2% in stage II they are above the international average and in particular in stage III with 44.3%; for all treated cases the 5 year survival rate was 64.4%. The advantages and radiobiological peculiarities of HDR-AL are summarized in Table XVI. The advantages are: protection of staff from exposure, short treatment times, external fixation of the applicators, constant and reproducible irradiation geometry, psychological and physical relief of the patients, elimination of primary treatment mortality by application without general anaesthesia, more exact and variable irradiation planning, in more than 40% of all cases HDR-AL treatm ent was carried out on an out-patient basis. The radiobiological peculiarities consist in the fact that HDR-AL breaks with the classic principle of protraction in brachytherapy. The fundamentally changed dose-time relationship in HDR-AL includes a decreased therapeutic scope as compared with LDR-BT. The comparatively higher radiobiological effectiveness of high dose rate irradiation and also of relatively higher single doses between 5 and 10 Gy per fraction requires a higher fractionation. It must be added that the overall treatment time should be long enough to allow for the repair and recovery processes of the healthy tissue. The dose distribution at HDR-AL must aim at isoeffects with an equivalent dosage to achieve the therapeutic target, namely equally good or improved survival rates and an acceptable proportion of side effects [13 — 15]. The facts presented prove that, considering the radiobiological peculiarities of the dose-time relationship with HDR-AL in addition to the recognized advantages, results can also be obtained in primary and post-operative irradiation of gynaecological tumours which are of the same value as those with conventional LDR-BT, and in some cases better, and that the frequency of side effects on the bladder, rectum and bowel can be significantly reduced.

REFERENCES

[1] O’CONNELL, D., JOSLIN, C.A.F., HOWARD, N., RAMSEY, N.W., LIVERSAGE, W.E., The treatment of uterine carcinoma using the Cathetron, Br. J. Radiol. 40(1967) 882. [2] JOSLIN, C.A.F., SMITH, C.W., The use of high activity cobalt-60-sources for intracavitary and surface mould therapy, Proc. R. Soc. Med. 63 (1970) 1029. [3] ELLIS, F., Nominal standard dose and the ret, Br. J. Radiol. 44 ( 1971) 101. [4] LIVERSAGE, W.E., Radiotherapy of carcinoma of the cervix: Relevance of dose-rate, Proc. R. Soc. Med. 66 (1973) 940. [5] NOTTER, G., “Clinical comparison of different regimes of fractionation and dosage”, Summary and Conclusions — Radiobiology Session, Special Rep. No. 17, Br. J. Radiol. (1978) 190. [6 ] SNELLING, M.D., LAMBERT, H.E., YARNOLD, J., Clinical results and complications following treatment of carcinoma of the cervix and endometrium using the Cathetron at the Middlesex Hospital, Special Rep. No. 17, Br. J. Radiol. (1978) 32. 26 GLASER

[7] ROTTE, K., Das ferngesteuerte Nachladeverfahren (Remote-Controlled Afterloading) für die intrakavitäre Kontakttherapie (Brachytherapy) gynäkologischer Karzinome, Radiologe 23 (1983) 20. [8 ] GLASER, F.H., RAUH, G., GRIMM, D., SALEWSKI D., MUTH, С.-P., HEIDER, K.-M., KRAFT, M., Das DECATRON — remote afterloading mit hoher Dosisleistung in der Kontakt-Curie-Therapie, Radiobiol. Radiother. (Berlin) 18 (1977) 707. [9] GLASER, F.H., Decatron remote afterloading therapy with high active sources, Special Rep. No. 17, Br. J. Radiol. (1978) 5 1. [10] GLASER, F.H., RAUH, G., Neue Aspekte in der Kontaktgammatherapie. DECATRON— ferngesteuertes Afterloading mit hohen Aktivitäten, Dtsch. Gesundheitsw. 33(1978)2155. [11] GLASER, F.H., Klinische Erfahrungen bei der Afterloading-Therapie gynäkologischer Tumore mit hohen Aktivitäten im Kurzzeit-Verfahren, 4th European Congr. of Radiology, Hamburg, 4 —8 Sep. 1979. [12] GLASER, F.H., GRIMM, D., HÄNSGEN, G., RAUH, G., Fraktioniertes Kurzzeit- Afterloading bei der Behandlung von Uteruskarzinomen, 15th Int. Congr. of Radiology, Brussels, 24 June-1 July 1981. [13] GLASER, F.H., GRIMM, D., HÄNSGEN, G., RAUH, G., HEIDER, K.-М., KRAFT, M., SALEWSKI, D., SCHUCHARDT, V., Fraktioniertes Kurzzeit-Afterloading mit hohen Dosisraten bei der Behandlung von gynäkologischen Tumoren, Radiobiol. Radiother. (Berlin) 23(1982) 481. [14] GLASER, F.H., GRIMM, D., HEIDER, K.-М., Zum Einfluß der zeitlichen Dosisverteilung bei protrahierter und fraktionierter Brachytherapie gynäkologischer Tumore. Mathematisch formulierte Modellvorstellungen und klinische Erfahrungen bei fraktioniertem Kurzzeit- Afterloading mit hohen Dosisraten, Radiobiol. Radiother. (Berlin) 25 (1984) 23 1. [15] GLASER, F.H., GRIMM, D., HÄNSGEN, G., RAUH, G., SCHUCHARDT, V., Klinische Erfahrungen bei der Afterloading-Kurzzeittherapie im Vergleich zur konventionellen Brachytherapie bei der Behandlung gynäkologischer Tumore, Strahlen therapie 161 (1985) 459. IAEA-SM-290/13

HIGH DOSE RATE AFTERLOADING IN THE TREATMENT OF CERVIX CARCINOMA WITH EXTERNAL IRRADIATION AND BRACHYTHERAPY

T. K U IP E R S Rotterdam Radio-Therapeutic Institute and the Daniel den Hoed Cancer Centre, Rotterdam, Netherlands

Abstract

HIGH DOSE RATE AFTERLOADING IN THE TREATMENT OF CERVIX CARCINOMA WITH EXTERNAL IRRADIATION AND BRACHYTHERAPY. The main dose at the cervix is delivered by means of external radiotherapy without central shielding. A midplane dose of 4600 cGy is delivered in 23 fractions. This is followed by high dose rate brachytherapy applied as a booster, administered in two fractions, each delivering 850 cGy at point A: 425 cGy from intrauterine and 425 cGy from vaginal sources. A rigid stainless steel tripartite afterloading applicator is inserted and fixed at a tenaculum which is attached to the posterior lip of the cervix. A rectal retractor is placed against the posterior vaginal wall and fixed at the treatment couch together with the applicator and the tenaculum. Metal markers on a foam plastic sponge are placed against the anterior rectal wall. A catheter behind the sponge is used for filling the sigmoid colon with contrast medium later on. The small bowel is visualized by means of orally administered contrast medium. For bladder dosimetry a thin metal chain is inserted which serves as marker for the area of the bladder base where late reactions most frequently occur: the cranial part of the medial section of the bladder base and the adjacent part of the posterior wall. For dose determination at neighbouring organs, lateral views are covered with sets of iSodose curves prepared with the same magnification factor and representing separately the dose distribution of intrauterine and vaginal sources. Brachytherapy is applied in a non-protected room, therefore the patient is moved into a tunnel. The time necessary for irradiation may vary between 5 and 20 minutes. Survival rates, local recurrence rates and complication rates were similar to those achieved with low dose rate brachytherapy. A standardized treatment technique and a checklist facilitate training in this field. The short application times provide optimal conditions for dosimetry, make long lasting bedridden isolation unnecessary and enable out-patient treatment.

1. TREATMENT POLICY

The main dose at the cervix is administered by means of external radio therapy (ERT) without central shielding. In 23 daily fractions of 200 cGy each, a midplane dose of 4600 cGy is delivered at the area of the primary tumour and the regional nodes. High dose rate brachytherapy (HDB) is applied as a booster delivered in two sessions: the first two weeks after the end of ERT, the second

27 28 KUIPERS

two weeks later. In both applications vaginal ovoids are combined with an intrauterine tandem. In each session 850 cGy are delivered at point A: 425 from sources in the ovoids and 425 from the intrauterine sources. In that way a booster of 1700 cGy is added to the 4600 cGy from ERT. The result is a total dose of 6300 cGy at point A, which corresponds to a TDF of 126. The total dose at point В is 5200 cGy, which corresponds to a TDF of 82. One beneficial effect of ERT applied first and without central shielding is a shrinking of the primary tumour, which makes the geometrical conditions more favourable for later brachytherapy. Other merits of this policy are: inaccuracies inherent in the application of central shielding are avoided and only two applications are necessary.

2. APPLICATION TECHNIQUE

The first step is the preparation with the following items:

(1) Enema with laxative. (2) Oral administration of 300 cm3 of contrast medium one hour before treat ment, in order to enable dosimetry of the small bowel. (3) Epidural anaesthesia, which can be better adapted to the duration of treat ment than lumbar anaesthesia. (4) The legs are fixed in the gynaecological position.

Preparation is followed by the insertions. After dilation of the cervical canal a rigid stainless steel applicator is inserted with its flange against the cervix. Then a tenaculum is attached to the posterior lip of the cervix and its handles are fixed at the stem of the intrauterine tube. The ovoids are inserted separately and similarly fixed. Next a rectal retractor is placed against the posterior vaginal wall and also fixed at the applicator. The combination of the rigid applicator and the tenaculum has the following advantages:

(1) The applicator cannot slip out of the uterine canal. (2) In the case of a latero-deviation of the uterus, positioning of this organ with the uterine axis into the median plane can nearly always be achieved. . (3) The uterine applicator can be fixed in the optimal anteversion with regard to the distance from the sigmoid colon and the small bowel. Usually an angle of 35° is to be preferred. (4) To prevent the uterus from being moved upward and the sources coming too near to neighbouring bowel sections, the cervix is kept at its original craniocaudal level during the intracavitary applications.

In an earlier study it was shown that a firmly inserted vaginal gauze packing often caused an upward displacement of the uterus, especially in young, stage I IAEA-SM-290/13 29 patients with supple vaginal walls. This was associated with a high percentage of severe sigmoid colon radiation lesions.

3. DOSIMETRY

Particular attention is paid to clinical dosimetry, which involves accurate knowledge of:

(1) Extent of tumour growth; (2) Dose distribution, effects of dose rate and fractionation; (3) Position of neighbouring organs.

Concerning the first item it must be stated that in cases of cervix carcinoma the borders of the tum our cannot be determined accurately, even with a CT scan. Regarding the second item: since the availability of computers the dose distribution can be reliably determined. Radiobiologists have provided much information on the effects of dose rate and fractionation over the last decades; however, many problems still remain to be solved. The third item is important in order to prevent unexpected high doses to neighbouring organs. These have to be visualized, either with a contrast medium or by the insertion of radio-opaque markers. For rectal dosimetry a foam plastic sponge 8 cm long, 3 cm wide and 2 cm thick is inserted into the rectum by means of a pair of tweezers after lubrication with jelly. Its middle third is positioned behind the ovoids and one of its wide surfaces against the anterior rectal wall. This side of the plug is provided with a row of square metal markers along one edge and round markers along the other. The whole is covered with a condom in order to keep it clean. This may, however, cause air bubbles, which can disturb the interpretation of radiographs and also put a distance between the markers and the rectal mucosa. To prevent this, a connection is made with the air of the treatment room by means of a plastic tube protruding out of the anus. A disposable second covering condom makes it possible to use the plug for several applications. After insertion of the rectal sponge has been completed, a catheter is placed against the posterior rectal wall with its top 2 cm above the upper border of the rectal plug. This enables the sigmoid colon to be filled with contrast medium later on. The bladder is a difficult organ for dosimetry; its shape is dependent on its contents. At a transverse section of a filled bladder three compartments can be distinguished: two lateral pouches and a central section in between. If the bladder is emptied, the bladder base and the lower part of the posterior wall remain in place, the lateral walls move in the median direction and the vault comes down in plies and folds until it reaches the bladder base. A probe inserted for measurements at this section of the bladder wall will easily be caught between these structures and be deviated in the ventral direction. In such a case no 30 KUIPERS relation will be found between the measured dose and radiation sequelae at the bladder base. Therefore at our institute, before inserting something into this organ for dosimetry, the bladder is emptied and then filled with a standard quantity of 200 cm3 of sterile saline. In Asian countries 150 cm3 appears to be endured much better. In most hospitals the balloon of a Foley catheter filled with contrast medium is used as a marker. To ensure that the balloon is touching the bladder base, the catheter is pulled downwards as much as possible. However, in such a case the balloon is found at the bladder neck and does not indicate that area of the bladder base where radiation reactions are most frequently encountered. This area involves the cranial part of the medial section of the bladder base and the adjacent lower part of the posterior wall. At our Institute, the difficulty of recognizing this area on radiographs performed for dosimetry is solved by the insertion of a thin sterilized silver (or stainless steel) chain. The first section o f the chain is placed at the deepest lateral pouch, the second part crosses the median line, the third section is put into the other lateral pouch and the remainder is led through the urethra with its end protruding out of the external orifice. All insertions being completed, the patient’s legs are stretched and the applicator is fixed in the desired position at the treatment couch. On average, insertion of the applicator and auxiliaries takes 30 minutes, then 60 minutes are spent on dosimetry. First, radio graphs are taken in anteroposterior (AP) direction, next a lateral film is exposed in such a way that an image with a magnification factor of 1.4 is obtained. There after, AP stereographic radiographs are prepared. Then transparencies provided with 1.4 times magnified sets of isodose curves, separately representing the dose distribution from the intrauterine sources and from the sources in both ovoids together, are laid upon the lateral radiograph. This enables the maximal dose at the bladder, rectum and small bowel to be determined easily. Following the preparation of the first series of radiographs, the sigmoid colon is filled with contrast medium and a second series of films is similarly exposed. With these radiographs and transparencies with isodose curves the maximal dose at the sigmoid colon is also determined. The stereographs are routinely used for a check after treatment. The use of standardized source configurations enables working with precalculated isodose curves.

4. TREATMENT

Dosimetry being completed, the table top of the simulator with the immobilized patient and the fixed applicator is moved on a trolley. Intracavitary irradiation is applied in a non-protected room and, therefore, the patient is moved into a tunnel. The time necessary for irradiation may vary between 5 and 20 min. IAEA-SM-290/13 31

Results achieved with HDB were compared with those from the last years when low dose rate brachytherapy (LDB) was applied. Randomization was not done for various reasons. The corrected 5-year survival rates following HDB or LDB are: stage I 82-80%; stage II 74-66%; stage III 34-44%; without statistically significant differences between the two methods. In each stage the survival curves almost coincide. Local and regional recurrence rates also appeared to be similar. The incidence of grade 3 complications was similar in 142 HDB and in 152 LDB patients: bowel damage 6% and urological lesions 3.5%. Most HDB patients were treated before the currently applied standard method had been introduced. In 68 patients treated with the HDB standard method, grade 3 bowel complications occurred in 4.4% and grade 3 urological complications in 1.5%. The advantages associated with HDB applications seem to make this type of treatment very useful in hospitals with large numbers of patients and with restricted numbers of beds. Such conditions are often encountered in developing countries. The figures show that at least equal results can be achieved as with conventional LDB. The question remains whether the HDB technique described above is not too complicated for centres in developing countries. It must be pointed out that, because the HDB technique has been standardized, it can be learned within a relatively short time. In the author’s personal experience a detailed checklist mentioning every act involved in the application is an important aid in training. It appeared to be very useful to provide assistance from a colleague with special skills in this field for 2 or 4 weeks when starting this type of treatment. Emphasis should be laid on the problem of scarcity of experienced hospital physicists in some countries. The machine should therefore be delivered with extensively detailed data on dose distributions and application times. If the requirements are sufficiently met, developing countries can profit fully from the merits of HDB. This has been shown in Indonesia, Taiwan, Thailand and India.

5. CONCLUSION

Survival and complication rates are similar for LDB and HDB. However, the short application times of the latter permit a more reliable dosimetry, are not associated with a long bedridden isolation of the patient and provide the possibility for out-patient treatment. The two latter items mean a lower psychological burden on the patient and a reduction of the cost of treatment for the community.

IAEA-SM-290/29

PRELIMINARY REPORT ON THE TREATMENT OF CARCINOMA CERVIX UTERI WITH A COMBINATION OF CONVENTIONAL IRRADIATION AND MITOMYCIN С AS RADIOSENSITIZER*

S. PURIBHAT Division of Radiation Oncology, Department of Radiology, National Cancer Institute

C. CHOTIGAVANICH, S. TANGKARATT, P. PUNTUMCHINDA Department of Radiology, National Cancer Institute

S. SOMBOONCHAROEN National Cancer Institute

Bangkok, Thailand

Abstract

PRELIMINARY REPORT ON THE TREATMENT OF CARCINOMA CERVIX UTERI WITH A COMBINATION OF CONVENTIONAL IRRADIATION AND MITOMYCIN С AS RADIOSENSITIZER. Between 22 December 1983 and 4 March 1986, 79 cases of cancer of the cervix uteri were given mitomycin С intravenously in two doses of 20 mg each on Day 1 and Day 22-24 in com bination with conventional irradiation. Early complications in the GI tract were minimal. Late complications of radiation proctitis were noted in 7.6% and only 1/7 cases needed radical manage ment. Bone marrow suppression was noted in a fair amount of cases but was not serious. Twelve cases ( 15.2%) showed unsatisfactory results. The early response is promising but it is too soon to observe any long term response and benefit.

1. INTRODUCTION

Cancer of the cervix uteri is a health problem in Thailand as in many other developing countries in this part of the world. It accounted for 15.7% of all cancers, or 29.9% of female cancer in 1981 [1]. There is a shortage o f cancer centres in Thailand. Only ten centres exist in the whole country for a population of 63 million. Of these ten centres, seven are situated in Bangkok, which has a population of 6.7 million. Because of this,

* Supported by IAEA Grant No. 323-E3-RC-277.2.

33 34 PURIBHAT et al. up-country people have no proper health care, not to mention early cancer detec tion. As a result, most of the cervical cancer cases come to us in the late stages of the disease: 49% as stage II, 25% as stage III and 11.6% as stage IV [2]. It has been found that conventional irradiation yields unsatisfactory results in stage III cancer cases (central failure 33.3%, métastasés 17.6% and survival 40% after two years) [3]. Therefore the method of using chemicals as radiosensitizers in treating cervical cancer has been developed.

2. MATERIAL AND METHODS

All stage III cervical cancer cases were included in this study, provided they met the following criteria:

— Known pathology — Haemoglobin more than 10 g% or haematocrit more than 30% — White blood cell count more than 6000 — Platelet count more than 200 000 — Blood chemistry WNL — Chest X-ray WNL — IVP acceptable — No distant métastasés — Age not over 70 — Availability for follow-up — Kanofsky’s score more than 80%.

3. PROTOCOL FOR THE STUDY

All cases were examined thoroughly, including a pelvic examination, before the start of treatment. They were examined again at least once a week during the course of treatment, including CBC and platelet count. Examination under anaes thesia was performed at the time of brachytherapy. The patients were followed up, first, at six weeks post-irradiation and then every three months in the first year and every four months thereafter, with CBC and platelet count done at each visit during the first year. Then CBC and platelet count, urine examination, blood chemistry, Pap smear, chest X-ray and IV pyelo graphy will be performed annually.

4. CHEMICAL

Two doses of mitomycin С (MMC) of 20 mg each were given intravenously on Day 1, and on Day 22-24, each followed by D 5% ^NSS 1000 mL infusion IAEA-SM-290/29 35

TABLE I. AGE DISTRIBUTION

Age (years) No. of patients %

Youngest patient 29 Below 39 4 5.1 40-49 29 36.7 50-59 34 43.0 60-69 12 15.2 Total 79 100.0

including vitamin Bco ( 1 mL) and vitamin С (500 mg) IV in 8 hours. A dose of 10 mg of metoclopramide was given intramuscularly before IV MMC every 4 h as required.

5. IRRADIATION

The patients were given external irradiation using 60Co teletherapy or Linac 4 MeV X-radiation to AP + PA whole pelvis up to 50 Gy in 5-6 weeks (daily dose of 180-200 rad, 5 d/week) followed by pelvic wall irradiation to 4-6 Gy in 2-3 fractions. After a rest of 7-10 days the patients received brachytherapy using radium (Fletcher-Suit’s afterloading tandem and vaginal vault ovoids) for a tumour dose of 30 Gy to point A. Except for a few cases at the very end of the study, they received a dose rate to point A of 21 Gy using the 137Cs RALS (remote afterloading system). During the period from 22 December 1983 to 4 March 1986, 79 cases of cervical cancer underwent this regimen. The youngest was 29 years old. Most of them were in their 5th and 6th decade as shown in Table I. As shown in Table II, most patients suffered stage IIIB, poorly differentiated squamous cell carcinoma (51/79 cases or 64.5%), while moderately differentiated squamous cell carcinoma came next with 14 cases (17.7%), followed by well differentiated squamous cell carcinoma with 7 cases (8.9%) and others (adeno carcinoma, squamous cell carcinoma grading not specified, etc.) with 7 cases (8.9%). Thus the more aggressive the carcinoma, the more advanced is the stage.

6. COMPLICATIONS

A fair number of patients developed some degree of nausea and vomiting (31/79 or 39.2%), 23 cases needed no treatment (29%), 7 cases (8.9%) needed intra- 36 PURIBHAT et al.

TABLE II. PATHOLOGY AND STAGE

Pathology Stage III tí No. of patients %

Poorly differentiated squamous cell carcinoma 51 64.5 Undifferentiated squamous cell carcinoma Moderately differentiated squamous cell carcinoma 14 17.7 Well differentiated squam ous cell carcinoma 7 8.9 Others 7 8.9 Total 79 100.0

TABLE III. IMMEDIATE COMPLICATION, UPPER GI TRACT

Anorexia, nausea and vomiting No. of cases %

Mild, no treatment 23 29 Moderate, anti-emetics and intravenous fluid needed 7 8.9 Severe, medication needed and irradiation interrupted 1 1.3 Total 31 39.2

TABLE IV. IMMEDIATE COMPLICATION, LOWER GI TRACT

No. of cases %

Tenesmus, mild loose bowel movement, 7 8.9 no treatment Loose bowel movement, medication and 12 15.2 IV fluid needed Severe, RT interrupted - - Total 19 24.1

venous fluid up to 5 times but irradiation was carried on as usual and only 1 case (1.3%) needed an interruption in irradiation for a few days as well as intravenous fluid administration (Table III). Regarding the lower abdomen, a total of 19 cases (24.1%) experienced lower gut symptoms. Of these, 7 cases (8.9%) needed no treatment, 12 cases (15.2%) IAEA-SM-290/29 37

TABLE V. IMMEDIATE COMPLICATION, URINARY BLADDER

No. of cases %

Dysuria, no treatment 5 6.3 Dysuria, treatment needed 2 2.5

Severe complication - - Total 7 8.8

TABLE VI. BONE MARROW SUPPRESSION, WHITE BLOOD CELLS

No. of cases %

Score 0 = no change 47 59.5 Score 1 = decreased to 2000-4000 30 38.0 Score 2 = decreased to below 2000 2 2.5

Note: RT interrupted if WBC was below 2000.

needed medication such as Lomotil (diphenoxylate hydrochloride and atropine) or Imodium (loperamide hydrochloride) and/or intravenous fluid. None needed an interruption in irradiation because of lower gut symptoms (Table IV). A total of 7 cases (8.9%) experienced bladder irritation (dysuria) but no haematuria developed. Of these, 5 cases (6.3%) needed not treatment and 2 cases (2.5%) needed local anaesthetics in the form of oral medication, i.e. phenazopyridine hydrochloride. None of them were severe (Table V). Haemoglobin and haematocrit could not be used as indicators in determining bone marrow suppression because of many factors such as bleeding from tumours and blood transfusions to raise the haemoglobin above 10 g% or haematocrit above 30% in order to carry on irradiation as planned. A decrease in white blood cells was experienced in a total of 32 cases (40.5%). Of these, 30 cases (38%) had a WBC between 2000 and 4000. In 2 cases (2.5%) the WBC dropped below 2000 and irradiation had to be interrupted for about a week to allow bone marrow recovery (Table VI). Thrombocytes dropped in a total of 39 cases (49.4%); in 5 of these cases (6.3%) thrombocytes dropped below 100 000 so that irradiation had to be interrupted for about 1-2 weeks to let the platelet count return to over 100 000. None of these experienced petichia or ecchymoses or increased bleeding per vagina. 3 8 PURIBHAT et al.

TABLE VII. BONE MARROW SUPPRESSION, PLATELETS

No. of cases %

Score 0 = no changes 40 50.6 Score 1 = decreased to 100 000-200 000 34 43.0 Score 2 = decreased to below 100 000 5 6.3

Note: RT interrupted if platelet count was below 100 000.

No platelet transfusion was performed (Table VII). There was no case of severe bone marrow suppression, and all of the patients were able to complete the planned treatment.

7. LATE COMPLICATIONS

There were six cases (7.6%) who developed late complications in the form of bleeding with bowel movement, i.e. radiation proctitis. Up to now all of them have received only supportive treatment. Of these, 2 cases were free of disease at 1 year and 2 years. One case developed distant métastasés to the supraclavicular node at 1 year accompanied by bleeding with bowel movement. One case developed bone métastasés at 1 year, received systemic chemotherapy and then developed bleeding with bowel movement at the 2 year follow-up. One case with local recurrence at 4 months follow-up had a second radium application and at one year follow-up the tumour persisted locally with rectovaginal fistula. This last patient had additional systemic chemotherapy for persistent tumour (third MMC 20 mg IV). At 2 years she was living tumour free but with occasional bleeding with bowel movement but needed no definite treatment.

8. RESULTS

Up to the time of this report the patients had been followed up from 4 to 24 months. There were 12 cases (15.2%) with unsatisfactory results (Table VIII). Of these, 1 expired at the third week of treatment from fulminating infection. One expired with disease at eight months after completion of treatment. The rest were found to have tumours and all but one had chemotherapeutic consulta tion and refused or was lost to follow-up. One case was found to have recurrent disease at the 4-month follow-up. She received a second radium application and was free of disease at 1-year follow-up but with radiation proctitis, needing only IAEA-SM-290/29 39

TABLE VIII. UNSATISFACTORY RESULTS

No. of cases %

Recurrent tumour at 4 month follow-up, 1 1.3 had chemotherapy Rx NED 1 year Persistent tumour, refused further treatment 9 11.3 Expired after 3 weeks of treatment from 1 1.3 fulminating infection Expired with disease at 1 year 1 1.3 Total 12 15.2

conservative treatment. Of these unsatisfactory cases two cases had distant métas tasés, one with pulmonary metastasis and one with supraclavicular and subcutaneous métastasés. Both cases were under local control.

9. CONCLUSION

A total of 79 cases of cervical cancer were included in the trial. Complications were acceptable. There were 12 cases (15.2%) with unsatisfactory results. Other wise the early response has been promising, but the period of follow-up is too short to see whether there is any real benefit, i.e. whether the long term survival is as good.

REFERENCES

[1] Thai Cancer Statistics by the National Cancer Institute (1981). [2] PURIBHAT, S., et al., Thai Cancer 10 4 (1984). [3] CHOTIGAVANICH, C., et al., Thai Cancer 9 (1983) 27.

IAEA-SM-290/S9

THE ROLE OF INTRA-ARTERIAL CHEMOTHERAPY IN THE MANAGEMENT OF UTERINE TUMOURS WITH A COMPLEX MODALITY TREATMENT

J. STUMPF*, G. VADON**, G. NEMETH* * Oncoradiological Centre, E. Weil Hospital and Polyclinic ** Radiological Clinic of the Medical University Budapest*

Budapest, Hungary

Abstract

THE ROLE OF INTRA-ARTERIAL CHEMOTHERAPY IN THE MANAGEMENT OF UTERINE TUMOURS WITH A COMPLEX MODALITY TREATMENT. The number of advanced uterine malignancies has decreased in the industrialized countries in recent decades, but even screening processes are less effective in a certain population and, therefore, a considerable number of advanced cases is detected every year. In the developing countries the proportion of advanced carcinomas is higher owing to less effective screening. The most widespread method for controlling stage III, IV uterine tumours is radiotherapy but its effectiveness is unsatisfactory. The best five-year survival rates are over 40% but usually not even 30-35% are achieved for stage III cervical carcinomas. For this reason a new protocol was started: a complex treatment involving chemotherapy, irradiation and, where possible, surgery. The initial results have been promising. In the past few years 16 patients were treated according to this protocol. A complete treatment was possible in all cases, of which 8 became operable and all but 2 have remained free of diesease. The mean survival time of the 6 cured patients is at present 20 months. Eight patients were not operated. In one case surgery was omitted because of complications during the long term intra-arterial chemotherapy, but the patient has been in complete remission for 21 months. Two other patients have been in complete remission. One deceased and 4 are in partial remission. Nine of the 16 patients who received the complete treatment have been in complete remission. This seems to be promising, even taking into account that the complex treatment was started in 16 cases.

1. INTRODUCTION

Effective screening has led to a relatively low proportion of advanced uterine malignancies in those countries with a good oncological service. In contrast, in the developing countries, malignant tumours are more often diagnosed only in their advanced stages.

+ Present address: Radiological Clinic of the Medical University Szeged, Hungary.

41 42 STUMPF et al.

Both early cervical and endometrial uterine tumours can be highly effectively controlled by surgery, by irradiation or with a combination of these methods. The results in advanced cases are far less satisfactory. The five-year survival rate of stage III uterine tumours usually does not amount to 35%, while only very few patients with stage IV cervical tumours survive the five year period [1 ]. In recent decades there have been no reliable developments in the techniques and results of the radiation therapy of advanced tumours [2]. A complex therapy consisting of cytostatic chemotherapy, irradiation and surgery seems to be a promising method of improving tumour control [3].

2. CYTOSTATIC CHEMOTHERAPY METHOD

Different cytostatics, administered in various ways, have been tested for the control of advanced uterine tumours with differing success. The most commonly used cytostatics were methotrexate, bleomycin, adriamycin, cis- platinum and 5-Fu. As far as the administration route is concerned locoregional infusion has proved to be superior to general administration in many tumours. Unfortunately, no specially effective cytostatic has been found for the control of uterine malignancies (except methotrexate in choriocarcinoma) and, therefore, locoregional administration of cytostatics seems to be essential to achieving a good response. It is well known that after radiation therapy reduced vascularity in the tumour bearing area can be expected. This fact may be responsible for the reduced effectiveness of cytostatic treatment after irradiation. Therefore, intra-arterial chemotherapy (alone or combined with intracavitary irradiation) should be the first stage o f a complex treatment (Fig. 1). A maximal reduction of the tumour size can be expected owing to the intact vascularity of the tumour and to the high locoregional cytostatic concentration. In the case o f dangerous tumour bleeding simultaneous intracavitary irradiation is needed. This appears to be a radiochemotherapy method but, in fact, in cases of tumours involving the small pelvis, it is not. Only a very low proportion o f the irradiation dose given by the intracavitary source will reach the pelvic wall (or even the area of parametrial infiltration). Therefore, as far as parametrial infiltration is concerned, a reliable effect is only to be expected from a cytostatic treatment. There is no doubt about the merely transient control of most solid tumours with cytostatics. Usually the tumour free interval after cytostatic treatment does not last more than a couple of months or a year, even if complete regression has been observed. The working hypothesis lies close at hand: after a successful cytostatic or combined (chemotherapy + radiotherapy) treatment, surgical removal of the organ can extend the tumour free interval. Therefore, IAEA-SM-290/59 43

FIG. 1. Scheme of the complex treatment.

in every case where an originally advanced cervical tumour became technically operable, surgery was performed.

3; PATIENT MATERIAL

The complete locoregional cytostatic treatment could be carried out in 16, originally incurable, cervical tumour patients, of which 8 became operable (Fig. 2). Six of these are now tumour free. Two patients developed recurrencies. Three of the 8 non-operated patients are also tumour free. Usually fibrosis in the small pelvis and/or poor general condition prevented surgical intervention. Altogether 9 of the 16 cases showed no evidence of tumour during the follow-up periods of 3 to 48 months. All the cases had histologically verified squamous cell carcinoma.

4. DISCUSSION

Naturally, intra-arterial chemotherapy also involves certain hazards and its application requires extreme caution and precision. The first step is the introduction (via the femoral artery) of an appropriate polyethylene cannula into the artery supplying the tumour area (Seldinger’s 44 STUMPF et al.

Patients with complete treatment (complete intra-arterial chemotherapy)

Extension of disease Technical result Course of disease*

St. 111. 10 Surgery possible -1- NED 6 1 ^ 8 _[_ Recurrent tumour 1 I 1 Died in tumour 16 1 1 I Patients 1 1 Non operable 1- NED St. IV . 3 1 4- PR I 1 8 4 ! 4- Died in tumour I______J-

* Follow-up: 3 -4 8 months

FIG. 2. Treatment results (16 patients given complete treatment).

method) [4]. With the cannula in place and fixed the patient is returned to the oncological ward for infusion o f the cytostatic at predetermined time intervals, usually twice a week. The whole procedure lasts 4 to 28 d. Two systems of cannula implantation may be employed: the open system, in which the cannula passes through the skin and is provided with a stopcock of extracorporeal manipulation, and the closed system, which is based on subcutaneous implantation of a metal or polyethylene portal covered with a silicon membrane (Fig. 3). If the open system is used, the nurse is in charge of regularly checking the operation of the stopcock, since its inadequate closure could lead to a serious, theoretically even fatal, loss of blood. Rinsing of the cannula with a clotting inhibitor (heparin) every 6 h is also essential. With the totally implanted, closed system extracorporeal contact is limited to the period of drug infusion, performed with a special injection needle through the skin and the silicon membrane of the portal (Fig. 4). Because of the reduced hazard of infection, the closed system has been preferred whenever possible. A further advantage of portal implantation is that the interval between rinsings with clotting inhibitor can be extended to 24 h or more. Another advantage of this system is that the portal implanted under the skin of the thigh prevents any displacement of the catheter. Infectious agents may enter the infused region either via the cannula or along it. Rigorous sterility measures are therefore imperative. Isolating the patient from her wardmates and rigorous cleaning of her environment are essential. Sterile instruments should be used and sterile gloves worn for all manipulations of the implanted cannula. Daily wound dressing is required to prevent infection of the stab canal and the bandage should be exchanged immediately whenever the slightest bleeding or contamination is noticed. IAEA-SM-290/S9 45

FIG. 3. Totally implantable infusion system.

The periodic infusion of the cytostatic through the intra-arterial cannula is effected by means of an electric pump. The delivery rate is adjustable according to the requirements (usually 5 mg/h). A portable pump ensures greater comfort for the patient and thereby also improves the success of the therapy. The patients are not confined to bed in the intervals between infusions, but the latter may last as long as 12—14 h. Proper attendance during infusion is therefore of particular importance. The serious hazards of intra-arterial chemotherapy must be taken into con sideration. Thrombosis of the cannulated arteries (usually arteria femoralis at the puncture site) occurs surprisingly rarely. Its frequency basically depends on the material and quality of the cannula inserted. This serious complication can be controlled either by surgical thrombectomy or by thrombolysis (intervention radiology method). Infection through or along the cannula may lead to sepsis and therefore its prevention is imperative. If a totally implantable infusion system is available and is implanted under the usual conditions of surgical sterility, the hazard of infection is relatively low. The general side effects of cytostatics such as myelosuppression or hair loss can be managed in the usual way. 46 STUMPF et al.

FIG. 4. Portal implanted under the skin o f the thigh.

Intervention radiology methods have developed dynamically in recent decades. They usually require only some accessories to the usual angiographic laboratories. Suitable experience in invasive radiological methods and teamwork between the gynaecological surgeon, oncologist and specialists in intervention radiology and radiotherapy are essential.

REFERENCES

[1] NEMETH, G., et al., Gynekol. 100 (1978) 1123. [2] FLETCHER, G.H., Textbook of Radiotherapy, Lea and Febiger, Philadephia (1973). [3] WALLACE, S., in Intervention Radiology (VEIGA-PIRES, J.A., Ed.), Excerpta Medica, Amsterdam-Oxford-Princeton ( 1980). [4] SELDINGER, A., Acta Radiol. (Sweden) 39 (1953) 368. IAEA-SM-290/23

INTRODUCTION OF ADVANCED REMOTE AFTERLOADING BRACHYTHERAPY TECHNIQUES INTO DEVELOPING COUNTRIES, CONSIDERING HIGH ACTIVITY SOURCES, MAINTENANCE AND CARE OF EQUIPMENT

U.M.F. ALTEMARK, S.M. QUANDT Buchler GmbH, Braunschweig, Federal Republic of Germany

Abstract

INTRODUCTION OF ADVANCED REMOTE AFTERLOADING BRACHYTHERAPY TECHNIQUES INTO DEVELOPING COUNTRIES, CONSIDERING HIGH ACTIVITY SOURCES, MAINTENANCE AND CARE OF EQUIPMENT. Remote afterloading equipment was developed to replace radium-226 in the treatment of gynaecological cancer and to give full radiation protection. Originally used by gynaecologists, afterloaders have now become more and more the tool of radiotherapists. The high sophistication of the equipment raises the question of whether this technique can be introduced into developing countries or whether it is needed at all. The isodose distributions necessary for successful brachytherapy can be created by simple equipment. To be able to treat the high number of cervix cases the HDR afterloading technique should be applied rather than LDR manual after loading. Dosimetry and treatment are similar or equal in the two modalities; however, HDR offers distinct advantages in the clinic. Simple HDR equipment offers manageable service and maintenance to hospital staff and no highly trained and paid engineer is required. Support from the manufacturer on a semi-regular basis improves the situation.

When the first commercially available remote afterloading equipment was developed in the 1960s the main intention was to replace radium-226 in the treat ment of gynaecological cancer and to give full radiation protection to medical staff. Recently, afterloaders have become more and more the tool o f radio therapists/radio-oncologists rather than o f gynaecologists. Advanced technology and equipment such as personal computers, semi conductors, miniaturized sources and, last but not least, the re-introduction of hyperthermia have revived abandoned brachy/curietherapy modalities. Mamma, nasopharynx and brain tumours (to name just a few) are already being successfully treated with these methods. New treatment modalities seem to ask for advanced equipment such as Afterloading BUCHLER me, GammaMed Hi and Selectron. All of them have one task in common, namely to transfer by remote control a source or sources under radiation protection into an applicator inside a patient, keep it there for a pre selected length of time and withdraw it into the safe after treatment. Depending

47 48 ALTEMARK and QUANDT on the skill or philosophy of a manufacturer, the source(s) is/are handled pneumatically, mechanically and/or computer controlled. The grade of sophisti cation seems almost unlimited. The only limiting factor is the computer/micro processor used (Figs 1, 2). Do sophistication and expensive equipment improve the economic and medical situation in developing countries, knowing that the isodose distribution — one of the key factors for successful brachytherapy — does not differ very much from equipment to equipment? My answer is a clear NO. This statement is based on the proceedings of the Meeting of the Inter national Working Party and International Conference on Diagnosis and Treatment of Carcinoma o f the Cervix in Developing Areas, held in Thailand, 28 January — 2 February 1985, and, in particular, the statements made by Alfred Goldson, Saroj Gupta, Richard Mould, Margaret Snelling and others [1—4]. Does the patient in developing areas have to renounce treatment methods used in advanced countries? For carcinoma of the cervix the answer is NO, provided the HDR remote afterloading method is used and suitable equipment is chosen. This answer needs to be explained: ( 1 ) Medically it is now accepted that HDR afterloading is safe and useful in the treatment of cancer of the uterus. It has some advantages compared with the LDR technique:

Short application times, better dosimetry and more accuracy, out-patient treatment possible and, therefore, more economic, increased comfort and psychologically better for the patients [5]. (2) The LDR technique allows the treatment of a maximum of one patient per 24 hours regardless of whether manual or remotely controlled afterloading is applied. What is the sociological advantage of successfully treating a few patients with the LDR technique when hundreds more are not treated at all? A simple calculation shows the cost effectiveness of HDR versus LDR equip ment (manual), not considering the buildings which are required in both cases. HDR Remote afterloading treatment 10 treatments/day 50 treatments/week 2500 treatments/year 5 fractions/patient = 500 new patients/year LDR Manual afterloading treatment 1 treatment/day (24 h modality) 5 treatments/week 250 treatments/year 3 fractions/patient = 83 new patients/year HDR US $ 100 000 equipment cost LDR =» US $ 36 000 equipment cost [6] IAEA-SM-290/23 49

Pneumatic drive M icroprocessor control Source trains Back-up com pressor

Cable drive Computer control Stepwise moving source Back-up battery/hand crank

Cable drive Electrom echanical control O scillating source G r a v i t y

Cable drive Computer control O scillating source Back-up battery

FIG. 1. Afterloading systems. 50 ALTEMARK and QUANDT

Oscillating source

Moving source

Source train

Source chain ГГШ ~ИН нрия ft

FIG. 2. Afterloading sources.

(3) To operate sophisticated equipment hospitals need technicians and/or trained physicists and they are not always freely available [7]. Preference must be given to simple equipment manageable by the available staff. Long treatment preparation time and frequent down-times reduce the number of patients treated. (4) Established treatment modalities require also a few dose distributions [2, 7] which can be produced by simple, safe and reliable equipment. Let me pick just one safety aspect as an example — the return of source(s) into shielding in the event of power failure. Existing afterloading equipment offers back-up compressors, back-up batteries, hand cranks and gravity. -The choice must be gravity (Fig. 3).

Now, what is a compromise for the developing countries? To allow the treatment of a high number of cervix patients and — if so needed — treatment of other carcinomas as well?

(1) HDR remote afterloading equipment with three 137Cs 2 Ci (74 GBq) sources, one oscillating in the tandem or uterine tube, two fixed in the ovoids. (2) Standard dosimetry as for the Egypt project but for HDR sources. An atlas similar to the Amersham Atlas [8], but individual dosimetry if so required. Treatment times of about 30 minutes per fraction. (3) A standard set of gynaecological applicators (ф 6 mm) which allow standard geometry [2]. Applicators for other carcinomas as an option. (4) Electromechanical controls and gravity to retract the source(s). (5) Rectal dose controlled by a dosimeter. (6) Price < US $100 000 (assuming US $1 = DM 2).

The above mentioned equipment is not a new invention but has been ‘tested’ in the People’s Republic of China since 1981 under difficult conditions. The IAEA-SM-290/23 51

Back-up compressor

Back-up battery

Hand-crank

SL 10Kg G r a v i t y

FIG. 3. Emergency — power failure. number of cases treated during the last five years outnumber by far the total cases treated in the Federal Republic of Germany over the last decade. The Afterloading BUCHLER performed well. Occasional failures and repairs were mostly caused by operational or human errors, which confirms the basic idea of the Egyptian project, namely to offer suitable training. This leads to the main concern of users in developing countries — service and maintenance. Regular maintenance and good care o f equipment is still the cheapest way to guarantee smooth operation and a long life, also in the developing countries. Three stages o f maintenance and care have to be considered: (1) by the hospital/end user (2) by the local agent/service station (3) by the manufacturer.

The Afterloading BUCHLER as described above and also suitable for rural areas (district hospitals), e.g. in India, permits the three stages of maintenance as described (Fig. 4). — The hospital engineer/technician can run routine checks on a regular basis (daily, weekly, monthly) and perform minor repairs or simply use the oil can as described in the manual. Electronic boards can be exchanged (but not repaired). 52 ALTEMARK and QUANDT

□ □ □ □ □ HOSPITAL XL ? 4 B A SIC MAINTENANCE □ U u □ D AILY; WEEKLY; MOWniLY Q 1ECK

STOCK □ □ □ □ □ 1 ANNUAL aiECK I STOCK (SMALL) II DIAG'JOSIS TX REPAIR AGEOT

EXPERIENCE STOCK (LARGE) REPAIR QUALITY ASSURANCE i â _ TRAINING ANNUAL SERVICE MANUALS

FIG. 4. Service concept.

— The local agent/service station undertakes the annual prescribed service, performing adjustments where needed. — The factory is available for major repairs or 5-year service where certain components may have to be exchanged. — The factory sends ‘check service engineers’ on a semi-regular basis to reconfirm the qualification of engineers in the service stations and observe the per formance of equipment even outside the warranty period. Training is certainly not the only factor. The down-time of equipment also depends on the availability of spare parts. Therefore, every unit will be supplied with sufficient spare parts for ‘hospital maintenance level’ — ‘service station level’. This may increase the initial cost of equipment but will provide the hospital with more security for a longer period of time. IAEA-SM-290/23 53

There are gaps between theory (or wishful thinking) and reality. Buchler donated an Afterloading BUCHLER to Ocean Road Hospital in Dar Es Salaam, Tanzania, which is closely connected to Prof. Ulrich Henschke’s ideas and which is represented at this symposium by Dr. J. Luande. Service has to be carried out by the hospital itself, every now and then supported by occasional visitors from the Deutsche Krebsforschungsgesellschaft in Heidelberg, Federal Republic of Germany. It remains to be seen how the unit will maintain itself.

Summary

The number of cervix carcinoma cases in the developing countries is high, while the number of trained operators for sophisticated remote afterloading machines is low. However, simple afterloading units permit the treatment of most standard cases, thus not depriving patients in developing countries from privileges patients enjoy in industrialized and rich countries. The Egyptian project can be used to introduce HDR afterloading and standard dosimetry in countries such as India, Pakistan, etc. where a vast number of patients are queuing up for treatment. The total cost related to the number of patients treated by an HDR installation is lower than the initial investment for an LDR manual afterloading department. A good service concept on three levels guarantees the equipment a longer life.

REFERENCES

[1] GOLDSON, A.L., et al., “Remote afterloading (RAL) cobalt-60 technique for carcinoma of the cervix in Haitian women”, Diagnosis and Treatment of Carcinoma of the Cervix in Developing Areas (Proc. Int. Working Party Bangkok, 1985) (ISBN 0-85274-561-3). [2] GUPTA, S., “The role of high dose rate brachytherapy in developing countries”, Diagnosis and Treatment of Carcinoma of the Cervix in Developing Areas (Proc. Int. Working Party Bangkok, 1985) (ISBN 0-85274-561-3). [3] MOULD, R.F., et al., “Aspects of physics and statistics in the training course of the IAEA/WHO project on intracavitary radiation therapy for cancer of the uterus in Egypt”, Diagnosis and Treatment of Carcinoma of the Cervix in Developing Areas (Proc. Int. Working Party Bangkok, 1985) (ISBN 0-85274-561-3). [4] SNELLING, M.D., “Aims of the WHO/IAEA Egyptian project for earlier diagnosis and afterloading brachytherapy of carcinoma cervix in the rural hospitals of developing countries”, Diagnosis and Treatment of Carcinoma of the Cervix in Developing Areas (Proc. Int. Working Party Bangkok, 1985) (ISBN 0-85274-561-3). [5] VAHRSON, H., et al., Int. Symp. High Dose Rate Afterloading in the Treatment of Cancer of the Uterus, Giessen, July 10-12, 1986, Statement of the Panel (oral communication, to be published). [6] TAYLOR, C.B.G., PACT (III) for IAEA/WHO Project EGY/6/004 (Egyptian Cancer Project) (1985). ALTEMARK and QUANDT

MOULD, R.F., “Dosimetry for the Amersham caesium-137 manual afterloading system with special reference to the WHO/IAEA brachytherapy project for Egypt, 1983—6”, Diagnosis and Treatment of Carcinoma of the Cervix in Developing Areas (Proc. Int. Working Party Bangkok, 1985) (ISBN 0-85274-561-3). MOULD, R.F., et al., “Radiation Dosimetry for the Amersham Caesium-137 Manual Afterloading System for Gynaecological Brachytherapy” (ISBN 0-901 259 08 X). DISCUSSION

(Summary of discussion held on Papers IAEA-SM-290/81, 13, 29, 59 and 23)

Concerning Paper IAEA-SM-290/81, a comment from the floor stated that the main problem of developing countries was advanced cases of carcinoma of the cervix for which the only treatment was external radiotherapy. Therefore, the suggestion o f the speaker to give intracavitary application is not feasible and should not be recommended when it is not possible in clinical practice. In operable cases surgery could be practised in District Hospitals and be freely available. M. Snelling noted that she had highlighted a new idea which anticipated the transfer o f suitable treatment technology to small hospitals. However, to provide surgery to the poor women in middle sized or small villages was, according to M. Snelling, a great problem without evident universal solution. In reply to a question on the possibility of increasing radiotherapy centres in developing countries to help in the management of carcinoma of the cervix of stages I—IV, M. Snelling expressed her doubt regarding the expediency o f treating stage I carcinoma of the cervix in large hospitals because during the wait for treat ment the carcinoma could become stage II or even III. If patients with carcinoma o f the cervix were cured in small hospitals near their homes, the capacity of the big hospitals would be used in a more rational way and the number of treated patients would be increased. In response to a participant, T. Kuipers (Paper IAEA-SM-290/13) mentioned that with epidural anaesthesia total immobilization of the lower half of the body was obtained, which together with fixation of the applicator at the treatment couch provided optimal conditions for reliable dosimetry. In the case of technical impossibility or counter-indications to epidural anaesthesia, a mild type of general sedation could be used. However, as during the treatment with high activity sources no other person than the patient can stay in the treatment room, many of the anaesthesiologists did not prefer general sedation because the condition of the patient could not be controlled sufficiently. Regarding Paper IAEA-SM-290/29, a speaker expressed her doubt that an effective drug for the treatment of cervical cancer was available at present. Another speaker noted that in his study with bleomycin, cervical canal fibrosis had occurred in 40% of cases at the time of intracavitary application. S. Puribhat replied that in his study morbidity had been acceptable and, with regard to the efficiency of the treatment protocol, long term survival is needed. He stated that all patients could undergo brachytherapy according to the protocol and there was no fibrosis at all. In response to various questions, S. Puribhat stated that tolerance of patients to treatment had been good and only in a few cases had radiotherapy been inter rupted because of cytopenia or other complications, to be resumed as soon as the

55 56 DISCUSSION

condition of the patients had improved. He said that packed red cells had been transfused when haemoglobin dropped to 10 g% owing to blood loss, so that patients could better tolerate the treatment, without intending to improve the tumour response. In response to a question, J. Stumpf (Paper IAEA-SM-290/59) pointed out that locoregional intra-arterial chemotherapy provided more high intercellular fluid concentration of the cytostatics in the region compared with intravenous adminis tration of the drugs, that the method permitted at the same time use o f labelled macro-aggregates to study the real distribution of the cytostatics, which is very important for treatment planning, and that surgery was usually carried out three weeks after completion o f intra-arterial chemotherapy and intracervical irradiation. In response to another question, J. Stumpf confirmed that patients with bilateral parametrial infiltration could be treated with the intra-arterial chemo therapy technique, as in those cases the amount of cytostatics would be shared according to the measure of the involvement of the small pelvis, e.g. 70% to the side infiltrated totally, and 30% to the side where only a small part of the parametrium was involved. Discussion on Paper IAEA-SM-290/23 focused on the advantages and dis advantages of low dose rate (LDR) manual and high dose rate (HDR) remote afterloading treatment techniques, taking into account the specific circumstances of developing countries. M. Snelling was of the opinion that staff in all centres should first learn the conventional dose rate manual afterloading system. J. Luande believed that in any case an LDR set or unit should be available in a big centre too for initial and refresher training of people from the periphery. N.T. Racoveanu believed that an LDR manual afterloading technique was preferable when the patient load was small. A speaker stated that because of the cost 137Cs and not 60Co sources should be used for intracavitary applications. Arguing that statement, V. Shanta informed the audience that in Madras, India, the cost of the set including the cost of 60Co sources for manual afterloading treatment was only about US $400, which permitted their wide use in district hospitals. Several speakers emphasized the advantages o f HDR remote afterloading treatment compared with LDR manual afterloading treatment. According to a speaker, quality assurance of HDR treatment could be maintained much more easily than LDR treatment because it was easier to check the position of the applicator during several minutes in the case of applying the HDR technique than during 20 hours or more in the case of applying the LDR technique. The worse quality assurance was, the higher was the rate o f complications. To treat the complications one should spend some money, which increased the cost of the LDR treatment technique. Therefore, according to the opinion of a speaker, the use of HDR units would be justified from the-economical point ofview. U.M.F. Altemark drew attention to the fact that HDR units provided full radiation protection to medical staff, which was very important as people became worried about the problem. DISCUSSION 57

Contrary to the above opinions, a speaker thought that insufficient radiobiology data required great precautions in the implementation of the HDR technique. M. Snelling believed that the HDR treatment technique should be used when there was a heavy patient load, good infrastructure and expertise. In that case cure rate and complications might be equal to LDR and 226Ra treat ment results. Summarizing the discussion, H.M. Mahfouz noted that speakers tried to compare two different methods which required two different philosophies, which did not seem to be a right approach.

DIFFERENT APPROACHES IN RADIATION THERAPY

IAEA-SM-290/84

Invited Paper RADIOPROTECTION OF THE IMMUNE SYSTEM DURING RADIATION THERAPY

H. ALTMANN, H. TUSCHL Institut für Biologie, Österreichisches Forschungszentrum Seibersdorf, Seibersdorf, Austria

E. KUN Department of Pharmacology and the Cardiovascular Research Institute, University of San Francisco, San Francisco, United States of America

Abstract

RADIOPROTECTION OF THE IMMUNE SYSTEM DURING RADIATION THERAPY. In cancer radiotherapy modulators of poly(ADP-ribose)-synthesis can have positive effects at different levels. These substances can have an antitransforming capacity. They could act as radiosensitizers and enhance differentiation processes generating a balanced lymphocyte subset system. С 5 7 Ы mice were used as an animal model since they have a high incidence of malignant lymphoma. Pretreatment of the mice with 3-methoxybenzamide resulted 3 days after 1 Gy 7 -irradiation in a rebalanced lymphocyte subset system. After an initial decrease, PAR-synthesis was clearly augmented 18 h after-drug and radiation treatment.

1. INTRODUCTION

Cancer affects millions of people, and radiotherapy fails in many cases to establish either tumour control or control of the development of métastasés because of the occurrence of immunological imbalances after radiation treatment. Enthusiasm over early results must, therefore, be tempered by considering the impact of radiotherapy and its potential long term consequences. The combination of chemotherapy and irradiation has often resulted in an increase in secondary neoplasms. The regulation of an induced immune response occurs by the interaction of different lymphocyte subsets. In the absence of one of these necessary cell subpopulations the circuit is interrupted and a deficient immune response will take place [1 ]. The opinions on immunological consequences during radiation therapy are still controversial because a multiplicity of lymphocyte subsets and factors are required for immunoregulation. The radiosensitivity of the helper T cell subpopulation has been determined, but that of suppressor T cells remains an open question [2]. In cancer therapy tumour radiosensitization by

61 62 ALTMANN et al.

chemical compounds is still a developing field, but little is known about the effect of these chemicals on the immune response and T lymphocyte subsets. The low toxicity and good tumour radiosensitization by inhibitors of poly(ADP-ribose) (PAR)-polymerase suggests that they could play a role in future radiotherapy [3]. The mechanism by which these inhibitors produce radiosensitization remains unclear, but may be related to an initial transient inhibition of PAR-polymerase with the induction of replicative DNA synthesis. A second mechanism may be due to an increase in the diffusion distance of oxygen as a result of the inhibition of cellular oxygen utilization. Benzamide derivatives, especially 3-nitrobenzamide, can act as a post-irradiation (electron) sensitizer in anoxic cells and has a synergistic effect in combination with its PAR-polymerase inhibitory properties [4]. PAR-polymerase represents a chromatin-bound enzyme and an enzyme associated with free cytoplasmic mRNA-protein particles [5]. The chromatin-bound PAR-polymerase is involved in structural and regulatory phenomena connected to DNA metabolism and the enzyme activity depends on DNA as a cofactor and can be activated by DNA strand breaks. In some laboratories the enzyme has been purified in association with DNA fragments, which eliminates the need for exogenous DNA to achieve activation. These frag ments, designated sDNA, are smaller than bulk DNA and are some 150-300 basepairs (bp) long [6]. In another study, a hexadeoxynucleotide and an octadeoxynucleotide in double stranded form activated the enzyme 30% more effectively than highly polymerized calf thymus DNA [7]. This is in accordance with electron micrographs, which show that several molecules of PAR-polymerase can be associated with a single sDNA molecule, suggesting that only a portion of each DNA is required for enzyme binding and activation [8]. We have identified short dispersed repetitive DNA sequences isolated from the zones of initiation of DNA synthesis as ALU elements [9]. A DNA of approximately 300 bp, 5-10 fold enriched in ALU sequences, activated the enzyme in crude chromatin more effectively than calf thymus DNA [10]. PAR-polymerase seems to have two active sites. The DNA binding site can be inhibited by benzamide derivatives and the NAD+ cleaving site by nicontinamide. These inhibitors of PAR-polymerase can induce differentiation processes [11]. Differentiation inducers can also suppress oncogene expression [12]. On the other hand, oncogenes can also suppress differentiation. Oncogenes are highly conserved genes whose normal function contributes to the regulation of the cell cycle. Antagonists of oncogene expression products are tumour inhibitory factors. Exposure of cells in the early S-phase to a DNA damaging agent and benzamide prevented transformation and also augmented poly(ADP-ribosyl)ation above the controls [13]. Mechanisms regulating malignant transformation may be associated with changes in gene regulation and expression caused by alterations in poly(ADP-ribosyl)ation [14]. The studies of Farzaneh et al. [15] using 3-aminobenzamide have shown an inhibition of in vitro muscle differentiation as monitored by cell fusion. In a similar system we could show that at low DNA IAEA-SM-290/84 63

synthesis PAR-synthesis achieves a maximum and this step precedes myoblast fusion during differentiation [16]. Cells are under multiple regulatory controls and poly(ADP-ribosyl)ation is certainly one important part in this system. It is therefore not surprising that ADP-ribosylation can interfere with both the transformation and the differentiation steps. In the following study we investigated the influence of methoxybenzamide (MBA) on ADP-ribosylation of proteins and the reconstitution of imbalances in lymphocyte subsets of mouse spleens after in vivo irradiation of C57bl mice. We have chosen spleen cells, because spleen serves two functions in protecting the host against bloodstream bacterial infections and tumour cells by its phagocytic filter and by its lytic activity of cytotoxic T lymphocytes. It is also a major antibody producing organ. Natural killer activities directed against tumour cells are a poorly understood component of the immune system [17]. The fate of irradiated tumour cells depends on the extent of injuries to radiation-sensitive targets, mainly DNA and the repair of DNA lesions, but also on the potential tumour killer activity of a balanced immune system. Several types of human tumour cells have been shown to have a short doubling time of only a few days (lymphomas ca. 3 d). Imbalances in lymphocyte subsets, which can occur as a result of radiotherapy, should be rebalanced within such a short time. The Cs7bl/6 mice used in our experiments have a high incidence of malignant lymphomas and the immune functions are well characterized in relation to the age of the mice [18].

2. MATERIAL AND METHODS

C57bl/6 mice, 8-10 weeks old (supplied by the Forschungsinstitut für Versuchstierzucht Himberg, Austria), were used for the experiments. Feed (altromin diet 131 ff) and bedding material were 7 -irradiated with 10 kGy for sterilization. The animals were divided into 4 test groups:

(1) Controls (administration of the carrier DMSO), (2) Administration of 3-MBA (100 mmol in a volume of 0.3 mL = approximately 1 mmol/23 g mouse) by stomach tube for 5 consecutive days, (3) Administration of the drug carrier and whole-body irradiation with 1 Gy 60Co with a dose rate of 0.2 Gy/min, (4) Administration of the test substance, followed by whole-body irradiation, given 1 h after the last drug application.

Animals were sacrificed 18 or 72 h after termination of the treatment, the spleens removed and cell suspensions prepared at 4°C [19]. The cell suspensions were washed twice with medium 199. Living cells were counted by Trypanblue staining. Lymphocyte subpopulations in the spleen cell suspensions were identified by immunofluorescence with the monoclonal antibodies. 106 cells/mL suspended in a cytotoxic medium (Cederlane, Becton-Dickinson) were incubated at 64 ALTMANN et al.

37°C for 60 min to reduce the background fluorescence from unspecific binding.

After centrifugation, the cell sediment was incubated with 10 ¡jlL anti-Thy 1.2 FITC at 4°C for 30 min. After twice washing in PBS, the cells were resuspended in PBS and transferred to a solution of polyvinylalcohol in glycerol. 10 juL of the cell suspension was placed on slides and examined for the percentage of fluorescent cells. Suppressor/cytotoxic cells were differentiated by anti-Lyt 2 antibody (Becton-Dickinson) with an indirect test. Biotinylated anti-Lyt 2 was incubated with the cells for 30 min. After two washings of cells Avidin FITC (1 Mg/105 cells) was added and the cells incubated for 30 min. Further preparation was the same as for direct tests applied for total T cells.

2.1. PAR-synthesis

Immediately after preparing the cell suspensions raw chromatin was isolated by lysis in a solution of 0.1 M EDTA, 0.002M Tris and 0.5% Triton X-100 for 5 min at 20°C. After washing the raw chromatin once with distilled water, 1 mCí 3H-NAD (NEN, NET-443, 1.3 X 10n Bq/nmol specific activity) was added in a mixture of 0.15M sucrose, 0.00IM EDTA, 0.003M DTE, 0.04M Tris, 0.04M MgCl2 and then incubated at 37°C for 30 min. The reaction was stopped by the addition of 40% perchloric acid (final concentration 20%) in the cold. The pellet was hydrolysed at 90°C in 6% PCA for 30 min. The DNA content (optical density at 260 nm and 290 nm) and the radioactivity (liquid scintillation counting) of each sample was measured in aliquots of the supernatant, obtained after centrifugation. The radioactivity (counts/min) of the samples was calculated as specific radioactivity (counts-min-1-jig-1 DNA).

3. RESULTS

To compare the concentration of 3-MBA in organs at 18 or 72 h after the last treatment, pharmacokinetic studies using 3H-labelled MBA were made. Eighteen hours after treatment the drug concentration in spleen was 0.093 ppm and after 3 days 0.040 ppm, compared with 63 ppm 1 h after application. During this time range the commitment phase of differentiation may occur. Since DNA synthesis is necessary for the recovery of some immunological functions, we did some studies on the incorporation of 3H-thymidine into the DNA of cells. Eighteen hours after treatment with drugs and/or irradiation the incorporation of 3H-thymidine was low, but 3 days after treatment the number of S-phases was more than doubled in both irradiated groups (data not shown). Figure 1 shows PAR-synthesis at 18 and 72 h post-treatment. At the 18 h point the two tailed t-test gives significant values for3>l,4>l,4>2,4>3. For the 72 h experiments only 1 > 2 and 4 > 2 was found significant. IAEA-SM-290/84 65

1 c o n t r o l 18 h after irradiation 2 3 M B A □

3 7 irradiation 72 h after irradiation A 3 M B A + 7 irradiation

n=36 n=20 n=30 n«17

FIG. 1. ADP ribosylation o f proteins in spleen cells at 18 and 72 h after 7 -irradiation (results in % o f controls).

A clear drug effect can be seen only in the groups where irradiation was also applied to the animals 18 h before the determination was made. After 3 days no more effects were visible in these combined groups. A small but significant decrease can be seen in the group where MBA was applied alone. To prove whether the high PAR values obtained in group 4 18 h after the treatment correlates with DNA strand breaks, we performed nucleoid sedimentation studies in neutral sucrose (data not shown). With 3-MBA and irradiation we obtained a faster sedimentation in group 4 than in group 3 (irradiation alone). This means that the increase in new PAR-synthesis is not dependent on DNA strand breaks. The results on lymphocyte subpopulations were introduced to a variance analysis. Both a parametric model and non-parametric analysis according to Kruskal and Wallis were performed. For Pan T cells (Fig. 2) groups 3 < 1 and 4 > 3 showed significant differences. ALTMANN et al.

1 control

2 3 M B A

3 7 irradiation

4 3 M B A + y irradiation

FIG. 2. Pan T-lymphocyies of Ы 6 mouse spleen 12 h after y-irradiation.

1 COntrOi

2 3 M 8 A

3 7 irradiation

4 3 M B A + 7 irradiation 'O

FIG. 3. Suppressor/cytotoxic T-lymphocytes 72 h after y-irradiation. IAEA-SM-290/84 67

1 control

2 3 M B A

3 7 irradiation

4 3 M B A + 7 irradiation

FIG. 4. B-lymphocytes 72 h after -¡-irradiation.

For suppressor/cytotoxic lymphocytes in the 4 experimental groups only 3 < 1 was significantly different (Fig. 3) and the В cells (Fig. 4) did not show any significant changes.

4. DISCUSSION

The high PAR-polymerase activity present in the lymphocytes 18 h after 7 -irradiation (in both the 7 -irradiated and the combined treatment groups) along with the low replicative DNA synthesis fits well the general picture of the commitment phase of differentiation. The low PAR-polymerase activity and high DNA synthesis belong to the final step of differentiation, also called the expressional and replicative phase. Gene products expressed on the surface of the murine T lymphocytes, such as Thy and Lyt antigens, which we determined during our experiments, are designated as differentiation antigens because they are considered to be expressed in the differentiation pathway from lymphoid progenitors to 68 ALTMANN et al.

T cells. Much of the regulation of lymphocyte differentiation is unknown. The reason for this is due to the inability of developing good in vitro assays and long term cultures of lymphocytes. We should not exclude from our discussion that 3-MBA could induce the migration of fully differentiated T lymphocytes to the spleens of the irradiated animals. Lymphocytes circulate through the body and there is also a constant movement of lymphocytes directly from the blood to the spleen. Another factor is the enhanced radioresistance of activated suppressor T cells compared with memory suppressor T cells and precursor T cells [20]. 3-MBA could also affect the expression of T cell receptors. It was demonstrated that radiation not only causes the death of these cells, but leads to a shedding of the receptors from the cell surface, causing an increase in the number of pseudo null cells [21 ]. An increase in null cells was found with a combination of radiation with PAR-synthesis inhibitors [22]. Imbalances in the normal distribution of helper and suppressor T cell subsets, with an excess or deficiency of T cell help or suppression, lead also to abnormalities in В cell immunoglobulin production. Although we did not find changes in the В cell subset after different treatments, the humoral immune defence mechanism can be disturbed, too. Suppressor T cell activity is also decreased in systemic lupus erythematosus (SLE). Autoantibodies to suppressor T cells may cause the decreased suppressor T cell activity in SLE and this in turn may result in increased autoantibody production [23]. In SLE lymphocytes imbalances and low values of PAR-synthesis could be detected [24]. The regulation of the immune response in tumour patients, especially after radiotherapy, is a delicate homeostasis between defined subsets of T lymphocytes, and the modulation of T cell subsets can be of therapeutic value. The mechanism of PAR-synthesis modulating drugs as antitransforming agents is not very well understood and seems to be strongly dependent on the treatment schedule. The same is true for the radiosensitizing effect of these drugs. Tumour development and also tumour destruction after radiotherapy are multistep processes and the use of a drug group with multistep action mechanisms is possibly one answer in the treatment of cancer patients.

ACKNOWLEDGEMENT

This work was supported by AFOSR Grant No. 84-0390.

REFERENCES

[1] GREEN, D.R., FLOOD, P.M., GERSHON, R.V., Ann. Rev. Immunol. 1 (1983) 439. [2] DORIA, G., AGAROSSI, G., ADORINI, L., Immunol. Rev. 65 (1982) 23. IAEA-SM-290/84 69

HORSMAN, M.R., BROWN, D.M., LENNON, M.J., BROWN, J.M., LEE, W.M., Chemical Modifiers of Cancer Treatment (Proc. Symp. Clearwater, 1985), Abstr. 4/8. GEORGE, A.M., LUNEC, J., CRAMP, W.A., BRENNAN, S., LEWIS, P.D., WHISH, W.J.D., ibid., Abstr. 6/8. THOMASSIN, H., NIEDERGANG, C., MANDEL, P., Biochem. Biophys. Res. Commun. 133 (1985) 654. NIEDERGANG, G., OKAZAKI, H., MANDEL, P., Eur. J. Biochem. 102 (1979) 43. BERGER, N.A., PETZOLD, S.J., Biochemistry 24 (1985) 4352. DeMURCIA, G., JOUNGSTRA-BILLEN, J., ITTEL, M.E., MANDEL, P., DELAIN, E., EMBO J. 2(1983) 543. ANACHKOVA, B., RUSSEV, G., ALTMANN, H., Biochem. Biophys. Res. Commun. 128 (1985) 101. ALTMANN, H., TOPALOGLOU, A., BRKIC, G., IGEGM Symp. Vienna, December 1985 (in press). KULLICH, W., KLEIN, G., BRUGGER, P., ALTMANN, H., Z. Rheumatol. 44 (1985) 108. MÜLLER, R., CURRAN, T., MÜLLER, D., GUILTBERT, L., Nature 314 (1985) 546. KUN, E., KIRSTEN, E., MILO, G.E., KURIAN, P., MUKARI, H.L., Proc. Natl. Acad. Sei. 80(1983) 7219. BOREK, C., MORGAN, W.F., ONG, A., CLEAVER, J.E., Proc. Natl. Acad. Sei. 81 (1984) 243. FARZANEH, F., SHALL, S., ZALIN, R., in Novel ADP-ribosylations of Regulatory Enzymes and Proteins (SMULSON, M.E., SUGIMURA, T., Eds), Elsevier North Holland - New York (1980) 217. TÖRÖK, O., ALTMANN, H., Growth, Cancer and the Cell Cycle (SKEHAN, P., FRIEDMAN, S.J., Eds), Humana Press Inc., Clifton, NJ (1984) 27. BROOKS, C.G., Nature 305 (1983) 155. HIROKAWA, K., UTSUYAMA, M., GETO, H., KURAMOTO, K., Gerontology 30 (1984) 223. ALTMANN, H., DOLEJS, I., DNA-Repair, Chromosome Alterations and Chromatin Structure (NATARAJAN, A.T., OBE, G., ALTMANN, H., Eds), Elsevier Biomedical P. (1982) 167. GILL, H.K., DHALIWAL, Y.S., SUKUMARAN, K.D., LIEW, F.Y., Immunology 53 (1984) 669. SZCZYLIK, C., WIKTOR-JEDRZEYCZAK, W., Int. J. Radiat. Biol. 39 (1981) 253. TUSCHL, H., KOVAC, R., IGEGM-Symp. Vienna, December 1985 (in press). WALDMANN, T.A., BRODER, S., Prog. Clin. Immunol. 3 (1977) 155. ALTMANN, H., SCHERAK, O., TOPALOGLOU, A., Verh. Dtsch. Ges. Rheumatol. 7 (1981) 561.

IAEA-SM-290/5

CARCINOMA OF THE OESOPHAGUS - PALLIATION BY INTRACAVITARY IRRADIATION

C.G. ROWLAND, K.M. PAGLIERO Royal Devon and Exeter Hospital, Exeter, Devon, United Kingdom

Abstract

CARCINOMA OF THE OESOPHAGUS - PALLIATION BY INTRACAVITARY IRRADIATION. Previous attempts at oesophageal brachytherapy over the last 60 years have failed because of the lack of safety both to patients and the staff. At Royal Devon and Exeter Hospital, modern afterloading techniques to treat oesophageal carcinoma have been applied. An applicator that accepts 48 caesium-137 sources delivered by the Selectron (Nucletron of Holland) low dose rate afterloading machine was designed. The appliactor is inserted under fluoroscopic control and fixed in position for the duration of treatment during which 1500 cGy at 10 mm off central axis are delivered. This pilot study was conducted with patients whose disease was too advanced to be resectable or in patients considered unfit for major surgery. Seventy-two patients with a median age of 76 years were submitted to the study. Sixty-nine patients were able to tolerate the treatment. More than four-fifths of the patients showed improvement in swallowing, squamous cell carcinoma and adenocarcinoma responding equally well. There was a zero hospital mortality and 70% of the patients were discharged home within 3 days. Side effects were minimal and in our hands this experience has proved superior to intubation, bougienage or palliative resection. Other treatment options were not precluded and indeed one-fifth of the patients subsequently went on to intubation. Some of the fitter responders were subsequently treated with external beam radiotherapy. Side effects were minimal and infrequent. The survival rate appears to have improved. Whereas most intubated patients do not survive six months, one-third of the patients treated with brachytherapy were alive and well after the same interval. As well as being advantageous to the patient, this treatment, by virtue of short term hospitalization and infrequent réadmissions, results in considerable worthwhile reductions in financial demands on health services.

1. INTRODUCTION

The symptoms of cancer of the oesophagus are amongst the most unpleasant that our patients have to endure. There is nothing worse than the sight o f an old person, choking on his food, spitting out his saliva, suffering from malnutrition and denied his one remaining pleasure in life — eating. The disease itself presents an almost insurmountable challenge. Belsey [ 1 ] describes ‘cure’ in this condition as a fortunate accident. Earlem [2] has estimated that only 4% of the cases can expect to survive 5 years. Intubation has been the mainstay in palliation for many years [3]. Even the limited leparotomy in these malnourished patients, however, proved too much for many and a 25% operative mortality is a common experience. Endoscopic intubation has reduced the mortality and duration of hospitalization. However,

71 72 ROWLAND and PAGLIERO intubation does nothing more than create an imperfect passage through the tumour, doing nothing for the tumour itself, and some 50% of the patients will experience unpleasant side effects such as blockage, dislocation, and bleeding. A minority of surgeons prefer, therefore, to perform palliative resection or bypass often at considerable mortality and for those who survive there is little prospect of a pleasant life between the period of convalescence and the inevitable decline. External beam radiotherapy has little useful effect on adenocarcinoma of the oesophagus. It has been beneficial in cases of squamous cell carcinoma but the systemic effects and the effects on the lung fields have precluded its use in the more debilitated patients. Recent interest in brachytherapy in other fields made us consider its use in the oesophagus. The idea is not new, having been described as long as 60 years ago with radium bougienage. The technique was uncontrolled and proved highly dangerous to the patients and staff around and was therefore abandoned. However, with modern afterloading techniques we have been able to introduce brachytherapy safely into the management of oesophageal carcinoma.

2. TECHNIQUE

We have used the remote afterloading system developed by Nucletron of Holland. This is a six channel low dose rate machine using 48 caesium-137 sources each of 40 mCi activity.1 A microprocessor controls the pneumatic transfer of the sources down a flexible tubing into previously inserted ‘applicators’, withdrawing them automatically into a special safe whenever staff have to enter the treatment room. We have designed a special applicator (Fig. 1) of 8 mm external diameter that is introduced under fluoroscopic control over a previously inserted guide wire. The sources are transferred along an insert tube that is locked within this external tube. During placement of the applicator an insert of non radioactive, radio-opaque marker pellets are used for precise localization. During treatment the insert tube transfers sources to the terminal 13 cm of the applicator so that, if it is appropriate to irradiate the entire oesophagus, this can be achieved in two applications at the same session.

3. INSERTION

Using either simple sedation and local anaesthesia or general anaesthesia according to preference, the lesion is viewed through a fibreoptic endoscope.

1 1 Ci = 37 GBq. IAEA-SM-290/5 73

FIG. 1. Selectron applicator containing dummy marker pellets and guide wire.

Using fluoroscopy, the upper and lower extent of the tumour is defined endoscopically and on the screen demonstrating where to position the applicator precisely. A guide wire is then passed into the stomach and the endoscope removed. The applicator can then be threaded over the guide wire into the appropriate position straddling the tumour (Fig. 2). Where a double application is proposed, the lowermost position is treated first. Having positioned the applicator, it is easily maintained by fixation to a simple face mask strapped to the patient’s head (Fig. 3). The patient can then be transferred back to the treatment suite without fear of dislocation of the applicator. In the treatment room the patient is continuously observed on closed circuit television. If staff need to enter, the sources are automatically removed and the Selectron automatically reprogrammes itself with regard to treatment times. 74 ROWLAND and PAGLIERO

FIG. 2. Applicator containing dummy marker pellets introduced into an oesophageal tumour over a guide wire.

4. DOSE

In our pilot study we chose to use all 48 caesium sources in a fairly high single dose fraction to try and gain tumour response, rapid palliation and sound economy, both financially to the health service and in time to the patient, with respect to length of hospitalization. A dose of 1500 cGy at 10 mm off central axis was chosen (giving approximately 3500 cGy at the surface). This was given in 1.14 h as a 12 cm line source. The effective target to line is thus represented by a cylinder 2 cm in diameter and 13 cm in length. We justified the risk of problems on the basis that these dying patients required rapid effective palliation but in the event were pleased to observe very few undesirable tissue side effects with this dose. IAEA-SM-290/5 75

5. PATIENT SELECTION

We believe that the only prospect of ‘cure’, however small the incidence, is resection. All patients who appeared to have localized disease suitable for resection and were considered fit for major surgery were offered this option. All other patients — those with unresectable disease, those who declined surgery and those too old and frail for major operation — were offered intracavitary irradiation by the Selectron. The only exclusions were patients with tracheobronchial mucosal involvement, following one case so treated who developed a fistula, and patients clearly so ill that they were not expected to live more than two or three days.

6. RESULTS

Seventy-two patients were treated with a median age of 76 years. The sex distribution was equal. Sixty-five per cent were considered unsuitable for major surgery and 35% were assessed as having unresectable disease. Three patients could not tolerate the treatment and 12 found it unpleasant. The hospital stay was significantly short, 70% of the patients having left hospital within 3 days. We consider this an important economic advantage over 76 ROWLAND and PAGLIERO

Weeks

FIG. 4. Squamous cell carcinoma o f the oesophagus. Relief o f dysphagia following intracavitary irradiation. Grades: 0 = Total dysphagia. Straight line terminating before 27 weeks = death. 1 = Liquids only. Dotted line = patient alive. 2 = S o ft food. 3 = N orm al diet.

Weeks

FIG. 5. Adenocarcinoma of the oesophagus. Relief o f dysphagia following intracavitary irradiation. Grades: 0 = Total dysphagia. Straight line terminating before 27 weeks = death. 1 = Liquids only. Dotted line = patient alive. 2 = Soft food. 3 = Normal diet. IAEA-SM-290/5 77 other treatment options as well as being socially beneficial in patients whose life expectancy is so short. Hospitalization is considerably less than that experienced with pulsion or traction intubation [4]. Patients were assessed for dysphagia as follows:

0 Unable to swallow saliva or liquids 1 Able to swallow liquids only 2 Able to swallow puréed or minced food 3 Eating normally.

The benefits to dysphagia are illustrated in Figs 4 and 5. In one case recurrent dysphagia resulted from a benign stricture most likely owing to irradiation. It responded to simple dilatation. All other cases of worsening or recurrent dysphagia were due to the tumours. The single most important advantage of the Selectron was in the manage ment of patients whose mental state was impaired to a point that prevented compliance with advice about diet. We have found with indwelling tubes that intelligent patients with good teeth can virtually eat normally. Many less intelligent patients, however, develop frequent bolus obstruction, often requiring readmission. Following Selectron treatment this has not occurred, which allows us to discharge our patients earlier and results in a reduced incidence of readmission. We were surprised that the responses of squamous cell carcinoma and adenocarcinoma were virtually identical, unlike our experience with external beam radiotherapy where, as a rule, adenocarcinoma shows a less favourable response. Selectron intracavitary brachytherapy for the group described is now our first line therapy. It does not preclude other forms of palliative therapy if there is no response or dysphagia returns. Indeed one-fifth of our patients needed subsequent intubation. In some cases the treatment has been repeated with

Weeks

FIG. 6. Survival following intracavitary irradiation. 78 ROWLAND and PAGLIERO useful effect. In a few fitter individuals treatment has been combined with external beam radiotherapy. At present we believe that this form of therapy is superior to intubation because of the side effects of the latter [4]. We believe it superior to laser vaporization [5] or simple bougienage [6], where nothing is done to halt the progress of the tumour. We have already noted a useful increase in prognosis. Survival following intubation is rare after 6 months, whereas following brachy therapy one-third of the patients survive (Fig. 6). Since this is coupled with zero mortality we feel it would be unjustified to conduct a controlled trial. Side effects were few; of the 69 patients undergoing full treatment 2 suffered sore throats, 2 oesophagitis, 5 epigastric pain and 1 nausea and diarrhoea. One patient developed a radiation stricture. One patient, as previously mentioned, developed tracheo-oesophageal fistula resulting in death and such a patient with tracheobronchial involvement would not now be treated. In addition to improving a patient’s well-being, the significant reduction in hospitalization, out-patient transport costs and use of community services has resulted in considerable financial savings to our health service.

REFERENCES

[1] BELSEY, R., “Is it possible to talk of cure of carcinoma of the oesophagus”, Cancer of the Oesophagus in 1984, Malsine S.A., Ed. (1984) 382. [2] EARLEM, R., CUNHA-MELO, J.R., Oesophageal squamous cell carcinoma: A critical review of surgery, Br. J. Surg. 67 (1980) 381. [3] CELESTIN, L.R., Permanent intubation in inoperable cancer of the oesophagus and cardia, Ann. R. Coll. Surg. Engl. 25 (1959) 165. [4] UNRUH, H.W., PAGLIERO, K.M., Pulsion intubation versus traction intubation for obstructing carcinomas of the oesophagus, Ann. Thorac. Surg. 40 4 (1985) 337. [5] FLEISCHER, D.F., et al., Endoscopic Nd: YAG laser therapy for carcinoma of the oesophagus. A new palliative approach, Am. J. Surg. 143 (1982) 180. [6] HEIT, H.A., et al., Palliative dilatation for dysphagia in esophageal carcinoma, Ann. Intern. Med. 89(1978) 629. IAEA-SM-290/47

RESULTATS D’IRRADIATIONS HEMICORPORELLES CHEZ 92 PATIENTES PORTEUSES DE CANCER DU SEIN POLYMETASTATIQUE TRAITEES A L’INSTITUT CURIE

D. JULLIEN Service d’hygiène radiologique, Institut de protection et de sûreté nucléaire, Fontenay-aux-Roses

J.-R. VILCOQ, F. CAMPANA Institut Curie, Paris

France

Abstract-Résumé

HALF-BODY IRRADIATION RESULTS FOR 92 PATIENTS WITH POLYMETASTATIC BREAST CANCER TREATED AT THE CURIE INSTITUTE. Between 1980 and 1985, 92 patients with polymetastatic breast cancers were treated with half-body irradiation: 34 had isolated bone métastasés, while 58 had métastasés at other ■ secondary sites. A total of 136 treatments were carried out on the 92 patients, of whom 33 received lower half-body irradiation, 15 upper half-body irradiation and 44 both. All 136 treatments used 5.5 MeV photons from a linear accelerator in two different ways: (1) for 111 treatments, 15 Gy were given in 10 fractions of 1.5 Gy with five fractions per week, and (2) for 25 treatments, 8 Gy were given in two fractions of 4 Gy 48 hours apart. Of the 87 patients with bone discomfort before irradiation, 60(69%) had complete and lasting relief from pain in 75% of cases. Nine (82%) of the 11 taking opiates, were able to stop them. Fifty patients with walking problems received lower half-body irradiation: 43 (86%) were able to walk normally again, and ten of those had been bedridden. The conclusion can be drawn that half-body irradiation can achieve a complete response, both analgesic (69%) and functional (86%). Immediate tolerance is excellent and side-effects are slight. The effect on survival is still to be demonstrated.

RESULTATS D’IRRADIATIONS HEMICORPORELLES CHEZ 92 PATIENTES PORTEUSES DE CANCER DU SEIN POLYMETASTATIQUE TRAITEES A L’INSTITUT CURIE. Entre 1980 et 1985, 92 patientes présentant des cancers du sein polymétastatiques ont été traitées par irradiation hémicorporelle: 34 présentaient des métastases osseuses isolées alors que, chez 58 patientes, ces métastases étaient associées à d’autres localisations secondaires. On a réalisé 136 traitements chez ces 92 patientes dont 33 ont reçu une irradiation hémicorporelle inférieure, 15 une irradiation hémicorporelle supérieure et 44 les deux. Les 136 traitements ont été réalisés grâce aux photons de 5,5 MeV issus d’un accélérateur linéaire selon deux protocoles: 1) pour 111 traitements, 15 Gy en 10 fractions de 1,5 Gy avec 5 fractions par semaine et 2) pour 25 traitements, 8 Gy en 2 fractions de 4 Gy à 48 h d’intervalle. Sur les 87 patientes qui présentaient des douleurs osseuses avant l’irradiation, 60 (69%) ont eu une réponse antalgique complète durable dans 75% des cas. Sur les 11 malades qui prenaient des opiacés,

79 80 JULLIEN et al.

9 (82%) ont pu les arrêter. Une irradiation hémicorporelle inférieure a été effectuée sur 50 patientes présentant des troubles de la marche: 43 (86%) ont retrouvé une marche normale, dont 10 malades grabataires. On peut conclure que l’irradiation hémicorporelle permet d’obtenir une réponse complète: antalgique (69%) et fonctionnelle (86%). Sa tolérance immédiate est excellente et ses complications peu sévères. L’effet sur la survie reste à démontrer.

1. MATERIEL

De janvier 1980 à septembre 1985, à l’Institut Curie (Paris), 92 patientes porteuses de cancer du sein polymétastatique ont été traitées par irradiation hémicorporelle. Le délai moyen entre le traitement initial de la tumeur mammaire primitive et l’apparition des métastases était de 42,8 mois. Parmi les patientes, 14 étaient métastatiques d’emblée. L’âge moyen lors de l’irradiation était de 55 ans. Les 92 patientes présentaient des métastases osseuses qui étaient isolées chez 34 et associées à d’autres localisations métastatiques chez 58. L’irradiation hémicorporelle a été indiquée dans trois situations: 1 ) l’échappement thérapeutique après un traitement par chimiothérapie et/ou hormonothérapie (80/92); 2) l’existence d’une contre-indication aux traitements médicaux (8/92); 3) la consolidation après l’obtention d’une rémission clinique et biologique par les thérapeutiques médicales (4/92). Sur le plan fonctionnel, 19 patientes étaient grabataires, 40 marchaient avec des cannes et 33 se déplaçaient normalement. Sur 92 patientes, 87 présentaient des douleurs osseuses avant l’irradiation et chez 11 malades la prise d’opiacés était obligatoire.

2. METHODE

L’irradiation a été réalisée grâce aux photons de 5,5 MeV issus d’un accélérateur linéaire, avec un débit de dose de 60 cGy par minute à 2 mètres de la source et à mi épaisseur (Si e = 30 cm). La patiente est placée au sol en décubitus latéral droit et gauche, alternativement un jour sur deux. Les hémicorps supérieur et inférieur, dont la limite est déterminée par l’ombilic, sont traités successivement à 6-8 semaines d’intervalle. Le premier hémicorps irradié est le plus douleureux. Au total, 136 traitements ont été réalisés chez les 92 patientes: — irradiation hémicorporelle inférieure (HCI): 33/92; — irradiation hémicorporelle supérieure (HCS): 15/92; — HCI + HCS: 44/92, soit, au total, 77 HCI et 59 HCS. Deux protocoles ont été utilisés: — 111/136: 15 Gy en dix fractions de 1,50 Gy avec cinq fractions par semaine; — 25/136: 8 Gy en deux fractions de 4 Gy à 48 heures d’intervalle. IAEA-SM-290/47 81

3. RESULTATS

3.1. Réponse antalgique

3.1.1. Réponse antalgique immédiate

Son évaluation a été réalisée à la fin de la dernière irradiation hémocorporelle. Sur 92 patientes, 87 présentaient des douleurs osseuses avant l’irradiation. La réponse antalgique a été: — complète dans 69% des cas (60 sur 87); — incomplète dans 18,5% des cas (16 sur 87). Au total, on a donc eu une réponse dans 87,5% des cas. Les opiacés ont pu être arrêtés dans 82% des cas (9 sur 11 ).

3.1.2. Réponse antalgique à 6 mois

Parmi les 60 cas de réponse immédiate complète, 43 patientes étaient évaluables six mois après l’irradiation: — dans 75% des cas (32/43), il n’existât aucune douleur, — dans 25% des cas (11 /43), des douleurs étaient réapparues.

3.2. Réponse fonctionnelle

Une HCI a été réalisée chez 50 patientes présentant des troubles de la marche. Parmi elles, 12 étaient grabataires. La réponse fonctionnelle observée a été: — complète avec un retour à une marche normale pour 86% des cas (43 sur 50) dont 10 sur 12 (83%) des malades grabataires; — incomplète chez 3 malades sur 50; — absente chez 4 patientes sur 50.

3.3. Tolérance

On a réalisé 51 traitements en hospitalisation et 41 en ambulatoire. Devant l’absence de signe d’irradiation aiguë, aucune prémédication, ni aucune réanimation n’a été nécessaire.

3.3.1. Tolérance des 59 irradiations hémicorporelles supérieures (HCS):

On a observé les résultats suivants: — 17% des malades ont présenté des signes digestifs (nausées, vomissements); — une hyposialie transitoire est apparue dans 61% des cas; — l’alopécie a été constante (100%) et est apparue 2 à 3 jours après la fin de l’irradiation; — aucune modification de la température, de la tension artérielle ou de la fréquence cardiaque n’a été observée. 82 JULLIEN et al.

3.3.2. Tolérance des 77 irradiations hémicorporelles inférieures (HCI):

Les résultats ont été les suivants: — 28,5% des patientes ont présenté des manifestations digestives (diarrhées, nausées, vomissements); — aucune Symptomatologie urinaire n’est apparue.

3.4. Complications

3.4.1. Complications hématologiques

Après l’irradiation, 27 malades ont développé une hypoplasie médullaire définie par moins de 1000 leucocytes par millimètre cube et/ou 50 000 plaquettes par millimètre cube. Les perturbations hématologiques se répartissaient de la façon indiquée au tableau I. La toxicité hématologique a été différente selon les territoires irradiés et les séquences thérapeutiques comme indiqué au tableau II.

TABLEAU I. REPARTITION DES PERTURBATIONS HEMATOLOGIQUES

Leucocytes Plaquettes Leuco. < 1000/mm3 et < 1000/m m 3 <50 000/mm3 Plaq. <50 000/mm3

Nombre de cas (total 27) 2 15 10

Délai moyen d’apparition 10 jours 11,5 jours - après la fin de l’irradiation

TABLEAU II. TOXICITE HEMATOLOGIQUE

Séquence Nombre Hypoplasie Accidents Accidents d’irradiation de cas médullaire hémorragiques infectieux

HCI - HCS 34 17 4/17 0/17 HCS - HCI 10 3 0/3 0/3 HCS seule 15 4 2/4 0/4 HCI seule 33 3 0/3 0/3

Total 92 27 6/27 0/92 IAEA-SM-290/47 83

Au total: — 21 patientes sur 27 ont eu une hypoplasie asymptomatique cliniquement; — aucun syndrome infectieux n’a été observé; — 6 malades ont présenté des syndromes hémorragiques peu sévères (épistaxis, gingivorragies); aucun cas n’a été mortel; — la séquence la plus hémato-toxique (17/34) est l’HCI suivie 6 à 8 semaines plus tard de l’HCS; le passé thérapeutique (chimiothérapie, radiothérapie) et la différence des volumes médullaires irradiés pendant l’HCI (30%) et l’HCS (70%) en sont probablement responsables.

3.4.2. Autres complications

Aucun cas de pneumopathie radique mortelle n’a été observé. L’irradiation selon un mode fractionné permet certainement d’éviter l’apparition de cette complication redoutable habituellement décrite pour les HCS en dose unique. Sur 59 malades, trois ont présenté des images radiologiques asymptomatiques et régressives spontanément et une seule une image radiologique régressive après antibiothérapie. Aucun cas de cataracte nécessitant une intervention chirurgicale n’est apparu. Deux patientes ont présenté une leucémie aiguë 10 mois et 27 mois après la fin de l’irradiation. La responsabilité de l’irradiation est à discuter car il s’agissait de malades polytraitées antérieurement.

3.5. Survies

Il est difficile d’étudier l’influence de l’irradiation hémicorporelle sur la survie car il s’agit d’une étude rétrospective n’appartenant pas à un essai randomisé. Les médianes de survie sont indiquées ci-dessous. a) Survie de la population globale: 11 mois. b) Survie en fonction des localisations métastatiques: — osseuses isolées: 17 mois; — os + foie ou cerveau: 6 mois; — os + autres: 11 mois. c) Survie en fonction de l’indication: — échappement thérapeutique: 9 mois; — contre-indication à un traitement médical ou irradiation de fermeture: 27 mois. Toutes ces différences sont statistiquement significatives. 84 JULLIEN et al.

4. CONCLUSIONS

4.1. Choix de l’irradiation en doses fractionnées

Le choix de l’irradiation à 15 Gy en 10 fractions de 1,50 Gy plutôt que celui de la dose unique de 6 à 8 Gy a été effectué pour plusieurs raisons. D’une part, le syndrome d’irradiation aigu est absent; en conséquence l’hospitalisation, la prémédication et la réanimation sont inutiles car la tolérance est excellente. D’autre part, le temps d’irradiation pour chaque séance est très court. Enfin, l’efficacité antalgique et fonctionnelle est équivalente à celle obtenue pour des irradiations en dose unique.

4.2. L’irradiation hémicorporelle est un traitement efficace des métastases osseuses hyperalgiques du cancer du sein

Dans la série étudiée, la réponse antalgique a été obtenue dans 87,5% des cas avec 69% de réponse complète. On note que l’arrêt des opiacés a été possible dans 82% des cas et que le retour à une marche normale a été obtenu dans 86% des cas, en particulier pour 83% des malades grabataires. On remarque qu’aucune compli cation hématologique mortelle n’a été observée et qu’aucun cas de pneumopathie radique n’est apparu. On doit aussi relever que l’effet sur la survie est à démontrer. L’irradiation hémicorporelle est donc une technique efficace dans le traite ment des métastases osseuses hyperalgiques des cancers du sein. On l’utilise pour les localisations osseuses des adénocarcinomes prostatiques, des épithéliomes bronchiques et dans la maladie de Kalher. Sa technique, bien que simple, nécessite cependant une dosimétrie fiable et des possibilités de surveillance en particulier hématologique stricte.

BIBLIOGRAPHIE

CHAOUL, H., LANGE, K., Veber-Lymphogranulomatose und die Behandlung mit Röntgen strahlen, Muench. Med. Wochenschr. 70(1923) 725-727.

FITZPATRICK, P.J., RIDER, W.D., Half-body radiotherapy, Int. J. Radiat. Oncol., Biol. Phys. 1 (1976) 197-207.

FITZPATRICK, P.J., RIDER, W.D., Half-body radiotherapy of advanced cancer, J. Can. Assoc. Radiol. 27 (1976) 75-79.

FRYER, C.J.H., FITZPATRICK, P.J., RIDER, W.D., POON, P., Radiation pneumonitis following a large single dose of X-rays, Int. J. Radiat. Oncol., Biol. Phys. 4 (1978) 931-936.

PRATO, F.S., KURDYAK, P., SAIBIL, E.A., CARRUTHERS, J.S., RIDER, W.D., ASPIN, N.. The incidence of radiation pneumonitis as a result of single fraction upper half-body irradiation, Cancer 39 (1976) 71-78. IAEA-SM-290/47 85

RUBIN, P., Radiation toxicology: quantitative radiation pathology for predicting effects, Cancer 39 (1976) 729-736.

SALAZAR, O.M., RUBIN, P., BROWN, J.C., FELDSTEIN, M.L., KELLER, B.E., The assessment of tumor response to irradiation of lung cancer: continuous-vs-split course regimes, Int. J. Radiat. Oncol., Biol. Phys. 1 (1976) 1107-1118.

SALAZAR, O.M., RUBIN, P., KELLER, B., SCARANTINO, C., Systemic (half-body) radiation therapy: response and toxicity, Int. J. Radiat. Oncol., Biol. Phys. 4 (1978) 937-950.

SALAZAR, O.M., RUBIN, P., HENDRICKSON, F.R., et al., Single dose half-body irradiation for palliation of multiple bone métastasés from solid tumors, Cancer 58 (1986) 29-36.

IAEA-SM-290/10

TRATAMIENTO MULTIDISCIPLINARIO DEL TUMOR DE WILMS Experiencia de trece años

T. LANCHE, R. BORREGO, G. MARTINEZ, R. RIVERA Instituto Nacional de Pediatría, Secretaría de Salubridad y Asistencia Pública, México, D.F., México

Abstract-Resumen

MULTIDISCIPLINARY TREATMENT OF WILMS’ TUMOUR: 13 YEARS’ EXPERIENCE. The authors describe the results obtained by them personally in their multidisciplinary treatment of Wilms’ tumour. In the National Paediatric Institute, Mexico City, 116 cases of Wilms’ tumour in children were studied between January 1971 and December 1983. Of these, only 57 were evaluated as only they had completed their multidisciplinary treatment and had been followed up for over two years. Wilms’ tumour is the solid abdominal tumour most frequently found in Mexican children. It is the fifth most frequent malign tumour after leukaemia, tumours of the central nervous system, Hodgkin’s disease and non-Hodgkin’s lymphomas, and retinoblastoma. The multidisciplinary treatment included: radical surgery; radiotherapy (site and dosage by group and by age of the child) and chemotherapy (drugs according to the group and histology of the tumour). In 82% of cases, the tumours occurred before the age of five, predominantly in girls. The average growth time was three months. Where tumour histology was favourable, 78% survived; 45% survived when the histology was adverse. For the various groups, survival was 100% in group I, 83.5% in group II, 71.5% in group III and 25% in group IV. The survival of all groups was 67% and the actuarial survival was 83%.

TRATAMIENTO MULTIDISCIPLINARIO DEL TUMOR DE WILMS: EXPERIENCIA DE TRECE AÑOS. Los autores muestran los resultados obtenidos de su experiencia personal en el tratamiento multidisciplinario del tumor de Wilms. En el Instituto Nacional de Pediatría, de la ciudad de México, en el período comprendido entre enero de 1971 y diciembre de 1983 se estudiaron 116 casos de niños con tumor de Wilms. De éstos únicamente 57 entraron a valoración por haber completado su tratamiento multidisciplinario y haber sido seguidos subsecuentemente durante más de dos años. El tumor de Wilms es el tumor sólido abdominal más frecuente en el niño mexicano. De todos los tumores malignos ocupa el quinto lugar en frecuencia después de las leucemias, de los tumores del sistema nervioso central, de la enfermedad de Hodgkin y linfomas no Hodgkin y del retinoblastoma. El tratamiento multidisciplinario incluyó: cirugía radical; radioterapia (sitio y dosis de acuerdo al grupo y edad del niño), y quimioterapia (drogas de acuerdo al grupo e histología del tumor). En el 82% de los casos, los tumores se presentaron antes de los 5 años de edad, con predominio en el sexo femenino. El tiempo de evolución promedio fue de 3 meses. La sobrevida en los tumores de histología favorable fue del 78% y en los de histología desfavorable del 45%. La sobrevida en los diferentes grupos fue del 100% en el grupo I, del 83,5% en el II, del 71,5% en el III y del 25% en el grupo IV. La sobrevida global de todos los grupos fue del 67% y la actuarial del 83%.

87 88 LANCHE et al.

1. INTRODUCCION

El tumor de Wilms fue descrito por primera vez como entidad clínica en 1899 por el cirujano alemán Max Wilms [ 1 ]. Se le han asignado otros nombres diferentes, siendo el más conocido el de nefroblastoma. El tumor de Wilms es un tumor embrionario altamente maligno que afecta al riñon del niño. Se manifiesta con mayor frecuencia durante los primeros 5 años de la vida, alcanzando un pico máximo entre los 2 y los 5 años de edad. Ocasional mente se presenta en el adolescente y muy raramente en el adulto. Es el tumor sólido abdominal más frecuente en el niño mexicano; esto contrasta con las estadísticas norteamericanas, donde el neuroblastoma ocupa el primer lugar [2, 3]. Se ha reportado una frecuencia de bilateralidad que varía del 3 al 13%. La diferencia señalada en estas cifras puede deberse a que hay casos en que cuando el tumor bilateral es pequeño en uno de los riñones, éste puede ser erradicado con el tratamiento quimioterápico, de modo que sólo un porcentaje pequeño del 3% sea el que se manifieste como tumor bilateral [4, 5]. Conforme han pasado los años se ha venido reportando un aumento de la sobrevida global. Antes de 1950, la sobrevida apenas alcanzaba un 20%. Gross y Neuhauser [6] reportaron un incremento al 47% con la introducción de la radioterapia post-nefrectomía. Años después, con la adición de la quimioterapia (primero la actinomicina [7] y más tarde la Vincristina [8, 9]) mejoró aún más la sobrevida. En esa época Färber reportó una sobrevida del 81% a 2 años. En la actualidad se espera una sobrevida hasta del 90% o más, gracias a la introducción de la Adriamicina para aquellos tumores con histología desfavorable.

2. MATERIAL Y METODOS

El Instituto Nacional de Pediatría de la ciudad de México es un hospital relativamente joven, ya que cuenta con tan sólo 15 años de vida. La presente serie comprende la experiencia obtenida por el grupo de oncología (un radioterapeuta, un cirujano y dos quimioterapeutas) en el período comprendido entre enero de 1971 y diciembre de 1983. Se atendieron 116 casos de tumor de Wilms, de los cuales solamente 57 fueron valorables ya que recibieron tratamiento multidisciplinario completo y tuvieron control subsecuente por más de 2 años después de su diagnóstico; 28 pacientes no puedieron ser clasificados por haber sido atendidos previamente fuera del Instituto (todos con nefrectomía no radical y quimioterapia incompleta); 31 casos no entraron a valoración por haber abandonado su tratamiento sin haberlo completado. En la gran mayoría de los casos el abandono se debió a falta de recursos económicos para continuar con la quimioterapia. IAEA-SM-290/10 89

3. PROTOCOLO DE ESTUDIO

El niño portador de un tumor abdominal es canalizado al Servicio de Oncología ya sea por la consulta externa or por el Servicio de Urgencias. Se le considera como una urgencia relativa tanto para efectuar su diagnóstico definitivo como para llevar a cabo su tratamiento. El niño entonces entra de inmediato al siguiente protocolo de estudio:

1) Elaboración de historia clínica completa.

2) Solicitud de los siguientes estudios de laboratorio: — citología hemática completa — química sanguínea — pruebas de funcionamiento hepático — pruebas de coagulación — examen general de orina con urocultivo, y — ácido vanil mandélico.

3) Estudios de gabinete: — radiografía de tórax posteroanterior y lateral — urografía excretora con venocavografía — gammagrafía hepática, y — tomografía pulmonar en caso de sospecha de metástasis.

4. DIAGNOSTICO

De los estudios de gabinete, el más importante fue la urografía excretora, que mostró generalmente una masa intrarrenal parenquimatosa dentro de una pseudo- cápsula, produciendo estrechez y atrofia de la corteza renal. Fue común encontrar distorsión del sistema pielocalicial o amputación de los cálices. En pocos casos se observó extensión tumoral hacia la vena renal y cava antes de la cirugía. Desde hace algunos años ya no se practica la arteriografía selectiva, porque con la urografía excretora es suficiente para llegar al diagnóstico; se ha reservado para los casos de tumor bilateral. Con los estudios antes mencionados pudimos hacer diagnóstico diferencial de hidronefrosis, riñon poliquístico, neurablastoma, hepatoma y esplenomegalia. En todos los casos estudiados, excepto en uno, se hizo el diagnóstico preoperatorio de tumor de Wilms.

5. TRATAMIENTO

El manejo multidisciplinario incluyó: cirugía radical; radioterapia, y quimioterapia. 90 LANCHE et al.

5.1. Cirugía radical

Generalmente se llevó a cabo de 48 a 72 horas después de la admisión del niño [10], procediéndose a lo siguiente:

— Se efectuó incisión transversa supraumbilical en el lado afectado. — Se examinó el abdomen y el retroperitoneo, con especial atención del riñón contralateral, el hígado y el resto de los órganos intrabdominales. — Se valoró el pedículo renal para determinar si había infiltración renal. En el caso de encontrarse libre de tumor, se ligó el pedículo renal antes de manipular el riñón y la masa tumoral; esto se hizo con el objeto de cerrar en forma temprana la vía de diseminación hematógena. — Cuando hubo extensión tumoral dentro de la vena renal o de la vena cava inferior, el tumor fue removido antes de la ligadura con el objeto de evitar una emboliza ción tumoral posterior. — Se trató de evitar en lo más posible una manipulación excesiva para no favorecer la ruptura del tumor. — Se efectuóla nefrectomía junto con el tejido perirrenal con ureterectomía y linfadenectomía'del hilio renal hasta nivel de la bifurcación de la aorta. — Se dejaron clips de plata en el lecho renal y se cerró por planos.

5.2. Radioterapia

Generalmente se inicia de 24 a 48 horas después de la cirugía. No hubo necesidad de esperar la cicatrización de los tejidos. En términos generales, el niño toleró bien la irradiación en el postoperatorio inmediato. En ocasiones hubo necesidad de esperar 72 horas para contar con el resultado histopatológico de la pieza operatoria [11]. La estadificación para la presente serie se basó en la clasificación por grupos del Estudio Nacional № 2 del Tumor de Wilms [12] de los EE UIJ, que a continuación se detalla:

Grupo I. Tumor limitado al riñón, completamente resecado. Cápsula intacta. No hay tumor residual aparente más allá de los límites de resección.

Grupo 11. El tumor se extiende más allá del riñón. Es resecado en forma completa. Infiltra la cápsula, tejidos perirrenales, ganglios linfáticos y vasos renales.

Grupo III Hay residual no hematógeno confinado al abdomen. El tumor se biopsió o hubo ruptura tumoral durante el acto quirúrgico. Presencia de implantes peritoneales. Infiltración de ganglios linfáticos más allá de las cadenas paraórticas. Tumor macroscópicamente irresecable por infiltración de estructuras vitales.

Grupo IV. Metástasis hematógenas a pulmón, hígado, hueso, cerebro, etc.

Grupo V. Tumor bilateral inicial o subsecuente. IAEA-SM-290/10 91

La conducta desde el punto de vista radioterápico fue la siguiente: a) En el grupo I tumores con histología favorable. No recibieron radioterapia los niños menores de un año de edad. b) Todos los niños con tumores de histología desfavorable, independientemente de la edad, recibieron tratamiento radioterápico. A partir del grupo II todos los niños recibieron radiación independientemente de la edad y de la histología; la única variación fue en el tamaño de los campos. En los grupos I y II los pacientes recibieron radiación al hemiabdomen correspon diente, cuyo límite superior llegó al diafragma y el inferior hasta la cresta iliaca correspondiente; se incluyeron los cuerpos vertebrales en su total anchura. En el grupo III los campos de radiación incluyeron a todo el abdomen, en su límite inferior llegaron hasta la base de los agujeros obturadores. Siempre se protegió el riñón contralateral. La dosis de radiación fue calculada al plano medio del abdomen y varió de acuerdo con la edad del niño — de 1800 a 4000 rads en dos y media a cuatro semanas. En todos los casos se empleó radiación de supervoltaje. En el grupo IV, pacientes con metástasis pulmonares recibieron 1200 rads en 2 semanas a ambos pulmones y radiación a todo el abdomen. En pacientes con metástasis hepáticas la dosis de radiación fue de 3000 rads en 4 semanas, además de la radiación correspondiente al abdomen. Ningún paciente presentó metástasis cerebrales. En el grupo V el tratamiento consistió en efectuar nefrectomía radical del riñón más afectado con resección parcial del riñón contralateral. Todo el abdomen recibió radiación con protección del riñón remanente.

5.3. Quimioterapia

Para el tumor de Wilms la quimioterapia es tan indispensable como el resto de las modalidades de manejo, sobre todo cuando se trata de etapas avanzadas o de tumores definidos como de mal pronóstico desde el punto de vista histológico. Esta modalidad se inicia generalmente al mismo tiempo que la cirugía (o dentro de las primeras 24 a 48 horas postnefrectomía). Se utilizaron las combinaciones de Actinomicina D, Vincristina, Adriamicina y Ciclofosfamida. La selección de la droga, la dosis y el tiempo de exposición estuvieron condicionados tanto a la etapa como a la variedad histológica, como se observa en el Cuadro I (histología favorable). Los grupos de histología desfavorable fueron tratados con el mismo esquema que se usó para el grupo IV de histología favorable detallado en dicho cuadro.

6. CONTROL SUBSECUENTE

Durante el primer año, el control se llevó a cabo con examen clínico y radiografía de tórax mensuales y urografía excretora trimestral; en el segundo 92

CUADRO I. QUIMIOTERAPIA EN EL TUMOR DE WILMS DE HISTOLOGIA FAVORABLE LANCHE et al. et LANCHE

S: sulfato de Vincristina; 2 mg/m2 de superficie corporal; vía intravenosa; dosis máxima de 2 mg por dosis. SS: sulfato de Vincristina; 2 mg/m2 de superficie corporal; vía intravenosa; el primero y quinto día por ciclo AM D. A: Actinomicina D; 15 gammas/kg/día/5 días; dosis máxima de 500 gammas por día. AD: Adriamicina; 10 mg/m2/día/3 días; dosis máxima acumulable de 450 mg. C: Ciclofosfamida; 10 mg/kg/día/3 días. IAEA-SM-290/10 93

№ de pacientes 30-

20-

10- П П п п 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 1 10 Edad en años FIG. 1. Frecuencia del tumor de Wilms.

FIG. 2. Sobrevida con respecto a la histología.

año, el control fue bimestral y la urografía semestral; en el tercero hubo control trimestral y radiografía de tórax semestral; en el cuarto se realizó el control clínico semestralmente, y del quinto año en adelante la revisión fue anual.

7. RESULTADOS

En el período comprendido entre enero de 1971 y diciembre de 1983 se estudiaron 116 casos de tumor de Wilms, de los cuales solamente 57 fueron valorables. 94 LANCHE et al.

ETAPA I: 100%

FIG. 3. Sobrevida con respecto a las diferentes etapas.

FIG. 4. Sobrevida global. IAEA-SM-290/10 95

1,0-

0,75- o < о m 0,50'

0,25-

A IÍ0 S DE SOBREVIDA 0,0 т------1------1------1 Г 1 2 3 4 5 FIG.5. Sobrevida actuarial de todos los grupos: 83% (1979-1983).

El tumor de Wilms ocupó el quinto lugar en frecuencia después de las leucemias, de los tumores de sistema nervioso central, de la enfermedad de Hodgkin y linfomas no Hodgkin y del retinoblastoma. El 82% de los tumores se presentaron antes de la edad de 5 años; sólo hubo dos pacientes adolescentes (Fig. 1). Hubo mayor predominio en el sexo femenino: 67 casos, contra 49 del sexo masculino. De la serie de 116 casos sólo hubo 2 tumores bilaterales. El tiempo de evolución varió desde 1 día hasta 18 meses, con un promedio de 3 meses. Las principales manifestaciones clínicas en orden decreciente fueron: presencia de tumor abdominal, ataque al estado general, fiebre, dolor, hematuria, cuadros enterales, hipertensión arterial, disuria y polaquiuria. El 21% de pacientes tuvo metástasis pulmonares al tiempo de la primera consulta. El número de pacientes de acuerdo con los grupos fue el siguiente: grupo I: 4 casos; grupo II: 18 casos; grupo III: 21 casos; grupo IV: 12 casos y en el grupo V: 2 casos. Definitivamente, en nuestros pacientes la histología fue un factor pronóstico de mucha importancia. En los tumores de histología favorable la sobrevida fue del 78%. En cambio en los tumores de histología desfavorable fue del 45% (Fig. 2). La sobrevida de acuerdo con la etapa fue la siguiente: en la etapa I: 100%; en la etapa II: 83,5%; en la etapa III: 71,5% y en la etapa IV: 25% (Fig. 3). La sobrevida global de todas las etapas fue del 67%, con una media de 4,8 años (Fig. 4). La sobrevida actuarial de todas las etapas es del 83% (Fig. 5). 96 LANCHE et al.

8. DISCUSION

El tumor de Wilms es el tumor sólido más frecuente en el niño mexicano. Con el manejo multidisciplinario es altamente curable en las etapas tempranas. Aun en la etapa III, la sobrevida todavía es alta. El componente histológico es un factor pronóstico muy importante; los tumores con histología favorable con componente epitelial, clásico o trifásico (blastomatoso, tubular) y rabdomiomatoso tienen una sobrevida mucho más alta que los tumores de histología desfavorable (con componente anaplástico). A partir de enero de 1984 adoptamos la clasificación del Estudio Nacional de Wilms № 3, en donde no hay variación respecto a la cirugía, pero sí en cuanto a la radiación y a la quimioterapia. Es importante señalar que en nuestro país los niños acuden a la primera consulta en etapas avanzadas. Tenemos la esperanza de que cuando podamos hacer una detección oportuna y temprana, podamos entonces alcanzar ese tan deseado 100% de curación.

REFERENCIAS

[1] W ILMS, M., Die Mischgeschwülste der Nieren, Arthur Georgi, Leipzig (1899) 1—90. [2] MARSDEN, H.B., STEWARD, J.K., “Wilms' tumors”, Tumors in Children, Springer, New York (1968). [3] YOUNG, J.L., MILLER, R.W., Incidence of malignant tumors in U.S. chüdren, J. Pediatr. 86(1975) 254-258. [4] FAY, R., BROSMAN, S., WILLIAMS, D.I., Bilateral nephroblastoma, J. Urol. 110 (1973) 119-126. [5] RAGAB, A.H., VIETTI, T.J., CRIST, W., et al., Bilateral Wilms’ tumors. A review, Cancer 30 (1972) 983-988. [ 6] GROSS, R.E., NEUHAUSER, E.B.D., Treatment of mixed tumors of the kidney in childhood, Pediatr. 6 (1950) 843-852. [7] FÄRBER, S., Chemotherapy in the treatment of leukemia and Wilms’ tumor, JAMA, J. Am. Med. Assoc. 198 (1966) 826—836. [8] VIETTI, T.J., SULLIVAN, M.P., HAGGARD, M.E., HOLCOMB, T.M., BERRY, D.H., Vincristine sulfate and radiation therapy in metastatic Wilms’tumor, Cancer 25 (1970) 12-20. [9] SUTOW, W.W., Proceedings: Chemotherapy in Wilms’tumor: An appraisal, Cancer 32 (1973) 1 1 5 0-1 153. [10] EHR LICH, R.M ., GOODW IN, W.E., The surgical treatment of nephroblastoma, Cancer 32 (1973) 1145-1149. [11] CASSADY, J.R., TEFT, M., FILLER, R.W., JAFFE, N.. HELLMAN, S., Considerations in the radiation therapy of Wilms’ tumor, Cancer 32 (1973) 598—608. [12] SUTOW, W.W., VIETTL, T.J., FERNBACH, D.J., “Wilms’tumor”, Clinical Pediatric Oncology, Second Edition, Mosby Company (1977) 547. IAEA-SM-290/19

RADIOTERAPIA PREOPERATORIA EN EL TRATAMIENTO DEL CANCER DE RECTOSIGMOIDES

C. GONZALEZ-MIRANDA, L. BADINEZ, J. RAJEVIC Servicio de Radioterapia, Hospital Militar, Santiago, Chile

Abstract-Resumen

PRE-OPERATIVE RADIOTHERAPY FOR RECTOSIGMOID CANCER. Between 1977 and 1982, 25 cases of rectosigmoid cancer were treated by the radiotherapy service of the Military Hospital using pre-operative pelvic cobalt teletherapy at doses of 50 Gy over five weeks, with a view to increasing the ease of resection of tumours which were initially fixed and to reducing the capacity for local and distant implantation of the tumour cells released during surgery, thus reducing the risk of local relapse and the spread of the disease. Twenty-one patients were operated on, 15 radically; in one case no residual tumour was found during exploration and five were not considered resectable (three because of abdominal métastasés and two because of local extension). Sixteen of the 21 patients operated on were considered to have resectable tumours; 69% of these had history of tumour fixation at deep pelvic levels. Local control was obtained in 86.7% of cases, with two local relapses. Ten of 15 patients who had tumours removed are alive with no evidence of disease after 18 to 64 months of follow-up. In four cases there was complete disappearance of the tumour after radiotherapy (16%), in nine cases there was a reduction by at least 50% in tumour volume (36%), and in four cases there was a reduction of less than 50% in the initial tumour volume (16%). The indications for this technique are discussed in the light of the results obtained.

RADIOTERAPIA PREOPERATORIA EN EL TRATAMIENTO DEL CANCER DE RECTOSIGMOIDES. De 1977 a 1982 se trataron en el Servicio de Radioterapia del Hospital M ilitar de Santiago de Chile 25 casos de cancer de rectosigmoides mediante telecobaltoterapia pelviana preoperatoria con dosis de 50 Gy en 5 semanas, con la intención de aumentar la resecabilidad de tumores inicialmente fijos y reducir la capacidad de implantación local y a distancia de las células tumorales diseminadas durante la cirugía, reduciendo el riesgo de recidiva local y diseminación de la enfermedad. Fueron intervenidos 21 pacientes (15 con cirugía radical); en un caso no se encontró tumor residual en la exploración y 5 se consideraron irresecables (3 por metástasis abdominales y 2 por extensión local). Se consideraron resecables 16 de los 21 pacientes intervenidos (76%); el 69% de éstos presentaba antecedentes clínicos de fijación tumoral a planos pelvianos profundos. Se obtuvo control local en el 86,7% de los casos y se registraron 2 recidivas locales. Diez de los 15 pacientes resecados están vivos, sin evidencia de enfermedad transcurridos entre 18 y 64 meses de seguimiento. En 4 casos se obtuvo desaparición completa del tumor después de la radioterapia (16%), en 9 casos hubo reducción de por lo menos el 50% del volumen tumoral (36%) y en 4 casos reducción menor al 50% del volumen tumoral inicial (16%). Se discuten las indicaciones de esta técnica a la luz de los resultados obtenidos.

97 98 GONZALEZ-MIRANDA et al.

1. INTRODUCCION

La radioterapia preoperatoria en el tratamiento del cáncer de rectosigmoides continúa siendo un tema de controversias. Hay numerosos trabajos que sugieren la obtención de un mejor control de la enfermedad local en lesiones precoces А у В de pacientes irradiados en comparación con pacientes no irradiados. No se ha comprobado efecto deletéreo con la radioterapia, comparado con la mortalidad operatoria aunque haya retardo en la cicatrización perineal postoperatoria. Desde los trabajos de Tepper (1968), se reconoce una importante reducción tumoral del tumor primario con el régimen de radioterapia preoperatoria, e incluso se describe la desaparición completa de algunos, tanto del punto de vista clínico como patológico. Ciertos trabajos señalan también una disminución del compromiso ganglionar regional en comparación con pacientes no irradiados (un 16% contra un 43%), lo que indicaría esterilización de los ganglios comprometidos microscópicamente (Combes y Gary-Bobo, 1983). Otros trabajos han demostrado incluso aumento de la sobrevida en estudios prospectivos y al azar, especialmente utilizando dosis bajas y cirugía inmediata (Rider, 1977). En este trabajo se presentan los resultados clínicos e histopatológicos de control local obtenidos en estudios retrospectivos de pacientes con cáncer de rectosigmoides tratados en el Hospital Militar de Santiago de Chile. Al mismo tiempo se señalan resultados preliminares de sobrevida de los mismos pacientes.

2. PACIENTES Y METODOS

De 1977 a 1982 se trataron 25 casos de cáncer de rectosigmoides con cobalto 60 en el Servicio de Radioterapia del Hospital Militar de Santiago de Chile. 24 pacientes recibieron dosis de 40-50 Gy en 4 a 5 semanas (20 recibieron 50 Gy, 2, 45 Gy y 2, 40 Gy) y 1 no completó la terapia programada por presentar disemi nación en el curso de ésta. 11 casos fueron tratados con técnicas de 4 campos AP РА y laterales, 2 casos con campos opuestos anterior y posterior y 2 pacientes con cicloterapia pendular a 120°. Todos los tratamientos se efectuaron con fraccionamiento de 2 Gy diarios, 5 veces por semana; en todos los casos se efectuó simulación radiológica con medio de contraste baritado. Posteriormente se hizo verificación del centraje y de los límites del área. De los 25 pacientes, 12 eran varones y 13 mujeres, la mayoría en la sexta y séptima década de la vida con un rango de 69 y 76 años de edad. No hubo diferencias en la distribución etaria en ambos sexos. La cirugía se realizó entre 1 y 8 semanas después de finalizada la radioterapia (en el 66,6% de los casos a las 4 a 6 semanas, un 16% a las 3 semanas y un 16,6% a las 8 semanas). IAEA-SM-290/19 99

Se analizó la respuesta clínica del tumor primario a la radioterapia y, de acuerdo a los hallazgos quirúrgicos e histopatológicos, se estudió si se produce aumento de los índices de resecabilidad de tumores inicialmente fijos y si aumenta el control local reduciendo el porcentaje de estadios avanzados.

3. RESULTADOS

3.1. Cirugía

Se operó a 21 de los 25 pacientes, uno se consideró inoperable por contrain dicación médica, otro presentó diseminación antes del término de la radioterapia y 2 no se presentaron a completar el procedimiento quirúrgico. En 15 de los 21 casos se efectuó cirugía radical. Dos fueron resecciones anteriores en tumores de 10-11 cm, desde el margen anal respectivamente, ambos con reducción de un 50% de la masa original después de haber efectuado la radio terapia (estadios clínicos B1 y B2 de Astler y Coller modificadamente, respectiva mente), sin signos de recidiva ni diseminación en el seguimiento posterior a los 18 y 24 meses. En 7 pacientes se efectuó la operación abdómino-perineal de Miles. En 3 de estos casos se comprobó una reducción clínica del tumor a menos del 50%, uno de los cuales recidivó en el periné y en la pelvis a los 17 meses de la intervención —previamente había sido catalogado como E.B.3 en el estudio postoperatorio, pero se consiguió una disminución de la fijación del tumor al sacro. Un paciente falleció a los 6 meses por enfermedad no relacionada con su cáncer (C2 en el estudio histopatológico) y uno se encuentra sin evidencia de enfermedad 36 meses después de la radioterapia, habiendo sido catalogado como C2 en el estudio de la pieza operatoria. En un caso de tumor sigmoideo alto, en el que no se estudió la respuesta clínica, el paciente se encontraba sin evidencia de enfermedad después de 5 años y 4 meses del seguimiento postradioterapia. En otro caso, inicialmente clasificado como B2, se comprobó lesión esterilizada en el acto quirúrgico, y el estudio histopatológico reveló una zona ulcerativa de 3 cm, sin signos de cáncer y con fijación al plano pelviano posterior; no evidenció enfermedad residual a los 27 meses de seguimiento. En 2 pacientes se observó respuesta de un 50% a la radioterapia (previamente habían sido clasificados como B2); uno de ellos está sin evidencia de enfermedad a los 39 meses del seguimiento y el otro presentó metástasis pulmonares a los 29 meses de seguimiento. En 6 pacientes hubo exenteración pelviana posterior, y los hallazgos quirúrgicos demostraron reducción de menos del 50% en 2 casos (estadios B2), los que a los 17 y 47 meses de control no evidenciaron signos de recidiva. Tres casos tuvieron una respuesta mayor del 50%; uno de los casos fue clasificado E.C.2 en la pieza operatoria y presentó metástasis óseas y recidiva local a los 7 meses; un enfermo falleció a los 12 meses por diseminación pulmonar y 100 GONZALEZ-MIRANDA et al. hepática sin evidencia de recidiva pelviana (B3 en la pieza operatoria); otro caso está sin evidencia de enfermedad local a los 42 meses de control, habiendo sido catalogado como B1 en la pieza operatoria. Finalmente, un caso presentó una respuesta completa en un tumor que medía inicialmente 6 cm, adherido a los planos vecinos y sólo mostraba rectitis actinica en la pieza operatoria, sin signos de enfermedad neoplásica a los 33 meses de seguimiento. Otro paciente no mostró signos de cáncer residual en la laparatomía con rectosigmoidoscopía preoperatoria, por cuyo motivo no se efectuó extirpación alguna, y hoy se encuentra sin evidencia de enfermedad a los 4 años de seguimiento. En 5 casos la laparatomía reveló lesiones irresecables: 3 por metástasis abdominales y 2 por extensión local. Cuantro de estos pacientes tuvieron una escasa o nula respuesta a la radioterapia preoperatoria y en el otro la lesión se redujo en más de un 50%, pero permaneció fijo al sacro. En un caso se descartó la operación por tratarse de un paciente con deterioro senil avanzado que hacía imposible manejar la colostomía; presentó una respuesta clínica completa al tratamiento radiante y falleció después de 7 años por otra enfermedad. Por último, 2 enfermos no se presentaron al tratamiento quirúrgico post radioterapia; en ambos casos la reducción tumoral fue más de un 50% del volumen inicial. A pesar de las citaciones para su control se ignora su estado actual.

3.2. Resecabilidad

Se consideraron resecables 16 de 21 casos, o sea un 76% de los pacientes intervenidos, 11 de los cuales tenían antecedentes clínicos de fijación a los planos pelvianos profundos antes de la radioterapia (68,8%).

3.3. Control local

Se obtuvo control local del cáncer en 13 de los 15 pacientes resecados (86,7%); un enfermo recidivó a los 17 meses de seguimiento y otro paciente a los 7 meses, después de 3 meses de haber presentado diseminación ósea múltiple.

3.4. Sobrevida

De los 15 pacientes resecados 10 se encuentran sin evidencia de enfermedad tras 18 a 64 meses de seguimiento, lo que equivale a un 66,6%. Un paciente falleció por otra causa, sin evidencia de enfermedad a los 8 meses de evolución, y 4 pacientes fallecieron por su cáncer tras 8 a 31 meses de seguimiento. IAEA-SM-290/19 101

3.5. Respuesta a la radioterapia

Se analizó la respuesta al tratamiento radiante preoperatorio de 17 pacientes, de los que se había consignado el tamaño tumoral por endoscopía, enema baritada y examen físico; el control posterior se ejerció por análisis clínico exploratorio y/o examen anatomopatológico de la pieza operatoria.

3.5.1. Respuesta completa

Se consiguió respuesta completa en 4 pacientes (16%); dos de los tumores medían 2 y 6 cm en el estudio histopatológico; un paciente no fue resecado por no haber tumor residual, y otro paciente no fue intervenido por contraindicación médica; el cáncer de éste medía 4 cm de diámetro mayor y falleció a los 7 meses de seguimiento por otra enfermedad, sin evidencias clínicas de cáncer.

3.5.2. Respuesta parcial

La respuesta parcial fue: a) Mayor o igual al 50% del volumen tumoral inicial: 9 pacientes (36%); 5 se encuentran sin evidencia de tumor en el último control; 2 han fallecido por metástasis; 1 paciente no se presentó a completar el tratamiento quirúrgico y el otro paciente fue irresecable. b) Menor que un 50% del volumen tumoral inicial: 4 enfermos (16%), 2 de los cuales se encuentran sin signos de cáncer, otro se consideró irresecable y el último presentó recidiva local a los 17 meses de seguimiento.

4. CONCLUSIONES

Los resultados de este análisis concuerdan (véanse los Cuadros I—III) con las publicaciones de otros centros, aunque la casuística es pequeña y el seguimiento no es completo como para postular conclusiones definitivas en cuanto a control local (86%) y a la resecabilidad (76%). El efecto citoreductor es de más de un 50% en los casos de respuesta completa, o al menos cercano al 50% del volumen inicial, confirmando el valor dé la radioterapia preoperatoria. Cabe destacar que la radioterapia preoperatoria, a pesar de los síntomas y signos desagradables que pueda producir, al final resulta una terapia beneficiosa ya que reduce los efectos de la masa inicial, disminuye el síndrome doloroso llegando a veces al semibloqueo de la evacuación, reduce o elimina el sangramiento, y muchas veces regula el tránsito digestivo a niveles normales. De todas formas es necesario precisar con là mayor certeza posible la extensión local del tumor primario, así como determinar las metástasis y su magnitud a fin de mejorar el criterio de operabilidad con el fin de obtener máximo rendimiento terapéutico. 102 GONZALEZ-MIRANDA et al.

CUADRO I. RESPUESTA CLINICA A LA RADIOTERAPIA PREOPERATORIA EN EL TRATAMIENTO DEL CANCER DE RECTOSIGMOIDES

Respuesta Casos (25) %

Completa 4 16 Parcial 50% o más 9 36 menos del 50% 4 16 Sin respuesta 1 4

No evaluable 7 28

CUADRO II. CONTROL LOCAL CON RADIOTERAPIA PREOPERATORIA Y CIRUGIA RADICAL EN EL TRATAMIENTO DEL CANCER DE RECTOSIGMOIDES

Situación local Casos (15) %

Control 13 86,6

Recidiva local 1 6,6 Diseminación 1 6,6

CUADRO III. SOBREVIDA Y FALLECIDOS TRAS TRATAMIENTO CON RADIOTERAPIA PREOPERATORIA Y CIRUGIA RADICAL DEL CANCER DE RECTOSIGMOIDES

Casos (15) %

Vivos S.E.E. (18-48 meses) 10 66,6 Fallecidos por cáncer 4 26,6 Fallecidos por otra causa 1 6,6 IAEA-SM-290/19 103

Creemos que es posible seleccionar un grupo compuesto de pacientes que presentan un pronóstico de mayor riesgo, de pacientes que responden escasamente a la radioterapia preoperatoria, y de pacientes en quienes persiste el compromiso metastásico ganglionar evaluado en el acto operatorio. Todos ellos podrían beneficiarse con otro tipo de terapia complementaria.

BIBLIOGRAFIA

BAYER, I., TURANI, H., LURIE, H., CHAIMOFF, CH., “The sandwich approach irradiation- surgery in rectal cancer — four years’ experience”, presented at the meeting of the American Society of Colon and Rectal Surgeons with the Section of Colo-Proctology, Royal Society of Medicine, and the Section of Colonic and Rectal Surgery, Royal Australasian College of Surgeons, New Louisiana (1984).

BOGNEL, C., PRADE, M., Etude anatomo-pathologique de 40 cas de cancers du rectum irradiés en pré-opératoire, Bull. Cancer 71 2 (1984) 149-152.

COMBES, P.F., GARY-BOBO, J., Cancers du rectum, radiothérapie pré-opératoire, Bull. Cancer 70 4 (1 9 8 3 )3 1 7 .

DERDEL, J., Is dose/time fractionation important in treating rectal cancer, Int. J. Radiat. Oncol., Biol. Phys. 11 (1984) 579-582.

G A STR O IN TE STIN A L TUM OR STUDY GROUP, Prolongation of the disease-free interval in surgically treated rectal carcinoma, New Engl. J. Med. 312 23 (1985) 1466-1472.

GUNDERSON, L.L., “Radiation therapy of colorectal carcinoma”, Digestive Cancer, Thatcher, N. (Ed.) (Proc. 12th Int. Cancer Congr., New York), Pergamon Press, Oxford (1979) 38.

GUNDERSON, L.L., COHEN, A.M., WELCH, C.E., Residual inoperable or recurrent colorectal cancer. Surgical radiotherapy interaction, Am. J. Surg. 139 (1980) 5 18-525.

MARTINEZ, A., EDMUNDSON, G., Combination of external beam irradiation and multiple- site perineal applicator (m upit) for treatment of locally advanced or recurrent prostatic, anorectal, and gynecologic malignancies, J. Radiat. Oncol., Biol. Phys. 11 (1984) 391-398.

MELLA, О., DAHL, О., Radiotherapy and resection for apparently inoperable rectal adenocarcinoma, Dis. Colon Rectum 27 (1984) 663-668.

N O H IU D D IN , М ., DERDEL, J., Results of adjuvant radiation therapy in cancer of the rectum, Cancer 55 (1985) 350-353.

ROBERSON, S., HERON, H., Is anterior resection of the rectosigmoid safe after preoperative radiation, Dis. Colon Rectum 28 (1985) 254-259.

SALMON, R., G U ILL E T , J., Cancers opérables du rectum: radiothérapie pré-opératoire, Masson, Paris (1983) 430-433.

SCHEIN, P., W OOLLEY, P., “Carcinoma de Colon”, Seminarios de Oncología, Editorial Médica Panamericana S.A., Buenos Aires (1978).

WEISS, D., RICH, T., Sphincter preservation in patients with low rectal cancer treated with radiation therapy with or without local excision or fulguration, Radiology 156 (1985) 527-531.

IAEA-SM-290/8

THE ROLE OF RADIOTHERAPY IN LOCALIZED NON-HODGKIN’S LYMPHOMAS

M.B. PATRICIO, R. CABRAL, M. NEVES, A. de PONTE, M. VILHENA Department of Radiotherapy of the Instituto Portugués de Oncologia de Francisco Gentil, Lisbon, Portugal

Abstract

THE ROLE OF RADIOTHERAPY IN LOCALIZED NON-HODGKIN’S LYMPHOMAS. Between 1970 and 1982, ninety-one patients with stages I and II non-Hodgkin’s lymphomas (NHLs) submitted to radiotherapy were reviewed. After presenting data regarding histology and clinical presentation and distribution by age and sex, the authors retrospectively analysed the results according to different factors. Although 60Co teletherapy was the primary treatment for the majority of these cases, chemotherapy was also given to 65% of the patients. The conclusions are the following: ( 1 ) From a total of 261 patients with non-Hodgkin’s lymphomas, 91 cases were clinical stages I and II. (2) 61.2% of the patients had extra nodal lesions, mostly of the head and neck. (3) Mediastinal involvement was present in 3.8%. However, mediastinal invasion had a higher incidence (31.7% ) in children. (4) In the early stages of N H L unfavourable histological types showed a higher incidence (71.7% ) than favourable ones. (5) Radiotherapy played the major role in the treatment of stages I and II, although a combination with chemotherapy was applied in 65% of the cases. There was no significant difference in survival rates between the groups (p < 0.05). ( 6) Adults showed better disease free survival rates than did children (60 versus 40%). (7) The best results were observed in the group of patients with extralymphatic involvement of the head and neck (76.5%). ( 8) Response to therapy depends mainly on the histological aspects. Therefore, the 2-year relapse free survival rates are 75 and 58.9% for low and high grade malignancies, respectively. (9) Our results are in agreement with the current literature and demonstrate how NHL prognosis is connected with various factors.

1. INTRODUCTION

In recent years there has been considerable progress in studies of non-Hodgkin’s lymphomas (NHLs), especially with regard to the introduction of the new histopathological classification [1]. It was proposed in 1982 by the International Working Formulation and followed by a simple modification in favourable or low grade and unfavourable histology, including intermediate and high grade malignancies, both methods showing good prognostic correlation. These histological features are at present being considered for studies of therapeutic trials.

105 106 PATRICIO et al.

Traditionally [2, 3], radiotherapy has been considered an important modality in the management of localized NHL with 5-year survival rates ranging from 39 to 59%. However, there has been some controversy about the extension of the fields and/or the need for combined chemotherapy. These aspects were studied in successive protocols within the EORTC [4]. Its recent published results indicate that for all favourable and for small bulk disease with diffuse histology stage I NHL, regional radiotherapy alone is the proposed standard treatment with mini-mantle for supradiaphragm presentation or inverted Y field + splenic hilum and spleen for infradiaphragm cases. The same conclusion from the trials was made relative to stage II2 NHL with favourable histology and non-bulky disease only. Other teams [5-7] have suggested the same standard treatment because there was no significant difference in the total survival rates of patients treated with radiation therapy with or without combination chemotherapy. Radiotherapy was, therefore, considered an important modality in the management of localized NHL, with the advantage of being a relatively short and less toxic treatment avoiding the unnecessary hazards attendant on chemo therapy. However, different chemotherapy protocols are under study in combination with radiotherapy to improve the results of stage II3, II4 bulky disease and/or unfavourable histology. This paper presents a retrospective analysis of clinical stage I and II non-Hodgkin’s lymphoma patients referred for radiotherapy at the Instituto Portugués de Oncología de Francisco Gentil (IPOFG) over a period of twelve years.

2. MATERIAL AND METHODS

Between 1970 and 1982 a total of 261 patients with NHL were treated at the Department of Radiotherapy of IPOFG, Lisbon. Of these, 91 cases were classified as clinical stage (CS) I and II (34.9%), according to the Ann Arbor Clinical Staging Classification. The distribution of CS I,II patients by age was a bimodal curve exhibiting a peak for children that represents 11%, and another, higher peak in the age group between 60 and 70. Forty patients were male and fifty-one female (ratio = 1.2); however, the sex ratio was not constant over all the age ranges, showing a predominance of males in the groups under 50 years of age. Mediastinal involvement and lower torso nodal presentation were uncommon in localized NHL, being observed in only 8.5%. In children, however, mediastinal invasion had a higher incidence (31.7%). All the biopsy specimens, with the exception of the cases studied outside the IPOFG, were histologically reviewed and classified according to the modified working formulation. IAEA-SM-290/8 107

Thus, in CS I, II patients a relatively small proportion of cases with favourable histology (28.3%) was observed, whereas the majority of cases (71.7%) had a diagnosis of unfavourable malignancy, with a similar sex distribution in the two groups.

3. TREATMENT

We only considered 80 cases in the therapeutic analysis of the CS I,II patients as we excluded patients who had received the greater part of their treatment outside the Institute. Although 60Co teletherapy was the primary treatment, chemotherapy was also given in 65% o f the cases, especially in stage II; 68.8% received irradiation with limited fields and 31.2% with extended ones. The majority of CS I patients were irradiated with limited fields (80%). With regard to CS II with nodal presentation, 78.6% of the cases received regional radiotherapy (mantle, mini-mantle or inverted Y fields according to the presentation of the disease); in 21.4% the limited fields technique was applied. The mediastinum was irradiated only when involved. In CS II E, the majority of the patients (82.1%) was irradiated with limited fields. Tumour doses varied between 40 and 50 Gy in 82% of the patients, while 18% received 25 to 30 Gy. Twelve patients were also submitted to various kinds of surgery.

4. RESULTS

When reviewing the survival rates, the five patients lost to follow-up or who died from intercurrent disease (ID) were considered as having died from NHL; however, we have excluded these 5 cases from the study of relapse free survival. The time of survival was calculated from the beginning of radiotherapy. In the statistical study we used the chi squares test. Assessment of the results of CS I, II patients according to different groups showed, relative to presentation, that the 2-year survival rates were 59.4 and 68.8% for nodal and extranodal localizations, respectively. In the group of extranodal disease, patients with head and neck lesions had better survival rates than those with gastrointestinal localization (76.5 versus 50%) (p < 0.05). With regard to the treatment modalities, the rates were similar with 63.6% for radiotherapy alone and 67.8% for the combination of radiotherapy and chemotherapy. For all the localized NHL patients, the 2-year survival and disease free survival rates were 65% and 57.5%, respectively; the 5-year survival rate was 63%. NHL in adults had better prognosis than in children as the 2-year relapse free survival rates were 60 versus 40%. However, this difference was not 108 PATRICIO et al.

10 years

8 years

2-year relapse free survival • Favourable histology: 75.0 % »-•« Unfavourable histology: 58.9 %

12 24 36 4 8 months

FIG. 1. NHL (CS I, II). Duration of initiai complete remission according to histology.

statistically significant (p < 0.05). The rates of complete remission, two years after treatment, were 70.8% for CS I and 51.8% for CS II patients. Figure 1 shows the duration of initial complete remission according to histology. As can be seen, favourable histological types had better rates than unfavourable ones. Tolerance to treatment was good as the incidence of early and late complications was very low.

5. CONCLUSION

We are confident that, with the present enhancement of clinical staging accuracy, especially in upper torso presentation (using abdominal CT scan + either lymphangiogram or ultrasound, bone marrow smears and biopsy), in addition to the knowledge of other factors such as age, tumour bulk and strict histological types according to the Working Formulation Classification, we will be able to provide guidelines for selecting NHL patients with high possibility of cure using radio therapy alone. Furthermore, the recognition of more advanced cases where combination with chemotherapy regimes is necessary will prevent the risk of failure with radiation alone and will improve the long term results with these CS I and II NHL patients. IAEA-SM-290/8 109

ACKNOWLEDGEMENTS

The authors wish to express their gratitude to Drs J.A. Ricardo and C. Siqueira for their histological and statistical assistance.

REFERENCES

[1] Classification of Non-Hodgkin’s Lymphomas. Working Formulation, Cancer 49 (1982) 2112. [2] TU B IA N A , M., P O U ILLA R T, P., H A Y A T, M., et al., Resultats de la radiothérapie dans les stades I et II, Bull. Cancer 61 (1974) 93. [3] BUSH, R.S., GOSPODAROWICZ, M., STURGEON, J., ALISON, R., Radiation therapy of localized non-Hodgkin’s lymphoma, Cancer Treat. Rep. 61 (1977) 1129. [4] CARDE, P., BURGERS, J.M.V., et al., Combined radiotherapy-chemotherapy for early stages non-Hodgkin’s lymphoma: The 1975-1980 EORTC controlled lymphoma trial (RTO 0079), Radiat. Oncol. 2(1984) 301. [5] HORWICH, A., PECKHAM, M., Bad risk non-Hodgkin’s lymphomas, Semin. Hematol. 20 (1983) 35. [ 6] SUTCLIFFE, S.B., GOSPODAROWICZ, M.R., BUSH, R.S., et al., Role of radiation therapy in localized non-Hodgkin’s lymphoma, Radiat. Oncol. 4 (1985) 21 1. [7] HOPPE, R.T., The role of radiation therapy in the management of the non-Hodgkin’s lymphomas, Cancer 55 (1985) 2176.

IAEA-SM-290/33

NEOADJUVANT CHEMOTHERAPY AND HYPOFRACTIONATED IRRADIATION IN THE TREATMENT OF HEAD AND NECK CANCERS

J.-M. DENEUFBOURG Radiotherapy Department (Prof. Closon), University Hospital, Liège, Belgium

Abstract

NEOADJUVANT CHEMOTHERAPY AND HYPOFRACTIONATED IRRADIATION IN THE TREATMENT OF HEAD AND NECK CANCERS. A study has been initiated to assess the feasibility and efficacy of combining chemo therapy with irradiation in head atfd neck cancers. A total of 151 consecutive patients were enrolled, all recently diagnosed and previously untreated. There were 118 males and 33 females, ranging in age from 27 to 91 years. The predominant sites were: oropharynx (58), oral cavity (31), larynx (29) and hypopharynx (18). Most tumours were locally advanced (21 T b 40 T 2, 54 T 3, 34 T4) with frequent lymph node involvement (77 N 0, 23 N b 5 N 2, 44 N 3). Squamous cell carcinoma was present in 144 cases. The chemotherapy consisted of a low dose combination of bleomycin (10 mg), etoposide (100 mg) and cis-platinum (15 mg) given on days 1, 3, 5 and 15, 17, 19. A major response rate of 70% was obtained (11% complete response + 59% partial response). Primary tumours regressed in 86% of cases and nodes in 58%. The response rates differed according to tumour site, tonsil and floor of the mouth showing maximum sensitivity. Side effects were minimal: 85% nausea, 50% vomiting, 10% mild haematologic depression, 20% alopecia. Ancillary support consisted of anti-emetics but no hyperhydration programme or mannitol diuresis was used. This chemotherapy was carried out on an out-patient basis. A total of 122 cases received exclusive radiotherapy. The treatment was initiated with a mean interval of 14 days. A split-course modality was used, consisting of two treatment periods separated by a 15 day rest interval: each irradiation sequence comprised 6 fractions over 2 weeks. The tumour dose per fraction amounts to 4 Gy, the total dose being 48 Gy with a TD F of 103. Eighty-eight per cent of primary tumours and 54% of lymph nodes had completely regressed at the end of irradiation. Acute side effects remained acceptable and patient compliance amounted to 100%. Late complications were infrequent and no cumulative toxic effect was observed. Two year survival rates for 36 stage III and 64 stage IV patients are 57 and 50%, respectively. Results at 3 years indicate 48 and 31% survival. Preliminary comparison with historical controls only shows trends in favour of neoadjuvant chemotherapy. The combination of chemotherapy with a hypofractionated split irradiation also possesses economic advantages and social benefits: lower staff occupation, better output of treatment machines, reduction of patients’ travel and hospital stay.

I l l 112 DENEUFBOURG

1. INTRODUCTION

A study has been initiated to evaluate the effects of chemotherapy given before radical treatment in head and neck cancers with bad prognosis. The planned treatment consisted of exclusive radiotherapy or surgery followed by irradiation. We used a course of hypofractionated radiotherapy which has previously proved to be quite effective with minimal acute and late side effects. The feasibility and efficacy of the neoadjuvant chemotherapy were assessed and a possible benefit on the patients’ early survival was tested. Effectiveness and harmlessness constitute the requirements of neoadjuvant chemotherapy. The main purpose is to obtain a significant reduction of the tumour volume in order to increase radiosensitivity, especially in hypoxic areas, and to ensure a better and easier surgical resection. A lethal effect on occult métastasés already present at the time of diagnosis is also anticipated. Side effects should be minimal with good patient compliance. Pre-treatment with chemo therapy must not unduly delay the planned radical treatment nor add peculiar complications to irradiation or surgery.

2. TREATMENT POLICY

All stages of oropharynx, hypopharynx and nasopharynx cancers are treated with chemotherapy before radiotherapy. T3N3 and T4 larynx cases receive the same treatment modality. Most tumours o f the oral cavity, whatever the stage, and salivary gland localizations are treated with chemotherapy, surgery and post operative irradiation. According to a previous protocol, patients are then eligible for long term maintenance chemotherapy.

3. PATIENT MATERIAL

A total of 151 consecutive patients were enrolled without case selection, all recently diagnosed and previously untreated. There were 118 males and 33 females (sex ratio = 3.58) ranging in age from 27 to 91 years (mean age = 58). The Karnofsky index was higher than 70 in most cases. The minimum follow-up period amounted to 6 months. The predominant sites of the disease were the oropharynx (mainly tonsil), oral cavity (mainly floor of the mouth and tongue); larynx cancers were all supraglottic and other tumour localizations were involved to a lesser extent (Table I). IAEA-SM-290/33 113

TABLE I. TUMOUR SITES

Oropharynx Tonsil 46 5 8

Base of tongue 11

Soft palate 1

Oral cavity Floor of mouth 12 31

Tongue 12

Gingiva 4

Cheek mucosa 2

Hard palate 1

Larynx 29

Hypopharynx 18

Nasopharynx 4

Nasal cavities 3

Salivary glands 3

TABLE II. UICC TNM CLASSIFICATION AND AJC GROUPING

N0 N, n 2 N3 Total

T, 15 2 0 4 21 I 15 T 2 22 6 1 11 40 II 18 T 3 21 • 7 2 24 54 III 41 T„ 19 8 2 5 34 IV 73

Total 77 23 5 44

TABLE III. HISTOLOGY

Squamous cell carcinomas 144 Well differentiated 67 Moderately differentiated 26 Poorly differentiated 25 Undifferentiated 3 Not otherwise specified 23

Adenocarcinomas 6

Synoviosarcoma 1 114 DENEUFBOURG

Staging refers to the UICC TNM classification and AJC grouping. Most tumours were locally advanced: T3 and T4 with frequent lymph node involve ment. About 50% of cases belonged to AJC group IV. No metastatic patient was included (Table II). As regards histology there were essentially squamous cell carcinomas of various degrees of differentiation (Table III).

4. CHEMOTHERAPY PROTOCOL

A single course chemotherapy regimen combines bleomycin (10 mg), etoposide (100 mg) and cis-platinum (15 mg) in a 3 h IV1 perfusion of 1 L normal saline. This chemotherapy is given 6 times over a 3 week period, on days 1, 3, 5 and 15, 17, 19. Total doses are low — 60 mg for bleomycin, 600 mg for etoposide and 90 mg for cis-platinum. Ancillary support consists of anti-emetics given as required (domperidone, metoclopramide, alizapride). No hyperhydration programme or mannitol diuresis was used. This treatment is carried out on an out-patient basis.

TABLE IV. RESPONSE RATES ACCORDING TO SITE

'j'ä N a Global

Tonsil 98 70 78 Larynx 66 54 62

Hypopharynx 76 50 61 Floor of mouth 100 40 75 Tongue 92 25 75

Base of tongue 64 33 50

a Per cent of major response rates (CR + PR).

TABLE V. RESPONSE RATES ACCORDING TO VOLUME

T, T 2 T 3 T 4 N i N 3

% CR 26 14 19 3 5 0 % PR 53 69 72 69 53 58

% CR + PR 79 83 91 72 58 58

1 IV = intravenous. IAEA-SM-290/33 1 15

TABLE VI. RESPONSE RATES ACCORDING TO HISTOLOGY

'j'â N a

Well differentiated 82 56

Moderately differentiated 83 47

Poorly differentiated 92 59

a Per cent of major response rates (CR + PR).

5. CHEMOTHERAPY RESPONSE

The results are clinically assessed 8 days after the last injection. The response criteria are those in standard use: complete response (CR) in cases of tumour disappearance, partial response (PR) when the product of two diameters is reduced by more than 50%, minor response (MR) when the reduction is less than 50%, no response (NR) in cases of stable disease or even tumour progression. An overall response rate of 92% was obtained with a rate of major response amounting to 70% (11% CR + 59% PR). Primary lesions proved sensitive in 86% (14% CR + 72% PR) and nodes in 58% (6% CR + 52% PR). Multiple nodes in the same patient are considered as a unique target. The response rates differed according to site, primary tumours of the floor of the mouth and tonsil as well as node satellites of tonsillar cancers showing the maximum sensitivity (Table IV). The major response rates are little dependent on tumour volume but a complete response is more likely to occur when the lesion is of limited extent (Table V). Histological differentiation seems to exert no influence (Table VI).

6. SIDE EFFECTS

The side effects remained acceptable without any life threatening complications. Eighty-five per cent of the patients experienced nausea (mild to severe) and 50% had vomiting. Digestive intolerance was maximal on days 1 and 15. A mild haematological depression - leukopenia and thrombopenia - usually occurred. The treatment had to be protracted in 5 cases by doubling the time interval; the etoposide dose was reduced by half in 10 cases and bleomycin with drawn in another. No renal dysfunction was observed as measured by rise of serum levels of creatinine. No pulmonary side effects were encountered but 116 DENEUFBOURG

6 fractions / 2 weeks

□□□□□□□ □□□□□□□

2 weeks rest interval

FIG. 1. Irradiation modality.

2 transient bronchospasms occurred at the time of injection. Alopecia has been noted in 20% of cases, more frequently among women. Two patients experienced mild cutaneous effects from bleomycin.

7. RADICAL TREATMENT

The planned radical treatment consisted of exclusive radiotherapy (122 cases) or surgical resection followed by irradiation (18 cases). Eleven non-respects of planning included patients who refused treatment and cases where clinical deterio ration prevented further treatment. The radical treatment was always initiated without undue delay (mean interval of 15 days for radiotherapy and 18 days for surgery). Seventy-four patients were eligible for further chemotherapy with monthly courses of high dose methotrexate (3 g) and leucovorin rescue.

8. IRRADIATION MODALITY

For the last 12 years we have been routinely using an original irradiation modality in the radiotherapy of head and neck cancers. The course consists of 2 treatment periods separated by a rest interval of 2 weeks. Each sequence comprises 6 fractions distributed over 15 days. The tumour dose per fraction amounts to 4 Gy and the TDF2 is 103. The biological equivalent dose in con ventional fractionation would be 63 Gy in 30 fractions over 6 weeks (Fig. 1). The treatment parameters such as radiation quality, portals, loading, target volumes were those in current use. Combining hypofractionation with a split course allows to benefit by the effectiveness of large doses per fraction without undergoing their poor tolerance.

2 TDF = Time-dose fractionation. IAEA-SM-290/33 117

TABLE VII. EVALUATION OF TOLERANCE

Compliance Late complications

Respect of prescribed dose Laryngeal oedema

Extension of rest interval Muscle fibrosis

Cutaneous sequelae Acute side effects Mucosa/bone/cartilage necrosis Mucosal congestion Asialia and dysgeusia Transient oedema

Degree of skin reaction Delay in wound healing

TABLE VIII. SOCIAL AND ECONOMIC ADVANTAGES

1 2 X 4 Gy 30 X 2.1 Gy

Patient’s travel 12 times 30 times -6 0 %

Hospital stay 4 weeks 6 weeks -3 3 %

Staff occupation 3 days/week 5 days/week -4 0 %

Therapy courses 390/year 312/year + 25%

TABLE IX. TUMOUR CONTROL AFTER IRRADIATION

All Good Poor cases responders responders (%) (%) (%)

Primary tumour Complete regression 88 95 67

Partial regression 12 . 5 33 Stable disease 0 0 0

Nodes

Complete regression 54 60 33

Partial regression 40 40 54 Stable disease 6 0 13 118 DENEUFBOURG

% STAGE III STAGE IV 1 T| N 2 2 T2N2 9 t3 n 2 2 T, N3 5 T2 N 3 7 I3N3 19 T4 N 0 8 T t N, ‘‘h n2 ± 4 N3 64

1 2 3 4 years

FIG.2. Actuarial survival.

As part of a study on quality of life, tolerance is at present under investigation (Table VII). Compliance amounted to 100%; acute and late side effects remained within the normal range. A preliminary comparison seems to indicate that neoadjuvant chemotherapy at low doses does not change local tolerance and has no cumulative toxic effect on normal tissues. Owing to the low number of fractions, this modality of irradiation presents several social and economic advantages: reduction of patients’ travel and hospital stay; the work load for the staff is reduced and the annual output o f the treat ment machines increases (Table VIII).

9. TUMOUR CONTROL

Immediate assessment indicates that high rates of tumour control were achieved after exclusive radiotherapy. At the very end of irradiation 88% of primary tumours had totally regressed and the complete regression rate of lymph nodes amounted to 54% (multiple nodes in the same patient being considered as a unique target) (Table IX). Good responders to chemotherapy presented more complete regressions of their primary tumour and satellite nodes.

10. SURVIVAL

The possible long term benefit of a neoadjuvant chemotherapy is still putative in the treatment of head and neck cancers. Actuarial survival curves have been IAEA-SM-290/33 119

TABLE X. POSSIBLE LONG TERM BENEFIT

Study Controls

Survival No. at risk Survival No. at risk (%) (%)

Larynx stage IV 1 year 73 22 63 36 2 years 47 12 46 22 3 years 33 4 32 16

Tonsil stage III

1 year 78 9 69 13 2 years 78 6 61 9 3 years 61 5 46 8

Tonsil stage IV

1 year 67 22 60 20 2 years 54 10 50 11 3 years 27 4 30 10

established for 36 stage III and 64 stage IV patients. Correction was made for duly documented deaths from intercurrent disease in cancer controlled cases, any uncertain fatal issue being considered as a therapeutic failure (Fig. 2). Results at 2 years seem promising with rates of 57 and 50%, respectively. Owing to the low number of patients at risk, values obtained at 3 years should be taken with caution. Stage III cases then enter a plateau phase with 48% survivors, whereas the survival of stage IV still decreases to 31%. A comparison of survival rates with historical controls treated in the same way but without neoadjuvant chemotherapy is being made. Preliminary results do not allow any clear cut conclusion up to now. Trends exist for a benefit in advanced carcinomas of the larynx and tonsil. However, more cases and time are needed (Table X).

11. CONCLUSIONS

(1) Neoadjuvant chemotherapy with bleomycin, etoposide and cis-platinum at low doses has shown significant anti-tumour activity and minimal toxicity. This chemotherapy is suitable for out-patient treatment. Irradiation and surgery are possible without undue delay and their tolerance is unchanged. A maintenance chemotherapy is not precluded. 120 DENEUFBOURG

(2) High rates of immediate tumour control are observed after hypofractionated split-course radiotherapy. This irradiation modality is well tolerated and, besides, possesses social and economic advantages. (3) More cases and time are needed to ascertain a possible beneficial effect on the long term prognosis. IAEA-SM-290/67

TECHNICAL PROBLEMS ASSOCIATED WITH RADIOTHERAPY FOR NASOPHARYNGEAL CARCINOMA Experience at the Department of Radiotherapy, Kenyatta National Hospital, Nairobi, Kenya

J.N. ONYANGO, A.M. BABU, N. TOLE Department of Radiotherapy, Kenyatta National Hospital, Nairobi, Kenya

Abstract

TECHNICAL PROBLEMS ASSOCIATED WITH RADIOTHERAPY FOR NASOPHARYNGEAL CARCINOMA. EXPERIENCE AT THE DEPARTMENT OF RADIOTHERAPY, KENYATTA NATIONAL HOSPITAL, NAIROBI, KENYA. Nasopharyngeal carcinoma, the fifth commonest in Kenya, is discussed. The clinical presentation, duration of symptoms, histological differentiation and the treatment method are outlined. The paper discusses the problems experienced in the Department of Radiotherapy at KNH owing to insufficient facilities in the country and late presentation of the patients to the oncologists. One year local primary tumour and neck node métastasés control is presented with suggestions for improving the overall results in the future.

1. INTRODUCTION

Nasopharyngeal carcinoma is the fifth commonest malignant tumour in Kenya. In men it is second to skin cancer. The male to female ratio is about 3:1. The peak age at diagnosis is 40—49 years in both males and females. It is not uncommon in children aged 7—12 years. There is a strong statistical association between nasopharyngeal carcinoma and infection with Epstein-Barr virus (EBV). The commonest histological diagnosis is that of poorly differentiated squamous cell carcinoma with heavy lymphocytic infiltration (86%), previously referred to as lymphoepithelioma. The vast majority of our patients present with locally advanced disease. The commonest clinical presentation is cervical lymphadenopathy. More than half of the patients present with neck nodes 5 cm in diameter or larger. About 10—15% present with clinical and radiological evidence of erosion of the base of the skull and cranial nerve involvement. The reasons for late presentation at the radiotherapy department are delays in seeking medical help and delays in referral after presenting at the peripheral hospitals.

121 122 ONYANGO et al.

The technical problems are, therefore, those encountered in trying to deliver a carcinocidal radiation dose to advanced nasopharyngeal carcinoma with large neck nodes without exceeding the tolerance dose to the local critical organs. Although the poorly differentiated carcinomas are more radioresponsive than the well differentiated ones the tumoricidal doses are basically the same.

2. RADIOTHERAPY AT THE KENYATTA NATIONAL HOSPITAL

Radiotherapy has become a very important speciality in medicine in other parts of the world since the discovery of ionizing radiation over three quarters of a century ago. In tropical Africa as in most of the developing countries, however, radiotherapy has taken a long time to be adopted as a medical discipline. The reason for this is to be found in the costs of both the equipment required and man power training. The simplest ideal radiotherapy department in a developing country would consist of one or more cobalt-60 machines, a simulator and intra cavitary gynaecological treatment equiment. The Department of Radiotherapy at the Kenyatta National Hospital was set up in 1968 jointly by the Kenyan Ministry of Health and the Swedish Government through the Karolinska Hospital. It was equipped with a 60Co teletherapy unit (activity 3 kCi). A bigger 60Co teletherapy unit (4 kCi) and a treatment simulator were installed in 1984. The workload is very large for such a small department — about 800 patients per year. There are frequent mechanical breakdowns which sometimes take long to repair as spares have to be purchased from abroad. There are no mouldroom facilities. Treatment shells and shielding blocks cannot be made to suit every patient. The absence of such facilities places limitations on precise treatment planning and reproducibility of the set-up at each treatment session. The large workload also means that less time can be spent in setting up each patient for treatment, with a consequent loss in accuracy.

3. INITIAL TREATMENT PORTAL SET-UP

After staging work-up, which includes examination under anaesthesia by ENT surgeons and X-rays, the initial treatment is planned for both the primary tumour and neck nodes. As a policy we treat the neck nodes and the supraclavicular nodes even if they are not clinically involved. This is done because the vast majority of our patients present with advanced disease. Radiotherapy treatment planning has to take into account direct tumour spread and regional lymphatic métastasés. Direct tumour spread occurs in several directions. Superiorly the tumour spreads through the foramen ovale; erosion of the base of the skull may or may not be present. Posterior spread is limited by IAEA-SM-290/67 123 the prevertebral fascia and the first and second cervical vertebrae. Anteriorly and anterosuperiorly the tumour spreads to involve the nasal cavity and the orbit. The symptoms may be nasal obstruction, epistaxis and proptosis. Lateral spread involves the eustachian tube and presents with tinnitus, deafness and serous otitis media. Inferior spread involves the soft palate and pharyngeal wall. This is, however, rare. The first nodes to be involved are the lateral retropharyngeal lymph nodes, the upper lateral jugular nodes and the jugulodigastric nodes. The mid and lower cervical nodes are involved later. Our routine treatment for the primary tumour is through parallel opposed lateral facial portals. The lateral retropharyngeal nodes are included in the treat ment volume for the primary tumour. The neck nodes are treated through parallel opposed AP and PA portals with shielding of the larynx and spinal cord. The local critical tissues are the brain stem, the cervical spinal cord and the eyes and the tolerance doses of these critical tissues should be taken into account when planning the treatment. In addition, high dose irradiation of the temporo mandibular joints leads to trismus. Pituitary insufficiency can result from high dose irradiation to the pituitary gland. The initial target volume for the primary tumour includes the whole pituitary fossa, and varying volumes o f brain stem and cervical spinal cord, depending on the size and distribution of the enlarged neck glands. The eyes are not irradiated. The anterior margin of the lateral facial field is set at least 1.5 cm posterior to the outer canthus. We have not been able to take routine verification films of these treatment volumes. We plan to do this as soon as adequate facilities are available. When there is evidence that the disease extends anteriorly to the nasal cavity we add an anterior facial field shaped like an inverted T with shielding of both eyes. If one orbit is involved it is irradiated to tumoricidal doses to achieve local tumour control. In such cases the lacrimal gland is shielded if possible to avoid a painful post-irradiation dry eye reaction. The tolerance dose of the brain stem and cervical spinal cord is taken as 4000 cGy in 4 weeks. After a dose of 4000 cGy in 4 weeks to 4500 cGy in 4 \ weeks the treatment volume is reduced to exclude the brain stem and spinal cord. Where there is no evidence of erosion of the base of the skull or cranial nerve involvement the pituitary fossa is also excluded from the new treatment volume. The reduced volume is irradiated to a dose of 60 to 65 Gy in 6 to 6j weeks. The neck nodes are irradiated to a target absorbed dose of 50 Gy in 5 weeks. A booster dose of 10—15 Gy is given to enlarge the remaining nodes through direct single portals. Before the installation of our simulator in 1984 all these plans were made using external landmarks as a rough guide. There has been an improvement in the adequacy and uniformity of dose delivery to the target volumes but a lot remains to be achieved with improved facilities, especially mouldroom facilities. 124 ONYANGO et al.

4. LOCAL TUMOUR CONTROL

The neck nodes are accessible to measurement. Our initial total regression rate of 32% for the neck nodes could be improved. Local control of the primary tumour is higher at 51% but this too could be improved.

5. SUGGESTIONS FOR IMPROVING LOCAL TUMOUR CONTROL

— Mouldroom facilities for making treatment shells, metal blocks and accessories — Early diagnosis of tumour and institution of treatment — More accurate target volume definition so that the high dose irradiation can be given to the smallest tissue volume necessary — Electron beam therapy equipment — Combining irradiation with chemotherapy — Assistance with facilities and places for manpower training.

6. CONCLUSION

Nasopharyngeal carcinoma patients present in this centre with locally advanced disease. Our local tumour control rates are satisfactory but could be improved with better facilities.

BIBLIOGRAPHY

FLETCHER, G., Text-book of Radiotherapy, Lea and Febiger, Eds, Philadelphia.

GUNVEN, P., KLEIN, G., HENLE, G., HEN LE, W., CLIFFORD, P, EBV in Burkitt’s lymphoma and nasopharyngeal carcinoma, Nature (London) 228 (1970).

INTERNATIONAL ATOMIC ENERGY AGENCY, Cobalt-60 Teletherapy - A Compendium of International Practice, IAEA, Vienna (1984) 695 pp.

LARSSON, L.-G., SEELIG, I., Malignant nasopharyngeal tumours. Result of radiation therapy, Acta Radiol. Ther. Phys., Biol. 15 (1976).

MITELMAN F., KLEIN, G., ANDERSON-ANURET, M., FORSBY, N., JOHANSSON, B., 14q+ marker chromosome in an EBV-genome-negative lymph-node without signs of malignancy in a patient with EBV-genome-positive nasopharyngeal carcinoma, Int. J. Cancer 23 (1979) 32. IAEA-SM-290/39

INTRACAVITARY IRRADIATION FOR NASOPHARYNGEAL CARCINOMA

R. UZEL, S. КОСА, S. OKKAN, A. ERTEM, S. TURKAN Department of Radiotherapy, Cerrahpa^a Faculty of Medicine, University of Istanbul, Istanbul, Turkey

Abstract

INTRACAVITARY IRRADIATION FOR NASOPHARYNGEAL CARCINOMA. The probability of local recurrence of nasopharyngeal carcinoma following radiation therapy is highest at the base of the skull. The findings from a historical group treated in the Cerrahpasa Medical Faculty confirms this experience. These failures may be related to under dosage in the superior region of the nasopharynx and the base of the skull owing to factors inherent in external irradiation fields and the anatomic peculiarities of the region. For this reason a compensating booster dose of 5—10 Gy,through small fields to the base of the skull and nasopharynx is usually advised. This also increases the dose to the adjacent portion of the brain and the pituitary. The dose in this region may also be increased by internal irradiation, which may consequently enhance the local control rate. Therefore, the possibility of replacing external irradiation booster fields with intracavitary radiotherapy is being contemplated. An experimental model consisting of a bolused skull is used to determine the dose distribution to the critical areas around the nasopharyngeal cavity. After placing the applicators of a high dose rate remote afterloader in the nasopharyngeal area, the dose con tribution to the anatomic sites chosen as critical points was measured by LiF dosimetry. It was observed that it is possible to deliver an effective booster dose to the nasopharynx and local tumour extension pathways without increasing the dose to critical organs which might produce complications. In 1983 clinical applications started in the Cerrahpasa Faculty of Medicine. Patients with nasopharyngeal carcinoma have since been treated with external and intracavitary irradiation. The modification of the radiotherapy programme is described. Intracavitary irradiation is given with a high dose rate remote afterloader to obtain 24 Gy at the reference isodose in 3 fractions at one week intervals. So far 18 cases were treated with this method and 7 of them were followed up for more than 2 years. The local control rates were compared with a historical series of 66 cases. It was found to be 71%, after 2 years, in the intracavitary irradiation group, compared with 47% in the historical group. No significant differences in complications between the two groups were observed.

1. INTRODUCTION

Nasopharyngeal tumours, because of their inaccessible localization, may remain unnoticed for long periods of time. They are usually quite extensive when discovered. Rich lymphatics drain the nasopharynx into the neck nodes. In many instances the presence of neck nodes directs the attention of the.examining

125 126 UZEL et al. physician to the nasopharynx. A good proportion of these tumours are poorly differentiated. Distant métastasés are not infrequent. Because of the generally advanced stage of the disease, overall 5-year survival rates used to be in the vicinity of 10%, although these tumours are considered to be radiosensitive and theoretically curable [1,2]. Recent publications, however, report higher survival rates of 44—59% [3—5], but improvement in the local control rate by more effective radiotherapy and development of adjuvant chemotherapy to decrease the rate o f distant métastasés is desirable. Nasopharyngeal cancer in Turkey is not as frequent as in the Eastern Asiatic countries, but the incidence is higher than in Western Europe. At the Cerrahpasa Medical Faculty of the University of Istanbul nasopharyngeal tumours constituted 1% of all malignant tumours and 15% of all head and neck malignancies, based upon the patient material available to us during the years 1978-1985. We have examined the outcome of these cases and decided on a modified radiotherapy scheme with the hope of improving the local control rate. Findings from this historical group of patients and early results obtained from patients treated with the modified radiotherapy scheme will be discussed.

2. MATERIAL AND METHOD

During the 8 years from 1978 to the end of 1985, 152 patients with nasopharyngeal tumours were seen. The youngest patient was 9, the oldest 78 years old, with a median age of 42. The male/female ratio was 2.4/1. Of these patients 18 were treated with external radiotherapy and intracavitary boost. In 19 of the remaining 134 patients radiotherapy either was not found indicated or the patients failed to complete the required course. There was no follow-up in 16 cases and 23 patients were not at risk for one year. All these cases were excluded from the evaluation, which left us with 76 patients. The histological distribution, stage, and T and N classifications of these patients are shown in Tables I-IV . The radiotherapy given to the historical group was fairly standard. Parallel opposing fields were used to irradiate the nasopharynx and upper neck. The superior border of this field passed from the upper surface of the celia turcica. The lower neck and both supraclavicular fossae were treated with an anterior field. Treatments were given with telecobalt. The lower neck field received 50 Gy in 5 weeks, the upper neck 60 Gy in 6 weeks. Nasopharynx and, if present, persisting neck nodes were given a 10 Gy boost in one week. If there was anterior extension of the tumour, the nasopharynx was treated with 3 fields by the addition of an anterior field. In the 18 patients who were treated with intracavitary boosts, the external radiotherapy to the nasopharynx did not exceed 60 Gy. Ten days after the completion of external radiotherapy 2 plastic tubes loaded with dummy sources IAEA-SM-290/39 127

TABLE I. HISTOLOGICAL DISTRIBUTION OF NASOPHARYNGEAL TUMOURS (1978-1985)

No. of patients %

Transitional 36 47 Anaplastic3 18 24 Epidermoid 13 17

Epithelial tumoursb 4 5

Lymphomas 5 7

Total 76 100

a So-called lympoepitheliomas are classified as anaplastic tumours.

b No histological differentiation.

TABLE II. STAGE DISTRIBUTION OF NASOPHARYNGEAL TUMOURS (1978-1985)

Stage I II Ill IV

Epithelial tumours 1(1.4) 5(7) 14(19.8) 51(71.8)

Lymphomas 1 - 1 3

Per cent figures are given in brackets.

TABLE III. T DISTRIBUTION OF NASOPHARYNGEAL TUMOURS (1978-1985)

T x To T i T 2 T 3 T 4

Epithelial tumours 3(4.2) 2(2.8) 7(9.9) 27(38.0) 20(28.2) 12(16.9) Lymphomas 1 1 2 1

Per cent figures are given in brackets. 128 UZEL et al.

TABLE IV. NODAL STATUS IN 76 PATIENTS

No N, N j N 3

Epithelial tumours 16(22.5) 12(16.9) 18(25.4) 25(35.2)

Lymphomas 2 - - 3

Per cent figures are given in brackets.

were placed in the nasopharynx through each nostril. After X-ray films had been made and isodose curves drawn by the computer, the tubes were connected to the high dose rate remote loading Curietron cobalt unit. Sources 21 mm in length were driven into each tube. The dose rate was 39 cGy/min at 2 cm distance from the axis of the tube laterally at mid-point. Eighteen patients who were treated with a modified programme by intra cavitary boost and external irradiation did not differ considerably from the historical group in age, sex, stage and histological distribution.

3. FINDINGS

Patients belonging to the historical group were examined according to the site of the primary tumour and local extension. Lymphoma cases made up 6.6% of the series with 5 patients. It was thought that analysis of this small group might lead to erroneous conclusions and so it was excluded from the investigation. From 71 epithelial tumour patients 5 had also to be excluded with Tx and T0 tumours since they were not suitable for primary site analysis. Of the 66 remaining cases, 10.6% were stage I, 40.9% stage II, 30.3% stage III and 18.2% stage IV. Analysing the site of involvement it was observed that nasopharyngeal structures and boundaries were not usually singularly involved but intermixture was present in the majority of cases. The incidence of involvement of nasopharyngeal locations is shown in Table V. As can be seen in this table, the roof and posterior wall were involved in 52 instances (79%), followed by 35 instances (53%) of anterior wall or choanae and 33 instances (50%) of lateral wall. These findings were based upon posterior rhinoscopy. When X-ray findings from the base of the skull views and, in the later patients, CT examinations were added to the clinical observations, the extension of the tumour to the neighbouring structures was depicted more accurately. In 10 instances the tumour extended to the structures of the oropharynx. This was followed by cranial nerve involvement in 8, and destruction IAEA-SM-290/39 129

TABLE V. SITE OF NASOPHARYNGEAL INVOLVEMENT IN 66 EVALUABLE EPITHELIAL TUMOURS3

Site No. of patients

Entire nasopharynx 12 Hemi-nasopharynx (roof, posterior, lateral walls and choanae) 15

Roof, posterior, lateral walls 15

Roof, posterior wall and choanae 4

Roof and posterior wall 6

Lateral wall and choanae 2

Lateral wall 4

Choanae 2 No visible tumour but positive blind biopsy 6 a 3 T x and 2 T 0 tumours are excluded.

TABLE VI. COMBINATION OF FAILURES ACCORDING TO SITE

Failure site No. of % patients

Nasopharynx3 only 16 37.2

Nasopharynx and neck nodes 9 20.9

Nasopharynx and distant métastasés 2 4.6

Nasopharynx, neck nodes and distant métastasés 2 4.6

Neck nodes only 7 16.4

Neck nodes with distant métastasés 2 4.6

Distant métastasés only 5 11.7

a Failures in neighbouring structures are included. 130 UZEL et al.

TABLE VII. SITES OF FAILURE IN THE NEIGHBOURING STRUCTURES OF THE NASOPHARYNX

No. of % patients

Base of skull 17 73.9

Etmoid sinus 3 13.0 Nasal cavity 2 8.7

Orbit 1 4.3

Oropharynx 2 8.7

of the base of the skull in 6 instances. The nasal cavity was involved in 4 instances and the orbit in 1 instance. It can be summarized that, at first presentation, in 4 out of 5 patients the tumour was seen to involve the roof and the posterior wall. About half of the tumour extensions to the neighbouring structures go towards the base o f the skull, either to involve the cranial nerves, or to produce bone destruction. With the availability of CT scans it is possible that involvement of the base of the skull will be demonstrated more frequently. Failure after radiotherapy was also investigated in this group of patients. They were followed up from 1 to 8 years. Failure sites are summarized in Table VI. It was seen that nasopharynx and adjacent structures were involved in 67.3% of the failures, while neck nodes could be found responsible in only 46.5%. Distant métastasés contributed to a lesser degree with 25.5%. The base of the skull was involved in 3/4 of the failures in sites neighbouring the nasopharynx (Table VII). Failure in the neck was 6.3% in N0 cases, while it was 18.2, 35.3 and 50% in N j, N2 and N3 cases, respectively. Failure at the site of the primary tumour also increased with T number from 14.3 in Ti to 33.3, 40 and 75% in T2, T3 and T4) respectively. Failures usually occur at the original site of involvement. In only a small proportion of T j, T2 and N0, Nj cases failure occurred at another site. Late cases almost always recurred at the original site of involvement. The overall actuarial 5 year relapse free survival rate was 30%. No difference could be detected between Tb T2 and T3, T4 cases. This rate was slightly higher with 36% in transitional cell cancers than in epidermoid cancer with 25%. This difference, however, was not statistically significant. In T i-T 2, Nq- ^ cases this rate was found to be 59%, while it was 27% in T3-T 4, N2—N3 cases. From the findings described above it can be summarized that: (1) it is rare to find and treat early stage nasopharyngeal cancers; (2) the majority of the IAEA-SM-290/39 131 nasopharyngeal tumours are located at the roof and posterior wall at first presentation (79%); (3) tumours extend equally to the base of the skull and the oropharynx and other adjacent structures; (4) the roof of the nasopharynx and the base of the skull are the major sites of treatment failure (67.3%), the base of the skull being responsible in 3/4 of the cases; (6) recurrences usually occur at or near the site of original involvement, especially in late cases; (7) when involve ment of the base of the skull or cranial nerves occurs, the possibility of a cure becomes remote. Failure at the base of the skull is observed by many radiotherapy centres [6]. This is attributed to underdosage in this area, resulting from lower doses at the edge of the field, dense bone throughout the base of the skull resulting in differential absorption, which may be at a level of 3.5% per centimetre of the bone for cobalt, the contour taken through the upper lip and the external auditory canal which is 1 cm thinner than the contour passing through the zigomatic arch, which may contribute another 5% decrease in the dose. This difference is claimed to result in 10—20% less dose at the base of the skull [1,5, 6]. To overcome the dose deficiency, a booster dose of 5—10 Gy through small lateral fields is added to the roof of the nasopharynx in many institutions. For this purpose it was decided to investigate the possibility of using intra cavitary irradiation to increase the dose at the roof of the nasopharynx and the base of the skull in patients with nasopharyngeal cancer.

4. INTRACAVITARY IRRADIATION FOR NASOPHARYNGEAL CANCERS

Intracavitary irradiation has been used to eliminate or prevent nasopharyngeal recurrences in the form of interstitial implantation with 1251 seeds or by intro ducing tubes containing 192Ir [7—10]. To devise a simple, reproducible method, it was decided to use 2 tubes loaded with 60Co. The dose distribution from these tubes placed in the nasopharynx was worked out by drawing isodose curves with a computer. TLD measurements were also performed. For this purpose a sectioned skull was used. Intracranial structures were simulated by filling the space over the base with paraffin. LiF rods were placed in the critical points. The points chosen were foramen lacerum, spinal cord, orbital fissure, pituitary gland, and lens (Fig. 1). Ten measurements were made for each point. The mean values of the doses measured coincided closely with the calculated doses. Table VIII summarizes the accumulated doses at the critical points when a dose of 24 Gy was given at a distance of 2 cm from the sources. When the dose from the intracavitary irradiation was added to the external radiotherapy dose, it was found that the dose to the spinal cord was raised from 3000 to 3706 cGy, to the lens from 300 to 602 cGy and, when anterior fields are added, from 600 to 902 cGy, and to the pituitary gland from 1200 to 1834 cGy. 132 UZEL et al.

FIG. 1. Isodose distribution for intracavitary irradiation related to the base o f the skull constructed from TLD dosimetry. The dark line near the numbers 1 -6 shows the sites of lithium fluoride rods. J - F. lacerum, 2 - spinal cord, 3 - orbital fissure, 4 - pituitary gland, 5 - arbitrary reference point for computer, 6— lens.

TABLE VIII. CORRELATION OF PHANTOM MEASUREMENTS (TLD) AND CALCULATED DOSES AT CRITICAL POINTS

Reference 5 Gy at 2 cm 24 Gy at 2 cm points

TLD Calculated (cGy) (cGy) (cGy)

Foramen lacerum 282 ± 17 312 1354

Orbital fissure 230 ± 2 1 253 1104

Spinal cord 147 ± 12 142 706

Pituitary gland 132 ± 8 136 634

Lens 63 ± 4 65 302 IAEA-SM-290/39 133

TABLE IX. LOCOREGIONAL CONTROL RATE ACCORDING TO TREATMENT IN PATIENTS WITH NASOPHARYNGEAL CARCINOMA

Stage External RT ( 66) External intracavitary RT (7)

I 1/ 1( 100) -

II 2/5(40) 1/2(50)

III 7/13(54) -

IV 21/47(45) 4 /5(80)

Total 31/66(47) 5/7(71)

Per cent figures are given in brackets.

These doses are not at a level to produce myelitis or pituitary insufficiency. The lens dose from external irradiation is over the threshold dose for cataract formation, therefore the addition of 302 cGy will not affect the outcome.

5. RESULTS AND CONCLUSION

Based on these calculations and measurements it was decided that intra cavitary irradiation in the above mentioned doses should not contribute adversely to the ill effects of external irradiation and the radiation programme in patients with epithelial tumours of the nasopharynx was modified as described above to include an intracavitary boost. In this programme 18 patients have been treated so far. Of these, 7 patients were followed up for more than 2 years. Two patients were in stage II and had T2 tumours, 5 were in Stage IV, 2 o f whom had T2 tumours and the remaining 3 T3 tumours. Early results obtained from the 7 cases were examined and compared with the historical group. Local control rates in comparison with the historical group are shown in Table IX. Although the number of patients is too small to draw any conclusions, these preliminary local control rates are encouraging. No ill effects of treatment have been observed in these patients over the first two years. The number of patients and the time elapsed since the treatment are not sufficient to perform survival analyses. The treatment of patients using this method will be continued to accumulate more data. 134 UZEL et al.

REFERENCES

[1] FLETCHER, H., Textbook of Radiotherapy (3rd Edn), Lea Febiger, Philadelphia (1980) 364. [2] TOKARS, R.P., GRIEM, M.L., Int. J. Radiat. Oncol., Biol. Phys. 5 10 (1979) 1741. [3] HOPPE, R.T., GOFFINET, D.R., BAGSHAW, M.A., Cancer 37 (1976) 2605. [4] MES1C, J.B., FLETCHER, G.H., GOEPFERT, H., Int. J. Radiat. Oncol., Biol. Phys. 7 4 (1981) 447. [5] M IL L IO N , R.R., CASSISI, N.J., Management of Head and Neck Cancer, J.B. Lippincott Co., Philadelphia ( 1984) 445. [6] BEDWINEK, J.M., PEREZ, C.A., KEYS, D.J., Cancer 45 (1980) 2725. [7] BUSCH, M., ALBERTI, W., Strahlentherapie 160(1984) 77. [ 8] HUANG, S.C., Int. J. Radiat. Oncol., Biol. Phys. 6 4 (1980) 401. [9] V1KRAM, B., HILAR1S, B., Int. J. Radiat. Oncol., Biol. Phys. 10 1 (1984) 153. [10] WANG, C.C., BUSSE, J., GITTERMAN, M„ Radiology 115 (1975) 737. IAEA-SM-290/56

CLINICAL APPROACHES OF CONVENTIONAL RADIOTHERAPY - EMPHASIS ON AFRICA

J. LUANDE Tanzania Tumor Centre, Dar es Salaam, Tanzania

Abstract

CLINICAL APPROACHES OF CONVENTIONAL RADIOTHERAPY - EMPHASIS ON AFRICA. Cancer management is a rapidly evolving science. Whereas there has been relatively ample time for effecting this evolution in the industrialized countries, the situation in Africa is quite different. Here circumstances demand that this urgently needed service be developed in a crash programme type of approach. Such an approach calls for a concerted strategy modification which involves many medical disciplines. One such discipline - radiotherapy - has a unique status in Africa and appears to command priority in development. Radiotherapy in Africa calls for special attention to the selection of equipment and radiation sources and training of personnel, among other things. As facilities exist in only a few African countries, many countries will be developing such facilities from scratch. It will be some time before each country has a single fully functional unit and even longer before more than one unit exists in every country. A minimum complement of radiotherapy equipment is suggested. Isotopes seem preferable as radiation sources. Two levels of personnel trained in radiotherapy, medical physics and radiotherapy technology may form a more practical option, at least in the initial stage of development. Simultaneous encouragement of research into epidemiology and relevant clinical cancer types, and a strong public health education programme will hasten a better understanding of existing cancer problems. A frontal attack deploying all these facets is best modelled on a comprehensive cancer centre type of programme, and regionalized personnel training is suggested.

Treatment o f cancer has remained an ever changing science since the disease was first recognized. Chronologically, surgery led the field for many years, but major strides in the art and science of cancer surgery, and in general surgery for that matter, only took place after the seconed half of the nineteenth century. Asepsis, antisepsis, haemostasis, blood transfusion, antibiotics, pre-operative and post-operative homeostatic medication all widened the scope o f surgery and reduced operative mortality drastically. Radiotherapy joined surgery in the management o f cancer early in this century. Again, as with surgery, it was the change from orthovoltage to megavoltage teletherapy in the early 1950s which very materially widened the scope of this therapy modality. Chemotherapy joined surgery and radiotherapy in the 1940s, but here also the effective use of drugs in cancer treatment was dramatically enhanced only after the concept of combination cytotoxic chemotherapy was developed in the early 1960s. A balanced combination of these three modalities — surgery, radiotherapy and chemotherapy — is today the backbone of cancer management.

135 136 LUANDE

This orderly evolution of the science and art of cancer treatment occurred simultaneously with many other aspects of our knowledge of the disease. Clinicians improved their knowledge of early cancer recognition, as did the general public to a certain degree. In both quarters this enhanced knowledge lowered the threshold o f disease suspicion, making earlier diagnosis possible. Improved investigational tools in the field of diagnostic imaging and other laboratory tests facilitated disease detection and the definition of its extent. Improvement in the pathological categorization of the various cancers made it possible to select the appropriate therapies more intelligently. Other areas of improvement included facilities for patient monitoring and support while treatment was being taken. The total effect of all these various areas of improvement has been the improved cancer cure rate reported in the cancer literature today. General survival figures of 40 to 50% are already being quoted in certain areas of the globe. The orderly evolution of the various parameters of cancer knowledge as depicted above did not take place uniformly throughout the world. In fact, most of this development was only true for a specific segment of the global population, the developed countries. In the developing countries the situation has been much different. Whereas in the developed countries cancer knowledge evolved sequentially, the developing countries have experienced what could more fittingly be termed a revolution in cancer practice. Much of what is currently practised with respect to cancer in this sector was imported ready made. In such a situation many disadvantages may arise. Typically, the speed of importation of ready made knowledge and its practice is often variable and may lack sequential logic. Variability and lack of sequential logic bring about certain inherent handicaps, some of which will be pointed out below. One of the most troublesome handicaps inherent in imported practice is the attitudinal conflicts often experienced in the recipient. People have a reason for and a sound background to whatever they practise culturally and this is tru'e of cancer also. The popular belief that cancer is not an important health problem in developing countries is still held by many people in these areas, spanning a wide spectrum including the general public, legislators, policy makers and the medical profession. Another handicap is the desire to import cancer practice neat and original. This is another inherent human trait. If one has to acquire a new item, whatever it may be, one prefers the best available and unmodified. A tremendous amount of restraint is often required to rearrange considerations and select items for their practicality and suitability. Considerations of practicality and suitability are dependent on a multiplicity of factors, all of which are obvious to us all. Imported equipment, for instance, has to have appropriate users and maintainers. Again the crash programme approach to cancer facility development imposes a further handicap owing to its speed. While it may take decades to perfect a piece of equipment following its invention, the importation of the same item may simply take the duration of a cargo airline flight to reach a destination - measurable IAEA-SM-290/56 137 in hours in many cases. Such a speed of acquisition may work against logical planning, adequate training of personnel, the development of support facilities and the preparation of adequate housing and maintenance of the equipment. These and many other handicaps will always pose a great challenge to the developing countries when they want to revolutionize their cancer science and practice. The 40 to 50% cancer cure rates quoted in some industrialized countries may take a long time to reach in developing countries. How fast they achieve these cure rates and what strategies prove most appropriate are very profound issues and do not form the subject of this paper. I will now, in the light of the points mentioned above, leave the general topic of cancer practice worldwide and concentrate on radiotherapy in Africa. Radiotherapy has a unique position with regard to cancer management in Africa. Of the three conventional modalities of cancer treatment mentioned above the cheapest is radiotherapy. In a recent WHO report a course of radio therapy in the developing countries was quoted as costing US $77 as against a course of chemotherapy costing over US $700, cost being a significant con sideration in Africa. The relevance of this factor ought to be reflected in our practice. Thus for cancer sites where the therapy results with radiotherapy are comparable to those with chemotherapy the former modality should get preference. Such cancer sites include Kaposi’s sarcoma, cervical carcinoma, laryngeal carcinoma and other head and neck tumours, certain types of non- Hodgkin’s lymphoma and even early Hodgkin’s lymphoma. Another factor to be considered in African radiotherapy practice concerns therapy intent. More patients present in the palliative stage of cancer in Africa than in the more affluent industrialized societies. For example, at our institute 90% of all head and neck cancers are advanced beyond cure. Palliation by means of radiotherapy is clearly cheaper than the use of the other two modalities. Thus, proportionately, more patients qualify for palliative radiotherapy on presentation in Africa. Such patients will require less elaborate treatment planning than those treated for cure. For these two reasons — more patients for radio therapy and less need for planning precision — a separate palliative radiotherapy machine might be appropriate. Such a machine will be chosen for sturdiness and simplicity rather than extreme beam sharpness and complicated manoeuvrability. This machine will then offload the patient burden from a second machine with more precision and manoeuvrability to be used for curative purposes. A further feature of radiotherapy seen in Africa is the overrepresentation o f cervical carcinoma in the patient population. Cancer of the cervix comprises one-quarter to one- third of all cancer patients in many of our institutions. This fact places intra cavitary brachytherapy on the priority list. Because of the large number of patients a high dose remote afterloading machine seems ideal. Obviously this will also require facilities for anaesthesia, among other things. So, in summary, an ideal radiotherapy unit in Africa ought to have as minimum equipment a sturdy, simple teletherapy machine for palliative work, a good high precision teletherapy 138 LUANDE machine for elaborate curative work and a cervix brachytherapy machine. Such a unit will most likely be situated at a major referral centre, most of which will host a medical school. Such a centre might later be supported by satellite brachy therapy units at a few regional or zonal medium size hospitals. Isotope radiation sources are more suitable in Africa than X-ray or other artificial radiation sources. Quite a few countries still have fluctuating power supply, which makes the smooth operation of man-made radiation sources cumbersome. On the other hand, isotopes have to be replaced from time to time. This aspect can prove a major limiting factor as isotopes have to be imported from overseas. In this regard the long half-life sources should be considered for some machines. For instance, a caesium source for brachytherapy may prove more convenient than a cobalt source, although the latter has superior physical properties. Apart from equipment issues, emerging African radiotherapy units will also be faced with personnel problems. Basic radiotherapy personnel such as radio therapists, physicists and technologists are currently very rare in Africa. Although they are rare, it will be neither economic nor appropriate for each unit to evolve a local training programme for such staff. So far most training has been offered in overseas institutions. This brings about several disadvantages such as the temptation o f graduates not to return to their home countries and inappropriate or irrelevant emphases in the training programmes. Therefore, regional training programmes should be started in Africa. Two levels of candidates for each discipline ought to be considered in the initial stage of such training. As mentioned above, up to 90% of the patients may require only simple palliative treatment, and thus a cadre of graduates with qualifications below the full MD, if well trained in routine radiotherapy work, will relieve radiotherapists tremendously. The latter may then have ample time to deal with more elaborate clinical, administrative and developmental aspects of the service. The same subcategorization of medical physics and therapy technology training may help streamline these areas. I have pointed out the uniqueness of African needs with respect to equipment, radiation sources and training facilities. Additional requirements crucial to the smooth running of a radiation service are : housing for such equipment, maintenance facilities, the regulation of electrical power supply, the training of dosimetrists, radiation safety officers, etc. For an overall improvement of radiation therapy results other services and strong surgical and medical oncology units, public health units and pain management units, among others, are crucial. Along with service efforts a modest cost effective cancer research programme addressing specific African cancer problems is essential. Thus it is clear that a multidisciplinary approach to establishing cancer facilities in Africa is mandatory. Such services are best modelled on the concept of a comprehensive cancer centre. Such a centre, by definition, is designed to incorporate most of these facets of development within a common descision making unit. IAEA-SM-290/6

HEAD AND NECK CANCER IN SAUDI ARABIA

B.S. CLUBB*, C.A. QUICK**, M.H. AMER*, E. MAHBOUBI***, M.A. EL-SENOUSSI*, H.P. SCHULTZ*, S.M. EL-AKKAD* *Department of Oncology **Department of Surgery ***Department of Biological and Medical Research

King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia

Abstract

HEAD AND NECK CANCER IN SAUDI ARABIA. Between July 1975 and December 1985, 1301 patients with histologically confirmed cancer of the head and neck were referred to the King Faisal Specialist Hospital and Research Centre. The commonest tumour site was the nasopharynx. Patients were analysed according to age, sex and geographical distribution. Possible aetiological factors are discussed. Stage analysis revealed a predominance of locally advanced stage III and IV tumours. For these patients a short course of radiotherapy, 1800 rad over 3 days, compared favourably with more protracted courses for palliation. Difficulties of reporting end results and future trends are discussed.

1. INTRODUCTION

The King Faisal Specialist Hospital and Research Centre (KFSH) is a 450 bed tertiary care medical centre in Riyadh and is the main cancer referral centre in Saudi Arabia. Over the last decade it has developed extensive radiological, diagnostic, laboratory and treatment facilities and a tumour registry. Experience gained from the KFSH gives insight into the situation of head and neck cancer in Saudi Arabia and is helpful in planning for the future.

2. MATERIALS AND METHODS

2.1. Tumour registry

The tumour registry at the KFSH has registered every patient with cancer referred according to age at the time of initial diagnosis, sex, geographical location and histological tumour type. Because of the wide geographical pattern of referral

139 140 CLUBB et al.

TABLE I. ICDO CATEGORIES FOR VARIOUS HEAD AND NECK CANCERS

ICDO Site

140 Lip

141 Tongue

142 Salivary gland

143 Gum

144 Floor of mouth

145 Other/unspecified

146 Oropharynx

147 Nasopharynx

148 Hypopharynx 149 Pharynx, ill defined

160 Nasal cavity, middle ear, paranasal sinuses 161 Larynx

TABLE II. PHOTON AND PARTICLE IRRADIATION AVAILABLE AT THE KFSH

Equipment Photon Electron Neutron (M eV) (M eV)

Co-60 1.17 1.33

Phillips - SL 75-10 4 linear accelerator 8 6 8 10

CGR - Therac 20 12 6 linear accelerator 18 9 13 17 20

Cyclotron 11 MeV average energy p(26) + Be IAEA-SM-290/6 141 to the KFSH, statistics from its registry are considered representative of Saudi Arabia as a whole. The registry used the International Classification of Diagnoses, Oncology (ICDO), 9th revision. This study of head and neck cancers is defined by ICDO categories 140 to 149 and 160 and 161 as outlined in Table I.

2.2. Combined head and neck clinic

Patients are usually seen first at a combined, multimodality clinic. Appropriate diagnostic and staging procedures are performed and all pathological material is reviewed before deciding on optimum treatment.

2.3. Radiotherapy

Radiotherapy was used in the majority of the patients, either alone or in combination with surgery or chemotherapy. The equipment available provides for a wide spectrum of photon and particle irradiation as outlined in Table II.

2.4. Simulation and planning

For patients with head and neck tumours, individual facio-cervical impressions are made on the patient using an alginate mix and plaster of Paris. The impressions are used to design a polyvinyl chloride mask. With the patient in the supine position, the mask is fitted over the face and neck and fixed to a head rest on the table, thus fixing the position of the patient. Simulation is done using a Phillips Simulator and treatment fields delineated with lines drawn on the cast, thus avoiding unsightly marks on the patient and the risk of their removal by the patient. Dosimetry was done using the GE RT/PLAN computer radiotherapy treatment planning system.

2.5. Fractionation schedules

(a) Conventional: used for patients with potentially curable tumours either alone or as an adjuvant prior to or following surgical excision, or in conjunc tion with chemotherapy. Tumour doses vary from 5000 to 7000 rad at a daily rate of 180—200 rad, 5 days per week, depending on the extent of the tumour.1 (b) Short course: often used for palliation. 1800 rad in 3 fractions of 600 rad each. Depending on the response this can be supplemented by further radiotherapy after a 4 to 6 week interval.

1 1 rad= 1.00 X 10~2 Gy. 142 CLUBB et al.

TABLE III. FREQUENCY OF ICDO CATEGORIES, SEX DISTRIBUTION AND M-F RATIO

ICDO Site No. cases Male Female M-F ratio

140 Lip 34 19 15 1.27

141 Tongue 172 107 65 1.65

142 Salivary gland 55 23 32 0.72

143 Gum 84 49 35 1.4

144 Floor of mouth 43 20 23 0.87

145 Other/unspecified 107 63 44 1.43

146 Oropharynx 32 25 7 3.57

147 Nasopharynx 399 291 108 2.69

148 Hypopharynx 137 65 72 0.9 149 Pharynx, ill defined 14 8 6 1.33

160 Nasal cavity, middle ear and 69 40 29 1.38 paranasal sinuses

161 Larynx 155 132 23 5.74

Total 1301 842 459 1.83

Tj N0 M0 T2 N0 M0 T3 N0 H0 T4 N0 ОГ N1 M0 Tl or T2 or Any T N2 or N3 Mo T 3 N1 Mo Any T Any N Mi

FIG. 1. Percentage of stage distribution of head and neck cancers. IAEA-SM-290/6 143

3. RESULTS

Between January 1975 and December 1985, 11 107 patients were entered into the KFSH tumour registry. Of these, 1301 (11.7%) were classified as head and neck tumours according to the ICDO categories 140 to 149 and 160 and 161. Table III shows the frequency of each ICDO category with the corresponding sex distributions and ratios. The majority of patients (88%) presented with advanced locoregional disease as shown in Fig. 1, which gives the percentage distribution of head and neck tumours according to stage. Most of the patients (96%) were Saudi nationals with the majority (80%) coming from the central, western or south-western regions of the country. Figure 2 shows the geographical distribution of the patients with head and neck cancer referred to the KFSH.

REGION 1 - CENTRAL - RIYADH REGION 2 - WEST COAST - JEDDAH REGION 3 - WEST INLAND - MADINA REGION 4 - EASTERN - DAMMAM

FIG. 2. Geographical distribution of head and neck cancers in Saudi Arabia. 144 CLUBB et al.

______= Nasopharyngeal group (total patients: 399)

...... * A ll groups except (to ta l p a tie n ts : 902) nasopharyngeal

FIG. 3. Distribution o f head and neck cancers according to age groups.

An analysis of age distribution for each ICDO category was similar except for nasopharynx, which tends to occur in a young population. This is shown diagrammatically in Fig. 3.

4. DISCUSSION

As yet, there is no national cancer registry in Saudi Arabia. However, over the last decade, as the KFSH has been the main referral centre for Saudi nationals with head and neck cancer, statistics from its tumour registry are felt to reflect the national trend. In this series, the commonest site for cancer was the nasopharynx, accounting for 33% of all head and neck cancers. Nasopharyngeal carcinoma is also common in Southern China, and South East Asia [1,2]. As in other parts of the world, Epstein Barr (EBV) virus titres are often elevated, suggesting a viral aetiology [3]. This, together with other observations including a trend to an earlier age of onset, a predominance of undifferentiated carcinoma with frequent lymphoid infiltration IAEA-SM-290/6 145 and a relatively high responsiveness to radiotherapy despite often advanced loco regional disease, are all factors requiring further investigation. The geographical distribution suggesting a higher incidence of head and neck cancer in the central, western and south-western regions could be due to a number of environmental or social differences across the country [4]. Such differences could be due to a greater use of tobacco, chewing of shamma or nutritional factors such as vitamin or mineral deficiencies [5—8]. Amer et al. [7] have shown a prevalence of oral cancer among shamma users in the Southern Asir region. A major problem has been the large number of patients with advanced stages of tumour at presentation. This is often due to a lack of patient or family awareness, although other factors including travel distance or reluctance to leave home to seek medical advice have to be considered. Similar factors have resulted in a number of patients becoming lost to follow-up to the tumour registry and this has made survival statistics difficult to obtain. However, a number of studies have been initiated to try to compile such statistics and these will be published later. A major finding has been the successful palliation of many patients with locoregionally advanced tumours of the head and neck treated with a short course of radiotherapy involving 3 fractions of 600 rad each [9]. This short course can be supplemented after 4 to 6 weeks by further radiation depending on the initial response. The degree of palliation obtained with this short course compared very favourably with a more conventional protracted course of treatment involving 5000—6000 rad irradiation over 5 to 6 weeks.

5. FUTURE TRENDS

(1) Epidemiological studies and basic research have been initiated and will continue at the KFSH in the hope of finding correctable aetiological factors for patients at risk for head and neck cancer. (2) A continuation of efforts to compile survival data through the tumour registry for comparative analysis with the results of other institutions around the world. (3) The development of programmes to increase public awareness of the early signs and symptoms of cancer for early detection and so to decrease the high number of advanced tumours. (4) The KFSH is about to commence fast neutron therapy using its cyclotron for head and neck cancer. As many patients are referred to KFSH, it should be possible to define the usefulness of neutron beam therapy for patients with head and neck cancer and this should act as a model for the entire Middle East and surrounding countries. 146 CLUBB et al.

REFERENCES

[1] JUNG, P.F., YU , C., Nasopharyngeal carcinoma in China, Postgrad. Med. 33 (1964) A77. [2] M U IR , C.S., Nasopharyngeal carcinoma in non-Chinese populations with special reference to South-East Asia and Africa, Int. J. Cancer 8 (1971) 351. [3] DE-THE, G., GESER, A., Nasopharyngeal carcinoma: Recent studies and outlook for a viral aetiology, Cancer Res. 34 (1974) 1196. [4] EL-A K K A D , S., A M ER, M .H., et al., Pattern of cancer in Saudi Arabs referred to King Faisal Specialist Hospital, Cancer (accepted publication, Sep. 1986). [5} KELLER, A .Z., TERRIS, M., The association of alcohol and tobacco with cancer of the mouth and pharynx, Am. J. Public Health 55 (1965) 1578. [6] M AHBOUBI, E., H A N N A N , M., et al., Oral cavity cancer and aetiological factors in Saudi Arabia, Proc. Am. Assoc. Cancer Res. 25 (1984) 221. [7] AMER, M., BULL, C.A., et al., Shamma usage and oral cancer in Saudi Arabia, Ann. Saudi Arabian Med. 5 (1985) 135. [8] R E D D Y , B.S., COHEN, L.A ., et al., Nutrition and its relationship to cancer, Adv. Cancer Res. 31 (1980) 237. [9] BULL, C.A., M cA RTHUR , P., et al., Treatment of advanced head and neck squamous cell carcinoma using a three day course of palliative radiotherapy, Ann. Saudi Arabian Med. 5 3 (1985) 143. IAEA-SM-290/11

ACCELERATED FRACTIONATION

V.H.J. SVOBODA Department of Radiotherapy and Oncology, St. Mary’s Hospital, Portsmouth, Hampshire, United Kingdom

Abstract

ACCELERATED FRACTIONATION. Fourteen years of clinical experience with accelerated fractionation is summarized. Regimes using two or three sessions every treatment day over a short period of time offer not only practical advantages but also fast palliation and a chance of improved therapeutic ratio, namely in radiotherapy of quickly repopulating tumours. This method can be modified for almost every indication of external beam photon treatment and is especially advantageous for centres which use their machines for long hours and have difficulty with daily transport of their out-patients.

1. INTRODUCTION

The history of radiotherapy has gone hand in hand with developments in technology. Machines offering better and more penetrating beams and larger output are becoming increasingly sophisticated. Today we can treat any volume in the human body to an accuracy of a few millimetres with a homogeneous depth dose. Over the years progress in radiotherapy has become linked in our minds with the progress of physics and with increasingly more expensive equipment. None theless, we only have to look at our present understanding of the effect of distri bution of the dose in time to realize how vague our knowledge is, yet no investment is required to study the time factor in clinical work.

2. HISTORY

Between the wars, Schinz [1] and Couttard [2] established the principles of flexible fractionation with 180 keV machines. The output was 2-3 R/min. Very long exposures had to be split into two daily sessions which were abandoned as soon as the dose rates improved. Eventually, 2 Gy daily emerged as the best increment for most clinical situations. Later, Paterson [3] reduced the overall time of his six week long courses and increased the dose per session for practical reasons. His institute has used 21 day courses ever since. It is claimed that the results are comparable with longer regimes. Others economized later by two or three sessions a week, but some evidence of increased late damage after

147 148 SVOBODA

TABLE I. ABBREVIATIONS USED IN THE TEXT

AF = Accelerated fractionation: NF > 3 6 , imin - 4 h CF = Conventional fractionation

DF = Dose per fraction gap = cuts the course into uneven parts

HF = Hyperfractionation: NF > 36, DF < 2 Gy

ID = AF started with a large introductory dose ( ID )

^ i n = minimal interval between the two sessions

Jm a x = maximal interval MFD = Multiple fractions every treatment day

MFDs = at least 2 sessions per day, ОТ similar to CF, half way split (example: Edsmyr [9]) M FDgl = gap after 1 /3 of TD (EORTC protocol 22851)

MFDg2 = gap after 2/3 of TD (EORTC protocol 22811)

NF = Number of fractions NSD = Nominal standard dose

NST = Normal soft tissue

ОТ = Overall time in days between first and last session split = after half TD

TD = Total dose

TR = Therapeutic ratio

48 24/10 - 4800 cGy TD given in 24 fractions; overall time 10 days (8 treatment days).

large fractions has appeared. Sambrook’s split technique successfully reduced the acute reactions [4]. None of these changes were designed to improve the therapeutic ratio. Only in the early seventies did Pierquin [5] return to the technique of long sessions and short overall time, using a low dose rate cobalt machine. Bäckström et al. [6], in the hope of bypassing the protective effect of hypoxia by smaller increments, treated the first patients three times a day. In 1963 Fowler et al. [7] showed that the size of the fractions is more important than the overall time, and the same fact was obvious from Ellis’s exponential formula. Yet, practical attempts in 1972 to treat with 2 Gy increments three times every treatment day were looked upon with disbelief and suspicion [8]. Accelerated fractionation (AF)1 has been introduced in many countries over the last 14 years. Edsmyr et al. [9] continued with hyperfractionated treatment

1 For abbreviations see Table I. IAEA-SM-290/11 149 of advanced bladder tumours and recently claimed superior results when compared with conventional courses. Norin et al. [10] reported improvement in the treatment of Burkitt’s lymphomas, which were resistant to daily regimes, and the EORTC Radiotherapy Group demonstrated the feasibility of AF in a head and neck pilot study [11]. A dose of 48 Gy was delivered in 30 fractions over 12 days. The tumour was then boosted after a pause of 4 weeks. An international randomized trial on this principle was completed in 1984 and the results are now expected. Many authors used large numbers of small fractions, often with a split or gap, so that the overall time was not very different from conventional regimes. The results were never inferior when compared with conventional courses. After more than 50 years we are perhaps better equipped to return to the question posed by Couttard [2], namely how to match the most suitable fractionation to an individual tumour. However, the introduction of frequent sessions could claim only moderate success and the statistical evidence of improved therapeutic ratio is not yet available. Such demonstrations are difficult in palliative or prophylactic regimes, or in tumours of a very poor prognostic category. The main reason, however, is that the individual cytodynamics of the tumour were disregarded by grouping the tumours only anatomically. Only some patients in each treated group are then likely to benefit.

3. EXPERIENCE

3.1. Rationale

In a busy department with limited resources HF is a luxury which would have to be justified by considerable benefit to the TR. The conventional range of 15—36 sessions offers enough opportunity for clinical research. Our main interest was the function of time and most of our 14 years’ experience is based on accelerated regimes, where we merely shortened the intervals to less than 24 hours. This saves beds, transport, and patients’ time at no extra cost of setting-up and machine time. However, is the difference in the intracellular damage recovery and in the repopulation between the tumour and NST sufficiently large to be exploited by accelerating the increments?

3.2. Breast cancer

In 1972 we started to treat advanced breast cancer by regimes of the type 40 Gy in 15—20/5, or 45—47 Gy in 30/12. With cobalt the skin reactions were brisk only if we built up the skin dose. The intensity of such reactions predicted well the late damage. The threshold of the late changes was steep: 40 15/5 caused no late changes at all, 42 15/5 led to slight thickening, but 45 15/5 caused 150 SVOBODA leathering and rib necrosis. A dose of 47.5 Gy was tolerated well only if given in 24—30/10—12. We call the plot of our experience of late changes caused by different fractionation the breast tolerance map [12]. The response of the tumours compared well with daily hyperfractionated regimes (50 X 130 cGy over 70 days), which produced complete regression of 30% of inoperable breast carcinomas. After AF, which took 5-12 days, 56% of similar tumours showed complete regression. The local recurrence rates in the two groups were comparable [12]. In slowly proliferating tissues we can deliver an efficient dose by a relatively small number of larger fractions.

3.3. Head and neck cancer

A group of advanced or frail patients suffering head and neck squamous carcinoma were treated with 50—55 Gy in 24—30/10—12. I published my early experience with neck tumours in 1975 and 1978 [8, 13]. I observed that 50 Gy did not always sterilize the tumour. This had two consequences: Peracchia and Salti [14] applied similar fractionation to larger volumes in advanced tumours of the oral cavity with disastrous results, and the clinicians concluded that improvement in the TR in head and neck cancers demands smaller increments and a gap. Up to 1985 we treated 95 patients with head and neck cancer either with AF (72 patients) or with MFDg2 (23 patients). The latter treatment was initially well tolerated, but 10 out of 23 patients suffered severe late damage - twice as many as after simple AF with a comparable NSD. Patients with large, infected tumours and tracheostomy had an increased risk of damage as did patients sensitized by misonidazole. The damage did not depend on the NSD or the size of individual increments within the material studied. Early tumours of the larynx were treated with 54 30/12 with excellent late tolerance. Ten patients, who were followed up for a minimum of 3 and a maximum of over 10 years, had a normal voice and no laryngeal changes. In the whole group of 95 patients 92% of primary tumours regressed completely within 3 months. Of the large regional métastasés 72% regressed. The MFDg2 regimes led to slightly better and more permanent tumour response than AF but 7 out of 14 laryngeal tumours so treated suffered serious late damage. After AF only 4 out of 45 patients were so affected. In extralaryngeal sites the figures were 3/9 for MFDg2, and 7/27 for AF. The better response in tumours boosted after a gap was therefore at the cost of increased late damage. The good acute tolerance did not mean improved TR. The figures for survival were similar to those in historical controls treated with CF.

3.4. Cancer of the bronchus

We started using AF for carcinoma of the bronchus in 1973. By 1980 we had treated 45 patients with different types of AF two, three or four times every IAEA-SM-290/11 151 treatment day, ОТ 5-21 days. The acute tolerance was excellent. Some patients improved symptomatically even during the course. The palliation was effective and fast. Sixteen patients, who were treated with the ID type o f AF, also responded well. In 20 patients complete, and in a further 5 almost complete regression was obtained. Long survivals were observed among patients with localized tumours who received a radical dose [15]. Twenty-four patients with localized squamous cell carcinoma not larger than 6 cm in diameter were therefore treated on a radical three dimensional plan, 48 Gy in 30/12. The volume was reduced after 2/3 of the dose. Of these only one patient was staged asT ^o, the rest were staged as follows: T2 N0 = 9, T3 N0 = 9, T2 Nj = 3, T3 N2 = 2. The crude survival by 8 June 1986 was 79% at 1 year, 50% at 2 years, 33% at 3 years, 25% at 4, and 17% (4 patients) lived more than 5 years. One died at 70 months, three are alive without recurrence at 92, 84 and 68 months. The last one had successful pneumonectomy for recurrence at 36 months. Pneumonitis was the limiting toxicity. We observed it clinically in 6 patients. Its seriousness depends on the treated volume, NSD, condition of the lung and site.

3.5. Cancer of the abdomen and pelvis

A range of advanced pelvic and abdominal tumours was treated either with small accelerated increments, or with the same regime, but after a larger ID [15]. Pelvic volumes of about 1 L tolerate 40 Gy in 30/12 without late damage. A dose of 45 Gy will produce cystoproctitis, which depends on the length of time intestine and bladder are irradiated, but the dose is probably at the limit of late tolerance. We treated 18 patients with advanced carcinoma of the bladder, 3 patients with cancer of the rectum, 3 ovarian, 5 prostatic and 3 colonic carcinomas, 3 tumours of the anus and 3 carcinomas of the cervix. AF was also used for parametrial boost between two uterovaginal insertions and before an anal implant with very good results. Recently we treated more patients with advanced carci noma of the stomach, colon and rectum. The acute reactions were mild.

3.6. Other conditions

In the seventies we applied AF to several hundred primary or metastatic skin lesions. Recently we also treated cranial contents and soft tissue tumours of the extremities, carcinoma of the oesophagus and lymphatic system before renal graft. The acute and late tolerances corresponded to our predictions but experience is still limited. 152 SVOBODA

4. DISCUSSION

In designing our accelerated courses, Ellis’s formula [16] provided us with good guidance. Recently the linear-quadratic model has helped us to understand better the specific role of dose per fraction in different tissues. We had to draw up practical courses for a busy department and we tried not to change many factors at the same time, avoiding the splits and gaps. Such pauses may improve acute reactions, which could be well treated, but do not reduce the risk of late damage. The variations in the tumour cell turnover and extent prevented an improved TR, but the responses were impressive and the advantage of conven tional fractionation was never lost. We mapped the tolerance of several normal tissues [11], demonstrated the steep threshold of the late damage and prepared the ground for a more flexible application of the time factor in radiotherapy. Small increments may offer some advantage in hypoxic tumours but we do not know how much better a fraction of 1 Gy is compared with 2 Gy. Acceleration presumes a different speed of intracellular recovery and repopulation dynamics in tumour and normal tissue. Its rational use will eventually require knowledge o f individual tumour cell turnover. Treatment of the breast without skin buildup does not require fractions much smaller than 3 Gy, applied in a short ОТ. Late damage after post-operative treatment can be avoided. There is no need for HF or split. Similarly, early tumours of the larynx can be treated rapidly with increments of a conventional size. The strategy must be modified for large tumours of the neck, which require irradiation of extensive volumes, and for the oral cavity, with its large area and rapid mucosal turnover. The AF tolerance can be increased by prolongation of imin to 4—6 h, or by using two sessions a day only. Such a frequency is probably enough to prevent rapid tumour repopulation in the intervals. Treatment during weekends is tempting but the logistics are not easy. The alternative is to reduce the increments and use up to 36 fractions. Brumm [17] documented well the importance of fractionation frequency for the radiotherapy of bronchus carcinomas. By simple extrapolation AF should offer the best option. Our experience confirms the need for fractions smaller than 3 Gy, unless we use a single dose of 6 Gy on Fridays, followed on Mondays by fast fractionation with small increments (ID). The limited pelvic volumes do not pose difficulties with tolerance, but their merit and TR remain to be explored. Trott [18] found that both squamous carcinomas and colorectal adenocarcinomas repopulate rapidly. We know that the whole abdomen can be treated with single exposure of up to 8 Gy. It is there fore not surprising that our clinical experience also finds ID for treatment in the pelvis and abdominal cavity feasible. AF with radical intent must use NSD values, approaching the normal tissue tolerance. Because of the steep onset of late damage, good dosimetry is mandatory. IAEA-SM-290/11 153

In slowly proliferating normal tissues, such as the breast and chest wall, we recommend increments between 200 and 270 cGy. For instance, no late damage is caused by prophylactic treatment after local excision of the type 38-40 Gy 15/5. Inoperable breast tumours respond well to 48 Gy in 30/12, or 42 Gy in 21/9. Squamous carcinomas in the neck can be either treated twice a day, giving 52 Gy in 26/17, imin= 6 h, or by three daily sessions, 50-52 26/11, if the volume is small. Large volumes including the oral cavity would be better treated with smaller increments (DF 170 cGy), 52 Gy 30/12, imin = 4 h. Another modification of AF has already been mentioned. The ID is started by a 6 Gy exposure on Fridays, followed on Mondays by 1 — 1.3 Gy incre ments, three times every treatment day. This method saves machine time and combines the advantages of AF and HF. ID kills the oxygenated cells in the sensitive phase of the cycle. The partly synchronized and repopulating tumour stem cells are then sterilized by fast small increments. An almost ideal type of AF was recently reported by Saunders [19]. She obtained good results with treatment in three sessions on 12 successive days (imin= 6 h, imax = 12 h). This regime allows no time for tumour recovery but uses a sufficient interval for complete repair of the sublethal intracellular damage in the normal tissues. Among many other applications, AF can be used for parametrial boost in carcinoma of the cervix, or to treat stage III before the uterovaginal insertion. The short course takes less of the patient’s time and saves hospitalization. Several sessions can be influenced by one dose of sensitizer. AF is therefore economical. We have now learned something about normal tissue reaction to AF. We saw that the response of tumours of the breast, lung and larynx was fast and good. We hope that labelling individual tumours will help us to adjust fractionation to the particular lesion and its site. Such flexibility of fractionation strategy is aimed at a better TR. Daily increments are unlikely to be the best answer for every situation. We hope that this simple method will lead to improved results. It opens a large field of clinical research to young teams who use their machines for more than eight hours a day and have limited facilities for transport, but are keen to take an active part in the progress of their speciality.

REFERENCES

[1 ] SCHINZ, H.R., Strahlentherapie 37 (1930) 31 ; Strahlentherapie 58 (1937) 373. [2] COUTTARD, H., Lancet 227 (1934) 1. [3] PATERSON, R., The Treatment of Malignant Disease by Radium and X-Rays, London (1984) 10, 232. [4] SAMBROOK, D.K., Clin. Radiol. 13 (1962) 1. [5] PIE R Q U IN , J., J. Radiol. (Paris) 51 (1970) 533. [6] BÄCKSTRÖM, A., et al., Acta Radiol. (Ther.) 12 ( 1973) 401. 154 SVOBODA

[7] FOWLER, J.,e ta l., Br. J. Radiol. 36(1963) 188. [8] SVOBODA, V.H.J., Br. J. Radiol. 48(1975) 31; Br. J. Radiol. 51 (1978) 363. [9] EDSMYR, F., et al., Radiother. Oncol. 4 (1985) 197. [10] NORIN, T., et al., Int. J. Radiat. Oncol., Biol. Phys. 2 (1977) 399. [11] VAN DEN BOGAERT, W., et al., Radiother. Oncol. 3 (1985) 139. [12] SVOBODA, V.H.J., Proceedings, Varian’s 4th European Clinac Users Meeting, Malta (1984) 70. [13] RESOULY, A., SVOBODA, V.H.J., Progress in Radio-Oncology II (KÄRCHER, K.H., et al., Eds), Raven Press, New York (1982) 339. [14] PERACCHIA, G., SALTI, C., Int. J. Radiat. Oncol., Biol. Phys. 7 (1981) 99. [15] SVOBODA, V.H.J., “Radiotherapist’s new look at bronchogenic carcinoma”, Progress in Radiooncology (Proc. 3rd Int. Meeting Vienna, 1985). [16] ELLIS, F., Int. J. Radiat. Oncol., Biol. Phys. 11 (1985) 1685. [17] BRUMM, P., Br. J. Radiol. 56 (1983)215. [18] TROTT, K.-R., et al., Radiother. Oncol. 3 (1985) 139. [19] SAUNDERS, M.I., DISCHE, S., Br. J. Radiol. 59 (1986) 523. IAEA-SM-290/7

FAST NEUTRON THERAPY FOR THE DEVELOPING WORLD: IS IT WORTHWHILE?

A. AISSI, J.M. FEOLA, B.S. CLUBB King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia

Abstract

FAST NEUTRON THERAPY FOR THE DEVELOPING WORLD: IS IT WORTHWHILE? The King Faisal Specialist Hospital and Research Centre (KFSH & RC) in Riyadh incorporates the only neutron therapy facility of its kind in the developing world. The facility consists of a cyclotron based isocentric neutron therapy system which uses one of the seven beam lines from the cyclotron. A few patients were treated in January 1984 after extensive work had been carried out to overcome mechanical and electrical problems inherent in the equipment and to collect the necessary dosimetric data of the machine. This work was complex and involved technical personnel who are not commonly available in developing countries. Moreover, the great distance separating the KFSH & RC from other neutron therapy centres made it difficult to establish collaborative studies and to compare physics data and radio biological parameters. These personnel and logistic considerations were more pronounced in the case of neutron therapy than with conventional radiotherapy modalities. Additionally, the current status of neutron therapy is such that any attempt to establish this therapeutic modality in a developing country should be carefully investigated taking into consideration the level of practice and the experience gained in the field of conventional radiotherapy in that particular country. In the case of Saudi Arabia, the implementation of the neutron therapy programme was influenced very much by the controversy that surrounded the merit of neutron therapy in the industrialized world. Nevertheless, the programme is well advanced and the country is now ready to benefit from it, thanks to the great effort and resources invested in it.

1. IN T R O D U C T IO N

Renewed hopes of the efficacy of neutrons to treat cancer, reported in the seventies by Catterall [1,2], were instrumental in the decision to establish a neutron therapy centre at the King Faisal Specialist Hospital and Research Centre (KFSH & RC) in Riyadh. This modality involved the installation of more sophisticated equipment than is usually needed in conventional radiotherapy. The neutron facility was and still is the only one of its kind in the developing world and gave Saudi Arabia the possibility to participate in the developmental and investigational stages of this radio therapeutic modality. However, the great complexity involved in implementing such a programme combined with the

155 156

rife. IS t al. et AISSI

FIG. 1. Layout of cyclotron and neutron therapy facilities. Linear accelerators have been installed in the two adjacent rooms on the left. IAEA-SM-290/7 157 dubious value of neutron therapy as reported in later publications [3, 4], made it difficult to implement the programme as initially planned. The project of assessing the feasibility of neutron therapy to treat cancer is still in progress but is limited to a few centres located in industrialized countries. KFSH & RC is just about to resume its programme after a limited and inconclusive attem pt in 1984. Should a developing country embark or have embarked in a programme to investigate the efficacy of neutron therapy? This paper attempts to answer this question, while discussing other important aspects based on KFSH & RC’s experience in this area. Many of these factors are common to any service or technology when newly introduced into a third world country, and special emphasis will be laid on those directly related to neutron therapy.

2. DESCRIPTION OF THE FACILITY

The neutron facility consists of a TCC (The Cyclotron Corporation, no longer in business) Isocentric Neutron Unit, Model 4000, connected to a TCC 30Cs cyclotron located in an adjacent room. The cyclotron has seven beam output lines, some of which are dedicated to tasks such as production of radioisotopes and physics research. Figure 1 presents the layout of the cyclotron facility, which was described in more detail by Barrall et al. [5]. The cyclotron has a unique feature since it is capable of accelerating 26 MeV protons in addition to 15 MeV deuterons. This feature was added during the construction phase after it was realized that neutrons produced from 26 MeV protons might have a better therapeutic quality than those generated from 15 MeV d e u te ro n s. After the charged particles are accelerated to their maximum energy they interact with a beryllium target located in the head of the isocentric unit. This unit can rotate 240° around a horizontal axis to optimize dose delivery to patients.

3. TECHNICAL CONSIDERATIONS

Even though the neutron therapy unit was installed and accepted by the end of spring 1982, it was not until January 1984 that patient treatment could be started. During that period a great deal of time and effort was needed to characterize the beam physically and biologically. The presence of technical experts in neutron physics, radiobiology, and cyclotron operations facilitated the task, but most of these had limited experience in clinical neutron therapy. Moreover, at that time the neutron dosimetry protocols were fairly recent and additional dosimetry equipment had to be acquired to comply with the newly introduced guidelines and methods. A committee was established to oversee all aspects of the preparation of the facility for clinical use. 158 AISSI et al.

3.1. Physics data

It is generally recognized that neutron dosimetry is more complex and less practised than conventional radiation dosimetry. The fact is that neutrons are always accompanied by gamma radiation, and the quantification of each of these two components is a very involved process requiring special instrumentation. Delays were experienced in having the dosimetry equipment calibrated and ready for evaluating the physical characteristics of the machine. At the initial stage a major decision had to be made: whether to use the p(26)+Be (26 MeV protons on a beryllium target) or the d(15)+Be (15 MeV deuterons on a beryllium target) neutron beam. Studies of these two beams indicated, as expected, that the proton produced neutron beam has more clinical advantages than the deuteron produced one [6 ]. The main advantage of the proton produced neutron beam resides in its higher average energy and thus better tissue penetration. Collection of the necessary physical data took almost one year, mainly because of:

( 1) the limited cyclotron time allotted to carry out physics measurements; (2 ) the unexpected electrical and mechanical problems encountered with the isocentric unit; and (3) the fact that all data had to be taken manually because of the low dose rate of the unit, which led to an unacceptably high noise level when using an automated data acquisition system.

Finally, the physical characterization of the neutron beam was achieved and separation of the gamma and neutron components was implemented in the treatment planning computer to provide the clinicians with all the information necessary to treat patients. It must be stressed that a separate evaluation of the components is an important aspect in neutron therapy, since neutrons and gamma rays have different biological effects on tissue.

3.2. Biological data

Determination of the relative biological effectiveness (RBE) of the neutron beam was also essential since it served as the basis of the assessment of the dose to be prescribed to patients. This parameter can be used to relate the neutron dose of one centre to another when following similar clinical protocols. The RBE of the KFSH & RC neutron beam was obtained using several in vivo and in vitro biological systems, since at that time most neutron therapy centres used different systems.

3.3. Other considerations

By the end of 1983 the personnel involved in neutron therapy felt the need to establish collaborative work with international centres to gain further IAEA-SM-290/7 159 experience in the field. Dr. Duncan from Western General Hospital, Edinburgh, Scotland, was invited to visit the KFSH & RC to discuss the feasibility and merit of neutron therapy as a modality to treat cancer. It is now well known that Duncan [3], in an attem pt to reproduce results obtained by Catterall at Hammersmith, concluded that there was “no qualitative advantage of neutrons in relation to tumor responses” and that increased late radiation morbidity resulted in his clinical trials. His conclusions were clearly stated during his visit and had considerable repercussions on the neutron therapy programme at the KFSH & RC. Medical staff from other departments in the hospital became less enthusiastic, and even the hospital administrative authorities started questioning the merit of this modality. Nevertheless, a clinical protocol for advanced squamous carcinoma of the head and neck regions using neutron therapy was established and the first patient received his first treatment in January 1984. From the 35 patients expected in the protocol, only 2 out of 5 were eligible after a 6 month trial period, mainly because many of the referring physicians were less convinced of the efficacy of this new modality. In addition, there was a pressing need to compare the physics and radio biological data with an independent institution to assess their accuracy. A decision was taken to stop patient treatm ent until some kind of assurance was obtained with regard to the clinical utilization of these data.

3.4. Intercomparison experiments

Organizing an intercomparison study in physics and radiobiology between our centre and another centre active in neutron therapy seemed at one time an unreachable objective. Over eighteen months of continuous effort elapsed before this objective was finally realized. The reasons were numerous but the major delays were caused when interacting with the centres involved and setting a date for the site visit. This visit was made by an EORTC (European Organization for Research on Treatment of Cancer) group from Université Catholique de Louvain, Brussels, in November 1985. The results of this intercomparison indicated that both physics and radio biological data were in good agreement and made it possible to relate the RBE of our beam to those of other centres in the world. Additionally, the KFSH & RC was elected a member of the EORTC high LET Group and to participate fully in their activities.

4. CLINICAL CONSIDERATIONS

The merit of neutron therapy as a modality to treat cancer has been subject to controversy. It started in 1938 in Berkeley, California, where patients were 160 AISSI et al. treated using a neutron beam from research cyclotrons. The results of the clinical trials, published a few years later [7], were unfavourable. The main conclusion was that neutron therapy should be discontinued. Twenty years later, and after extensive research into neutron radiobiology, clinical trials were resumed at the Hammersmith Hospital and Royal Postgraduate Medical School of the University of London. The results of 800 patients treated between 1970 and 1978 were summarized as follows [8 ]. “In the great majority of cases, treatm ent resulted in the complete regression of tumours with very few recurrences. Successful treatments were particularly noteworthy in advanced tumours of the following sites: oral cavity, salivary glands, paranasal sinuses, breast and anus. Sarcomas of soft tissues also regularly disappeared and the recurrence rate was much lower than those previously obtained with other forms of treatm ent.” This conclusion was a confirmation of the earlier results reported elsewhere [1,2]. At that time, considering the incidence of advanced head and neck cancers in the Kingdom of Saudi Arabia, it was decided that the country might benefit from neutron therapy. This was in line with the plan of having a very specialized tertiary care hospital, and neutron therapy would be an extension of the cyclotron service to be provided in the hospital. In 1982, with the installation of the neutron unit close to completion, Duncan [9] stated in a publication that: “We have found, therefore, no evidence of an increase in therapeutic ratio which would indicate that neutron therapy is of clinical advantage. The effectiveness of the techniques employed may have been limited by the comparatively poor penetrating quality of the neutron beam in Edinburgh. The next generation of hospital based high energy cyclotrons now being manufactured and installed will better be able to make a definitive assess ment of fast neutron therapy compared to results achieved by good megavoltage X-ray therapy.” The KFSH & RC neutron beam has a higher energy than that of Edinburgh since it uses the p(26) + Be rather than the d( 15) + Be reaction, and our unit can be qualified as a medium energy machine. If one accepts Duncan’s conclusion, the KFSH & RC neutron facility can be expected to yield better results than those achieved at Edinburgh. Furthermore, the results of a randomized RTOG stu d y [ 10 ] have demonstrated a “significant advantage for neutrons when compared to photon radiation therapy in a group of patients with very advanced squamous cell carcinomas of the head and neck.” This was confirmed in another study from Europe [11] and the current trend of thinking is that neutron therapy has been proven to be superior to other conventional radiotherapeutic modalities, at least for certain types of tumours. Clinical trials are still in p rogress [ 1 2 ] to assess the merit of this modality to treat other cancers. IAEA-SM-290/7 161

5. CONCLUSIONS

To answer the question of the value of a fast neutron therapy programme, a thorough evaluation must be made of the country’s needs and its material and human resources as well as of other available radiotherapeutic modalities. While this evaluation may be straightforward in the industrialized world, our experience has shown that there are other considerations that need to be taken into account when it involves third world countries. These considerations are related to two major areas: justification of neutron therapy in relation to the incidence of certain types of cancer in the country or region concerned, and availability of logistic support for the programme. Careful evaluation should be made regarding the following logistic aspects:

(1 ) Availability of reliable and integrated neutron therapy units with adequate beam penetration, since complications are more prominent when using low energy machines. (2) Training of all staff involved in neutron therapy. (3) Development of research programmes to investigate the use of neutron therapy in the management of certain tumours that may be more common in the country. (4) Establishment of a well geared system of patient follow-up to optimize the evaluation process of this therapeutic modality. (5) Collaboration with international centres to exchange information and experience.

Since neutron therapy clinical trials are still being conducted only in a few centres worldwide, and considering the fact that many developing countries have yet to satisfy their basic needs in conventional radiotherapy machines, the acquisition of neutron therapy units by these countries is not recommended at present. However, countries like Saudi Arabia, which has strong financial and technical capabilities and a commitment to medical research, should embark in a programme to investigate the efficacy of neutron therapy and to make use of it to improve the quality of patient care.

ACKNOWLEDGEMENT

The authors are grateful to Dr. Salah El-Akkad, Department of Oncology, for reviewing this manuscript.

REFERENCES

[ 1 ] CATTERALL, M., SUTHERLAND, I., BEWLEY, D.K., First results of a randomized clinical trial of fast neutrons compared with X or gamma rays in treatment of advanced tumours of the head and neck, Report to the Medical Research Council, Br. Med. J. 2 (1975) 653. 162 AISSI et al.

[2] CATTERALL, M., First randomized clinical trial of fast neutrons compared with photons in advanced carcinoma of the head and neck, Clin. Otolaryng. 2 (1977) 359. [3] DUNCAN, W., ARNOTT, S.J., “An interim assessment of the experience of fast neutron therapy in Edinburgh”, Progress in Radio-Oncology II (KÄRCHER, K.H., et al., Eds), Ravenpress, New York (1982) 57. [4] DUNCAN, G.T., ARNOTT, S.J., ORR, J.A., KERR, G.R., The Edinburgh experience of fast neutron therapy, Int. J. Radiat. Oncol., Biol. Phys. 8 (1982) 2155. [5] BARRALL, R.C., MERENDINO, K.A., FETEIH, N., A medical cyclotron and program at the King Faisal Specialist Hospital and Research Centre, Riyadh, Proc. IEEE, Trans. Nucl. Sei. NS-30 2(1983) 1777. [6] NAIR, R.P., AL-SIARI, A., SKAGGS, L.S., Dosimetric considerations ofd(15)+Be and p(26)+Be neutron beams from an isocentric cyclotron facility, Med. Phys. 13 2 (1982) 207. [7] STONE, R.S., Am. J. Roentgenol. Radium Ther. 59 (1948) 77 1. [8] CATTERALL, M., BEWLEY, D., Fast Neutrons in the Treatment of Cancer, Academic Press, New York (1949). [9] DUNCAN, W., “Clinical experience of fast neutron therapy in Edinburgh”, 13th Int. Cancer Congr., Part D (Research and Treatment) (1983) 261. [10] GRIFFIN, T.W., DAVIS, R., HENDR1CKENSON, F.R., MAOR, M.H., LARAMORE, G.E., Fast neutron radiation for unresectable squamous cells carcinoma of the head and neck: The result of a randomized RTOG Study, Int. J. Radiat. Oncol., Biol. Phys. 10 (1984) 2217. [11] WAMBERSIE, A., GUEULETTE, J., VYNCKIER, S., Survey of clinical data and recent trends in neutron therapy, submitted for publication to Ann. Saudi Med. (1986). [12] LARAMORE, G.E., GRIFFIN, T.W., Potential applications of clinical neutron beam generators in cancer management, submitted to Appl. Radiol. (1984). IAEA-SM-290/78

Invited Paper

TREATMENT OF CARCINOMA OF THE THYROID GLAND WITH IODINE-131

R. H Ö F E R Abteilung für Nuklearmedizin der II. Medizinischen Universitätsklinik Wien, and Ludwig Boltzmann Institut für Nuklearmedizin, Vienna, Austria

Abstract

TREATMENT OF CARCINOMA OF THE THYROID GLAND WITH IODINE-131. For an optimal effectiveness of radioiodine therapy of differentiated thyroid carcinoma a straightforward strategy must be observed: Surgical ablation of the gland and of regional lymph nodes; histological verification of differentiated thyroid carcinoma including immuno- histochemistry; ablative radioiodine therapy; methodologically optimized follow-up including radioiodine retention measurements and thyroglobulin to determine whether there are recurrences and/or métastasés; determination of further therapeutic measures, i.e. further radioiodine therapy, surgical removal of primary and secondary tumour manifestations, conventional radiotherapy, chemotherapy or a combination of such measures; suppressive thyroid hormone therapy between the follow-up (therapy) studies; life-long control of patients. Thus the follow ing 5 year probabilities of survival (life table method) have been obtained between 1973 and 1982: 134 patients with differentiated papillary thyroid carcinoma showed a survival proba bility of 0.96 ± 0.02, and 133 patients with differentiated follicular thyroid carcinoma 0.92 ± 0.03.

1. INTRODUCTION

Thyroid cancer is a rare tumour. Approximately 10 300 cases of thyroid cancer are discovered annually in the United States of America [1]. In the United Kingdom about 1 in every 200 cancer diagnoses is a thyroid tumour, and in 22 European cancer registries 0.4% of males and 0.9% of females of all cancer diagnoses are thyroid cancer [ 2 ]. However, though uncommon as a clinical entity and as a cause of death, thyroid cancer has received high clinical attraction because radioactive iodine used by the thyroid epithelial cells to produce thyroid hormone can be used to treat most of the differentiated forms of thyroid cancer. It is generally accepted that differentiated thyroid cancer, the primary tumour, the local spread and distant métastasés can effectively be treated with radioactive iodine.

163 164 HÖFER

There is, however, considerable disagreement as to the question of whether the therapeutic use of radioactive iodine is truly necessary or whether it would not be better to substitute other forms of treatment that are less cumbersome with respect to the handling of the patient and to contamination problems and which have, it is claimed, fewer side effects.

2. RADIOACTIVE IODINE: APPRAISAL OF THE PROS AND CONS

Arguments on the pros and cons started already with the advent of radioiodine therapy and to some extent are still going on. The first and most commonly used argument against radioiodine treatment is the good life expectancy of differentiated thyroid cancer patients in general and the decrease in death rate in patients with differentiated thyroid cancer given thyroid hormone alone in suppressive doses compared with groups that did not receive thyroid hormone suppression after surgery. This effect can mainly be demonstrated in patients with papillary cancer. Crile [3], for example, argued that few, if any, patients with papillary carcinoma derive any benefit from 131I therapy and there are many others who would support his view . I feel, however, that today we have enough evidence that patients with differ entiated thyroid cancer will benefit from radioiodine therapy. With the growing number of patients treated with radioactive iodine and with the prolonged periods of observation of such patients evidence has been mounting that indicates increased survival rates and decreased tum our recurrence in patients receiving radioactive iodine therapy. In 1981, in a series of 576 patients with papillary carcinoma of the thyroid, Mazzaferri and Young [4] found significantly improved prognosis after post-operative treatm ent with iodine-131 and the same authors reported in 1980 [5] that post-operative recurrence rate decreased in 214 patients with pure folli cular thyroid carcinoma after radioiodine and thyroid hormone therapy. Samaan e t al. [6 ] in 1983 presented an analysis of the results of the treatm ent of 706 patients with differentiated thyroid tumours and showed that the frequency of recurrence was significantly less in those patients who were given an ablative dose of iodine-131 after thyroidectomy. In addition to this statistical evidence, we should not forget the individual cases who may dramatically improve on radioiodine therapy even with extensive métastasés. The second argument used against radioactive iodine therapy in differentiated thyroid cancer are the side effects of this therapy. Acute side effects are symptoms of radiation sickness, radiation cystitis, gastritis and sialoadenitis and, of course, swelling and radiation induced inflammation of thyroid remnants and tumour tissue. There may also be some bone marrow suppression. All these side effects are strongly dependent on radiation dose and can, furthermore, be well treated symptomatically, for example by corticosteroids, and cannot as such be used as an argument against radioiodine therapy. IAEA-SM-290/78 165

More attention must be paid to the long term hazards of the treatment of the thyroid cancer with radioactive iodine such as the carcinogenic hazard, and the increase in the incidence of leukaemia and infertility. In 1969 Pochin [7] reviewed a group of patients treated for thyroid carcinoma in this respect and Edmonds and S m ith [8 ] have updated this review very recently. They investigated 254 patients treated with high activity iodine-131 for thyroid cancer and found a small signifi cant increase in death rates by cancer of the bladder and from leukaemia. The considerable gonad irritation was compatible with apparently normal fertility. The obviously very competent authors find it reassuring to see that their observa tions show the relatively low risk of high activity iodine-131 therapy and they argue that the benefits of therapy do not appear to be outweighed by long term harmful effects. Summarizing the above discussion one can say that there is good evidence that a patient with differentiated thyroid carcinoma will benefit from the post operative use of radioactive iodine [9].

3. SELECTION OF PATIENTS FOR 131I THERAPY

The next and probably most important step is now to select patients or groups of patients for radioiodine therapy. Unfortunately, very few rules can be set up here on direct evidence and comparatively much of the planning of the thera peutic strategy will be speculative. We are on safe grounds only when selecting patients according to histology since only differentiated papillary and/or follicular thyroid carcinomas will take up radioiodine. At the time of selection of a patient for radioiodine treatm ent — that is immediately after surgery — in most of the cases unfortunately we cannot measure whether or not the tum our cells take up radioiodine. All we know is that up to about 75% of all differentiated carcinomas potentially concentrate radioactive io dine. A further selection criterion is the fact that follicular tumours are all poten tially functioning and, furthermore, may have already metastasized (without clinical signs or symptoms) by the time the patient seeks medical advice. In patients with purely papillary tumours the arguments in favour of radioiodine therapy are less cogent. The prognosis for these patients when they are young is especially good, provided the tum our has not spread too extensively and there are no regional or distant métastasés. Reiners [ 10] has recently published an interesting mathematical description of survival rates depending on risk factors in cases of thyroid carcinoma such as sex, age, histological tum our type, anaplastic cells, tum our size and distant métastasés using a Weibull function. It might well be possible in future to separate low risk patients from high risk patients with such an approach, a procedure which would greatly help in selecting patients for radioiodine therapy. 166 HÖFER

Finally, it should be pointed out that recently Juhasz et al. [11] have typed thyroid epithelial cell cancer for HLA association. They found that 10 of 12 patients with metastatic disease were DR1 positive compared with only 18 of 41 patients without métastasés. Whether HLA typing will become a selection crite rion, however, remains an open question. In conclusion to the selection problem there is a good case for ablative radio iodine therapy in all patients with follicular tumours. If a patient with a papillary tumour is young and the tum our shows little or no follicular structure, is small and not invasive and if there is no spread to regional lymph nodes, probably no treatm ent will be required after extensive surgery other than suppressive thyroid hormone therapy.

4. PREREQUISITES FOR I31I THERAPY

We have now to consider the steps preceding radioiodine therapy for an opti mal therapeutic effect. There is no question — as in all other cases of malignant disease — that one major point in optimal therapy is early diagnosis. We must not dwell long on the subject since — again as in most malignant diseases — early diagnosis remains a problem. Careful clinical evaluation of the individual case is still of primary importance. Radionuclide scintigraphy — detection of nodules, cold for iodine or technetium, hot for tum our seeking tracers — has improved early diagnosis but not solved the problem. The same is true for fine needle biopsy, ultrasound and specific tumour markers such as thyroglobulin in the first place. One aspect of the early diagnosis of malignant thyroid disease is the careful observation and examination of high risk populations such as people who have received radiation therapy to the upper body as infants, children or even young adults [12]. Another high risk group may be sorted out by HLA typing since there are recent reports on the association of thyroid malignancy with HLA/DR antigens by Panza [13], Sridama [14] and Juhasz [11]. The number of patients examined, however, is too small to give really relevant results and there are also contradictory views such as those published by Dralle et al. [15]. When there is suspicion of thyroid malignancy the next step is surgical inter vention, whatever further treatm ent is planned. While there is no general agree ment as to the extent of surgical therapy, the intention of post-surgical radioiodine treatment favours near total thyroidectomy, together with lymph node resection limited to the extent of lymph node involvement, with preservation of the sternal cleidomastoid muscle and the internal jugular vein. An experienced surgeon can certainly carry out this procedure with low morbidity, preserving endangered nerves and parathyroid glands. As already mentioned, the most important prerequisite of effective radio iodine therapy is exact histological differentiation of the operation specimen with regard to two questions: IAEA-SM-290/78 167

(1) Is the tumour a differentiated thyroid carcinoma or (2) Is it a papillary or a follicular tumour and to what extent is it a mixed type.

It is generally accepted today that the important question of whether the tumour is a differentiated thyroid carcinoma should not be answered without immunohistological examination demonstrating thyroglobulin in the specimen. Before we now return to details about radioiodine therapy itself one more prerequisite must be discussed. To obtain the optimal effect of radioiodine therapy together with minimal risk to the patient and minimal side effects of the radia tion therapy require the tumour tissue to be reduced to a minimum before radio iodine therapy so that the lowest possible dose may be used effectively. This has already been pointed out with the request for near-total thyroidectomy and the removal of affected lymph nodes. However, the maximum reduction of malig nant tissue is also required for distant métastasés and radical surgery before radioiodine therapy includes, at least in our strategy, surgically accessible métastasés too.

5. CALCULATION/ESTIMATION OF 131I DOSE

The efficacy of radioactive iodine therapy, however, is not only directly related to tumour mass but also to tumour uptake and retention. Effective tumour uptake is achieved if there is a concentration of 0.5% of the dose per gram of tum our tissue with a biological half-life of about 4 days and a tumour dose of approximately 25 000 rad (259 Gy) is delivered after administration of 150 mCi (5.5 GBq) 131I [16]. However, this is rough guessing and up to now the amount of radio active iodine selected for ablation of the thyroid remnant at the first therapeutic ‘ablative’ intervention and for therapy later on has been relegated to empirical decisions based on the extent of the disease and the sites of involvement. Doses of iodine-131 for ablation vary from 30 to 100 mCi (1.1-3.7 GBq) and for therapy from 100 to 200 mCi (3.7-7.4 GBq) per dose. However, efficacy can be demon strated in larger dose ranges and exact dosimetry would certainly be better than empirical decisions. The ultimate aim is to know the minimum administered activity necessary to eliminate the disease and to determine this activity from the administration of a small tracer dose before onset of treatment. At the moment, however, this aim is still not within reach. We hoped that with the advent of single photon emission tomography the quantitation of tumour uptake would become possible but recent experiments in our laboratory have shown that the collimators available are not adequate for quantitative measurements for the high energy o f 131I. To illustrate this fact, we have measured and compared the modu lation transfer function of 13II line sources with the high energy collimator of the large field of view (LFOV) camera and with the collimator (medium energy) available for SPECT [17]. As one can see, the modulation transfer function of the medium energy collimator is much narrower, which means a broader line spread 168 HÖFER

FIG. 1. Modulation transfer function \1 7 \

function and therefore much worse resolution than with the high energy collimator (Fig. 1). Furthermore, measurement of the ‘partial volume effect’, that is, the ratio of activity detected at the actual position of the radioactive sample compared with the activity measured in the surrounding camera field, shows clearly that the medium energy collimator of the SPECT is unfit for measuring 131I quantitatively. For practical purposes therefore it seems that for the time being we will have to stick to the empirical dose and in our hands the simple rule of using 80 mCi (3 GBq) as the ablative dose and 150 mCi (5.5 GBq) as the therapeutic dose has been quite effective. An interesting way of calculating the highest tolerable dose is used by Leeper [18], using the rationale of achieving a limiting dose of 200 rad (2 Gy) exposure to the blood. With this approach doses between 70 and 650 mCi (2.6 and 24 GBq) of radioactive iodine have been given without severe side effects but with the high efficacy of large dose ranges.

6 . THERAPEUTIC AND FOLLOW-UP STRATEGY

There is general agreement as to the ablative, therapeutic and follow-up pro cedure once it has been decided to give radioiodine to the patient. Once adequate uptake has been ensured after surgery by radioactive iodine imaging and uptake studies, the ablative dose of radioactive iodine is administered between 4 and 6 weeks after surgery. Thyroid hormone suppressive therapy is started immediately thereafter using levo-thyroxine. The suppressive dose of T4 completely inhibiting TSH secretion lies between 150 and 200 ¿ug daily. Inhibited TSH secretion must be controlled by using a TSH determination of the highest sensitivity or by performing a TRH test. Further therapeutic doses of radioiodine IAEA-SM-290/78 169 are usually given at intervals of 3 to 6 months until no residual thyroid or function ing tum our tissue is demonstrable. For these follow-up studies thyroid hormone suppression therapy with thyrox ine is stopped 4 to 6 weeks before follow-up. With tri-iodothyronine it is sufficient to interrupt treatment for only 14 days. The search for iodine retaining tumour tissue or métastasés can be carried out either by whole-body scanning with a test dose of between 2 and 10 mCi or by measuring the whole-body iodine-131 reten tion of a low tracer dose in the microcurie range. Finally, the presence or absence of differentiated thyroid carcinoma tissue may also be tested by determining the level of circulating thyroglobulin since thyroglobulin is produced by thyroid epithelial cells only and therefore is a highly specific tum our marker. It has been discussed frequently what is the best method for such follow-up studies and we have compared results obtained with measuring whole-body retention of radio iodine 72 hours after application of the tracer dose with a shadow shield whole- body counter with the results of thyroglobulin serum measurements — the reference method for both approaches was whole-body scanning after the therapeutic dose of radioiodine. Our data showed that while the sensitivity of the thyroglobulin serum concentration with respect to the detection of tumour tissue is higher than measurements of whole-body retention, the sensitivity is further increased by using both methods together [19]. Since, furthermore, the whole-body measure ment is the only one of the two methods that gives information on the uptake of radioiodine in tum our tissue, the ideal follow-up procedure is to use both methods. There are many authors who also advocate the simultaneous use of thyroglobulin measurements and some method for determining radioiodine uptake in tumour tissue. It must not be forgotten that the diagnostic reliability especially of the radio activity measurements must be documented by showing that the thyroid hormone level is low, the thyrotropin secretion high and that there is no iodine contamina tion present. Withdrawal of suppressive thyroid hormone therapy results in 4 weeks not only in the characteristic clinical picture of hypothyroidism but also leads to changes in biochemical parameters such as an increase in creatine-phosphokinase,an increase in transaminase and others [20]. The discomfort of the patient and such bio chemical changes make it desirable to do the follow-up examination while the patient is still on thyroid hormone therapy. Most authors agree, however, that while radioactive uptake tests are per se impossible, thyroglobulin measurement is also not reliable when performed while the patient is receiving suppressive doses of thyroid hormone. The use of thyroglobulin combined with thallium whole-body scanning has been suggested recently [21 ] as an alternative but further studies on this subject are necessary. In principle, as everybody agrees, the patient must be followed up for the rest of his life since late recurrences and the late appearance of métastasés are not uncommon. This poses a special problem with respect to what has just been said 170 HÖFER

FIG. 2. Survival probabilities for the four predominant histological tumour types [22].

about the side effects of taking the patient off thyroid hormone therapy and the general attitude for the long term observation of the patient is the following: Once total radioactive iodine ablation of tumour tissue has been achieved the patient will be followed up for another 3 to 5 years according to the follow-up scheme described above. Thereafter, one may try a follow-up strategy using con ventional methods together with thyroglobulin, however, while the patient is still on thyroxine therapy. The final outcome of the individual case will depend very much not only on the efficacy of diagnosis, surgery and follow-up therapy with radioiodine and other adjuvant therapy such as surgery of métastasés or external irradiation and chemo therapy, but also greatly on perfect organization of the periodical recall of the patient, the close co-operation with the family doctor and, of course, the co-operation of the fully informed patient himself. It is essential that a specialized team be concerned with the thyroid cancer patients and most of the follow-up should preferably be done on an out-patient basis. The team must include all specialists possibly concerned with the patient. Let me finish with the retrospective analysis of patients and results of treat ment of our Division of Nuclear Medicine in Vienna from 1973 to 1982 [22]. The following 5 year probabilities of survival using the life table method have been obtained (Fig. 2): 134 patients with differentiated papillary thyroid carcinoma showed a survival probability of 0.96 ± 0.02, and 133 patients with differentiated follicular thyroid carcinoma of 0.92 ± 0.03. These data are in good agreement with all other data reported from other centres and certainly demonstrate how reward ing radioiodine treatm ent in differentiated thyroid carcinoma is for the patient and for the treating physician. IAEA-SM-290/78 171

REFERENCES

[1] FREITAS, J.E., GROSS, M.D., RIPLEY, S., SHAPIRO, B., Semin. Nucl. Med. 15 (1985) 106. [2] POCHIN, E.E., in Rational Diagnosis of Thyroid Disease (HOEFER, R., et al., Eds), H. Egermann, Vienna (1977) 169. [3] CRILE, G., Jr, J. Am. Med. Assoc. 195 (1966) 721. [4] MAZZAFERRI, E.L., YOUNG, R.L., Am. J. Med. 7 0 (1981) 511. [5] YOUNG, R.L., MAZZAFERRI, E.L., RAHE, A.J., DORFMAN, S.G., J. Nucl. Med. 21 (1980) 733. [6] SAMAAN, N.A., et al., J. Clin. Endocrinol. Metab. 56 (1983) 1131. [7] POCHIN, E.E., in Thyroid Cancer (HEDINGER, E., Ed.), UICC Monogr. Ser., Vol. 12, Springer Verlag, Berlin (1969) 293. [8] EDMONDS, C.J., SMITH, T., Br. J. Radiol. 59 (1986) 45. [9] GOOLDEN, A.W.G., Clin. Endocrinol. 23 (1985) 81. [10] REINERS, C., REIM ANN, J., RENDL, J., BOERNER, W„ Nucl.-Med. 25 (1986) A21. [11] JUHASZ, F., BALASZ, G., STENSZKY, V., KOZMA, L., FARID, N.A., Cancer 58 (1986) 52. [12] DUFFY, B.J., Jr., FITZGERALD, P.J., Clin. Endocrinol. 10 (1950) 1296. [13] PANZA, N., et al., Tissue Antigens 20 (1982) 155. [14] SRIDAMA, V., HARA, Y., FAUCHET, R., DE GROOT, L.J., Cancer 56 (1985) 1086. [15] DRALLE, H., ROBIN-WINN, M., REILMANN, L., LAUE, A., TÖRÖK, M., Klin. Wochenschr. 64(1986) 522. [16] BLAHD, W.H., Semin. Nucl. Med. 9 (1979) 95. [17] BERGMANN, H., HOEFER, A., personal communication. [18] LEEPER, R.D., SHIMAOKA, K., Clin. Endocrinol. Metab. 9 (1980) 383. [19] WEISSEL, M., BERGMANN, H., HOEFER, R., Radioaktive Isotope in Klinik und Forschung (HOEFER, R., BERGMANN, H., Eds), H. Egermann, Vienna (1984) 681. [20] WEISSEL, M., KAINZ, H., HOEFER, R., J. Nucl. Med. 27 (1986) in print. [21] FRITZSCHE, H., HUGL, H., KARGL, M., WEISS, P., in Radioaktive Isotope in Klinik und Forschung (HOEFER, R., BERGMANN, H., Eds), H. Egermann, Vienna (1986) 585. [22] WEISSEL, M., BERGMANN, H., STEIDLER, B., Acta Med. Austriaca 1 0 (1983) 83.

IAEA-SM-290/77

CURRENT PROSPECTS FOR 12SI

IN TUMOUR THERAPY

L.E. FEINENDEGEN, W. BOECKER, J. BOOZ, E. POMPLUN, W. SCHADEN, F.H.A. SCHNEEWEISS, C.A. SONDHAUS, G. TISLJAR-LENTULIS Institut für Medizin der Kernforschungsanlage Jülich GmbH, Jülich, Federal Republic of Germany

Abstract

CURRENT PROSPECTS FOR 125I IN TUMOUR THERAPY. To evaluate 125I as a tool in tumour therapy, microdosimetric data, results from animal experiments and information obtained from clinical applications are presented. Numbers and energies of electrons and photons emitted were calculated to determine the energy deposition in critical biological targets (spheres of 20 nm diameter) after incorporation of 125IUdR. The total energy deposited per decay includes the energy imparted to Auger electrons (0.96 keV) and the ionization potential ( 1.07 keV) which is built up in consequence of the electron cascade. The latter will be locally absorbed. Thus, 125I acts like a high-LET radiator. In animal experiments tumour bearing mice were continuously injected intravenously with cold IUdR and I25IUdR. Tumour regression improved from 10% for the controls to 25% after infusion of 2 X 10”4 mol/L IUdR and to 31% after infusion of 200 ¿tCi of 12SIUdR per mouse, respectively. In clinical applications two patients with advanced squamous carcinomas of the tongue and buccal cavity were locally intra-arterially perfused with 125IUdR. To the first patient 50 pCi 12sIUdR were administered via the arteria lingualis every 12 h followed at day 12 by 500 juCi 125IUdR per 12 h period. In the second patient ‘interport’ catheters were implanted in the arteria carotis externa. Twenty-five pCi were infused 10-18 h daily for 44 days, to a total of 1080 pCi. Retained activities remained below the expected levels in both patients and the treatments were discontinued. To render more promising the application of 12SI in tumour therapy it appears necessary to specifically augment n sIUdR uptake in tumour DNA.

1. IN T R O D U C T IO N

Iodine-125 is a powerful toxic agent when incorporated in DNA [1]. It may thus one day become an effective tool in tumour therapy when applied in the right manner. Microdosimetry provides the basis for dose evaluations [2]. Encouraging results from animal experiments with 125lUdR suggest the pursuit of its application in medicine [3]. Data on clinical experience are presented which provide information for improvements of 1251 application for the benefit of the patients.

173 174 FEINENDEGEN et al.

2. MICRODOSIMETRY

The isotope 1251 decays by electron capture and internal conversion subsequently followed by Auger electron cascades. To determine the energy deposition in microscopic biological volumes after incorporation of 12SIU d R , Monte-Carlo simulations of the decay and the subsequent Auger cascades have been performed, yielding the numbers and energies of emitted electrons and photons. By using a new method of calculating the kinetic energies of the emitted particles [4] based on a relativistic Dirac-Fock computer code [5], it is possible to distinguish between two terms of the total energy deposited per decay: (i) the energy imparted by Auger and Coster-Kronig electrons; and (ii) the ionization potential which has been built up on average at the end of the cascade as a consequence of the multiple ionization. This charge will be neutralized in condensed phase like biological tissue and the corresponding energy of 1.07 keV will be locally absorbed. The results for the energy deposited in spherical target volumes of different size homogeneously loaded with 1251 isotopes demonstrate that in small targets the ionization potential is the dominant contributer to the total energy deposition (Fig. 1). In an assumed critical target o f 20 nm diameter (section of basic chromatin fibre) the mean energy deposition of Auger electrons (0.96 keV) plus the ionization potential (1.07 keV) result in

FIG. 1. Energy imparted by homogeneously distributed 12SIatoms in spheres as a function o f sphere diameter; dashed line: energy imparted by the emitted Auger electrons; solid line: condensed phase. The constant gap between the curves represents the ionization potential remaining on the atom after emission o f the electron cascades. The ostensible convergence o f the curves with increasing target size results from the logarithmic scale o f the ordinate. IAEA-SM-290/77 175

FIG. 2. Device for long term intravenous infusions of mice. Top: fixation of the mice in standard macrolon cages; bottom: section o f the plastic tube casing the infusion catheter fixed in a vein o f the mouse tail.

a total value of 2.03 keV per decay. When characterizing the radiation quality by the microdosimetric quantity mean lineal energy, yp, it can be shown th a t 1251 incorporated in a critical target of 20 nm diameter is a high-LET radiator (L = yp = 270 keV-jum "1 ). These findings support experimental radiobiological results on the toxicity of 125I in DNA.

3. ANIMAL EXPERIMENTATION

Cold IUdR and 125IUdR were infused into sarcoma 180 bearing female NMRI mice for a period of 10 days in order to study the effects of these substances on tum our growth and regression when infused continuously. A device was built which makes it possible to intravenously infuse mice with solutions of cold or radioactive IUdR for periods up to several weeks (Fig. 2). 176 FEINENDEGEN et al.

10 20 30 40 50 DAYS AFTER INOCULATION OF 2«105 TUMOUR CELLS FIG. 3. Patterns o f growth and regression o f sarcoma 180 in NM RI mice; percentage distribution o f control tumours.

Four days before the onset of the infusions the mice were inoculated with 3 X 10s tumour cells in 0.1 mL solution into the right hind leg. Cold or hot IUdR was infused in concentrations ranging from 2 X 10 '10 to 2 X 1СГ4 m o l/L , 5 mL per mouse per day. The highest to tal1251 activity administered per mouse was 200 pC i. After termination of the infusions tumour sizes were measured every 4 days with a caliper until the animals died. The percentage of delayed growing and completely regressing tumours increased significantly with rising concentrations of both cold and 125IUdR in the infused solutions. Four patterns of tum our growth and regression were observed (Fig. 3): Group I followed a standard kinetic curve. The volumes of the tumours increased monotonously. After a steep rise, gradual transition to saturation occurred. In Group II an intermediate plateau was observed. Group III showed temporary tum our regression after a first maximum, followed by a minimum of volumes before another rise led to saturation. Group IV represents complete tumour regression. A pronounced maximum showed after a short steep rise followed by a decline to non-palpable tumours. The patterns of curves I to IV are highly significant and remained the same independent of the amount of infused IUdR. It was the percentage of delayed growing and of completely regressing tumours that increased significantly with rising amounts of IUdR. Percentage values represented by curves I to IV changed from 67, 17, 6 and 10% for the controls to 12, 44, 19 and 25% after infusion of 2 X 10 ~4 mol/L cold IUdR. After infusion of 20 or 200 juCi of 125IUdR per mouse the number of mice following curves I to IV changed from the control values to 63, 21, 12 and 4%, and to 23, 31, 15 and 31%, respectively (Table I). There is reason to assume that the data in Table I show the effects on tum our cell DNA of both IUdR per se and an additional effect of 125lUdR if the totally infused amount of radioactivity exceeds perhaps 10 0 ßC i. This is probably due to the toxicity of IUdR compared with TdR which it replaces in part and/or, IAEA-SM-290/77 177

TABLE I. PATTERNS OF GROWTH AND REGRESSION OF S-l 80 IN NMRI MICE; PERCENTAGE DISTRIBUTION OF TREATED TUMOURS COMPARED WITH CONTROLS AFTER LONG TERM INFUSIONS OF COLD IUdR O R 125IU d R

Growth patterns Tumour treatment I II III IV

Controls 67 17 6 10 2 X lCT4 mol/L IUdR 12 44 19 25 20/JCi 125IUdR (2 X 10“10 mol/L) 63 21 12 4 200 ßCi 125IUdR (2 X 10“9 mol/L) 23 31 15 31

in the case of 12SI application, of high LET resembling radiation and molecular decomposition owing to 125I decay in tum our cell DNA. Presumably, maximal amounts of IUdR which correspond to a thymidine replacement of about 50% could be incorporated into tumour DNA by long term infusions [6 ]. T his is ‘ also of interest in view of a possible application of cold IUdR in photon activation therapy [7—9].

4. CLINICAL DATA

Preliminary and limited clinical trials have been made to investigate the feasibility of locally intra-arterially perfused 12sIUdR as a tum our therapeutic agent. The main objective of these trials was to investigate tumour uptake and possible bone marrow toxicity. Two patients with advanced, incurable squamous cell carcinomas of the tongue and buccal cavity were treated. The plan was to proceed as long as tum our control or regression was evident but, if further growth occurred, to terminate the IUdR treatment and begin alternative therapy. Both patients had neck dissection to reduce the probability of lymph node spread and were cannulated beforehand. As a first approximation it was assumed that the tum our mass in the first patient was 100 ± 50 g, that the growth fraction might be 20%, and that a total of 2000 1251 decays per cell would be required for >99% probability of cell death. On this basis, it was decided to administer 10 ßC i per hour via the arteria lingualis continuously for the first 24 h, followed by 100 juCi every 24 hours thereafter. In an effort to increase tum our uptake, this schedule was changed 178 FEINENDEGEN et al. after 4 days to 50 ¡iCi 12SIUdR + 2 mg 5-fluorouracil (5FU) every 12 h, alternating with 12 h of 0.1% glucose perfusion. At day 12 the activity was increased to 500 juCi IUdR + 10 mg 5FU per 12 h period, but urine recovery and external measurement indicated that the retained activity still remained far below the expected level. At 14 days the total activity administered had reached 2190 ß d and treatm ent was discontinued, since the chosen limit based on possible bone marrow toxicity would be exceeded in 20 days without evident increase in the tumour uptake. No tumour growth was observed during the three month observation period which followed. However, growth resumed thereafter and the patient was operated at four months post-treatment. It was concluded that either the growth fraction had been too low or that the perfusion technique had not produced appreciable uptake in the tum our cells, since over 99% of the administered activity had been excreted at 30 days. Tumour biopsy material taken at the time of operation showed no detectable activity in the cell suspension. Based on the experience with the first patient, an attem pt was made to change the catheterization technique in order to permit perfusion for a longer period with a smaller amount of activity. ‘Interport’ catheters were implanted in the arteria carotis externa on both sides. The tumour mass was comparable to that in the first patient, and approximately 25 дСл were perfused 10—18 h daily for 44 d. By this method 1080 ß d were delivered, but only 870 pC i w ere collected in the urine. However, total body counting 14 d after the end of the perfusion showed that the uptake in the tumour had not exceeded 50 nCi. This was confirmed by impulse infusion using a double labelling technique combined with biopsy and DNA extraction (0.18 nCi/1.24 g). Based on the latter measure ments, it was estimated that an effective infusion would take too long to be practical, and it was therefore decided to stop the treatment. During the next four months no tumour growth was observed. However, tumour growth then recurred and the patient was treated with chemotherapy followed by surgery. At surgery an activity determination was made in the DNA extracted from representative tumour sites and none was detected. Chromo some analysis of peripheral blood lymphocytes did not indicate any radiation effect.

5. C O N C L U SIO N

The clinical data presented resulting from long term infusions of tumour patients with 125IUdR strongly suggest that ways and means must be found to increase the 1251 uptake considerably in tum our cell DNA before any final conclusion can be drawn with regard to the therapeutic value of 12s IU dR . Additional infusion of flourodeoxyuridine may prove to be useful since thymidine competes with iododeoxyuridine for incorporation into DNA, and FUdR blocks IAEA-SM-290/77 179

TdR synthesis. Further, additional infusion of cytidine may be helpful since IUdR suppresses CdR synthesis [6 ]. ,25I damage to the bone marrow would probably not be a prohibitive factor for 1251 perfusion therapy.

ACKNOWLEDGEMENTS

The authors wish to thank Mrs. M. Meurer and Mr. A. Wiertz for skillful technical assistance. The mouse infusion set-up was designed by F. Wirtz within the framework of his doctoral thesis.

REFERENCES

[1] FEINENDEGEN, L.E., Radiat. Environ. Biophys. 12(1975) 85. [2] CHARLTON, D.E., BOOZ, J., Radiat. Res. 87 (1981) 10. [3] SCHNEEWEISS, F.H.A., KRONENBERGER, P.H., TISLJAR-LENTULIS, G., FEINENDEGEN, L.E., Proc. 7th Int. Congr. Radiation Research (BROERSE, J.J., BARENDSEN, G.W., KAL, H.B., VAN DER KOGEL, A.J., Eds), Martinus Nijhoff Publ., Amsterdam (1983) D2. [4] POMPLUN, E., BOOZ, J., CHARLTON, D.E., submitted to Radiat. Res. [5] DESCLAUX, J.P., Comp. Phys. Commun. 6 (1975) 31. [6] COMMERFORD, S.L., personal communication. [7] FAIRCHILD, R.G., BRILL, A.B., ETTINGER, K.V., Rep. BNL-29616, Brookhaven Natl Lab., Upton, NY (1982). [8] FAIRCHILD, R.G., BOND, V.P., Rep. BNL-31401, Brookhaven Natl Lab., Upton, NY (1982). [9] TISLJAR-LENTULIS, G., FEINENDEGEN, L.E., BOND, V.P., Strahlentherapie 145 (1973) 656.

DISCUSSION

(Summary of discussion held on Papers IAEA-SM-290/5, 47, 10, 19, 8, 33,39,56,11 and 78)

In response to the questions of several speakers, K.M. Pagliero (Paper IAEA-SM-290/5) pointed out that the criteria for selection of patients for intra cavitary irradiation were exactly the same as would apply for intubation; that tubular stenosis was far more common in papillaries; that a laser beam was not used in the case of oesophagus tumours; that intubation was not given prophylactically but only in cases when dysphagia had returned subsequently; that, if adenocarcinoma of the lower part of the oesophagus was to be treated, a special method of application should be used; that a tracheoesophagic fistula was noted in one patient with tumour extension into the trachea, but no tracheoesophagic fistula developed when the tum our was limited to the oesophagus; that it was a pilot study and no combination of intracavitary and external therapy was made; that the tumour dose was defined at the distance o f 1 cm from the source; and that in cases where gastric nodes were positive, brachytherapy probably offered more than surgery. In response to questions from several speakers, D. Jullien (Paper IAEA-SM-290/47) said that half-body irradiation given in fractionated form had the following advantages over a single dose: (1) There was no acute radiation syndrome; (2) Hospitalization and resuscitation were unnecessary, and there was no medical mobilization; (3) The patient was treated on an out-patient basis; (4) The irradiation time was of the order of two or three minutes per session and twenty or thirty minutes in total, and there was a time gain compared with the single dose which mobilizes the equipment for several hours; (5) The results were equivalent, if not better; the supplementary irradiation was given immediately after the half-body irradiation if there was still residual pain; the mechanism by which half-body irradiation had an effect was probably different from that of localized irradiation; the results of the whole-body instead of the half-body irradiation in two stages were satisfactory, but some major haematological problems in those patients with a burdensome therapeutic prehistory occurred; half-body irradiation bèfore chemotherapy would be a very interesting line of research with expectation of less bone marrow suppression. Answering questions, T. Lanché (Paper IAEA-SM-290/10) pointed out that in the case of stage III, owing to the infiltration of lymph nodes, only half of the abdomen, that is to say the tumour bed, was irradiated; that the upper level of the field was restricted by the diaphragm and the lower level by the obdurator foramen and included vertebrae; that in stage III owing to massive residual disease the whole abdomen had to be irradiated with protection of the contralateral kidney and ovary; that survival in the favourable group of patients was 78% as against 45% in the unfavourable group.

181 182 DISCUSSION

In response to questions, C. Gonzáles-Miranda (Paper IAEA-SM-290/19) mentioned that he did not think that there were any difficulties in studying the correlation between the results of clinical assessment of the efficiency of pre operative radiotherapy and post-operative pathological evaluation, and that he did not use the technique of brachytherapy with 192Ir. In response to questions, M.B. Patricio (Paper IAEA-SM-290/8) pointed out that the trial had been done according to the protocol proposed by EORTC for clinical stage I, II of non-Hodgkin’s lymphomas (NHL) patients; that complete regression had been assessed in high risk patients; that the assessment was rather expensive and did seem to be reasonable if the risk of failure could be prevented with radiation alone. She also said that prophylactic cranial irradiation in children with clinical stage I and II NHL had not been used; that adjuvant chemotherapy in clinical stage II of NHL had been given in all cases with unfavourable histology and/or bulky disease or with two or more involved areas (II3, II4). In response to questions from several speakers, J.-M. Deneufbourg (Paper IAEA-SM-290/33) said that the neoadjuvant chemotherapy was given for all sites of head and neck tumours; that, according to the definition, the neoadjuvant chemotherapy was given before irradiation or surgery; that concomitant chemo therapy (bleomycin plus telecobalt therapy) had been tried at another time in tonsil cancer, but no benefit had been obtained and, moreover, tolerance had been very poor in the combined treatment group. He also said that the response to chemotherapy had been assessed very early, eight days after the last injection, but the rate of response might be higher with a longer delay; that chemotherapy did not necessarily induce more response to radiation but perhaps merely indicated a group of tumours (or patients) which were likely to be presenting complete regression under radiotherapy; that the poor responders to chemo therapy might need more aggressive, radical treatm ent and that the target volume of radiotherapy had not been reduced whatever the degree of chemotherapy response; and that the resection usually had encompassed the primary margins of the tumour but sometimes a less mutilating operation had been possible, e.g. salvage of mandibula. Answering questions, R. Uzel (Paper IAEA-SM-290/39) pointed out that a patient was treated in a sitting position in a chair which was equipped with a head fixation device; that the tubes inserted into the nasopharynx were individually fixed by mechanical clamps; and that it was not possible to have a post-treatment film and, therefore, objective proof of the setting up was not available. However, a mark was put into the nostril and thus a rough check could be made after treat ment. R. Uzel said that he was trying to stay away from using nasopharyngeal moulds as much as possible since they were usually not tolerated by patients when general anaesthesia was not used; that the reference point at 2 cm from the source axis was an arbitrary point which would hopefully correspond with the lateral walls and the roof of the nasopharynx; that the reference isodose DISCUSSION 183

curye passing from those points would receive 24 Gy and the doses to the other critical points could be computerized from isodose curves; that everything was done to avoid overdosage to nasopharyngeal structure; that usually 60 Gy in six weeks or 70 Gy in 7—7.5 weeks was given by external irradiation; that in order to control a reaction of the oral mucosa which prohibited the normal nutrition of patients a pause in irradiation sometimes had to be given for one to two weeks. He continued that in some cases TDF calculations were done to adjust the time- dose effect; that failure of the treatment regarding the neck nodes occurred in 6.3% (N0), 18.2% (Nj), 35.3% (N2), and 50% (N3) of cases. If the tumour in the nasopharynx showed a complete response but the neck failure was persisting two months after the treatment, resective surgery or chemotherapy was given. In response to questions, J. Luande (Paper IAEA-SM-290/56) stressed that he entirely agreed that in the case of advanced lesions of carcinoma of the cervix the combination of brachytherapy and external radiation therapy was required. In response to questions, V.H.J. Svoboda (Paper IAEA-SM-290/11) pointed out that more individual characteristics of the tumour were needed to select suitable ones for accelerated fractionation; that it would be better to study and to identify those tumours which proliferated fast; that slow proliferating breast cancers could also respond to accelerated fractionation and that this could be explained by the existence of a specific relationship between normal tissues and tumours in their response to accelerated fractionation. In response to the questions of several speakers, R. Höfer (Paper IAEA-SM-290/78) pointed out that the extent of surgery was always determined by the result of normal histology and if there was a contradiction between frozen section biopsy and normal histology the case was reoperated; that larger doses of radioiodine did seem to be preferable to repeated small single doses because they eventually became very high even in the case of ablation; that medulary thyroid carcinoma did not take up radioiodine and, therefore, usually extensive surgery had to be undertaken; that labelled monoclonal thyroid globulin anti bodies did not show good results in tumour detection, probably because of a very high content of thyroid globulin in the serum, and the results of the trial of labelled calcitonin antibodies were not available to make a definite decision; that according to the published data the radioiodine carcinogenesis lasted approximately 15 months for leukaemia and from 10 to 20 years for solid tumours (breast, bladder cancer).

HYPERTHERMIA

IAEA-SM-290/69

CURRENT STUDIES OF HYPERTHERMIA AND HYPOXIC CELL RADIOSENSITIZERS IN CLINICAL ONCOLOGY

N.M. BLEEHEN, G.C.W. HOWARD, H.F.V. NEWMAN Department of Clinical Oncology and Radiotherapeutics, Addenbrookes Hospital, Cambridge, United Kingdom

Abstract

CURRENT STUDIES OF HYPERTHERMIA AND HYPOXIC CELL RADIOSENSITIZERS IN CLINICAL ONCOLOGY. Two methods to overcome the resistance of hypoxic tumour cells to ionizing radiation are being investigated. Hypoxic cells have been shown to be as sensitive to hyperthermia as oxic cells. For this and other reasons there is a synergistic interaction between radiation and hyperthermia. Local superficial hyperthermia has been combined with low dose radiation in the treatment of superficial tumours, in a non-randomized study. The addition of hyperthermia resulted in a significant increase in overall tumour response. An increased incidence in normal tissue toxicity was also noted. The feasibility of pelvic regional hyperthermia using an annular phased array applicator has been assessed. Using this technique therapeutic temperatures have been achieved in the majority of treatments, but are difficult to maintain owing to the acute toxicity of the treatment. Both local and regional hyperthermia are now being further assessed in randomized studies. Hypoxic cell radiosensitization showed early promise from the studies using hyperbaric oxygen and metronidazole. Extensive evaluation of misonidazole showed that its usefulness was limited by toxicity. The newer agents Ro 03-8799 and SR 2508 are considerably less toxic; used initially individually, and more recently together, encouraging patient tolerance and tumour concentrations have been seen. These developments are discussed in detail.

1. INTRODUCTION

One of the reasons why radiation may fail to control localized tumours is that they contain radioresistant hypoxic cells. A number of different techniques have been investigated in an attempt to overcome this problem. Two methods currently being studied at Addenbrookes Hospital, Cambridge, are the use of hypoxic cell sensitizers and hyperthermia.

2. HYPERTHERMIA

The biological basis for hyperthermia has been reviewed by many authors [1]. In vitro experiments have shown that hypoxic cells are as sensitive to hyperthermia as fully oxygenated ones. Conditions often found in conjunction with hypoxia

187 188 BLEEHEN et al. such as low pH and poor nutrition have been shown to confer increased thermo sensitivity. Radioresistant ‘S’ phase cells are as sensitive to heat as cells in other parts of the cell cycle. All these factors suggest that hyperthermia may be cytotoxic to those cells which are less sensitive to radiation. The repair of potentially lethal and sublethal radiation damage is also inhibited by hyperthermia. Radiation and hyperthermia thus interact in a synergistic rather than purely additive way. Both localized superficial and regional hyperthermia are being investigated at Cambridge. Following initial feasibility studies using locally developed equipment, human treatments have been performed using the BSD 1000 clinical hyperthermia system.

3. SUPERFICIAL HYPERTHERMIA

Sixteen patients with one or more superficial malignant lesions have been treated with hyperthermia and radiotherapy. Where two or more lesions were assessable in one patient, selected lesions were treated with radiotherapy and hyperthermia, the others were treated with radiation alone. Twenty lesions received the combined modality, 21 radiotherapy alone. Where possible (in 36 of the 41 treated lesions), a radiation dose of 24 Gy was given in 6 twice weekly fractions. Lesions selected for hyperthermia were treated immediately (within 30 minutes) after each radiation fraction. The desired treatm ent was to raise the minimum tumour temperature to 43°C for 60 min. Heat doses were calculated for each treatm ent as minutes equivalent at 43°C (min eq. 43°C) [2]. In 57% of the 80 treatments a heat dose in excess of 30 min eq. 43°C was achieved.

TABLE I. TREATMENT LIMITING TOXICITY DURING 74 REGIONAL HYPERTHERMIA TREATMENTS

Toxicity No. of times recorded

Generalized patient discomfort 32 Pelvic discomfort 17 Systemic hyperthermia (core temperature >39.5°C) 9 Tachycardia (pulse >150) 5 Hypertension and hypotension 5 Excessive normal tissue temperature (>43°C) 2 None 2 IAEA-SM-290/69 189

Thirty per cent of treatments were unsatisfactory with a heat dose of less than 15 min eq. 43°C. Small lesions (less than 12 cm2) received better quality heat treatments and were more likely to achieve a complete response. The addition of hyperthermia to radiation significantly increased the overall (complete and partial) response rate (p < 0.01). The heated group did, however, have an increased incidence of severe skin reactions with moist desquamation occurring in 7 of the heated lesions compared with none of the radiation alone controls. If the lesions with in-patient controls are analysed separately, a significant increase in response (p < 0.025) at the expense of increased toxicity is maintained. The incidence of severe skin reactions may be related to the large radiation fraction size (4 Gy) and the close timing of the two modalities. Both of these factors have been shown to increase the skin reaction [3]. We are at present attempting to confirm these encouraging results in a randomized trial whilst comparing different numbers of hyperthermia fractions.

4. REGIONAL HYPERTHERMIA

Pelvic regional hyperthermia has been performed using the BSD annular phased array applicator (APA). Twenty-two patients have been treated in a pilot study. All patients had extensive pelvic tumours, most of which had received prior radiation and/or chemotherapy. Patients received radiation doses varying from 12 to 65 Gy depending upon remaining radiation tolerance. Hyperthermia treatments were given immediately after radiation (within 60 minutes) on a once or twice weekly basis. Seventy-four treatments have been performed. Where possible, intratumour temperatures have been measured by tracking thermistors along catheters placed under computerized tomographic control. In addition, temperatures have been monitored in the bladder, rectum and vagina. An oral thermistor was used to monitor core temperature. Temperatures in excess of 42°C were reached at sites within the pelvis in 78% of the treatments. Such temperatures were, however, poorly tolerated and accumulated heat doses were low. In only 41% of treatments was a maximum intrapelvic heat dose of 5 min eq. 43°C exceeded. Almost all treatments were limited by the acute toxicity of the procedure, which, although marked, was usually of short duration (Table I). Tumour response and late toxicity are difficult to assess owing to the variable nature of the prior treatments of these patients. We have, however, noted no excessive late toxicity, and palliation has been achieved in all patients with symptoms. We conclude that the APA is capable of inducing pelvic hyperthermia but that it is a time consuming and often poorly tolerated treatment. A randomized study is now under way to assess the value of pelvic regional hyperthermia with radical radiotherapy in the primary management of extensive rectal tumours. 190 BLEEHEN et al.

5. CHEMICAL RADIOSENSITIZATION

Following the demonstration of hypoxic cell radiosensitization using hyperbaric oxygen [4, 5], attention turned to the use of drugs as a more practical solution to this aspect of radioresistance. Although Urtasun [6 ] demonstrated that glioblastoma can be sensitized using metronidazole, drug toxicity limited its administration to a 9-dose regimen. On this schedule, metronidazole was only capable of improving suboptimal radiotherapy to the level of the best conventional fractionation. The widely tested successor misonidazole has not fulfilled initial expectations. Only 5 of 35 randomized trials have shown any benefit for this agent, though in several more studies there was a (non-significant) margin of benefit [7]. The potential of misonidazole is limited, as the maximum tolerated total dose is 12 g/m2, owing to the occurrence of peripheral neuropathy. The search for agents with a superior therapeutic potential has led to the development of two rather different 2-nitroimidazoles, which have potential both individually and in combination. The Roche agent Ro 03-8799 is a lipophilic compound, possessing a basic side chain. In a series of 52 patients [8 ], the dose limiting toxicity was an acute syndrome consisting of flushing, sweating, dizziness, nausea, taste abnormalities and affective mental changes, lasting 30 to 60 min. There was no chronic toxicity or peripheral neuropathy. The maximum tolerated single dose was 1 g/m2, and thrice weekly doses of 0.75 g/m 2 were readily tolerable. An attempt to reduce toxicity by splitting the racemic mixture into its enantiomers has not proved successful, both enantiomers demonstrating equal toxicity [9]. The mean plasma elimination half-life was 6.1 h, and this remained constant throughout multidose schedules. Ro 03-8799 was found to concentrate in tissue. In a series of 20 patients receiving 1 g/m 2 Ro 03-8799 before biopsies or lobectomy, mean tissue to plasma ratios for non-brain tumour, brain tumour and normal brain were 355, 329 and 381%, respectively. The mean absolute drug concentration in tumour was 62.6 jug/g (0.24 mmol), being 1.5 times greater than would be expected for the same administered dose of misonidazole. Furthermore, at least 1.5 times as many doses may be given. The prediction of sensitizer enhancement ratio (SER) is difficult, but in vitro evidence using V79 cells suggests that Ro 03-8799 is likely to be at least 2.1 times as efficient a sensitizer as misonidazole [10]. Growth delay experiments in mouse fibrosarcoma predict an efficacy of 1.8 times that of misonidazole at concentrations of 0.25 mmol. Overall, a single dose SER of at least 1.4 would be a reasonable expectation. The NCI agent SR 2508 is a hydrophilic 2-nitroimidazole. It produces few acute symptoms, and single doses of >3 g/m 2 are tolerable. Peripheral neuropathy limits the total dose to between 30 and 40 g/m2, the larger dose being tolerated over 5-6 weeks [11]. The plasma elimination half-life is around 5.5 h, and a tumour to plasma ratio of 60-100%, similar to that of misonidazole, is seen [ 11]. Single dose SER values of around 1.4 are expected. IAEA-SM-290/69 191

These two agents with differing toxicity spectra have been administered together in an effort to achieve greater SER without increasing toxicity. Ten patients were treated with escalating single doses and, using 0.75 g/m 2 Ro 03-8799, four patients experienced the known acute syndrome. This was not exacerbated by accompanying doses of SR 2508 up to 3 g/m2. Following this, multiple doses were administered, initially on a 9-dose schedule treating thrice weekly. A dose of 0.75 g/m 2 Ro 03-8799 was combined with 0.5, 1.0, 1.5 and 2 g /m 2 SR 2508, treating two patients at each dose level. The acute Ro 03-8799 syndrome was seen in all patients, but was not exacerbated at the higher doses of SR 2508. Proceeding to a 12-dose, thrice weekly schedule, 0 .75 g /m 2 Ro 03-8799 was combined with 2 and 2.5 g/m 2 SR 2508. Four of the five patients experienced Ro 03-8799 toxicity, and two of the three patients receiving 30 g/m 2 SR 2508 experienced peripheral neuropathy, WHO grade 1. In both patients the neuropathy had resolved 3 months after cessation of SR 2508. The mean plasma T values for Ro 03-8799 and SR 2508 were 5.2 and 5.4 h, respectively. They are similar to the values expected for either agent administered alone, and remained constant during the multiple dose schedules. The combined radio sensitizer concentrations achieved in a series of 6 su p er ficial tumours biopsied after dosing show encouraging results. Some of these patients received only 0.5 g/m 2 of each agent because of limited drug availability, but normalizing to a clinically relevant dose of 0.75 g/m 2 Ro 03-8799 with 2 g /m 2 SR 2508, the mean tumour concentrations over the first 30 min were 30.2 (±7.2, 2 SD) jug/g and 71.8 (±7.0, 2 SD) ¿ig/g for the two agents, respectively. These represent mean tum our to plasma ratios of 370% (Ro 03-8799) and 70% (SR 2508). The mean absolute tum our concentration of Ro 03-8799 is slightly lower than that seen in the previously mentioned larger series [ 8 ]. T h is m ay be a sampling effect owing to small patient numbers, but it is proposed to attempt an escalation of the Ro 03-8799 dose to 1 g/m 2 as part of the phase I assessment of the combination. The tum our levels of SR 2508 are in the range seen when this agent is used alone [11]. Accepting the limitations of SER prediction, a single dose ratio of 1.6 would seem possible, and the maximum tolerated multi dose schedule for the combination has not yet been reached.

6 . CONCLUSION

From the above discussions it is clear that hyperthermia and the newer radiosensitizers provide encouragement that the problem of tum our hypoxia may be overcome. Future directions of research may include the combination of these strategies. 192 BLEEHEN et al.

REFERENCES

[1 ] FIELD, S., BLEEHEN, N.M., Hyperthermia in the treatment of cancer, Cancer Treat. Rev. 6 (1979) 63. [2] DUNLOP, P., et al., “The use of thermal dose in the clinical application of localised hypeithermia”, Hyperthermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984), Vol. 1, (OVERGAARD, J., Ed.), Taylor and Francis, London and Philadelphia (1984) 187. [3] ARCANGELI, G., et al., Clinical results after different protocols of combined local heat and radiation, Strahlentherapie 151 (1983) 82. [4] HENK, J.M., “Long term results of hyperbaric oxygen and radiotherapy in head and neck cancer”, Chemical Modifiers of Cancer Treatment (Proc. Conf. Clearwater, 1985). [5] WATSON, E.R., et al., Hyperbaric oxygen and radiotherapy. Medical Research Council trial in carcinoma of the cervix, Br. J. Radiol. 51 (1986) 879. [6] URTASUN, R., et al., Radiation and high dose metronidazole in supratentorial gliomas, N. Engl. J. Med. 274 (1976) 1364 and Br. J. Cancer 46 (1982) 101. [7] DISCHE, S., Chemical sensitizers for hypoxic cells. A decade of experience in clinical radiotherapy, Radiother. Oncol. 3 (1985) 97. [8] ROBERTS, J.T., et al., A clinical phase 1 toxicity study of Ro 03-8799, plasma, urine, tumour and normal brain pharmacokinetics, Br. J. Radiol. 59 (1986) 107. [9] NEWMAN, H.F.V., et al., A comparative study of Ro 03-8799: racemic mixture and enantiomers, Br. J. Radiol. (1986) (in press). [10] WILLIAMS, M.V., et al., In vivo assessment of basic 2-nitroimidazole radiosensitizers, Br. J. Cancer 46 (1982) 127. [11] COLEMAN, C.N., et al., Initial report of the phase I trial of the hypoxic cell radio sensitizer SR 2508, Int. J. Radiat. Oncol., Biol. Phys. 10(1984) 17. IAEA-SM-290/46

THE ROLE OF HYPERTHERMIA IN RADIATION THERAPY IN DEVELOPING COUNTRIES

T. SUGAHARA Health Research Foundation, Kyoto, Japan

Abstract

THE ROLE OF HYPERTHERMIA IN RADIATION THERAPY IN DEVELOPING COUNTRIES. Extensive studies on the improvement of cancer therapy with conventional low LET radiation and chemical or physical means have been carried out for more than 30 years. Among them the combined use of hyperthermia and radiation therapy has been studied for more than 10 years. The improvement of cancer therapy by this combined treatment has been well established. However, the unavailability of appropriate heating equipment has limited the use of hyperthermia in developing countries. Hyperthermia will become one of the most useful methods to improve cancer therapy in developing countries when appropriate heating equipment is available at a reasonable cost. Capacitive type RF heating equipment for both superficial and deep seated tumours has been developed in Japan. From this standard equipment a simplified model for limited superficial and shallow seated tumours prevalent in developing countries has been developed.

1. IN T R O D U C T IO N

The improvement of cancer therapy with conventional low LET radiation combined with physical or chemical means has been studied for many years [ 1 ]. The study was initiated by the introduction of BrdUr and hyperbaric oxygen in the early 1960s. Since then, several agents for radiosensitization have been studied experimentally and then clinically one by one. The situation is illustrated in Fig. 1 as research life cycles. In the early 1970s two chemicals, misonidazole, a hypoxic cell sensitizer, and WR-2721, a very potent radioprotector, seemed to be very promising. Both of them failed to demonstrate any significant improvement in radiotherapy in randomized clinical trials, probably because of their high toxicity. At present, second generation hypoxic cell sensitizers such as SR-2508, Ro-03-8799 and RSU-1069 are under clinical trials in the United States of America and the United Kingdom. In Japan also hundreds of compounds have been synthesized and screened as hypoxic cell sensitizers in vitro and in vivo during recent years. Among them AK-2123, a nitrotriazole derivative produced by Kagiya, et al. seems promising [2 ]. In parallel with these second generation drugs, various kinds of non-hypoxic sensitizers have been introduced such as IdUr and PLD repair inhibitors.

193 194 SUGAHARA

>) Colony formation

+> U л! D >

■P ci (U

I I I I I_I_LJ_I I I I I I I I I I I ...... 1956 I9G0 1970 1980 1986 FIG. 1. Life cycle o f radiosensitization research.

TABLE I. TUMOUR HISTOLOGY AND PRIMARY EFFECTS

Note: CR (complete response): complete regression. PRa (partial response): partial regression of more than 80%. PRb (partial response): regression between 50 and 80%. NR (no response): regression of less than 50%. a Per cent overall response rate (CR + PR). IAEA-SM-290/46 195

TABLE II. TUMOUR SIZE AND PRIMARY EFFECT

Effect

Tumour CR PRa PRb NR Total diameter (cm )\

< 3 .9 39 7 4 18 68 (67%) 4 -5 .9 37 10 12 7 66 (71%) 6 -9 .9 22 18 22 5 67 (60%) > 10 13 4 7 11 35 (48%)

Total 111 39 45 41 236

Concerning hypoxic cells in tumours, the application of perfluorochemicals as oxygen carriers combined with the inhalation of carbogen (95% 0 2 + 5% C 0 2 ) resulted in significant sensitization in experimental tumours. Some clinical trials have been started using these chemicals. Compared with these chemical modifiers the progress of hyperthermic oncology is very remarkable. Probably because of the physical nature of the method, the radiosensitizing effect of hyperthermia, demonstrated in cultured cells and experimental tumours, has been confirmed in human tumours. Thus the only radiosensitizing method clinically available so far is hyperthermia.

2. CLINICAL FEATURES OF HYPERTHERMIA AS A RADIOSENSITIZER

According to the clinical experience reported so far, the radiosensitizing properties of hyperthermia seem to be quite appropriate for its application in developing countries. It has the following advantages:

( 1 ) Hyperthermia sensitizes radioresistant cells more efficiently than radio sensitive tumours when combined with low LET radiation. Table I [3] shows that the responses of shallow seated tumours were nearly the same, irrespective of their histology. 196 SUGAHARA

b ой Z О ой H о S ОЙ W X н и Œ Н X Н

с. сл 00I Цн ой Z о ей н о S ей ш К н и х н ь о Z о ьо 5 £ s о и

W CQ < н IAEA-SM-290/46 197

(2) Rather advanced tumours of about 5 cm diameter could be well treated with hyperthermia, as shown in Table II [3]. Hiraoka et al. [4] reported that heating was rather easy in deep seated tumours of more than 5 cm diameter with a capacitive type RF apparatus, the Thermotron RF-8 . (3) For alleviation, the local tumour response and pain release are marked when radiation therapy is combined with hyperthermia. (4) In heating tumours other than superficial ones, the subcutaneous fat layer is the major obstacle because of its high impedance. The layer is generally not so thick in developing countries as compared for instance with the Caucasus area. RF capacitive type heating can be easily applied in developing countries if the layer is less than 2 cm th ic k .

3. DEVELOPMENT OF HEATING EQUIPMENT

Sophisticated heating equipment has been developed in developed countries. In Japan capacitive type hyperthermia equipment has been developed with the co-operation of engineers, physicists, radiobiologists and radio-oncologists which can be used for both superficial and deep seated tumours. The equipment has been developed, starting from a very simple model through various kinds of improvements for heating deep seated tumours, into a rather expensive but sophisticated apparatus. However, since special types of cancers are prevalent in each developing country, such as oral, breast, nasopharyngeal, cervical, bladder and so on, more simplified models of hyperthermia equipment available only for limited but prevalent tum our sites should be developed, especially for superficial and shallow seated tumours. Table III compares in outline our simplified model “Thermotron RF -8 S P ” with the original model “Thermotron RF-8 ” for each component. The Thermotron RF -8 is useful in almost all cancers of the head and neck including oral cancers, breast and bladder cancers, many cancers of the prostate, stomach and cervix and limited cancers of the lung prevalent in some parts of the developing countries. Our simplified model will be useful in all oral, breast and bladder cancers and in a limited number of other cancers. The simplification reduces the cost by about one-quarter.

4. CONCLUSION

The application of hyperthermia as an adjuvant to radio- and chemotherapy may contribute greatly to an improvement in cancer therapy. It is especially appropriate in developing countries because at present it is the only agent clinically available for radiosensitization and inoperable advanced cases of cancer are more frequent in developing countries than in developed countries. We are expending every effort to provide developing countries with low cost, efficient heating equipment for clinical trials. SUGAHARA

REFERENCES

SUGAHARA, T., Modification of Radiosensitivity in Cancer Treatment, Academic Press, Tokyo (1984)3. SHEN, Y., et al., The radiosensitizing effect of Ak-2123 in experimental tumor MA-737, Radiosens. Newsl. S 3 (1986) 1. MATSUDA, T., personal communication. HIRAOKA, M., et al., Jpn. J. Hypertherm. Oncol. 2 (1986) 23. IAEA-SM-290/12

INTERACTION BETWEEN RADIOTHERAPY AND HYPERTHERMIA IN HUMAN SOLID TUMOURS

T. VAZQUEZ, P. KASDORF F a c u lty o f M edicine, University of Montevideo, Montevideo, Uruguay

Abstract

INTERACTION BETWEEN RADIOTHERAPY AND HYPERTHERMIA IN HUMAN SOLID TUMOURS. Between March 1983 and April 1985, 214 patients with advanced head and neck cancers, recurrent breast tumours in the chest wall, malignant melanomas and secondary adenopathies from unidentified primary tumours, soft tissue sarcomas and non-Hodgkin’s lymphomas were submitted to a combined treatment with conventional radiotherapy and hyperthermia (250 keV, 4 Gy per fraction, 2 fractions per week). A total dose of 40 Gy was given in five weeks plus hyperthermia produced by external microwaves (2450 MHz), in 10 sessions of 1 h each follow ing radiotherapy. Complete regression (CR), partial regression (PR), and non-responders (NR) were distributed according to the kind of tumour treated, as follows: head and neck cancers 78/127 (61.5%) CR, 37/127 (29.1%) PR, 12/127 (9.4%) NR; recurrent breast cancers 23/35 (65.7%) CR, 10/35 (28.6%) PR, 2/35 (5.7%) NR; melanomas 7/8 CR, 1/8 PR; secondary adenopathies 20/29 (69%) CR, 7/29 (24.1%) PR, 2/29 (6.9%) NR; soft tissue sarcomas 5/7 CR, 2/7 PR; non-Hodgkin’s lymphoma 8/8 CR. Our results are similar to those reported by other investigators and confirm the need for continuing research in this field. Combined treat ment can be a useful tool particularly in developing countries because of the relatively low cost equipment.

1. IN T R O D U C T IO N

Radioresistant tumours pose the most difficult problem in oncology and the results have not been stimulating. A wide variety of treatment methods have been and are being investigated. Only recently have radiobiological and clinical studies shown encouraging results in therapy with a combination treatment of hyperthermia and radiotherapy [1—5]. Hypoxic cells, the most radioresistant of a tumour population [6 ], are sensitive to moderate doses of heat in association with radiation. Furthermore, the ‘S’ phase of the cell cycle, which is known to be the least radiosensitive, is most sensitive to heat. In G, and G 2 the opposite phenomena are observed. A synergistic enhancement throughout the cell cycle has been demonstrated [6 , 7]. In addition, the repair process of radiation damaged cells is inhibited by hyperthermia [6 , 7].

199 200 VAZQUEZ and KASDORF

For this reason we believe that the combination of hyperthermia and radio therapy will be of great importance in the treatment of solid human tumours in the future. Much research is under way and much equipment has already been developed to control the radioprotective effect of hypoxia. Neutrons and heavy particles have been and are being utilized but the necessary equipment is too expensive for developing countries with restricted health budgets. The relatively low cost of hyperthermia equipment therefore constitutes a great advantage. At the IAEA Meeting in Cairo in September 1984 [8 ] we reported the results of a combination treatment with hyperthermia and radiotherapy in 50 patients.

2. MATERIAL AND METHODS

Between March 1983 and April 1985 we treated a total of 214 patients. A total of 127 patients had advanced head and neck epidermoid carcinomas, 35 had recurrent breast cancers in the chest wall, 29 had secondary adenopathies from unidentified primary tumours, 8 had metastatic or recurrent melanoma nodules, mostly located in the extremities, 7 had sarcomas and 8 h ad n o n -H o d g k in ’s ly m p h o m as. The patients were treated with a combination of radiotherapy and hyper thermia. The radiotherapy equipment consisted of a Müller 250 keV X-ray apparatus. The patients received two fractions per week of 4 Gy each, amounting to a total radiation dose of 40 Gy in five weeks (10 sessions) at a source to skin distance of 30 cm. Hyperthermia was given with the Siemens ‘Ultratherm’ 2450 MHz external microwave generator. It was usually started 15-30 min after the irradiation and lasted for 1 h. A temperature of 43°C was reached in about 10 min. The treat ment schedule comprised 10 sessions with a 72 h interval in order to avoid thermotolerance reported in the literature [9]. This schedule is also used in other cen tres [ 10]. The temperature at the heated field was monitored by a minimum of two thermistor probes encased in 24 gauge needles inserted at the depth of the tumour and one thermistor embedded in a plastic tube placed at the skin surface during each treatment session. They were continuously checked by the operator on a d isplay. Patients were controlled once a week during treatment, 15 days after its completion, and subsequently once a month for the period of one year. The classification of tum our response was as follows: when the tumour dis appeared clinically for more than four weeks we classified the case as complete regression (CR); when the regression was greater than 50% it was classified as partial regression (PR); lesser regressions, stabilizations or progressions were classified as non-responders (NR). IAEA-SM-290/12 201

Treatment complications were classified as follows:

(a) Early complications: 0 — no visible reaction; 1 — slight erythema; 2 — severe erythema; 3 — desquamation, 3.1 — dry, 3.2 — exudative; 4 — second degree burns; 5 — necrosis, 5.1 — muscular, 5.2 — bone. (b) Late complications: 1 - skin atrophy; 2 - ulceration; 3 - subcutaneous fibrosis.

3. R E S U L T S

Table I shows the treatm ent results for the 127 patients with epidermoid tumours of the head and neck. Table II refers to those with recurrent breast cancer in the chest wall. Treatment results for melanomas, secondary adeno pathies from unidentified primary tumours, sarcomas and non-Hodgkin’s lympho mas are shown in Tables III—VI, respectively. Sixty-four per cent of the epithelial tumours and melanomas attained continued complete tumour regression, ranging from 4 to 24 months. We confirmed that tumours located less than 2 cm deep in the head and neck usually reached the prescribed temperature of 43°C. This was more difficult to attain in the chest wall and in the extremities, where only 70% of the lesions reached 43°C. This difficulty was due to the larger area to be treated because of superficial tum our spread. Fifty per cent of the more deep seated lesions (2—4 cm) reached 42.5°C; the percentage dropped to only 20% whenever the tumour location was deeper. Tumour size is a determining factor in attaining complete regression. In head and neck locations only 50% of the tumours of maximal 2—4 cm in diameter achieved CR. With regard to complications, 318 sites were treated in the 214 patients. In 32 of the 318 sites (10%) complications of some significance developed: In 26 ( 8 .2%) tum our necrosis developed with severe, deep seated ulceration that did not heal spontaneously. In 6 (1.8%) second degree burns occurred from mal- positioned thermistor probes. The present results confirm those reported by us at the Meeting in Cairo in 1984 for 50 patients [8 ]. The greater patient population has not significantly modified any of our data. For head and neck lesions we reported 16/32 CR (50%), 10/32 PR (32%) and 6/32 NR (18%) as against a current 61.5% CR, 29.1% PR and 9.4% NR as stated in Table I. For chest wall locations, our report in 1984 stated 6/10 CR (60%), 3/10 PR (30%) and 1/10 NR (10%) as against current data of 65.7% CR, 28.6% PR and 5.7% NR (Table II). The complication rates reported in 1984 were also similar. We had 10% necrosis and 5.5% second degree bums. The results obtained are consistent with the literature. 202 VAZQUEZ and KASDORF

TABLE I. HEAD AND NECK CANCERS

Treatment No. of % result patients

CR 78 61.5 PR 37 29.1 NR 12 9.4

Total 127 100.0

TABLE II. RECURRENT BREAST CANCER IN THE CHEST WALL (23 adenocarcinomas; 7 ductal carcinomas; 5 scirrous carcinomas)

Treatment No. of % result patients

CR 23 65.7 PR 10 28.6 NR 2 5.7

Total 35 100.0

TABLE III. MELANOMAS

Treatment No. of result patients

CR 7 PR 1

Total 8 IAEA-SM-290/12 203

TABLE IV. SECONDARY ADENOPATHIES

Treatment No. of % result patients

CR 20 69.0 PR 7 24.1 NR 2 6.9

Total 29 100.0

TABLE V. SARCOMAS (SOFT TISSUES)

Treatment No. of result patients

CR 5 PR 2

Total 7

TABLE VI. NON-HODGKIN’S LYMPHOMAS

Treatment No. of result patients

CR 8

4. DISCUSSION

One of the technical difficulties relative to the adequate heating of larger lesions or of deep seated tumours is the inhomogeneity of the thermal dose; only small lesions or shallow seated tumours (less than 2 cm) uniformly reached the prescribed temperatures [11, 12]. This was due to the 2450 MHz external 204 VAZQUEZ and KASDORF

microwaves used which have a limited range. Furthermore, on irregular surfaces such as the head and neck, coupling of the applicators can be improved, resulting in better temperatures and decreased microwave leakage [10, 13]. However, in patients with extensive lesions such as recurrent carcinoma of the breast in the chest wall or melanoma in the extremities, it is sometimes necessary to use multiple applicators to reach the desired temperature. At present it is difficult to correlate the temperature at the depth of the tumours with response and tumour control because of the inadequate clinical thermometry systems available and the inhomogeneity of the heat distribution. It is apparent that only portions of the tumour reach ‘therapeutic temperatures’; yet it is possible that even temperatures in the range of 42.5°C may have a radiosensitizing effect. Stringent quality control programmes must be designed in hyperthermia to enhance the accuracy of temperature measurements and to assess equipment performance. Non-perturbing probes (high resistivity thermistors, gallium arsenide, etc.) and small probes that can be placed in catheters should be used to measure the temperature at multiple points within the tumour and in the normal tissues. Non-invasive thermometry techniques must be developed and tested in appropriate models so that in the near future they can be incorporated into clinical practice. From the data reported in Refs [4, 14—16], which are similar to ours, epidermoid carcinomas, adenocarcinomas, and sarcomas are equally sensitive to hyperthermia, whereas malignant melanomas seem to be more sensitive. The tolerance of normal tissues to heat, alone or combined with irradiation, has been satisfactory [5, 17]. It will be extremely im portant in prospective clinical trials to quantitatively assess the effects ofheat delivered at different temperatures or times in critical organs such as the brain, lung, liver, kidney, and intestines [18]. Further technological advances are necessary in order to optimize the application of heat in the treatment of patients with malignant tumours before biological predictions can be definitely confirmed or refuted in a clinical setting. Although the development of equipment and clinical thermometry techniques is still in progress it is possible to treat effectively a significant number of lesions. Clinical trials to enhance our knowledge of the combination treatment with radiotherapy and hyperthermia should be encouraged.

REFERENCES

[1] ARCANGELI, G., et al., Effectiveness of microwave hyperthermia combined with ionizing radiation: Clinical results on neck node métastasés, Int. J. Radiat. Oncol., Biol. Phys. 6(1980) 143. [2] ARCANGELI, G., et al., Tumor control of therapeutic gain with different schedules of combined radiotherapy and local external hyperthermia in human cancer, Int. J. Radiat. Oncol., Biol. Phys. 9 (1983) 1125. IAEA-SM-290/12 205

[3] BICHER, H., et al., “Results of a phase I/II clinical trial of fractionated hyperthermia in combination with low dose ionizing radiation”, Hyperthermia (BICHER, H., BRULEY, D., Eds), Plenum Press, New York (1982) 87. [4] FAZEKAS, J., NERLINGER, R., Localized hyperthermia adjuvant to irradiation in superficial recurrent carcinomas: A preliminary report on 46 patients, Int. J. Radiat. Oncol., Biol. Phys. 7 (1981) 1457. [5] KIM, J., et al., Treatment of superficial cancers by combination hyperthermia and radiation therapy, Clin. Bull. 9 (1979) 11. [6] DEWEY, W.C., et al., “Cell biology of hyperthermia and radiation”, Radiation Biology in Cancer Research (MEYN, R.E., WITHERS, H.R., Eds), Raven Press, New York (1980) 589. [7] OVERGAARD, J., “Time-temperature relationship for hyperthermic cytotoxicity and radiosensitization. Implications for a thermal dose unit”, Hyperthermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984), Vol. 1, Taylor & Francis, London and Philadelphia (1984) 191. [8] VAZQUEZ, T., KASDORF, P., “Interaction between hyperthermia and radiotherapy in 50 superficial solid human tumors”, Improvement of Cancer Therapy by the Combination of Treatment by Conventional Radiation and Physical or Chemical Means (Proc. Research Co-ordination Meeting Cairo, 1984). [9] LAW, M.P., AHIER, R.G., SOMAIA, S., FIELD, S.B., “Does thermotolerance increase during fractionated hyperthermia? ”, Hyperthermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984), Vol. 1, Taylor and Francis, London and Philadelphia (1984). [10] PÉREZ, C., EMAMI, В., VON GERICHTEN, D., “Clinical results with irradiation and local microwave hyperthermia in cancer therapy”, Hyperthermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984), Vol. 1, Taylor & Francis, London and Philadelphia (1984) 398. [11] DORBEC, L., CHABBLE, Y., RAULT, P., et al., A 13.56 MHz hyperthermia apparatus for local treatment of deep tumors, J. Eur. Radiother. 2 (1981) 157. [12] ISRAËL, L., BESENVAL, M., Phase I and phase II trials of the use of localized radio frequency hyperthermia in 49 cases of deep-seated cancer, Bull. Cancer 68 (1981) 296. [13] SCOTT, R., et al., Hyperthermia in combination with radiotherapy: A review of five years experience in the treatment of superficial tumors, Int. J. Radiat. Oncol., Biol. Phys. 9 (1983) 1327. [14] MARMOR, J., HAHN, C., Combined radiation and hyperthermia in superficial human tumor, Cancer 46 (1980) 1986. [15] STORM, F.K., ELLIOT, R.S., HARRISON, W., et al., Radiofrequency hyperthermia of advanced human sarcomas, J. Surg. Oncol. 17 (1981) 91. [16] KIM, J., et al., Combination hyperthermia and radiation therapy for malignant melanomas, Cancer 50 (1982) 478, [17] MANNING, M., et al., Clinical hyperthermia: Results of the phase I clinical trial combining localized hyperthermia with or without radiation, Int. J. Radiat. Oncol., Biol. Phys. 5 (1973) (abstract). [18] LUK, K.H., FRANCIS, M.E., PEREZ, C.A., JOHNSON, R.J., Radiation therapy and hyperthermia in the treatment of superficial lesions: Preliminary analysis: Treatment efficacy, and reactions of skin and subcutaneous tissues, Am. J. Clin. Oncol. (CCT) 6 (1983) 399.

IAEA-SM-290/60

HYPERTHERMIA AND RADIOTHERAPY OF LOCALLY RECURRENT BREAST CANCER

A.O. BADIB, S.A. KHALIL, M.Y. GOUDA, A.M. SALAMA Radiotherapy Unit, Faculty of Medicine, Alexandria University, Alexandria, Egypt

Abstract

HYPERTHERMIA AND RADIOTHERAPY OF LOCALLY RECURRENT BREAST CANCER. The role of hyperthermia as an adjuvant to suboptimal radiation doses was evaluated in 69 women with recurrent breast cancer after surgery plus radical radiotherapy. Patients were randomized into three groups with equal representation of number and volume of lesions. The first two groups received conventional daily fractionated radiation of 30 Gy in 4 weeks. Heat was added twice weekly in the second group. Equivalent twice weekly radiation fractions (295 cGy in 8 fractions) were given in the third group combined with hyperthermia in all frac tions. The results showed that combined thermoradiotherapy was effective and well tolerated. Higher tumour response rates were seen after such combinations and significantly so after hyper thermia and twice weekly radiotherapy. No significant differences were seen in the degree of acute radiation reactions in the three groups of the study.

1. IN T R O D U C T IO N

There has been great interest in the possibility of using hyperthermia as an adjuvant to radiotherapy in the treatment of cancer patients [1 ]. The rationale for such a combination treatment is partly based on the radiosensitizing effect of heat through its selective lethal effect on radioresistant cells in the ‘S’ phase and on acidic, hypoxic and nutritionally deprived tum our cells [2, 3]. Heat is also effective because of its inhibition of cellular repair systems [4]. In the management of cancer patients with local failure after radiation to full tissue tolerance, one has to restrict reirradiation to suboptimal dose levels to avoid excessive damage to normal tissues in the previously irradiated volume. Hyperthermia can be exploited in this situation to enhance the radiation response [5]. The aim of the present study was to evaluate the combination of hyper thermia and limited dose radiotherapy in the management of local recurrences of breast cancer following initial radical surgery and irradiation.

207 208 BADIB et al.

2. MATERIAL AND METHODS

The study included 69 female patients with locally recurrent adenocarci noma of the breast. All patients had radical mastectomy followed by post operative radiotherapy of 45 Gy in 4 weeks. The age ranged between 35 and 60 years, with a mean of 41 years. Recurrences were detected in the anterior chest wall 8 to 38 months after initial treatment. All lesions were superficial and measurable: 27 lesions were solitary and 42 were multiple, the thickness varying from 0.5 - 4 cm. Patients were randomized into three groups with equal representation of the number and thickness of lesions. The first group received radiation alone, while the second and third groups received hyperthermia plus radiation. Infrared heating was used in superficial lesions of 1 cm in thickness or less.- This was delivered using a 150 W source (Philips) at a distance of 30 cm with surface air cooling. Low frequency heat was used in lesions of 1-4 cm thick ness. A 0.5 MHz radiofrequency unit with metallic plate or pencil-like electrodes was used. The tumour temperature was kept at 42-44°C for */2 h to 1 h imme diately before RT. Temperature measurements of the skin and tumour were made using thermocouples. Radiotherapy was given using DXR (3 mm Cu HVT) or cobalt-60 units (with wax bolus). Conventional daily fractionated radiation to a total dose of 30 Gy in four weeks was given in the first and second groups. Equivalent twice weekly fractions (295 cGy in 8 fractions) were computed using TDF factors [6 ] and given in the third group. None of the patients received concomitant chemotherapy or hormonal therapy in this study.

3. R E S U L T S

The response of the tum our and the skin to the treatment was evaluated over a period of four weeks after treatment. The evaluation criteria for tumour response (Table I) included: complete response (CR) — complete tumour regres sion; partial response (PR) — 50% to less than 100% tum our regression; and minimal response (MR) — less than 50% tumour regression. Normal tissue reactions were determined according to the degree of erythema (mild to severe) and the development of wet desquamation ± blistering (Table II). Patients in the first group of radiation alone showed tumour CR in 19.0% and total response (CR + PR) in 47.6%. The use of hyperthermia twice weekly with conventional daily radiation in the second group improved the tumour CR rate to 30.4% and the total response to 65.2%. This increase is not statistically significant (P > 0 .1 ). When hyperthermia was given with all the twice weekly radiation fractions in the third group, a statistically significant improvement (P< 0.025) was seen in the rates of tumour CR (48%) and total response (76%). An antalgic effect was obtained in patients with painful lesions after complete or partial tum our regression. IAEA-SM-290/60 209

TABLE I. CLINICAL TUMOUR RESPONSE

Group 1 Group 2 Group 3 Tumour response No. % No. % No. %

CR 4 19.0 7 30.4 12 48.0 PR 6 28.6 8 34.8 7 28.0 MR 11 52.4 8 34.8 6 24.0

Total 21 100.0 23 100.0 25 100.0

T A B L E II. ACUTE RADIATION REACTION

Erythema Wet desqamation Group Total ± blisters Mild Mod./Severe ! 1 No. 21 15 3 3 % 100.0 71.4 14.3 14.3 2 No. 23 15 4 4 % 100.0 62.5 17.4 17.4 3 No. 25 13 6 5 % 100.0 52.0 24.0 20.0

Various degrees of acute radiation reactions were recorded. Mild to severe erythema developed in 81.2% of cases which healed within two weeks of the end of the treatment. Wet desquamation ± blister formation were seen in 12 patients (17.4%). They healed spontaneously within four weeks after treatment. No statis tically significant differences were seen in these reactions in the three treatm ent groups (P>0.25).

4. DISCUSSION

The early results of this study are consistent with other published reports [5, 7, 8 ]. However, variations in the techniques of hyperthermia, in the sequence of treatment and in the types and sites of tumour make comparison difficult. 210 BADIB et al.

Patients in this study had received tolerance levels of prior radiation that restricted radiation doses to recurrences to suboptimal levels. The result of such radiation alone was not impressive, similar to the results reported in previous publications [9, 10]. Hyperthermia combined with reduced radiation doses in this series resulted in an enhancement of tumour response, which was only significant when heat was given with all fractions of radiation in the twice weekly schedule. However, a wide spectrum of tum our response was reported from a variety of combinations and methods of fractionation of heat and radiation [ 11 ]. In this trial the use of long intervals of at least 72 hours between heat treatments and the sparing of most of the normal tissues around the tumour apparently resulted in no enhancement of the radiation response in normal tissues. Others [ 1 ] advocated sequential treatment with hyperthermia given four hours after radiation to avoid thermal enhancement of normal tissues. Results on delayed effects of the combination treatment are not available because of the short observation period of this study. However, it is expected that when heat is used as a modifier of radiation injury, the time of occurrence of that injury will be similar to that for radiation alone. Direct heat damage, whether in tumour or in normal tissues, appears rapidly and late effects from heat alone are unlikely to occur [ 12].

REFERENCES

[1] OVERGAARD, J., Cancer 48 (1981) 1116. [2] FIELD, S.B., BLEEHEN, N.M., Cancer Treat. Rev. 6 (1979) 63. [3] SUIT, H., GERWECK, L.F., Cancer Res. 39 ( 1979) 2290. [4] DEWEY, W.C., et al., Radiology 123 (1977) 463. [5] MANNING, M.R., et al., Cancer 49 (1982) 205. [6] ORTON, C.G., ELLIS, F., Br. J. Radiol. 46 (1973) 529. [7] BRENNER, H.J., YERUSHALMI, A., Br. J. Cancer 33 (1975) 91. [8] KIM, J.M., et al., Cancer 40(1977) 161. [9] CHU, F., et al., Cancer 37 (1976) 2677. [10] MAHMOUD, L.M., et al., Alex. Med. J. 23 ( 1981 ). 124. [11] BAGSHAW, M.A., Cancer Therapy by Hyperthermia, Drug, and Radiation (Proc. Int. Symp. Colorado, 1980), NCI Monograph 61, Maryland (1982) 431pp. [12] DENEKAMP, J., Cancer Therapy by Hyperthermia, Drug and Radiation (Proc. Int. Symp. Colorado, 1980), NCI Monograph 61, Maryland (1982) 4 3 1pp. IAEA-SM-290/73

COMBINED RADIOTHERAPY AND HYPERTHERMIA:

PROGNOSTIC VARIABLES AFFECTING TUMOUR RESPONSE

G. ARCANGELI Istituto Medico e di Ricerca Scientifica and Regina Elena National Cancer Institute

M. BENASSI, S. CARPINO Regina Elena National Cancer Institute

F. M A U R O ENEA, Casaccia

Rome, Italy

Abstract

COMBINED RADIOTHERAPY AND HYPERTHERMIA: PROGNOSTIC VARIABLES AFFECTING TUMOUR RESPONSE. The influence of tumour volume and isoeffect thermal dose, expressed as equivalent minutes at 42.5° С (Eq. 42.5), on initial and long term tumour response has been evaluated on 38 patients with a total of 81 multiple neck métastasés from squamous cell carcinoma of the head and neck and on 17 patients with a total of 38 cutaneous and nodal métastasés from malignant melanomas. In both groups of tumours the probability of complete response appeared to decrease with increasing tumour volume. However, the volume effect was more pronounced for the lesions treated with radiation alone than for those treated with a combined modality, indicating that the addition of heat is more beneficial to the large than to the small lesions, the latter being easily controlled by radiation alone. However, for response duration the volume effect was more pronounced in the lesions treated with radiotherapy alone. In this group of lesions local control was 0% between 6 and 16 months when the volume was larger than 4 cm3, in contrast to the lesions treated with the combined modality in which local control was 45% and 27% at 28 months for tumour volumes larger than 10 cm3. The variable of isoeffect thermal dose was analysed for its influence on response in head and neck cases. The probability of complete response appeared to increase with increasing the Eq. 42.5, thus describing an isoeffect thermal dose-response relationship. The response duration was also related to the isoeffect thermal dose in that the estimated local control distribution was better in the group of lesions that received more Eq. 42.5. Finally, from the analysis of the data on head and neck tumours it was possible to derive the amount of isoeffect thermal dose necessary to achieve a certain probability of tumour response for a given tumour volume.

1. INTRODUCTION

Although the potential of using hyperthermia in addition to radiation is now well documented from the accumulated experimental and clinical experience [1 -4 ], there are factors and variables that have been found to affect both tum our and

211 212 ARCANGELI et al.

normal tissue responses in pet animals and human beings, namely the number and size of hyperthermic and radiation fractions, the sequence and interval between the two modalities, fractionation schedules, tumour volume, tumour temperature, duration of heating, heating method, etc. [5-11]. We have extensively shown the influence of most of these variables on the hyperthermic enhancement of the radiation response and on the therapeutic effect in miscellaneous superficial lesions [ 12]. This paper reports the results on immediate response and response duration and their correlation with tumour volume and isoeffect thermal dose in two groups of patients homogeneous with respect to both tumour site and histology and treated under the same conditions.

2. MATERIAL AND METHODS

Among the patients treated by our group between 1977 and 1985 there were two groups represented by 38 patients with a total of 81 multiple neck node métas tasés from squamous cell carcinomas of the head and neck, and by 17 patients with a total of 38 cutaneous and/or nodal métastasés from malignant melanomas. The details of hyperthermic and radiation treatment have already been described elsewhere [13, 14]. Briefly, radiation was given with 6 MeV photons or with electrons of various energies by means of linear accelerators. The total dose and fraction size varied according to the treatment protocols reported below. In the first years the temperature was monitored at regular intervals, with the power off, by inserting an 18 gauge constantan-copper thermocouple wire inside the plastic lumen of a standard intracath probe previously placed at the base of the lesion. More recently we have used microthermocouples of 100 /um diameter (Medtra, Detroit, MI) and multisensor probes (IT 17 thermocouples; Bailey, Saddle Brook, NY) with 3 thermocouple sensors every 1.5 cm, which react poorly with the electromagnetic field, allowing us to monitor the tempera ture in different tumour locations continuously. In patients with head and neck tumours, radiation was delivered in 3 fractions/d of 1.5 to 2 Gy each, with 4 h intervals between fractions, to a total dose of 60 Gy; heat, at 42.5°C for 45 min, was applied every other day, immediately after the second daily fraction of radiation, for a total of 7 hyperthermic treatments. Patients with melanoma lesions were treated according to the following two protocols. In the first protocol, radiation was given as 2 weekly fractions of 5 Gy each to a total dose of 40 Gy; heat, at 42.5°C for 45 min, was applied either immediately (immediate treatment) or 4 h after (delayed treatment) each radiation fraction, for a total of 8 hyperthermic sessions. In the second protocol, radiation was delivered as 2 weekly fractions of 6 Gy each to a total dose of 30 Gy; heat, at 45°C for 30 min, was applied immediately after each radiation fraction for a total of 5 hyperthermic treatments. In this case the skin around the lesion was cooled by means of circulating cold water. All neck node métastasés from squamous cell carcinoma of the head and neck were treated between March 1977 and August 1979. At that time multi- IAEA-SM-290/73 213 sensor probes were not yet available and, therefore, the temperature was measured only at one point, which presumably represented the minimum tumour tempera ture. Because of poor matching it was not always possible to achieve the prescribed temperature. In this case treatments of 45 min at the maximum tolerable tempera ture were delivered, and the temperature data were converted to equivalent minutes at 42.5°C (Eq. 42.5), by applying an empirical isoeffect relationship recently employed by several investigators [15, 16] based on 1°C being equiva lent to a factor of 2 in heating time above the transition point (i.e. 42.5°C in our case), and a factor of 4 below. Although we do not believe that this is the best way to quantify thermal dose, this relationship can be used to compare responses to different time-temperature combinations. For this purpose Eq. 42.5 was simply derived from the average of the temperature taken every 5 min during the whole period of heating. The influence of the treatment modality on the time to first recurrence detection from the beginning of treatment was analysed by means of the Kaplan and Meier product limit method [17]. The difference between the curves was tested by means of the Mantel-Haenszel test [18]. Every patient had at least two lesions, each one assigned to a specific treatment modality. Patients who died without suffering the event of interest (i.e. local recurrence of the lesion treated with the specific modality under consideration) were designated as recurrence free. The probability of complete response was calculated by means of a non linear optimization method [19].

3. R E S U L T S

The results on head and neck tumours have been partially reported else where [20]. The initial overall complete response rates were 79% (30/38) and 42% (18/43) for the lesions treated with the combined modality and radiotherapy alone, respectively. The difference was statistically significant (P<0.05). A general idea of response duration is given in Fig. 1 which shows an actuarial analysis of local control according to the Kaplan and Meier product limit method [17]. The maximum follow-up was 28 months. At 2 years local control was 58% and 14% in thb combined modality and radiotherapy alone groups, respectively. The estimated local control distribution as a function of the tumour volume in the two treatment groups is shown in Fig. 2. The response duration seems to be significantly dependent upon tumour volume in both groups. However, the volume effect was more pronounced in the radiation alone than in the combined modality group. For lesions of small volume the response duration is similar in the two groups, in that there was no recurrence during the follow-up period. How ever, in lesions of larger volume local control after radiatiotj alone was 0% at 1 5 ,7 an d 6 months for increasing volumes. In the combined modality groups, 27 and 45% lesions of 11 to 25 cm 3 and larger than 25 cm3, respectively, were still controlled at 28 months. The difference between the two treatment groups was statistically significant for the curves corresponding to each class of volumes (P C 0 .0 5 ). 214 ARCANGELI et al.

MONTHS

FIG. 1. Estimated tumour local control distribution in lesions treated with radiation plus heat and radiation alone, according to the Kaplan and Meier product limit method. The difference between the curves is statistically significant (P< 0.05J (Reprinted from Ref. [6] with permission o f Taylor & Francisj.

RT R T + H T ------<10 cm^

MONTHS

FIG. 2. Estimated tumour local control distribution for different ranges o f tumour volumes in the two treatment groups. The difference between the two groups was statistically signifi cant for the curves corresponding to each class of volume (P<0.05). (Reprinted from Ref. [6] with permission o f Taylor & Francis).

Response duration was also analysed as a function of the isoeffect thermal dose (Fig. 3). The curve at 0 Eq. 42.5 is that obtained after radiation alone and is derived from Fig. 1. After 28 months, 56 and 92% lesions treated with 125 and 305 Eq. 42.5 were still controlled in comparison with 14% lesions controlled in the group treated with 0 Eq. 42.5. The curve at 0 Eq. 42.5 was not statistically different from that at 56 Eq. 42.5; however, when the former curve was compared with those at 125 and 305 Eq. 42.5, the difference was statistically significant (P<0.05), clearly indicating the presence of a threshold dose above which the response duration increased with increasing thermal dose. IAEA-SM-290/73 215

TABLE I. COMPLETE RESPONSE IN MELANOMA 3

Radiotherapy Temperature/ Fractionation RT + HT RT alone schedule time (°C/min)b

5 Gy X 8 42.5/45 2/week 10/13(77%) 5/9(55%) 6 Gy X 5 45/30 2/week 6/8(75%) 4/8(50%)

3 Modified from Ref. [21]. b Heat delivered in combination with all radiation fractions.

MONTHS

FIG. 3. Estimated tumour local control distribution for different Eq. 42.5. The lesions that received 0 Eq. 42.5 are those treated with radiotherapy alone. The difference between this latter curve and the curves at 125 and 305 Eq. 42.5 was statistically significant (P< 0.05). Reprinted from Ref. [6] with permission of Taylor & Francis).

The results on melanoma have also been partially reported elsewhere [21 ]. Table I shows that the addition of heat resulted in a similar enhancement of the radiation response for both treatm ent schedules. In both groups of tumours the variables of tum our volume and isoeffect thermal dose were analysed for their influence on the probability of tumour response by means of a non-linear optimization method [19], assuming that the probability of cure is proportional to the number of surviving cells [2 2 ]. Further assumptions were that the initial tumour cell number is proportio nal to the initial tum our volume and that the radiation tum our cell killing occurs following the linear-quadratic model [23, 24] for radiation, and a simple exponen tial function for hyperthermia. Under these circumstances, the probability of cure can be expressed by the function

p = e-Kx-e_7H-V-e-(“D +0Dd> (1) I 216 ARCANGELI et al.

TUMOUR VOLUME (% OF 60 cm3)

FIG . 4. Probability o f com plete response as a function o f tum our volum e in head and neck lesions.

FIG. 5. Probability of complete response as a function of tumour volume in melanoma lesions. IAE A-SM-290/7 3 217

THERMAL DOSE (% OF 400 Eq. 42.5)

FIG . 6. Probability o f com plete response as a function o f therm al dose fo r two sets o f tum our volum es in head and neck lesions. where Kx is a constant, H is the Eq. 42.5, 7 is a constant, V is the tum our volume, D is the total dose, d is the dose per fraction, and a and ß are the linear and the quadratic components of the model, respectively. The probability of complete response appeared to decrease with increasing tumour volume both in head and neck (Fig. 4) and in melanoma lesions (Fig. 5). However, the volume effect was more pronounced for the lesions treated with radiation alone than for those treated with the combined modality, indicating that the addition of heat was more damaging to the large than to the small lesions. The results are well represented by the function ( 1 ). The alpha/beta ratios resulting from this analysis were 9.5 in head and neck and 2.3 in melanoma lesions. These values were not modified by the addition of heat, in contrast to the constant K, which was shown to vary as a function of tumour volume. The relationship between the isoeffect thermal dose and the probability of complete response in head and neck tumours was examined by analysing the trend of response probability for different volumes as a function of the Eq. 42.5 (see Section 2). The data are well fitted for two different volumes, indicating that the probability of complete response increased with increasing isoeffect thermal dose (Fig. 6). The probability of complete response seems to depend upon four variables (initial tumour volume, isoeffect thermal dose, total and fractional radiation dose) and it is then defined within a five-dimension space. By extracting the 218 ARCANGELI et al.

TUMOUR VOLUME (cm3)

FIG . 7. Probability o f cure as a function o f both therm al dose and tum our volum e by giving constant values to other variables. Successes (SI and failures (F ) observed in head and neck l e s i o n s . values of alpha, beta, gamma and Kx from experimental results by means of a non-linear fitting, and by giving a constant value to the other variables, the probability of tumour response can be obtained as a function of the isoeffect thermal dose and the tumour volume (Fig. 7). The successes and failures observed in the head and neck lesions are indicated in the isobologram.

4. DISCUSSION

The results of this study on head and neck cancer and on melanoma show that both immediate response and response duration are influenced by the addi tion of heat. However, the initial and the long term response are affected by two important variables, namely tumour volume and isoeffect thermal dose. The volume effect was more pronounced for the lesions treated with radia tion alone than for those treated with the combined modality. This was clearly seen both in melanoma and in head and neck lesions by comparing the probabi lity of response in both treatment groups with and without heat. Furthermore, in head and neck tumours the estimated local control distri bution showed that the duration of response was less affected by tumour volume in the lesions treated with heat than those treated without heat. IAEA-SM-290/73 219

Tumour response has been positively correlated with mean temperature [10], lowest daily average temperature [7] and minimum temperature [9]. In head and neck lesions we have also correlated initial tumour response and response duration with the isoeffect thermal dose by converting the recorded temperature to equiva lent minutes [ 16]. The data show a striking thermal dose-response relationship, in that the probability of response for different tumour volumes appears to increase with increasing isoeffect thermal dose. The estimated local control distribution shows that the response duration was also correlated with the isoeffect thermal dose in that the duration of response was longer in the lesions that received more Eq. 42.5. Finally, the analysis carried out by means of a non-linear optimization method clearly shows that the probability of tumour response is a function of both isoeffect thermal dose and tumour volume. This analysis allows the evaluation of the amount of the isoeffect thermal dose necessary to achieve a certain probability of tumour response for a given tumour volume.

REFERENCES

[1] DETHLEFSEN, L.A., DEWEY, W.C., Proc. 3rd Int. Symp. Cancer Therapy by Hyperthermia, Drugs and Radiation, Natl. Cancer Inst. Monogr. 61 (1982). [2] OVERGAARD, J., “Rationale and problems in the design of clinical studies”, Hyper thermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984) (OVERGAARD, J., Ed.), Vol. 2, Taylor & Francis, London and Philadelphia (1984) 325. [3] OVERGAARD, J. (Ed.), Hyperthermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984), Vols 1 and 2, Taylor & Francis, London and Philadelphia (1985). [4] STORM, F.K., Hyperthermia in Cancer Therapy, G.K. Hall Medical Publishers, Boston (1983). [5] DEWHIRST, M.W., SIM, D.A., GROCHOWSKI, K.J., “Thermal influence on radiation and induced complications vs. tumor response in a phase III randomized trial”, Hyper thermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984) (OVERGAARD, J., Ed.), Vol. 1. Taylor & Francis, London and Philadelphia (1984) 313. [6] HIROAKA, M., JO, S., DODO, Y.,et al., Clinical results of radiofrequency hyperthermia combined with radiation in the treatment of radioresistant cancers, Cancer 54 (1984) 2898. [7] LUK, K.H., PAJAK, T.F., PEREZ, C., et al., “Prognostic factors for tumor response after hyperthermia and radiation”, Hyperthermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984) (OVERGAARD, J., Ed.), Vol.l, Taylor & Francis, London and Philadelphia (1984) 353. [8] OLESON, J.R., SIM, D.A., MANNING, M.R., Analysis of prognostic variables in hyper thermic treatment of 161 patients, Int. J. Radiat. Oncol., Biol. Phys. 10 (1984) 2231. [9] SIM, D.A., OLESON, J.R., GROCHOWSKI, K.J., ‘An update of the University of Arizona human clinical hyperthermia experience including estimates of therapeutic advantage”, Hyperthermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984) (OVERGAARD, J., Ed.), Vol. 1, Taylor & Francis, London and Philadelphia (1984) 367. 220 ARCANGELI et al.

[10] VAN DER ZEE, J., VAN RHOON, G.C., WIKE-HOOLEY, J.L., et al., “Thermal enhance ment of radiotherapy in breast carcinoma”, Hyperthermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984) (OVERGAARD, J., Ed.), Vol. 1, Taylor & Francis, London and Philadelphia (1984) 345. [11] OVERGAARD, J., OVERGAARD, M., “A clinical trial evaluating the effect of simul taneous or sequential radiation and hyperthermia in the treatment of malignant mela noma” , Hyperthermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984) (OVERGAARD, J., Ed.), Vol. 1, Taylor & Francis, London and Philadelphia (1984) 383. [12] ARCANGELI, G., NERVI, C., CIVIDALLI, A., et al., Problem of sequence and fraction ation in the clinical application of combined heat and radiation, Cancer Res. (Suppl.) 44(1984) 4857. [13] ARCANGELI, G., CIVIDALLI, A., LOVISOLO, A.G., et al., Clinical results after different protocols of combined local heat and radiation, Strahlentherapie 159 (1983) 82. [14] ARCANGELI, G., CIVIDALLI, A., CRETON, G., et al., Tumor control and therapeutic gain with different schedules of combined radiotherapy and local hyperthermia in human cancer, Int. J. Radiat. Oncol., Biol. Phys. 9 (1983) 1125. [15] SAPARETO, S., “The biology of hyperthermia in vitro”, Physical Aspects of Hyperthermia (NUSSBAUM, G.H., Ed.), American Institute of Physics, New York (1982) 1. [16] FIELD, S.B., MORRIS, C.C., The relationship between heating time and temperature: its relevance to clinical hyperthermia, Radiother. Oncol. 1 (1983) 179. [17] KAPLAN, E.L., MEIER, P., Non-parametric estimation from incomplete observation, J. Am. Stat. Assoc. S3 (1958) 457. [18] MANTEL, N., Evaluation of survival data and two new rank order statistics arising in its consideration, Cancer Chemother. Rep. 50(1966) 163. [19] WILDE, D.J., BEIGHTER, C.S., Foundations of Optimization, Prentice Hall, Englewood Cliffs, NJ (1967). [20] ARCANGELI, G., ARCANGELI, G.C., GUERRA, A. et al., Tumor response to heat and radiation: prognostic variables in the treatment of neck node métastasés from head and neck cancer, Int. J. Hyperthermia 1 (1985) 207. [21] ARCANGELI, G., “Radiotherapy with or without hyperthermia in the treatment of melanoma” , Proc. 1st Int. Congr. Skin Melanoma, Venice, 1985 (in press). [22] PORTER, E.H., The statistics of dose-cure relationships for irradiated tumors, Br. J. Radiol. 53(1980) 336. [23] CHADWICK, K.H., LEENHOUTS, H.P., A molecular theory of cell survival, Phys. Med. Biol. 18(1973) 78. [24] FOWLER, J.F., What next in fractionated radiotherapy? Br. J. Cancer 49 (Suppl. VI) (1984) 285. IAEA-SM-290/S2

DEVELOPMENT OF THERMAL RADIATION THERAPY FOR PATIENTS WITH RADIORESISTANT TUMOURS

A. CYB, A. KONOPLYANNIKOV Research Institute of Medical Radiology, Academy of Medical Sciences of the USSR, Obninsk, Union of Soviet Socialist Republics

Abstract DEVELOPMENT OF THERMAL RADIATION THERAPY FOR PATIENTS WITH RADIO RESISTANT TUMOURS. The authors describe results obtained in treating tumours with a combination of radiation and local hyperthermia. The hyperthermic effect was produced at 42-43°C for one hour on special devices designed for this purpose in the USSR (2450, 915, 460 and 40.68 MHz). Rational schemes were devised for combining radiation and heat treatment, taking into account the radiosensitizing and cytotoxic effect of hyperthermia. Experience was obtained in treating patients with cancer of the rectum, osteogenic sarcoma, cancer of the corpus and cervix of the uterus and tumours of the oral cavity, larynx and throat; and there was evidence of a beneficial adjuvant effect of hyper thermia in radiation therapy or combined therapy.

Additional localized hyperthermia is now considered to be one of the most promising methods for increasing the effectiveness of radiation therapy for cancer. Our own work [ 1, 2], and that of other authors [3, 4], shows that the use of hyperthermia in radiation therapy is based on the following premisses:

(1 ) Hyperthermia (> 42°C) inhibits the repair processes in irradiated tumour cells and thus increases radiation damage; (2) Hyperthermia itself has a cytotoxic effect which is most clearly expressed in radioresistant tumour components (cells in hypoxic zones, cells in S-phase and so on); (3) During hyperthermia, tumours tend to become hotter than the surrounding normal tissues because blood flow is slower and heat removal is therefore less effective, and the processes of heat sensitization and heat damage are more pronounced in tumour cells; (4) In tumour damage by heat radiation, a stronger reaction by the organism against neoplasia can be expected (immune response, enhancement of non specific resistance and so on).

2 2 1 222 CYB and KONOPLYANNIKOV

Weeks / 1 / 2 / 3 / 4 / 5 / First stage Second stage

Irradiation X X X X X X xxxxx xxxxx 4—6 Gy, twice weekly 2 Gy, five times weekly 2— 3 weeks 2 or more weeks

Hyperthermia H H H H Before irradiation Not performed 42— 43°C, 1 h

Additional treatments — chemotherapy, surgery, preliminary brachytherapy w ith fast neutrons ( 252Cf) — are only carried out for certain tumours

FIG . 1. General outline o f com bined heat and radiation treatm ent plan fo r cancer patients developed at the Research Institute o f M edical Radiology o f the Academ y o f M edical Sciences.

Analysis of the experimental data has enabled us to establish several important principles concerning the best ways of using combined heat and radiation treatment for several types of tumour:

(1) It is essential to achieve homogeneous heating of the tumour to no lower than 42 to 42.5°C. Heating should then continue for about 60 min; (2) Heat treatment should be combined with high and medium fractionation plans (4 to 6 Gy) once or twice a week. The total number of fractions of combined heat and radiation therapy ranges from 3 to 6 or more; (3) Heat should evidently be applied immediately before irradiation; (4) In radiation treatment of patients with various types, locations and sizes of tumour, additional heat treatment may be used at the beginning of the course and also about half way through if tumour regression is poor.

Clinical trials are currently in progress at the Research Institute of Medical Radiology of the USSR Academy of Medical Sciences to test combined heat and radiation therapy techniques on cancers of the rectum, osteogenic sarcoma, oral cavity, larynx and glottis tumours, cancers of the corpus and cervix uteri and others. A general combined heat and radiation treatment plan is shown in Fig. 1. It differs from the conventional daily irradiation plan of small fractions of about 2 Gy, which is in widespread use. For combination with heat treatment, doses of 4 to 6 Gy per fraction were normally delivered once or twice a week. Over the course, a total of 3 to 6 hyperthermia sessions were given followed by irradiation, and treatment then reverted to a conventional, small-fraction irradia tion plan. This treatment plan enabled the advantages of combined heat and radiation treatment to be demonstrated, both as a radical treatment and for pre-operative irradiation. For local heating of neoplasms to 42-44°C, UHF (2450, IAEA-SM-290/52 223

TABLE I. CHARACTERISTICS OF HYPERTHERMIA SYSTEMS FOR LOCALIZED TUMOUR HEATING

Microwave Applicators Power Temperature Hyperthermia system frequencies monitoring (MHz) (W) External Internal

Yacht-2 (Sail) 2450 200 + - Automatic Yacht-3 (Raft) 915 200 + + Automatic Yacht-4 (Skiff) 460 300 + + Automatic Hypertherm (Screen) 40.68 400 + Manual ф 3-15 cm; Length 5-15 cm

915 and 460 MHz) and VHF (40.68 MHz) radiation was used. The main character istics of the equipment used are given in Table I. The equipment can heat tumours with various locations and volumes using both external and specially designed intracavitary applicators, the latter for tumours of the rectum and corpus uteri. Tumour and normal tissue temperatures were measured using sensors inserted into the tissue with a needle or a special guide. Special attention was given to measuring temperatures at the boundary between tumour and normal tissue, with one of the sensors introduced under the base of the tumour. The readings from this sensor were taken as the basis for manual or automatic regulation of the heat output of the equipment. The temperature here was maintained at 42-43°C. Currently, along with thermistor and thermocouple devices (minimum diameter 0.6 mm) we are using non-metallic fibre optic glass sensors (minimum diameter approximately 1 mm). These are used in the first instance for tempera ture measurement in intracavitary hyperthermia. Hyperthermic temperatures (> 42°C) are reached in 10 to 15 min at most, irrespective of the type of device used, and a constant temperature is then main tained, generally for 1 h. Temperatures are constantly monitored throughout and microwave output regulated manually or automatically. The power needed to maintain hyperthermia, depending on tumour location and size, the type of applicator used and a number of other factors, is from 20 to 200 W. To avoid overheating superficial skin layers and other normal tissues, forced air or water cooling is applied. Circulating water in specially designed applicators make it possible, not only to cool normal tissues, but also to focus the electromagnetic field onto the proper site. Five years’ experience with the equipment, comprising 1100 UHF and VHF sessions, have demonstrated its reliability, ease of operation and safety. 224 CYB and KONOPLYANNIKOV

TABLE II. IMMEDIATE RESULTS OF COMBINED HEAT AND RADIATION THERAPY IN PATIENTS WITH VARIOUS TUMOURS

Regression No. Localization No. of patients NR PR TC

( 1) Rectal cancer 54 17 30 7 (31%) (56%) (13%)

(2) Cancer of corpus uteri 15 2 13 - (13%) (87%)

(3) Osteogenic sarcoma 38 22 16 - (58%) (42%) (4) Head and neck tumours (a) Primary 13 1 1 11 (8%) (8%) (84%) (b) Recurrent 18 2 9 7 ( 11%) (50%) (39%) (c) Lymph node métastasés 15 1 12 2 (7%) (80%) (13%)

The combined therapy department of our institute (headed by Professor B.A. Berdov) has had most experience over the longest time with combined heat and radiation treatment of rectal cancer. The effectiveness of localized UHF hyperthermia (915 MHz) was investigated in 83 patients with locally invasive rectal cancers, 55 of whom received combined heat and radiation therapy as a pre-operative treatment. The remaining 28, who had absolute contra-indications to surgery, were given the treatment as the first stage in a split course of radiation therapy. All the patients who received pre-operative combined heat and radiation therapy had localized, difficult-to-remove tumours invading various segments of vagina or prostate. Thorough investigation revealed no distant métastasés in organs or the various lymph node groups, i.e. the patients were at the T2.3MqNq stage. The pre-operative course consisted of 10 fractions of 4 Gy, 3 times a week. Heating to 42-43°C began from the third irradiation session in order to reduce the risk of possible metastasis from the tumour periphery, where cells are well supplied with oxygen and should be partially ‘sterilized’ by the first 3 fractions. After a total dose of 40 Gy is accumulated and before surgery there is a break of 3 to 4 weeks which is needed for the effects of the combined treatment to manifest themselves as a regression of the primary tumour. Table II shows that 13% of patients showed complete tumour regression (CR), which was never IAEA-SM-290/S2 225 * observed in the control group. Significant regression (>50%) was observed in more than half the patients receiving combined heat and radiation treatment and in only one-third of those receiving pre-operative irradiation alone. The differen ces observed are statistically significant and indicate that addition of localized hyperthermia to the course of pre-operative therapy enhances tumour damage. Radical surgery was attempted after the 3 to 4 week interval. The combined treatment made this possible in 61.2% of patients, compared to 22.2% for radia tion therapy alone; i.e. addition of hyperthermia to pre-operative irradiation tripled the operability of patients with locally invasive cancer of the rectum. A total of 31 patients underwent radical surgery, for whom follow-up periods range from two months to four years. Three patients died of the disease after three to four years. The remaining 28 are still alive with no evidence of local recurrences. In patients for whom surgery was contra-indicated, the combined therapy was supplemented with a radical programme of irradiation (daily superfraction ation up to a total focal dose of 75-80 Gy). Patients irradiated according to the same plan but without additional heat treatment served as a control group. During a 1 year follow-up period, survival in the main group was 21 of 28 (75%) and 8 of 19 (42%) for the control. Thus, additional heat treatment of patients with rectal cancer improved the results of both combined (radiation plus surgery) therapy and radiation treatment alone. There were good immediate results from combined heat and radiation therapy also in patients with other types of tumour (Table II). In treating patients with radioresistant tumours (stages I and II cancer of the corpus uteri), heat (intracavitary, to about 42°C) was applied once a week before intracavitary irradiation and the radiation dose was reduced to 3 Gy per fraction (from 4 Gy in the control group), which is the dose modification factor for the normal tissue, in order to avoid any damage to normal tissue. There was marked tumour regression in 87% of the patients, much higher than for irradiation alone. Treatment of this group began only recently, and longer term results will be published later. Combined treatment of patients with osteogenic sarcoma comprised localized VHF heating (tumour temperature from 42 to 47 °C), irradiation and chemotherapy (adriamycin and adriablastin). From 6 to 12 months later, radical surgery was performed in patients with no evidence of metastasis. The reaction of this radio resistant tumour to combined heat and radiation was very pronounced. Marked tumour regression (>50%) was observed in 42% of the patients, compared with 16% of the control group. A lack of significant tumour regression, or even continued growth, was observed in 10% of patients in the main group, compared with 44% in the control group. The addition of localized heat treatment to the combined treatment plan did not increase the frequency of métastasés; radiation damage to normal tissue in the irradiation field was no greater and in some cases even less. Morphological studies of tumours removed during surgery showed that combined heat and radiation treatment causes pronounced damage to osteogenic 226 CYB and KONOPLYANNIKOV

sarcoma cells. An effective combined treatment plan (including localized hyper thermia) for osteogenic sarcoma patients has now been developed for the first time and enables the growth of the primary radioresistant tumour to be controlled. The use of localized VHF hyperthermia in treating tumours of the oral cavity became possible due to the development of individual water cooled appli cators which could heat tumours selectively. The effect of combined heat and radiation treatment, generally carried out against a chemotherapy (bleomycin) background, is highly dependent on the nature of the neoplasm (Table II). By the end of therapy, 84% of patients with primary tumours showed complete regression and 8% showed marked (>50%) regression; of patients with lymph node métastasés, there was complete regression in 13% and marked in 30%, and patients with recurrent tumours occupied an intermediate position with complete regression in 39% of cases and marked regression in 50%. It is worth noting that a number of the patients in these groups were given doses of 3.5 to 5.3 Gy from a 252Cf source at the start of therapy using the implantation technique rather than gamma radia tion from a 60Co source. Where this was the case, localized VHF heating was given, from 5 to 7 days after removal of the 252Cf sources, in conjunction with chemotherapy and teletherapy at 5 Gy per fraction. The follow-up period in this group has also been rather short, but the immediate results indicate that combined heat and radiation therapy is a rather promising technique in patients with tumours of the tongue and oral mucosa, and also for those with métastasés in the lymph nodes of the neck; the probability of obtaining a good therapeutic effect with traditional methods is very low indeed. The institute has acquired some experience in using localized hyperthermia in the combined treatment of laryngeal cancer. A total of 37 patients have been treated, of whom 21 received 915 MHz UHF radiation and 16 received 40.68 MHz VHF. It has been found that localized heating in radical radiation treatment of this cancer increases the 3 year survival rate to 50%, compared to 40% for the control group. Even higher survival rates were achieved by adding hyperthermia to combined treatment by radiation and surgery, where 3 year survival was 87%, and 75% with no recurrences. The data above show how effective additional hyperthermia is in therapy with radiation and combined radiation and surgery. The electromagnetic localized hyperthermia equipment, developed in the USSR, is remarkable for its reliability, ease of operation and relative cheapness. The principles of the heat and radiation therapy method, based on previous radiobiological research, proved to be correct and made it possible to achieve better results in the treatment of cancer. Of course, much prolonged and painstaking work remains to be done in the development of this new and promising line of treatment, but the first stage can indeed be consi dered complete. The effectiveness of combined hyperthermia and radiation techniques, and also the fact that they are within the reach of most conventional cancer radiotherapy centres, indicates that they are to be recommended for general adoption in many countries, including the developing countries. This is all the IAEA-SM-290/S2 227 more urgent for the developing countries, as problems are frequently encountered there when patients suffer from neglected tumours which conventional radio therapy alone cannot cure. In this situation, additional hyperthermia is the best adjuvant for radiation or combined treatment.

REFERENCES

[1] KONOPLYANNIKOV, A.G., et al., Localized Hyperthermia in Radiation Therapy for Malignant Neoplasms (in Russian), Medical Information Service, Moscow (1983). [2] ALEXANDROV, N.N., et al., The application of hyperthermia and hyperglycaemia to the treatment of malignant tumours, Medicine, Moscow (1980). [3] HAHN, G.M., Hyperthermia and Cancer, Academic Press, New York (1982). [4] OVERGAARD, J. (Ed.), Hyperthermic Oncology (Proc. 4th Int. Symp. Aarhus, 1984), Vols 1 and 2, Taylor and Francis, London and Philadelphia (1985).

IAEA-SM-290/70

HYPERTHERMIA IN CANCER RADIOTHERAPY A clinical study at the Osaka City University Hospital

Y. ONOYAMA, T. NAKAJIMA, M. TSUMURA, A. KIDA, M. TANAKA Department of Radiology, Osaka City University Medical School, Osaka, Japan

Abstract

HYPERTHERMIA IN CANCER RADIOTHERAPY. A CLINICAL STUDY AT THE OSAKA CITY UNIVERSITY HOSPITAL. A clinical study of hyperthermia using a microwave and an RF capacitive type unit has been made under continuous temperature monitoring at several points in the tumour by multisensor probes. Eighty-seven patients with radioresistant tumours were treated between 1980 and 1985. Three cases of severe burns were observed in the early period of the study; however, no such complication was encountered later because of improvements to the applicator. Of 68 patients eligible to evaluate the primary tumour response, complete response (CR) was observed in 28%, partial response (PR) in 51% and no response (NR) in 21%. The overall response rate (CR + PR) was 90% for superficial tumours (less than 6 cm from the skin surface) and 53% for deep seated tumours. No definite correlation could be found between the degree of the primary effect and temperature. It is difficult to determine the exact relationship on the basis of the limited temperature data obtained by invasive thermometry. Better methods of estimating temperature distribution are necessary for a quantitative analysis of the efficacy of hyperthermia. Although several problems, such as improvement of heating and thermometric equipment, have to be overcome, we believe that hyperthermia can play an important role in cancer therapy in the near future.

1. INTRODUCTION

Attempts to treat cancer by elevated temperature have been made for many years; however, hyperthermia revived recently on the basis of modern experimental studies both in vitro and in vivo. The results of these studies elucidate the biological rationale for hyperthermia in cancer therapy, especially in combination with radiation: heat sensitivity of radioresistant ‘S’ phase cells and of cells with a low blood supply (nutritionally deprived, low pH and hypoxic), inhibition of repair from radiation damage, and selective heating of the tumour with decreased blood flow. Since the middle of the 1970s clinical trials of hyperthermia have been under taken especially for superficial lesions. Encouraging results have been reported [1—3]; however, many problems still remain to be solved before its routine use in clinical cancer therapy, such as the development of more efficient heating apparatus, thermometric systems and of methods to evaluate the measured temperatures.

229 230 ONOYAMA et al.

Since 1980 a clinical and experimental study of electromagnetically induced local hyperthermia has been carried out in the Department of Radiology of the Osaka City University Medical School. A total of 87 patients had been treated by the combined therapy by the end of 1985. This paper describes the outline of our treatment system and the clinical results with some considerations of thermometry.

2. EQUIPMENT FOR HYPERTHERMIA AND THERMOMETRY

Two types of heating machine are used in our department, a 2450 MHz microwave unit and a 6—13.56 MHz radiofrequency (RF) capacitive type unit. The clinical trial was first started using the microwave unit for physical therapy; however, the majority of patients were treated with the 8 MHz RF unit with a pair of water cooled electrodes, because of the limited penetration of microwaves. Figures 1 and 2 show the distribution of percentile rise of temperature, normalized to the maximum temperature rise on the vertical plane of the phantom containing 0.22% NaCl, 0.1% NaN3 and 4.0% agar agar [4]. Microwave units can heat only superficial tumours within 2—3 cm from the surface, whereas RF units can heat more deeply and the temperature distribution can be changed by a combination of electrodes with different diameters. Thermocouple probes were made by ourselves using very thin copper- constantan wire (0.08 mm in diameter) with sufficient electrical insulation by Teflon and RF filters such as ferrite coils and condensers. The temperature can be measured continuously by these probes during electromagnetic heating with the 2450 MHz microwave or the 6—13.56 MHz RF units. We developed a computerized thermometric system able to record temperatures at up to 23 points simultaneously by the combination of thermocouple probes, a digital temperature recorder and a microcomputer with floppy-disk data storage [5]. The temperature rise distribution in Fig. 1 was drawn by this.system using a subtraction and interpolation program.

FIG . 1. Tem perature distribution in the phantom heated with the 2450 M H z m icrowave unit,

150 W, fo r 10 min. IAEA-SM-290/70 231

(a)

(b)

FIG . 2. Tem perature distribution in the phantom heated with the 8 M H z R F unit, 300 W, for 10 min.

(a) Both electrodes are equal in size (21 cm diam eter),

(b) Com bination o f electrodes o f different size (18 and 21 cm diam eter).

In clinical thermometry, multisensor probes with 5 hot junctions at intervals of 1 to 2 cm are introduced into the prescribed position in the tumour through a catheter (18—19 gauge), which is inserted into the tumour under X-ray fluoroscopy or an ultrasound guide (Fig. 3). The temperature measured is continuously stored in the computer system and information, such as time-temperature curves, numerical data of temperature and patient data, is printed out by the plotter after the end of the treatment (Fig. 4). 232 ONOYAMA et al.

Copper (0-1 ттф enamel cover)

FIG. 3. Block diagram o f thermo metric system and multisensor thermocouple probe.

HYPERTHETRMIR 1 2 / 1 0 ^ 1 S B 5

U a 3 I- (Г к U Q- z и h

M I N U T E S 3

P it 1 ent n*m* Max Rve h*dl S t . d*v Eq . 11 me

Trettnent No. : 3002 2 40.7 39.96 40.2 . íes 6 . 967C-01 : 1

Cl tл le*t dIignot 1 s s RT-NECK TimOR 3 42.0 48.91 41.2 . 140 4 .244C+00

Hi «to log/ ! EPIDERMOID Cfl 4 43.? 42.47 43.0 . 195 Э.04ЭЕ+01 Tumor 11 te : RT-NECK 5 44.9 43.53 44.2 .245 9.347C+ei Oita «empfing interval C **cendt 1 : 60

43 6 45.5 43.87 44.5 .270 I.262E+02

FIG. 4. Example of printout from computerized thermometric system. Upper panel shows time-temperature curves at 5 points (Nos 2-6). Lower left is patient record. Lower right shows maximum, average, median temperature, standard deviation and equivalent time at 43°C for each measuring point. IAEA-SM-290/70 233

3. CLINICAL TRIAL AND RESULTS

Between 1980 and 1985 a total of 87 patients with locally advanced tumours or tumours histologically unfavourable to radiotherapy alone were treated with hyperthermia combined with radiation. A classification of patients according to primary disease is shown in Table I. Hyperthermia, aimed to raise the intratumoral temperature over 42°C for 30 to 40 minutes, was generally performed twice a week within about one hour after irradiation with a dose of 4 Gy. Conventional fractionation, 2 Gy 5 times a week, was also used for patients with large tumours or head and neck tumours. The total number of the heat sessions ranged from 1 to 11 (average 6.6), and the total radiation dose varied from 10 to 70 Gy, because a combination therapy was often used on the recurrent or resistant tumours showing an unfavourable response to conventional radiotherapy, as a booster treatment. Eighteen patients were treated with the microwave unit and 69 with the RF unit. The primary effect of the combined therapy was evaluated according to the degree of tumour regression. Complete response (CR) was defined as clinical disappearance of any measurable tumour, partial response (PR) corresponded to regression over 50% from the initial size of the tumour as the product of two dimensions, no response (NR) to regression under 50%. Nineteen patients, whose treatment was ceased before reaching 3 sessions of hyperthermia and a total radiation dose of 20 Gy or to whom it was given immediately before or after major surgery, were excluded from the analysis of the

TABLE I. CLASSIFICATION OF PATIENTS ACCORDING TO PRIMARY DISEASE

Site of primary disease No. of patients

Head and neck 26 Breast 13 Gastrointestinal tract 11 Liver 10 Soft tissue 7 Uterine cervix 4 Other miscellaneous 12 Métastasés from unknown origin 4

T otal 87 234 ONOYAMA et al.

TABLE II. PRIMARY RESPONSE ACCORDING TO MAXIMUM DEPTH OF TUMOUR

Tumour response Overall response No. of patients CR PR NR rate (%)

Superficial tumours 49 17 27 5 90 (less than 6 cm) Deep seated tumours 19 2 8 9 53 (more than 6 cm)

Total 68 19 35 14 79

primary effect. The most frequent cause of incomplete treatment was the poor general condition of the patient, followed by local pain in obese patients whose subcutaneous fat layer was more than 2 cm thick, and nausea and vomiting in those with head and neck tumours. The primary response was evaluated as CR in 19 patients (28%), PR in 35 (51%), and NR in the remaining 14 (21%). The overall response rate (CR + PR) was 79% for the evaluable 68 patients. The response rate varied with the depth of the tumour. The response rate for superficial tumours with a maximum depth less than 6 cm from the surface was 90%, whereas that for deep seated tumours was 53%, although excellent response or long term survivors were sometimes encountered among the patients with tumours of the lung, liver and rectum (Table II). The results are partly attributable to the fact that the heating equipment currently available cannot heat deep seated tumours sufficiently in all patients, and partly to the fact that the degree of shrinkage of the tumour is not always a good measure for evaluating large and deep seated tumours, which often contain a lot of necrotic tissue in their central portion. Different response rates were observed according to the site of lesions even among patients with superficial tumours. The CR rates for tumours in the neck and supraclavicle were nearly equal to those for tumours in the chest, back and other superficial regions. However, the overall response rate was 84% for the former and 100% for the latter, because 5 NR patients belonged to the neck and supraclavicle group (Table III). Although various reasons may be responsible, one of the most important causes may be attributed to the inhomogeneity of the temperature distribution due to the small size of electrode applicable in these regions and to the anatomical complexity, such as the existence of the airway, bony structure and great vessels, as discussed later. No significant difference in the response rate according to the histology of the tumour was observed, especially in superficial tumours. This was 92% for IAEA-SM-290/70 235 squamous cell carcinoma and 88% for adenocarcinoma and other miscellaneous tumours (Table IV). The result suggests the existence of a beneficial effect of hyperthermia, because adenocarcinoma is usually more radioresistant than squamous cell carcinoma. We have often experienced an unexpectedly good response in patients with fibrosarcoma, hepatocellular carcinoma, malignant fibrous histiocytoma, etc.

TABLE III. PRIMARY RESPONSE ACCORDING TO TUMOUR SITE

Tumour response Tumour site No. of patients LR PR NR

Superficial Neck 25 9 13 3 Supracla vicie 6 2 2 2 Chest, back 13 5 8 0 Abdomen 2 0 2 0 Extremities, etc. 3 1 2 0

Deep seated Abdomen, pelvis 16 1 7 8 Chest, back 2 1 0 1 Groin 1 0 1 0

TABLE IV. PRIMARY RESPONSE ACCORDING TO HISTOLOGY

Tumour response Histology No. of patients CR PR NR

Superficial Squamous cell carcinoma 25 9 14 2 Adenocarcinoma 16 7 7 2 Others3 8 1 6 1

Deep seated Adenocarcinoma 11 1 6 1 Squamous cell carcinoma 2 1 0 1 Others3 6 0 2 4 a Soft tissue sarcoma, malignant fibrous histiocytoma, malignant melanoma, malignant teratoma. 236 ONOYAMA et al.

Concerning the toxicity of hyperthermia, no conclusive information is available at present because of the limited number of patients and the short observation period. As an acute complication, 3 cases of severe burns, which required 3 to 9 months to heal, were observed in the early period of the trial when we were not so familiar with the use of the equipment. No such complication has been experienced after improvements to the cooling system of the applicator. It is difficult to determine the efficacy of combined hyperthermia on the basis of our non-randomized trial. However, the results strongly suggest that hyperthermia is effective in the treatment of radioresistant tumours, taking the unfavourable conditions of the patients into consideration. Hyperthermia treatment can be performed with minimal toxicity especially for superficial tumours of less than 6 cm maximum depth; therefore a large scale control study should be made to determine the efficacy of this treatment modality.

4. RELATIONSHIP BETWEEN TEMPERATURE AND TUMOUR RESPONSE

For a quantitative evaluation of the beneficial effect of hyperthermia in cancer radiotherapy it is important to know the relationship between intratumoral temperature and tumour response. Furthermore, a knowledge of the temperature distribution in both the tumour and the adjacent normal tissue is imperative to predict the beneficial and hazardous effects. In clinical hyperthermia by electromagnetic heating, the temperature distribution varies markedly according to time and space, as shown in Fig. 4. Therefore, it is difficult to define the representative temperature for the particular session of hyperthermia. As to spatial variation of temperature, we have

TABLE V. RELATIONSHIP BETWEEN RESPONSE AND TREATMENT TEMPERATURE

Response No. of patients Treatment temperature HTT (°C) LTT (“С)

CR 11 41.6 ±0.89 40.6 ±0.68 PR 23 42.0 ±0.81 40.7 ± 0.89 NR 8 41.5 ± 1.59 39.8 ±0.81 NEa 8 42.6 ± 0.61 40.5 ± 1.33

3 Not evaluable for response. IAEA-SM-290/70 237

О)E о>Е Ь Н О)с Е «о о К XО) X

Low Treatment Temperature (°С) Low Treatment Temperature C'C)

Low Treatment Temperature (°C)

FIG. 5. Relationship between high treatment temperature (HTT) and low treatment temperature (LTT). Scattergram for (a) neck and supraclavicle, (bj chest, back and other superficial regions, and (с) deep seated tumours.

temperature records for several points in the tumour, because a multisensor probe was used in most patients. We defined the arbitrary median value of the highest temperature measured for each session as high treatment temperature (HTT), and that of the lowest as low treatment temperature (LTT). HTT and LTT do not mean true maximum and minimum temperatures in the tumour, and are only arbitrary indicators of the treatment, because of the limited number of measuring points and of the poor reproducibility of the measuring position in each session of fractionated treatment. Table V shows the relationship between the 238 ONOYAMA et al.

FIG. 6. Calculated temperature distribution in a patient with pelvic tumour (malignant fibrous histiocytoma) heated with the 8 MHz RF unit. degree of primary response of the tumour and HTT or LTT in patients treated with the RF unit and whose temperature was measured at several points in the tumour. No positive correlation was found between HTT and the degree of response. LTT for the NR group was about 1°C lower than that for the other groups; however, the difference was not statistically significant. The relationship between HTT and LTT was analysed according to the site of the tumour (Fig. 5). Marked differences in scattergram patterns were observed between those for neck and supraclavicle and deep seated tumours (a and c) and those for other superficial tumours (b). LTT for the former did not rise proportionally to HTT and did not exceed 42°C in any patient, whereas LTT for the latter rose proportionally to HTT and exceeded 42°C. This result means that the temperature achieved with the heating equipment currently available depends largely on the condition of the patients, such as the anatomical structure of the relevant site and physiological factors including the amount of blood flow. Therefore, it is impossible to determine the relationship between temperature and effect on the basis of insufficient information on the temperature distribution which can be obtained by invasive thermometry with multisensor probes. A quantitative analysis of the relationship between temperature and effect should be made separately according to the site of the tumour, which shows a IAEA-SM-290/70 239 similar pattern of temperature distribution, to minimize the influence of an inhomogeneous temperature distribution. The other method is to get more detailed information on the temperature distribution by using new methods. Computer simulation combined with correction by the temperatures measured at several reference points is one of the promising methods. We are developing a computer program in collaboration with colleagues in the Engineering School to calculate the temperature distribution by a finite element method, which takes account of the electrical and thermal characteristics of the human body and of the state of blood flow in all tissues. Figure 6 shows the temperature distribu tion calculated by this program in a patient with a pelvic tumour heated by 8 MHz RF. This work is still in the initial stages at present; however, we believe that computer simulation in combination with invasive thermometry can play an important role in thermal dosimetry in the near future.

5. CONCLUSION

The results of our clinical study of hyperthermia combined with radiotherapy were encouraging not only in superficial but also in deep seated radioresistant tumours. Response rates were 90% for the former and 53% for the latter. No definite relationship between effect and temperature in the tumour could be determined in the present study, mainly because of insufficient information on the temperature distribution by invasive thermometry with multisensor thermocouples. The problem will be overcome by improving the thermometric system and involving a computer system. Hyperthermia has reached the stage where its efficacy has to be determined by a large scale control study. Such a randomized study is in progress at present with the collaboration of many Japanese institutions. We believe that hyperthermia can play an important role in cancer therapy in the near future, although many problems, such as the development of more efficient heating equipment and thermal dosimetric systems, have to be solved by close collaboration of engineers, biologists and physicians.

REFERENCES

[1] KIM, J.H., et al., Combination of hyperthermia and radiotherapy for cutaneous malignant melanoma, Cancer 45 8 (1978) 2143. [2] U ,R., et al., Microwave induced local hyperthermia in combination with radiotherapy of human malignant tumors, Cancer 45 3 (1980) 38. [3] LUK, K.H., et al., Clinical experiences with local microwave hyperthermia, Int. J. Radiat. Oncol., Biol. Phys. 7 5 (1981) 15. [4] ISHIDA, T., et al., Muscle equivalent agar phantom for 13.5 MHz RF-induced hyperthermia, Shimane J. Med. Sei. 4 2 (1980) 134. [5] NAKAJIMA, T., et al., “Clinical experiences with hyperthermia in cancer radiotherapy: special reference to in vivo thermometry”, Modification of Radiosensitivity in Cancer Treatment (SUGAHARA, T., Ed.), Academic Press, Tokyo (1984) 381.

DISCUSSION

(Summary of discussion held on Papers IAEA-SM-290/69, 46, 12, 60, 73, 52 and 70)

The seven papers were discussed together. Their main topics were the subject of the discussion. The first topic was related to technical problems of hyper thermia combined with radiotherapy. In response to a question, N.M. Bleehen noted that he did not reject a sensitizing effect of hyperthermia combined with radiotherapy, but had only tried to stress that the clinical evidence had not yet shown an improvement in the cure rate of previously untreated cancer. According to N.M. Bleehen, there was certainly evidence for improved local control for a short while in recurrent diseases, but it was not clear whether a small increase in radiation dose would do the same with an equivalent increase in normal tissue reactions. Controlled studies should be carried out on previously untreated diseases with curative intent using optimal radiation schedules with and without well conducted hyperthermia. Replying to the question on the interaction between irradiation and hyper thermia with regard to the process of spontaneous reoxygenation known to occur during conventional daily fractionation, N.M. Bleehen mentioned that clinically there was no answer. However, it was known from experimental investigations that heat had a considerable effect on the tumour microcirculation and that might interfere with the reoxygenation. That would not be the case with sensitizers. Experimental results and results in patients with recurrent or metastatic diseases had not led to the conclusion that it was a problem. The advantage of hyper thermia was that it should kill cells pretreated from access to oxygen and sensitizers, thus enhancing the overall cell kill. G. Arcangeli added that the effectiveness of hyperthermia combined with radiotherapy could not be evaluated irrespective of the mode of fractionation of radiation; that he preferred to use conventional fractionation and to heat once a week at 43°C for one hour, for instance in the case of breast cancer, head and neck tumours, including the lymph nodes; that melanoma was an exception which was known to show better response to larger fractions and a larger total dose than the conventional one; and that such an approach could permit evaluation of the therapeutic gain of hyperthermia and its influence on normal tissue response. In response to the question of when the clinical assessment of the effect of hyperthermia and radiotherapy was made, N.M. Bleehen pointed out that clinical assessment of the effect of hyperthermia and radiotherapy should be done as frequently as possible but usually it was reported during the treatment, at the end of the treatment, and at monthly follow-up; that the overall response reported was the highest seen and the duration of the response was also recorded.

241 242 DISCUSSION

Regarding the question, G. Arcangeli believed that at that stage of investiga tions, evaluation of the response of the combined modalities should be done as it was done in the case of applying radiation alone; that partial response could not be considered by him as an indication of better response irrespective of irradiation alone, but that complete response and the local control during the time were preferred. Addressing the rostrum, a speaker wondered whether it was possible at that moment to agree on a simple method to describe the heat dose to tissue as had been agreed in relation to radiation dose. Two authors, N.M. Bleehen and Y. Onoyama, answered and both were of the opinion that it was too early to standardize hyperthermia and to define any new unit of thermal dose in clinical practice. The main reasons were different methods of heating, insufficient thermometry, heterogeneous temperature distributions in vivo, etc. Many factors had to be accounted for and more experience on the relationship between the temperature of single and multiple heating and tumour response should be accumulated before starting to solve the problem. In response to the next question, N.M. Bleehen expressed his doubts on the possibility of using non-invasive techniques routinely in the near future. According to his opinion, nuclear magnetic resonance (NMR) would be difficult to apply for this purpose both on theoretical grounds of temperature resolution and also with regard to measurement in real time. Thermography, as had been described by scientists in Sheffield, might be easier to use in the clinic. In response to questions, Y. Onoyama pointed out that hyperthermia usually had been given one hour after irradiation, that a radiosensitizing effect of hyper thermia had been observed in animal experiments but it was too early to draw any conclusions from that non-randomized clinical study. In response to other questions, Y. Onoyama said that isotherm distribution had been calculated by the finite element method of two dimensions using a large-scale computer. For the computation, the transverse section taken from CT had been divided into many small areas and electrical and blood flow rate constants for each area had been fed into the computer. The results of the calculation had been compared with the real temperature measured at several points of the same patient. The measured temperature had corresponded fairly well with the calculated one. This work was only beginning then and it was time consuming, but more simple methods which could be applied to a microcomputer were being developed. In the opinion of Y. Onoyama, computer simulation in combination with invasive thermometry in a limited number of points of the human body could play an important role in clinical hyperthermia. A speaker addressed the rostrum about the experience available on hyper thermia of tumours with a depth of more than 3 cm. In answering the question, N.M. Bleehen noted that the depth of the heating depended on the frequency of the microwaves and radio frequency applied. In most situations, superficial heating only was possible and it would, even so, be rather inhomogeneous. DISCUSSION 243

Heating at a depth greater than 5 cm was possible with multiple applicators such as the microwave equipment produced by BSD or the Thermotron RF from Japan. However, even with this equipment, heating was not always possible and would depend on the depth of the tumour and the extent of its blood perfusion. A speaker commented that the presence of blood vessels did seem to be the most important reason for inhomogeneous temperature distribution and that experimental and theoretical studies indicated that it was very difficult if not impossible to heat the surrounding of a blood vessel to an adequate level. N.M. Bleehen agreed with the statement but added that, hopefully, well perfused areas would have fewer hypoxic cells, resistant to radiation therapy; that even such large vessels as arteries and veins would introduce cooling zones by con duction rather than perfusion; that it would create problems in computer modelling of heat distribution as well as of course in in vivo conditions. T. Sugahara also agreed that the temperature in the surroundings of blood vessels was lower but radiotherapy and chemotherapy, which were effective in the surroundings, when combined with hyperthermia, could permit an improvement in local response. The second topic of the discussion was related to the problem of combination of hyperthermia with radiotherapy, chemotherapy and chemical sensitizers. A speaker, referring to the curves of heating response of tumours shown by G. Arcangeli, suggested to start treatment of large tumours which were behind the curves with chemotherapy and, after reducing their volume for example from 600 to 60 cm3, which was within the radius of G. Arcangeli’s curves, to combine with hyperthermia, which would be minimal. In his reply, G. Arcangeli noted that the idea might be acceptable but his curves were taken from the old study when he had had no opportunity to heat large tumours. However, improved equipment and techniques permitted more effective heating of larger tumours than before. Answering the question regarding the mechanism of the action in man of hyperthermia alone and in combination with irradiation and drugs, N.M. Bleehen mentioned that it was only'possible to speculate at that stage of knowledge. It was likely, he added, that there would be an increase in damage to the micro circulation by hyperthermia. This would most probably be maximal in those areas where blood flow was slow and heating greatest. These areas were likely to contain hypoxic cells so synergism or at least additive effects in terms of long term tumour control should still be obtained. Access of drugs to the tumour might be a problem, but they were unlikely to get deep into the tumour anyway and would still be effective in the well vascularizing areas. In connection with the report from the USSR, a speaker asked the rostrum to comment on the existing experience and the mode of hyperthermia potentiation of chemotherapeutic response. Comments were made by A. Konopliannikov, who pointed out that according to 4—5 years’ experience of the study in the USSR, hyperthermia had significantly potentiated curative effects of radiotherapy. It 244 DISCUSSION had increased threefold the radical surgery in pre-operative treatment for rectum cancer, provided complete or partial response of 42% of cases of osteogenic sarcomas and most of the cases of tumours of the corpus uteri, and enabled radical treatment of patients with oral cancer who were previously thought to be incurable. Answering the question of another speaker, A. Konopliannikov noted that bleomycin in a single dose of 7.5 mg had been given two hours before hyper thermia. This had been repeated from six to eight times over a period of three to four weeks. T. Sugahara, referring to experience in Japan, mentioned that the com bination of hyperthermia and chemotherapy was being studied in the clinic, particularly on tumours of the intestinal tract which were very difficult to irradiate without involving irradiation of a great volume of normal tissue. The response was reported to be good in some clinical conditions, but statistical data were still not available. With regard to which chemicals should be combined with hyperthermia, chemotherapeutic drugs might have preference over hypoxic sensitizers because chemotherapeutic drugs were effective not only against primary tumours but also against métastasés. Thus, the combination of chemo therapy and local hyperthermia might be very promising. G. Arcangeli added that the combination of hyperthermia and chemotherapy was a problem even more complicated than the combination of hyperthermia with radiation. Additional factors influencing tumour response should have been taken into consideration, among them being the degree of thermo-activation of the drugs and the permeability of the cell membranes. The study of these matters was still at the experimental stage. Replying to the next question, N.M. Bleehen pointed out that clinically there was no antagonism between Ro-038799 and SR-2508 applied together, while pharmacologically there was no interaction between them. He remarked that it was too early to say whether there was tumour response enhancement but the increased tumour concentration should effect optimal enhancement when both drugs were given together and the least normal tissue damage when radio therapy was given before. Replying to the question of a speaker, N.M. Bleehen indicated that there was no evidence for a higher concentration of Ro-038799 and SR-2508 in lymph nodes than in the primary tumour. There were differences associated with the degree of fibrosis in a limited context. Thus, breast cancer which was very fibrostic might show lower concentrations than very cellular tumours. The highest concentration had been found in malignant melanoma. The amount of necrosis could also influence the drug concentration. However, one should remember that our biochemical analysis is based on a mean concentration in milligrams of tissue rather than in the microenvironment of the hypoxic cells. In connection with the possibility to combine chemical sensitizers, hyper thermia and radiotherapy, N.M. Bleehen replied affirmatively but added that DISCUSSION 245 much more should be learned about each of the modalities before applying them. An increased effect of that combination had already been shown in the laboratory. In the clinic it might be envisaged to have a sequence of sensitizers followed by radiotherapy at the optimum time, soon afterwards followed by hyperthermia to interact with both, but so far this could not be done routinely. In the course of the discussion, S. Krishnamurthi made comments on combined treatment modalities. The biological effect of each modality should be well understood because of the objective being to improve therapeutic efficiency but not to combine therapeutic toxicity. Unless the toxicity of the individual treatment and its individual mode of action are evaluated empirically, combined multiple modalities should be studied in randomized prospective trials. The third and last topic of the discussion touched upon the strategy of involving developing countries in solving the technical problem of hyperthermia and its implementation in clinical practice. A speaker, referring to the title of the symposium, asked whether the audience was in a position to recommend hyperthermia and randomized trials on hyper thermia to developing countries in view of their technical and social infrastructure. Three authors reported their point of view on the matter. S. Krishnamurthy believed that as far as cancer therapy was concerned, many of the ‘so-called developing countries’ of the third world did, in fact, have good know-how so that in this respect they could not be considered developing countries. For example, India, Egypt, Malaysia, Singapore, etc. would have no problem in carrying out the clinical trials. The only problem occurring in developing countries was the establishment of co-operative programmes because of the difficulty of getting people together and follow-up of patients. The purchase of equipment was another problem of an economic nature. The current meeting, according to S. Krishnamurthy, might contribute to solving the first problem by facilitating the setting up of a co-operative study. Y. Onoyama agreed that hyperthermia was still developing and was a time consuming method of cancer therapy. However, in his institute hyperthermia was administered only to patients with radioresistant tumours. This permitted treatment of from 15 to 30 patients per week. Y. Onoyama hoped that the technical and economic problems of hyperthermia would be solved in the near future and hyperthermia would be available in the leading institutions of developing countries. N.M. Bleehen thought that hyperthermia was a complex research procedure and, therefore, it should not be practised at centres in developing or even industrialized countries It was a labour intensive procedure which needed meticulous engineering and other support. Specific randomized studies which were needed to get important answers should be restricted to specialized centres with great expertise in hyperthermia, adequate access to patients and their follow-up. Because of sub-group diseases, a few centres in a co-operative study should be involved. 246 DISCUSSION

Another speaker once again referred to the question posed which emphasized the title of the symposium, and was of the opinion that hyperthermia could not be recommended to developing countries and contested some statements. He said that multi-institutional randomized trials could not be started without an agreement on heat dose description, and that the possibility of treating 16 patients per week would help large numbers of people. In his response N.M. Bleehen repeated that clinical trials on hyperthermia in developing countries should have a low priority but treatment centres should concentrate on providing standard treatment to as many as possible. N.M. Bleehen expressed his disagreement with the above statement, that detailed studies were not possible especially in academic centres and particularly in industrialized countries. As soon as a group of experts in hyperthermia could agree to the dose specification and standardization of other clinical parameters, multicentre studies must be carried out, otherwise researchers would continue to present anecdotes rather than unequivocal assess ment of the value of hyperthermia in curative treatment. It would require considerable resources of equipment, personnel and medical dedication, and schedules with and without well conducted hyperthermia. N.M. Bleehen believed that researchers were moving to the stage where formal randomized studies on patients, other than with superficial skin nodules, would have to be done. In his response, T. Sugahara agreed to the above statement, pointing out that if people waited until ‘rad’ or ‘gray’ were established, radiotherapy might still be at an infant stage. As the first approximation, researchers had had reproducible clinical data on hyperthermia at that time. According to T. Sugahara’s calculations, 25 sessions a week and 1250 sessions a year were possible. If three heatings for each patient were enough, it would mean several hundreds of patients being treated each year. If one heating was as good as several heatings, the number of treated patients would be many more. S. Sugahara agreed that hyperthermia was still time consuming and labour intensive, but what could one say about surgery or neutron therapy? As against those, a hyperthermia machine costing US $50 000 to 300 000 might be affordable for developing countries. CHEMICAL MODIFIERS

IAEA-SM-290/43

RADIOTHERAPEUTIC POTENTIATION BY CHEMICAL AND PHYSICAL MEANS IN ORAL SQUAMOUS CELL CARCINOMAS

S. KRISHNAMURTHI, V. SHANTA, N. GOPALAN, A. VASANTAN, N.M.S. REDDY Cancer Institute, Madras, India

Abstract

RADIOTHERAPEUTIC POTENTIATION BY CHEMICAL AND PHYSICAL MEANS IN ORAL SQUAMOUS CELL CARCINOMAS. Oral squamous cell cancers have been the commonest male and the second commonest female malignancy on the South Indian Peninsula since time immemorial. Nearly 93% of these cancer cases attend a hospital for the first time at an unimaginably advanced stage of the disease. Improved education, communications and health services have taught them to seek hospital aid but not early enough. While, therefore, prevention and early detection programmes are under way, mainly advanced cases are treated in the hospitals. Radiation is often the only possible treatment that can be offered, but the survival rate is poor — a mearge 19% at 5 years. Various combination treatments have been studied since 1958, in carefully designed concurrent randomized controlled clinical trials, in an effort to improve the rate of cure. The combination of radiotherapy and surgery raised the 5 year survival rate to 52%, where it could be practised, but its scope was limited. Chemical sensitization proved, after initial promise, disappointing. The only cytotoxic drug that, as a single agent, distinctly, reliably and consistently improved the radioresponse and survival was bleomycin — a CR (complete response) of 80%, a RFR (recurrence free rate) of 70% and a 5 year NED (no evidence of disease) of 60%. There was, however, a persistent failure rate of 30-40%. Three trials were undertaken to reduce this failure rate: ( 1 ) Hbo + BLM as radiopotentiators; (2) BLM + VCR + 5-Fu as radiopoten- tiating agents; (3) BLM + hyperthermia. The triple drug combination effected a CR of 76.3% against BLM + RT of 70.73% — an improvement that was not significant — but this was accom panied by unacceptable toxicities. The third trial is still going on.

1. INTRODUCTION

Oral squamous cell cancer is the commonest malignancy in the male (29%) and the second commonest in the female (18%) population on the South Indian Peninsula. Over 90% attend a hospital clinic for the first time at an unimaginably advanced stage of the disease. A relieving feature of these massive tobacco cancers, however, is that they tend to remain locoregional over much of their natural history, and remote métastasés are rare (0.75%) [1]. Even the nodal métastasés occur late and stay operable over a long period of time. The essential problem in these cases is, therefore, the control of the primary malignancy.

249 250 KRISHNAMURTHI et al.

The primary lesions, however, are so enormous that surgery, ab initio, is just impossible and the lifeline of therapy is only radiation. Radiation per se, however, yields only a meagre 19% 5 year survival NED (no evidence of disease) [2]. The following forms of combination therapies have been under trial at the Institute since 1958 in an effort to improve the survival rate:

( 1 ) Radiation + surgery, (2) Chemical sensitization with Synkavite, metronidazole and Hbo, (3) Cytotoxic drug potentiation with methotrexate (Mtx), CDDP, 5-Fu, VCR, bleomycin (BLM) and pepleomycin (PEP) in single and multidrug schedules, (4) Physical potentiation with hyperthermia alone or in combination with PEP.

Many of these trials have already been published [1—8]. A full radiation course followed by radical surgery did improve the 5 year disease free survival to 52%, but surgery could be practised in only 48% of the cases irradiated. Its applicability was, hence, limited [ 1 ]. Synkavite did induce an improved response, raising the 5 year survival NED to 38% - a modest achievement [3]. Metronidazole showed no positive results and Hbo was a disappointment [4, 8]. The only combination that did produce something spectacular was RT + BLM — a complete response (CR) rate of 80%, a recurrence free rate (RFR) of 70% and a 5 year survival NED of 60%. These results were consistent, reliable and reproducible. The clinical trials were repeated using PEP and an unselected series was also treated with the combination, obtaining almost identical responses. Nevertheless, there was also a persistent failure rate of 30—40%, which was not reduced by the addition of either Hbo or CDDP [4, 8]. Two other trials were, therefore, undertaken subsequently as part of the series.

2. TRIALS

2.1. Hypothesis

Barranco [9] demonstrated partial synchronization of dividing cells in vitro in the ‘S’ phase by BLM by a transient block in the S/G2 interphase. Nearly 4 times the usual number of cells were reported to have accumulated in the ‘S’ phase. ‘S’ is a radioresistant phase and hence it was hypothesized that an ‘S’ phase active agent, appropriately timed, could enhance ‘cell kill’. Anti metabolites and hyperthermia were both, reputedly, ‘S’ phase specific. Hence two trials were designed — the first using a combination of BLM, 5-Fu, VCR and RT and the second BLM + hyperthermia + RT. VCR was used essentially as an ‘M’ phase blocking agent to synergize with RT. Both were prospective concurrent randomized controlled clinical trials, both the randomization and evaluation being blind. IAEA-SM-290/43 251

2.2. Criteria of eligibility

(1) All histologically confirmed T 3 and T 4 cases (UICC classification) of squamous cell carcinomas of the buccal mucosa who had not had any form of specific treatment earlier. (2) Metastatic cervical N0, N[, N2 nodes, and N3 lymph nodes confined to the submandibular region were no bars but N3 nodes elsewhere were excluded. (3) Deep infiltration of soft tissues including skin, whether fungating or not, was no bar but extensive destruction of bone excluded the case. (4) Trismus resulting from infratemporal fossa invasion or extensive destruction of the face rendered the case ineligible. (5) Systemic métastasés excluded the case. (6) Age and disease such as diabetes mellitus, hypertension, tuberculosis, myocardial ischaemia and endocrinal disorders were no bars, but reasonably good general health was essential. (7) Neuropathies, pulmonary fibrosis or chronic respiratory diseases (emphysema, asthma), impaired renal or hepatic function excluded the case. (8) Only cases admitted to cobalt-60 teletherapy were eligible. (9) Informed consent was mandatory.

2.3. Construction of groups

The eligible cases were blindly randomized by the sealed envelope technique into coded groups (x, y, z, etc.) in the tumour registry. The coded protocols were sealed and deposited with the Director and Scientific Director, and broken only after statistical analysis. The drugs were administered by medical oncology as per schedule, the RT and hyperthermia were given by radiation oncology and the evaluation of response was done by the head and neck surgical oncology group.

2.4. Trial I ( 12th in the series)

2.4.1. Therapeutic protocols

RT: Cobalt-60 teletherapy at a SSD of 80 cm, dose rate > 100 rad/min, using two opposing wedged fields where necessary, on a 3 day/week schedule (Mondays, Wednesdays and Fridays), TmD of about 250 rad (cGy)/fraction, minimal total TmD of 5000-6000 rad/6-8 weeks.1 A 0.5 cm wax plate was used when skin was involved. RT was given about 30 h after the BLM injection.

2.4.2. Drug protocol

— BLM or PEP: 5 mg as IV bolus 24—30 h before RT, i.e. on Sundays, Tuesdays and Thursdays.2

1 TmD = tumour dose. 2 IV = intravenous. 252 KRISHNAMURTHI et al.

TABLE I. PATIENT MATERIAL IN TRIAL I

(X) Study 38 (Y) Control 41 79

TABLE II. SEX DISTRIBUTION

Male Female

Study 20 18 Control 21 20

TABLE III. AGE (IN YEARS)

Maximum Minimum Mean

Study 70 29 46 Control 72 31 45

TABLE IV. HISTOLOGICAL GRADE (BRODER’S)

Study (X) Controls (Y)

G I 34 37 G II 4 3 G III G IV Grading not possible — 1 IAEA-SM-290/43 253

TABLE V. CONCOMITANT DISEASES

Study (X) Controls (Y)

Hypertension 4 3 IHD 2 2 DM 2 2 PT 1 3 Hypothyroidism - 1

— 5-Fu: 250 mg as IV bolus about 24 h after the BLM injection (i.e. Mondays, Wednesdays and Fridays).

The average cell cycle time of an oral squamous cell carcinoma was assumed to be about 48 h on the basis of the literature, and of G[ 18-24 h. Unfortunately there is no cyto-flowmeter available at the Institute to measure cell cycle time. Serial biopsies to study LI and MI were not found practicable. A maximum ‘S’ phase buildup was calculated at about 24 h after the BLM injection on this basis and hence the chronology. VCR: 1 mg IV bolus on Mondays and Fridays only, at the same time as 5-Fu. The dose was limited because of the drug’s known neurotoxicity. As an ‘M’ phase blocking agent, it was hoped that it would maximize cycling cells in G2 and M at about 6 h after injection. The drug protocols ran concomitantly with the RT.

2.4.3. Patient material

Of the 97 eligible cases only 79 consented to the trial (see Tables I—V).

2.4.4. Criteria of response

(1) CR: Total clinical healing within the volume of irradiation at 8 weeks after end of RT. (2) RFR: Freedom from recurrence of tumour within the volume of irradiation at 5 years or death (whatever the cause and whichever was earlier). (3) 5 year survival with NED.

2.4.5. Results

The results are shown in Tables VI and VII. 254 KRISHNAMURTHI et al.

TABLE VI. RESULTS OF CR

Category No. of cases CR (No.) (%)

(X) Study 38 29 76.31 (Y) Controls 41 29 70.73

TABLE VII. RESULTS OF RFR AND TWO YEAR SURVIVAL NED

Category No. of cases RFR 2 year survival NED (No.) (%) (No.) Crude Corrected8 (%) (%)

(X) Study 38 20 52.63 17 44.73 47.36 (Y) Controls 41 22 53.65 18 43.9 43.9

3 Corrected for intercurrent deaths only. Note: Since the trial closed in late 1983 only the 2 year recurrence free and survival data are available at present.

2.4.6. Toxicities

There was no significant advantage in the triple drug combination, while the morbidities were excessive and unacceptable (Table VIII).

2.5. Trial II. Hyperthermia trial

2.5.1. Biological basis

As already mentioned, BLM or PEP induced a piling up of cycling cells in the ‘S’ phase around 24 h after injection in oral squamous cell cancers. Hyperthermia, an ‘S’ active agent, administered at that time, could enhance the tumour cell kill in the ‘S’ phase. Hyperthermia is believed to inhibit repair of SLD and PLD inflicted both by BLM and RT. It also damages the tumour cell membrane and promotes increased intracellular transport of certain cytotoxic drugs, including BLM, enhancing the latter’s lethal effect. The trial was conceived on the logic of this synergism. IAEA-SM-290/43 255

TABLE VIII. TOXICITIES

Morbidity Study Controls (No.) (%) (No.) (%)

Acute mucositis 22 57.8 15 36.5 Xerostomia 23 60.5 16 39.0 Alopecia 22 57.8 13 31.7 Abdominal cramps IS 39.4 4 9.7 Leucopenia (transient) 11 28.9 4 9.7 Peripheral neuritis (transient) 12 31.5 1 2.4 Tissue oedema and deep fibrosis 22 57.8 14 34.1

2.5.2. Criteria оf eligibility

As in section 2.2.

2.5.3. Construction o f groups

The eligible cases were blindly randomized into 3 coded groups (x, y, z), the trial groups were assigned to each code by the Director and Scientific Director and kept sealed in their offices. The three trial groups were:

(1) RT + PEP + hyperthermia (HT) (2) RT + placebo (PL) + hyperthermia (3) RT + PEP

It was proposed to enter 30 cases in each group, making a total of 90, over a 3 year period. The trial commenced in January 1984 and is scheduled to term inate on 31 December 1986.

2.5.4. Therapeutic protocols

PEP: 5 mg IV (intravenous) bolus in normal saline on the mornings of Sundays, Tuesdays and Thursdays. The placebo for the controls was only the normal saline. R T: High energy X-rays from a 6 MeV linear accelerator were used at an FSD of 100 cm. The dose rate was adjusted to around 80 rad (cGy)/min. 256 KRISHNAMURTHI et al.

A minimum TmD of 250 rad/fraction was administered on a 3 day/week schedule (mornings of Mondays, Wednesdays and Fridays) using two opposing lateral fields (wedged where necessary) to a total TmD of 5500 to 6000 rad in 7 - 8 weeks. HT: Deep tissue heating was obtained with the Thermotron RF-8. This is a capacitive unit generating RF radiations at a frequency of 8 MHz. The RF power is delivered through two coaxial cables to two corresponding metal disc electrodes, varying in size from 10 to 25 cm, mounted on two opposing fixed arms of a gantry. Only two opposing fields can be used. The metal discs are enclosed in plastic bolus bags through which cooled 1% saline is circulated. The saline temperature can be varied from 10 to 30°C, but we usually maintain a temperature of around 20 to 25°C. This induces surface cooling and reduces the risk of skin and subcutaneous fat burns. The bags also facilitate electrode-surface coupling. The transmitted power on this unit can exceed 800 W. Heating immediately followed radiotherapy. The interval between the end of radiation and the attainment of 42°C in the tumour was never permitted to exceed 20 minutes. The time taken to raise the tumour temperature from body temperature to 42°C never exceeded 12 minutes. This time sequencing was meticulously monitored to prevent the development of thermal tolerance and to synergize, to the extent possible (with the proviso that our hypotheses and understanding were reasonably correct), the biological effects of RT and HT. The tumour temperature was monitored by four flexible microprobes (0.64 mm dia) fitted with copper-constantan thermocouples, enclosed in fine plastic catheters (18 gauge). Two of these catheters were implanted in the tumours a few times in the initial stages, and the temperatures registered compared with those in the buccal grooves to standardize the thermometry. In subsequent sittings only the buccal groove temperatures were recorded. The temperatures were read on 4 digital thermometers. Hyperthermia was applied only twice a week — Mondays and Fridays — for 30 to 43 min per session, for fear of developing thermal tolerance. The HT course ran concomitantly with the RT schedule, the number of sessions averaging 12 to 16. The tumour temperature range was maintained at 42°C ± 5°C. The patients did not tolerate higher temperatures, and lower temperatures were not permitted as tumour stimulation and dissemination were feared, apart from thermal tolerance. All patients were sedated during HT. Valium (10 mg) and Fortwin (30 mg) were injected in an IV bolus at commencement of HT. This substantially mitigated patient discomfort during the 45 min immobilization.

2.5.5. Criteria of response

As in Trial I. IAEA-SM-290/43 257

2.5.6. Deficiencies in the trial

In the machine

The Thermotron RF-8 is ponderous, manoeuvrability is restricted, and the movements are coarse. The setting, in consequence, is both tiresome and needlessly prolonged, limiting the number of cases that can be treated in an 8 h day to a maximum of 8.

In the procedure

(1) Immobilization for 45 min is uncomfortable and the patients are restless towards the end. (2) About 15 min after commencement of HT systemic reactions appear - profuse sweating, tachycardia, and occasionally arrhythmias. Head cooling with ice caps has to be continued right through the treatment for fear of CNS morbidity. Patients feel exhausted at the end of HT. (3) Precise thermometry is unsure. With the type of invasive thermometry provided one is never confident of an adequate minimal homogeneous thermal dose distribution in the tumour.

2 .5 .7. Morbidities encountered

(1) First degree skin burns at the electrode periphery. This can be prevented by large bolus bags inflated 2/3. (2) Cardiac arrhythmias: Tachycardia and extrasystoles. None of the patients, however, exhibited any abnormal ECG patterns, either during treatment or at follow-up. (3) Acute oral mucositis: This could not be attributed to hyperthermia alone. (4) One case of BLM pneumonitis (patient recovered).

All patients received a special high protein-high calorie diet (20% protein and 2000 calories) during treatment. Most patients’ body weight stayed static during the course of treatment, rose in a few, but losses were always less than 2 kg. The patients were carefully monitored clinically daily. Haematology and biochemical parameters were checked weekly and a radiogram of the chest and ECG was done at the beginning of treatment, at 3000 rad TmD, at 6000 rad TmD, and at monthly intervals thereafter, for 2 months, subsequently at 3 monthly intervals for 6 months.

2.5.8. Results

The results are shown in Tables IX—XI. 258 KRISHNAMU RTHI et al.

TABLE IX. PATIENT MATERIAL IN TRIAL II

No. of cases entered in the trial 1/1/84-30/6/86 68 No. of cases currently under treatment 6 No. of cases just completed treatment and 3 first evaluation not yet done No. of cases not completing treatment 7 (10.3%) No. of cases evaluated at 8 weeks 52

TABLE X. DETAILS OF CASES NOT COMPLETING TREATMENT

Interstitial pneumonitis 1 Haemodynamic instability under HT 1 Discontinued treatment 4 Declined HT 1

TABLE XI. RESPONSE IN THE 52 CASES NOW AVAILABLE FOR EVALUATION

Complete response (CR) 37 (71.20%) Residual disease (NR) 15 (28.80%)

The overall response rate does not differ very much from those obtained in the previous trials. The differential response rates in the three groups are not available yet, for obvious reasons, but should be by the middle of 1987. The heating machines, however, need to be considerably improved without escalating their cost, their operations simplified and the thermal dosimetry system developed to greater precision. IAEA-SM-290/43 259

REFERENCES

KRISHNAMURTHI, S., SHANTA, V., Combined therapy of buccal mucosal cancers, Radiology 99 (1971) 409. KRISHNAMURTHI, S., SHANTA, V., Evaluation of treatment of primary and secondary gingival carcinomas, Br. Med. J. 1 (1963) 1261. KRISHNAMURTHI, S., SHANTA, V., Clinical sensitisation in radiotherapy of oral and cervical cancers, Cancer 20 (1967) 822. KRISHNAMURTHI, S., SHANTA, V., “Combination therapies in oral squamous cell carcinomas”, Prospective Methods of Radiation Therapy in Developing Countries, IAEA-TECDOC-266, IAEA, Vienna (1982) 101. SHANTA, V., KRISHNAMURTHI, S., Combined therapy of oral cancer, Gann Monogr. on Cancer Res. 19 (1976) 159. SHANTA, V., KRISHNAMURTHI, S., Combined therapy of oral cancer: Bleomycin and radiation: A clinical trial, Clin. Radiol. 28 (1977) 427. SHANTA, V., KRISHNAMURTHI, S., Combined Bleomycin and radiotherapy in oral cancer, Clin. Radiol. 31 (1980) 617. SHANTA, V., KRISHNAMURTHI, S., SHARMA, M., Irradiation, Bleomycin and hyperbaric oxygen in treatment of oral carcinoma, Acta Radiol. 22 (1983) 13. BARRANCO, S.C., HUMPHREY, R.M., The relevance of in-vitro survival and cell cycle kinetics data to the clinical use of Bleomycin: Fundamental and clinical studies of Bleomycin, Gann Monogr. Cancer Res. 19 (1976) 83.

IAEA-SM-290/15

METRONIDAZOLE IN THE TREATMENT OF NASOPHARYNGEAL CARCINOMA

A. HIDAYATALLA Radiation Oncology Academic Unit, Faculty of Medicine, University of Khartoum, Khartoum E. ABDEL RAHMAN, H.M.A. HAMAD Radiation and Isotope Centre, Ministry of Health, Khartoum Sudan

Abstract

METRONIDAZOLE IN THE TREATMENT OF NASOPHARYNGEAL CARCINOMA. Metronidazole (MET), though a relatively week hypoxic cell sensitizer, is currently the most tolerable nitroimidazole for clinical usage. Through its clinical and biological character istics nasopharyngeal carcinoma (NPC) is a favourable tumour for radiosensitization. A random ized prospective study investigating the use of MET in the radiation therapy of NPC was conducted at the Radiation and Isotope Centre, Khartoum, during 1984 and 1985. Eighty-five patients were randomly selected into one of two groups for radiation treatment with or without MET. Using conventional doses and schedules of radiotherapy the treatment was well tolerated except for moderately severe gastrointestinal disturbances and mucosal reactions in the group receiving MET. A favourable treatment response was observed in the MET group with quicker and more lasting regression of the primary and lymph node tumour masses and an improvement in both the actuarial and recurrence free survival at two years post-treatment.

1. INTRODUCTION

Metronidazole (MET), a 5-nitroimidazole compound well known in medical practice as a trichomonacide, was the first drug of its type found to be a highly potent radiosensitizer of hypoxic tumour cells [ 1 ]. The efficacy of the drug was confirmed in hypoxic tumour systems [2 ] and demonstrated by clinical trials in human tumours [3, 4]. However, the use of the drug was soon abandoned in favour of the 2-nitroimidazoles, and in particular misonidazole, which promised to be a much more efficient radiosensitizer of hypoxic cells [5]. However, clinical trials with misonidazole for over a decade proved to be disappointing [6], largely because of the unacceptable neurotoxicity associated with the high doses of the drug required for useful radiosensitization. Other hydrophilic compounds are currently undergoing Phase I toxicity studies [7-9] as radiosensitizers and their acceptability for routine clinical use is still undetermined. On the other hand,

261 262 HIDAYATALLA et al.

MET is widely available, inexpensive and, despite moderate gastrointestinal disturbances, is currently the safest hydrophilic compound for radiosensitization. As early clinical trials on it were cut short by the abrupt introduction of misonida zole, there is still a place for continuing those early trials on other tumour systems. The present study is one such trial where MET is used as a radiosensitizer for the treatment of nasopharyngeal carcinoma (NPC). NPC is a common malignacy found in the Sudan. It formed 5.8% of all cancers seen at the Sudan Cancer Registry [10] and 7.8% of tumours treated with radiation at the Radiation and Isotope Centre, Khartoum (RICK) [11]. The tumour has a male to female ratio of 3:1 and tends to occur in young age groups, with 54% of the patients below the age of 40. Almost all the neoplasms seen at RICK were of the undifferentiated type and usually presented with large primary masses and huge cervical lymph nodes. Despite rapid regression of the tumour masses following irradiation, locoregional recurrence is unacceptably high. This is commonly accom panied by distant métastasés [12]. Post-irradiation recurrence or continued evolu tion of the tumour and nodes is more common with the larger masses of the primary and regional lymph nodes, suggesting the regrowth of colonogenic anoxic cells in these masses which have been protected from the full cytocydal effects of radiation therapy. The use of hypoxic cell sensitizers in such a situation is called for and has been advocated by Ho [13], but has not been tried so far. The present study aims at evaluating the effect of MET on the radiation treatment of NPC using optimal doses of MET with conventional doses and schedules of radiation therapy.

2. MATERIAL AND METHODS

2.1. Material

During 1984 and 1985 (inclusive) a total of 108 cases of NPC reported for treatment at RICK. Of these, 85 cases satisfied the selection criteria for the study; 23 were excluded on the basis of the factors shown in Table I. The cases selected for the study were randomized into two groups for radiation treatment with MET (Group I) and without MET (Group II). Each group was further divided into two subgroups for treatment with either three fractions of radiation per week or five fractions, so that four subgroups were formed (Table II). The groups and sub groups were all well matched for age, sex and stage of disease. For the purpose of this presentation, consideration will be given primarily to the two major groups — I and II — for treatment of NPC with or without MET. The effects of fractionation will be referred to only where relevant. Group I comprised 44 patients, of which 34 were males and 10 females. The age distribution was 12 to 67 years, with a median age of 37.2. It included cases of stage 1(1), stage II (6), stage III (30) and stage IV (7) with carnial nerve involvement. Group II comprised 41 cases, of which 30 were males and 11 females. The age distribution was 12 to 70 years, with a median age of 39.3. It included cases I AE A-SM-290/IS 263

TABLE I. NPC CASES EXCLUDED FROM THE STUDY

No. of cases Cause for exclusion

2 Age below 12 years 3 Metastatic disease 3 Poor performance 4 Incomplete treatment 2 Other histology 4 Chemotherapy 3 Previous radiation elsewhere 2 Refused treatment

23 Total

TABLE II. DISTRIBUTION OF STUDY GROUPS

Group Subgroup Radiation MET No. of cases

I A 3 F/W a + 21 I В 5 F/W + 23 II С 3 F/W - 17 II D 5 F/W - 24 a F/W = fractions per week.

of stage I (2), stage II (3), stage III (28) and stage IV (8). The staging was based on that proposed by Ho [13].

2.2. Radiotherapy

Radiation treatment was given to all patients with 60Co units using two lateral fields which extended from the base of the skull downward to include the nasopharynx and the cervical lymph nodes to at least 2 cm below the lowest palpable nodes. The anteroposterior margin extended backward to include the most posterior groups of nodes. Shielding of the spine, part of the ear and the parotid gland was provided where possible. Anteroposterior fields were occasionally added for huge nodes to reduce the dose to the spine. 264 HIDAYATALLA et al.

A total dose of 5400 cGy (30 F/6 W) was given to patients (subgroups В and D) on the 5 F/W schedule1. An equivalent dose of 4800 cGy (18 F/6 W) was given to patients (subgroups A and C) on the 3 F/W schedule.

2.3. Metronidazole

The full course of treatment was given to all patients in the study. Patients in Group I were given MET orally on an empty stomach 2 h before the start of radiation treatment. The MET dose given was determined according to body surface area, i.e. 5 g/'m2 for each single dose with a maximum of 8 g per dose. In practice, tolerable doses varied from 6 to 8 g. A maximum of 9 doses with 6 given during the first 3 weeks of treatment and 3 during the last 3 weeks were optimal; however, a minimum of 6 doses was acceptable. MET was swallowed in tablet form with water, milk or a soft drink, according to preference. Anti emetics, including IM largactile were given whenever necessary.

2.4. Monitoring and follow-up

All patients were assessed during and at the end of treatment for tumour and lymph node regression as well as for side reactions from the radiation and MET treatment. Special note was made of gastrointestinal disturbances and of oropharyngeal mucosal reactions. Regular follow-ups were designed for all patients who were monitored for ( 1 ) primary and lymph node regression, (2) evidence of locoregional recurrence and distant métastasés and (3) long term side effects of radiation and MET tretment. Follow-up of patients was continued up to March 1986 so that patients had a minimum of 3 and a maximum of 24 months follow-up.

3. RESULTS

3.1. Side effects

Gastrointestinal disturbances and mucosal and skin reactions were the main early side effects of treatment. Gastrointestinal symptoms ranged from severe nausea with recurrent vomiting (severe), occasional vomiting (moderate) to loss of apetite only (mild). The incidence and intensity of these symptoms was much more pronounced in Group I patients than in Group II (see Table III). The severity of the reaction could be much relieved by using antiemetics and spacing or reducing the MET dose. Oropharyngeal mucosal reactions were common in the two groups, sometimes causing difficulty in swallowing and ulcerations of the mouth and pharynx. Bicarbonate and mycostatin mouthwashes proved useful, but relief usually followed ceasing radiation treatment for one or two days. Mucosal reactions are commonly associated with severe gastrointestinal reactions and are much more common in patients receiving MET.

1 F/W = fractions per week. IAEA-SM-290/15 265

TABLE III. EARLY TREATMENT REACTIONS

Gastrointestinal Oropharyngeal Group Severe Moderate Mild Severe Moderate Mild

Group I 12(27%) 22(50%) 10(23%) 17(39%) 28(41%) 9(20%) Group II 3(7%) 14(34%) 24(59%) 6(15%) 24(59%) 11 (26%)

Radiation skin dermatitis and hair loss at the site of treatment were noted in almost all patients. Both reactions were transient and tolerable. Later radiation reactions were confined to mouth dryness, particularly in patients where a large portion of the parotid gland was included in the field of treatment. This appeared to be more severe in patients receiving MET. Only one patient from Group I, subgroup A, showed transient Lhermitte’s Syndrome three months after the end of therapy. No acute or late neurotoxi city was otherwise noted, despite inclusion of part of the spinal cord in most treatment fields.

3.2. Tumour regression

Group I patients receiving MET exhibited earlier and more complete regression than Group II patients. In both groups there was good symptomatic relief by the end of the treatment period, except for the cranial nerve palsies which persisted after treatment. Residual nodes were observed in 6 cases in Group I. Five of these disappeared by 3 months post-therapy and only one progressed to clinical recurrence. On the other hand, 6 cases in Group II progressed to full- fledged recurrence within 3 months after treatment. MET appears to have protected Group I patients against early recurrence.

3.3. Therapeutic results

The therapeutic results of the treatment regarding actuarial survival and recurrence and metastasis free survival for Groups I and II patients for the period of the study are shown in Tables IV and V, respectively. Actuarial survival curves for the two groups are shown in Fig. 1, while recurrence free (RF) survival curves for both groups are shown in Fig. 2. Both tables and figures show a follow-up period ranging from 3 to 24 months. About an equal number of patients in each group (10 in Group I and 12 in Group II) failed to report for the whole period of follow-up. Although a few of these patients may be alive and well, they are considered as dead for the values reported in the figures and tables. The therapeutic results indicate an overall improvement in actuarial and recurrence free survival in Group I over that in Group II. The improvement is 266 HIDAYATALLA et al.

TABLE IV. THERAPEUTIC RESULTS (GROUP I)

Patients Survival Follow-up (months) Eligible Lost Dead RF Actuarial

3 44 1 - 42 43 6 38 3 1 28 34 12 30 2 6 18 22 18 15 2 3 8 10 24 10 2 2 6 6

TABLE V. THERAPEUTIC RESULTS (GROUP II)

Patients Survival Follow-up (months) Eligible Lost Dead RF Actuarial

3 41 1 4 30 36 6 36 3 5 21 28 12 28 3 4 13 20 18 14 4 - 6 10 24 6 1 2 2 3

noted throughout the period of follow-up and is more pronounced for recurrence free than actuarial survival. One year post-therapy recurrence free survival for Group I patients was 60%, while that for Group II was 46%. This indicates a better local control for patients treated with MET.

4. DISCUSSION

4.1. Rationale

It has long been shown experimentally that the proportion of hypoxic cells in a tumour is related to the tumour size and type [14] and that the proportion of viable hypoxic cells increases to a maximum as the tumour outgrows its blood supply [15]. The role of hypoxic cells as a primary cause of radioresistance and treatment failure has been well established by early experimental clinical studies IAEA-SM-290/15 267

MONTHS

FIG. 1. Actuarial survival curve.

MONTHS

FIG. 2. Recurrence free survival curve.

[16, 17]. Of the various approaches to the problem of hypoxia — high LET radiation, hyperbaric oxygen, unconventional fractionation and hypoxic cell sensitizers — the sensitizer approach appears to be the most promising on the grounds of economy and simplicity of application [18]. Most of the sensitizers currently under clinical trials have either proved to be too toxic for routine clini cal practice [5], or are still in their early investigational phase [6, 7]. MET, though a relatively weaker sensitizer, has the advantage of low toxicity, availability and possibility of usage in routine radiotherapy practice. With its previous partial successful use in clinical practice [4, 5] it was an obvious choice for use in a developing country. NPC, by virtue of its clinical presentation as a large tumour with locoregional spread and relatively late métastasés and high local rate of recurrence, is a good candidate for radiosensitization. The use of MET for NPC sensitization in this study has been different from the commonly used high dose-short time courses. Our approach of multiple small doses of sensitizer with conventional radiation was adopted for its practical applicability. However, the total cumulative dose of over 42 g was higher than total doses given with short courses. Moreover, 268 HIDAYATALLA et al.

Fowler [19] has suggested that many small fractions of sensitizer might (1) give a better ultimate cell kill, and (2) expose the hypoxic cells to longer exposure to the cytotoxic effects of the sensitizer. Both factors would justify the use of the present method.

4.2. Results

The early treatment results were available for evaluation in the study for all patients. Side effects during the treatment from gastrointestinal disturbances and oropharyngeal reactions were sometimes severe, but were transient and toler able. The initial therapeutic effect, tumour regression, was quicker and more complete with MET. This result is in line with that reported from -South Africa, where significant improvement has been demonstrated in invasive carcinoma of the larynx and pharynx treated with radiation and the radiosensitizer misoni dazole [20]. Longer term effects are more difficult to assess because of the short follow-up periods and of the relatively large number of patients lost to the study. However, these lost patients are about equal for the two study groups so that comparison of the results of the two groups can still be valid. With such an assumption it is clear that the sensitizer group fared much better in local control of the tumour and actuarial and disease free survival.

ACKNOWLEDGEMENTS

The authors wish to thank the staff of the Radiation and Isotope Centre, Khartoum, for their enthusiastic collaboration in conducting the study and also Mrs. Patricia Kelly for her assistance in preparing this manuscript.

REFERENCES

[1] ADQUITH, J.C., et al., Br. J. Radiol. 47 (1974) 474. [2] HALL, E.J., et al., Radiology 117(1975) 173. [3] URTASUN, R.C., et al., New Engl. J. Med. 194 (1976) 1364. [4] DEUTCH, G., et al., Br. J. Cancer 31 (1975) 75. [5] FLOWER, J.F., ADAMS, G.E., DENKAMP, J., Cancer Treat. Rev. 3 (1976) 227. [6] DISCHE, S., Rev. Radiother. Oncol. 1 (1980) 157. [7] DISCHE, S., et al., Br. J. Radiol. 54 (1981) 156. [8] SAUNDERS, M.I., et al., Br. J. Cancer 46 (1982) 706. [9] ROBERTS, J.T., et al., Br. J. Radiol. 59 (1986) 107. [10] MALIK, M.O.A., et al., J. Natl. Cancer Inst. 62(1979) 221. [11] HIDAYATALLA, A., et al., Sur. J. Cancer Clin. Oncol. 19(1983) 705. [12] CACHIN, Y., et al., Nasopharyngeal Carcinoma Etiology and Control, 1RAC Scientific Publications, Switzerland, 20(1978) 131. [13] HO, J.H.C., Nasopharynx in Treatment of Cancer (KIETH, E., Ed.), Hainan, London (1980) 247. IAEA-SM-290/15 269

[14] KALMAN, L.F., Radiology 105 (1972) 135. [15] THOMLINSON, R.T., Proc. Carmel Conf. Time and Dose Relationship in Radiation Biology, BNL Rep. 50203 (C-57) (1969). [16] GRAY, L.H., et al., Br. J. Radiol. 26(1953) 638. [17] CHURCHIL-DAVIDSON, I., et al., Br. J. Radiol. 39 (1966) 321. [18] ADAMS, G.E., et al., Progress in Radio-Oncology (KÄRCHER, K.H., Ed.), Thieme, Stuttgart and New York (1980) 84. [19] FOWLER, J.F., Progress in Radio-Oncology (KÄRCHER, K.H., Ed.), Thieme, Stuttgart and New York (1980) 95. [20] OVERGAARD, J., et al., Acta Oto-Laryngol. Suppl. 386 (1982) 215.

IAEA-SM-290/40

EFFECT OF ORNIDAZOLE ON FRACTIONATED IRRADIATION IN CARCINOMA OF THE CERVIX AND LARYNX+

S. OKKAN*, R. UZEL*, Z. YAZICI**, A. AKÇASU***, N. TURAN***, S. TURKAN* *Department of Radiotherapy **Division of Biostatistics ♦ ♦♦Department of Pharmacology

Cerrahpa^a Faculty of Medicine, University of Istanbul, Istanbul, Turkey

Abstract

EFFECT OF ORNIDAZOLE ON FRACTIONATED IRRADIATION IN CARCINOMA OF THE CERVIX AND LARYNX. The sensitizing effect of ornidazole was tested in a double-blind randomized study of patients with carcinoma of the cervix and larynx. A total of 50 patients with carcinoma of the cervix and 36 patients with carcinoma of the larynx were randomized and followed up from 12 to 42 months from December 1982. All cases with carcinoma of the cervix were treated with external and intracavitary irradiation. A dose of 7.5 Gy in two fractions was given in the 1st and 3rd week. Conventional daily fractionation of 2 Gy was given in the 2nd, 4th and 5th week. Intracavitary treatment was followed by giving 2 fractions of 10.83 Gy each at one week interval. Ornidazole or placebo was given 3 hours before irradiation in a dose of 2.5 g/m2 in the 1st and 3rd week and intracavitary treatment. In patients with laryngeal tumours conventional fractionation was used and 70 Gy was given in 7 weeks. Ornidazole or placebo was given in the 3rd, 4th and 5th week in a dose of 2.5 g/m2. The results showed that there was no difference in local control and the survival rates in the ornidazole or placebo groups with carcinoma of the larynx. In contrast, in patients with carcinoma of the cervix the local control rate seems somewhat better in the ornidazole group (64% as against 48%). Two- year survival rates amounted to 64.2 and 51.6% in the ornidazole and placebo groups, respectively. However, these differences are not statistically significant and may be due to the small number of patients. Ornidazole was well tolerated by the majority of the patients. No neurotoxic side effects such as peripheral neuropathy or convulsions were observed with a total dose up to 30 g. Therefore, increasing the total dose and the number of drug fractionations may be tried with the hope of better and higher levels of sensitization. For this purpose it is concluded that the study should be continued for cervical cancer cases with an increase in the total dose of ornidazole together with a change in the external radiotherapy schedule.

+ Supported by the International Atomic Energy Agency with Research Grants Nos 3350/RB, RB/R1 and RB/R2.

271 272 OKKAN et al.

1. INTRODUCTION

Patients with advanced cervical and laryngeal cancers have a poorer prognosis, mainly because of failure to control the local disease with radiotherapy. One of the causes of local failure is radioresistance owing to the presence of hypoxic cells within the tumour. Hyperbaric oxygen trials suggested that hypoxia was an important problem for tumour control in carcinoma of the cervix and larynx [1,2]. Chemical agents sensitizing the hypoxic cells to radiation in vitro and in vivo may diminish this resistance. However, clinical attempts to overcome radio resistance with chemical agents, particularly misonidazole, have shown no advantage. The reasons for this can be attributed to limitations on the sensitizer dose because of the neurotoxicity of the drug, resulting in an insufficient concent ration of the drug in tumour tissue, thus an inability to produce oxygenation levels higher than that materializing during fractionated irradiation [3-5]. The present study was devised to investigate the sensitizing effect of another 5-nitroimidazole compound, ornidazole, which has been used as a sensitizer as reported previously [6, 7]. In this study the sensitizing effect of ornidazole was tested with fractionated irradiation in carcinoma of the cervix and larynx with a double-blind randomized plan. An interim report is presented.

2. MATERIAL AND METHOD

2.1. Patient population

A prospective double-blind randomized study was started in December 1982. All patients admitted to the study had to have stage IIB, III, IVA cervical or T3, T4 laryngeal cancers. The diagnosis was histologically confirmed in all cases. Complete blood counts, electrolytes, blood urea nitrogen, creatinine and liver enzymes were determined before the treatment. Patients less than 75 years old and without any evidence of renal or hepatic insufficiency \yere eligible to participate in the study. All patients were randomly allocated to receive tablets of either ornidazole or placebo 3 h before radiotherapy. Tablets were dispensed by the statisticians who possessed the randomization codes. The study commenced in December 1982 and until June 1986 64 cases with carcinoma of the cervix were entered into the study. Fifty cases who completed the treatment and were followed up for more than 12 months (mean 20.8 months) were ruled as eligible. The remaining 14 cases who are still unpaired or had a follow-up period less than 1 year were not evaluated. The patients were stratified by stage (FIGO), age, Hb level and histopathological differentiation (Table I). In the second part of the study 44 cases with carcinoma of the larynx have been included since December 1982. All patients refused surgery or were found IAEA-SM-290/40 273

TABLE I. STRATIFICATION OF PATIENTS WITH CARCINOMA OF THE CERVIX

Ornidazole Placebo Total

Stage

n B 13 13 26 III в 11 11 22 IVA 1 1 2

Hb level <12 g 14 16 30 >12 g 11 9 20

Age 30-39 5 7 12 40-49 2 4 6 50-59 11 6 17 60-69 4 6 10 70 + 3 2 5

Histological differentiation Well differentiated 15 12 27 Moderately differentiated 7 7 14 Poorly or undifferentiated 3 6 9

to be inoperable. Eighteen paired cases who completed the treatment and were followed up for 12 to 40 months (mean 16.5 months) were evaluated. The patients were stratified by T and N status (T3, T 4 and N0—N3) and age (Table II).

2.2. Radiotherapy

(1 ) Carcinoma of the cervix. All cases were treated with external and intra cavitary irradiation. External pelvic irradiation was given with a linear accelerator with a photon energy of 18 MeV and 45 Gy was given in 5 weeks to the mid pelvis through 15 X 15 cm anterior and posterior parallel fields. Two fields were treated each day. The radiotherapy schedule was altered to maximize the potentiating 274 OKKAN et al.

TABLE II. STRATIFICATION OF PATIENTS WITH CARCINOMA OF THE LARYNX

Ornidazole Placebo Total

Tumour T3 10 10 20 T4 8 8 16

Lymph node

N0 9 11 20 Ni 5 3 8 n 2 1 1 2 N3 3 3 6

Age 30-39 1 0 1 40-49 4 3 7 50-59 8 6 14 60-69 4 5 9 70 + 1 4 5

effect of ornidazole. For this purpose, 3.75 Gy was given on Mondays and Thursdays in the 1st and 3rd week. A total dose per week of 7.5 Gy was delivered, which was estimated to be radiobiologically equivalent to a weekly dose of 10 Gy given by conventional fractionation. The conventional daily fraction of 2 Gy was given in the 2nd, 4th and 5th week, bringing the total dose to 45 Gy. No central shielding was used. In some stage IIIB patients a booster dose of 10 Gy in 5 fractions was given to the extensively involved parametrium. Intracavitary irradiation using a high dose rate Curietron 60Co remote after loader was started one week after completion of external irradiation in 2 weekly fractions. In each fraction 10.83 Gy was given to point A, totalling 21.66 Gy. The technique has been described previously [8]. (2) Carcinoma of the larynx. A course of external beam treatment with a 60Co unit was planned for all cases. Two lateral parallel fields of7X9tolOX12i were used to deliver 10 Gy in 5 fractions per week to a total dose of 70 Gy in 7 weeks. The field size was reduced to 6 X 8 to 7 X 10 cm after 50 Gy. IAEA-SM-290/40 275

2.3. Ornidazole

Ornidazole1 was given to a stratified group of patients while their parallel pairings received placebo. The drug was given orally, twice a week in a dose of 2.5 g/m 2 for three weeks and the total dose was limited to 15 g/m2. Radiation therapy was always given 3 h after administration of the drug. In cervical cancer cases ornidazole or placebo was given twice a week with an interval of 72 h for the 1st and 3rd week of external irradiation and for the intracavitary treatments. To patients with laryngeal carcinoma the drug was given before irradiation on Mondays and Wednesdays during the 3rd, 4th and 5th week. Thus the interval between two consecutive administrations was 48 h. Blood and cervical tumour samples were obtained before and after the ornidazole administration. Plasma and tumour concentrations of ornidazole were measured by high pressure liquid chromatography (HPLC). The detailed technique has previously been reported [7].

2.4. Statistical analysis

In this report local control, survival and complication rates were evaluated. The McNemar test [9] for paired observations was used to analyse the local control of the disease, distant métastasés and treatment complications. One tail probability was accepted in testing the hypothesis. Actuarial life-table methods [9] were used to calculate the 2-year survival rates. The Mantel-Haenszel test [10] was used to compare the survival curves.

3. RESULTS

3.1. Carcinoma of the cervix

Table III shows the local control rates in the ornidazole and placebo groups. Local control was achieved in 64%, of the patients receiving ornidazole, compared with 48% for the placebo group. However, this difference was not statistically significant. Most of the recurrences (91%) were observed in the first year after treatment. A two-year follow-up period was reached for 36 patients. The follow-up periods of the remaining 14 patients were 12-22 months after treatment. In 36 patients who were at risk for 2 years the local failure rates were 28 and 50% for the ornidazole and placebo groups, respectively. In these patients the proportion of stage IIB was higher than for the entire group.

1 Biteral® (Tiberal®, F. Hoffmann-La Roche and Co. Ltd., Basel, Switzerland) and placebo were supplied by Roche, Istanbul, Turkey. 276 OKKAN et al.

TABLE III. LOCAL CONTROL AND DISTANT METASTASIS RATES IN PATIENTS WITH CARCINOMA OF THE CERVIX

Local control Distant métastasés Stage p p Ornidazole Placebo Ornidazole Placebo

Ив 11/13(85) 8/13(62) NS 0/12 2/12 III в 5/11(45) 3/11(27) NS 0/10 1/10 IVA 0/1 1/1 0/1 0/1

Total 16/25(64) 12/25(48) NS 0/23a (0) 3/23 a(13) <0.05

Four patients could not be evaluated for distant métastasés. Per cent figures are given in brackets. NS = Not significant

TABLE IV. VESICAL AND RECTAL COMPLICATIONS IN PATIENTS WITH CARCINOMA OF THE CERVIX

Group No. of patients3 Complication P (No.) (%)

Ornidazole 23 8 35 NS Placebo 23 4 17

Total 46 12 26 a Four patients could not be evaluated for the complication rate.

TABLE V. LOCAL CONTROL RATES IN PATIENTS WITH CARCINOMA OF THE LARYNX

Group No. of patients Local control P (No.) (%)

Ornidazole 18 5 28 NS Placebo 18 4 22

Total 36 9 25 I. . uvvlrtsofptet ih acnm h cervix. the f o carcinoma with patients f o rates Survival 1. FIG. I. . SurvivalFIG. rates 2.patients of withcarcinoma thelarynx.of Survival (%) IAEA-SM-290/40 Months Months 277 278 O K K A N et al.

The 2-year actuarial survival rates amounted to 64.2 and 51.6% in the omidazole and placebo groups, respectively (Fig. 1). These figure^ were 73.7 and 66.9% for stage IIB and 51.2 and 29.6% for stage III. Although the survival curves of the ornidazole group seem to be somewhat better than for the placebo group, this superiority was not statistically significant. Because of the small number of patients, the results were not evaluated according to other prognostic factors such as stage, age, histological differentiation and haemoglobin level. The visceral and mucosal radiation tolerance of the rectum, small intestine and bladder were satisfactory in both groups. Moderate and severe complications occurring 8 to 12 months after treatment were seen in 12 patients (26%). Although an increase in the complication rate was observed in the ornidazole group, this difference was not statistically significant (Table IV). In three patients taking the placebo distant métastasés were observed in contrast to none in the omidazole group (P < 0.05). Local failure was also present in all the patients with distant métastasés.

3.2. Carcinoma of the larynx

Local tumour control rates are summarized in Table V. There was no difference in local control rates between the ornidazole and placebo groups. Laryngeal oedema resulting from treatment was seen in 38% of the ornidazole group compared with 50% for the placebo group. Thirty-two patients have completed the 2-year follow-up period. The actuarial survival rates were found to be 26.4 and 29.6% at 30 months for the ornidazole and placebo groups, respectively (Fig. 2).

3.3. Side effects of the treatment

Ornidazole was well tolerated by the majority of the patients. Of the patients receiving omidazole, 44% had one or more side effects such as somnolence, dizziness and nausea. Side effects were seen more frequently in patients with carcinoma of the cervix. However, no neurotoxic side effects were seen in patients receiving omidazole. The serum concentration of omidazole reached its maximum level 2 to 4 h after oral administration. The tumour concentration of omidazole ranged from 14 to 93% of the serum concentration at the time of irradiation. The details of toxicology and pharmacokinetics have previously been reported [7].

4. DISCUSSION

In the present study local control and 2-year survival rates were found to be 64 and 64.2% for the ornidazole group compared with 48 and 51.6% for the placebo group in cervical cancer cases. These differences are not statistically significant and may be due to the small number of patients. IAEA-SM-290/40 279

The local failure rate of 15% for the ornidazole group compared with 38% for the placebo group in stage IIB cases is interesting and seems encouraging. It is known that more than 75% of local recurrences occur within the first 2 years and 96% in the 3 years following treatment [11, 12]. In our series 77% of the patients (20/26) had a follow-up period of more than 2 years and 86% (6/7) of the recurrences were seen in the first year after radiotherapy. In view of this it could be stipulated that the relapse pattern will show smaller differences in the following years. The rate of distant métastasés is 0% for the ornidazole group compared with 13% for the placebo group. Although this difference is statisti cally significant, its implication is obscure because of the small number of patients. In a study by Sutherland and MacFarlane [13] about twice as many cells were killed by ornidazole than by misonidazole when combined with hyperthermia. Therefore this finding may be explained by a lower incidence of local recurrence or a cytotoxic effect of ornidazole. In contrast to the encouraging results obtained in cervical carcinoma, ornidazole does not have a sensitizing effect in patients with laryngeal carcinoma. This may be explained by the difference in the radiotherapy fractionation scheme. Ornidazole probably has no sensitizing effect when reoxygenation is optimized using conventionally fractionated irradiation. On the other hand, it is accepted that the unconventional irradiation regimens using large doses would limit the benefit of reoxygenation. Therefore any study using large doses with sensitizers may show some benefit compared with those with no sensitizers such as the glioblastoma trial by Urtasun et al. [14]. In the present study the major part of the external irradiation consists of conventional fractionation, whereas high dose rate intracavitary treatment may be accepted as an advantage for the use of sensitizers. A recent study by the Medical Research Council [15] failed to show any benefit from using misonidazole in advanced cervical carcinoma. However, in this trial conventional daily fractionation and low dose intracavitary treatment were usually used in treating the patients and only 2 centres out of 12 employed a high dose rate technique. Previous experience with high dose rate intracavitary radiotherapy has demonstrated that local control and survival rates are similar to those of low dose rate treated patients [16-19]. Therefore, modifications in the external radiotherapy schedule and the employment of high dose rate intracavitary irradiation may be more advantageous when using sensitizers. Nausea, dizziness and somnolence were the prominent side effects in patients with cervical carcinoma, seen in 70% of the patients receiving ornidazole [7]. Although this rate is higher than the results reported with lower doses for protozoal infections [20-22], no neurotoxic side effects such as peripheral neuropathy or convulsions were observed with a total dose up to 30 g. On the other hand, the measurements of serum and tumour tissue concentration of omidazole showed that this drug is capable of diffusing into tumours in sufficient quantities and gives a good concentration relative to the serum levels [7]. There fore, increasing the total dose and the number of drug fractions seems feasible and may produce a higher level of sensitization. 280 O K K A N et al.

In conclusion, this interim analysis showed no evidence of benefits achieved from ornidazole in patients with laryngeal carcinoma. Therefore this part of the study was closed at the end of 1985. In contrast, although the difference in local control and 2-year survival rates is not significantly superior, the results seem to be somewhat better in the ornidazole group of patients with carcinoma of the cervix. On the other hand, increasing the total dose and the number of drug fractions may be tried with the hope of a better and higher level of sensitization. For this purpose, it was decided that the study should be continued for cervical cancer cases with an increase in the total dose of ornidazole together with a change in the external radiotherapy schedule.

REFERENCES

[1] HENK, J.M., Int. J. Radiat. Oncol., Biol. Phys. 7 8 (1981) 1 125. [2] WATSON, E.R., et al., Br. J. Radiol. 51 (1978) 879.

[3] BROWN, J.M., Int. J. Radiat. Oncol., Biol. Phys. 1 0 3 (1984) 425.

[4] DISCHE, S., Radiother. Oncol. 3 1 (1985) 97. [5] FOWLER, J.F., Int. J. Radiat. Oncol., Biol. Phys. 11 4 (1985) 665. [6] OKKAN, S., UZEL, R., Int. J. Radiat. Oncol., Biol. Phys. 8 10 (1982) 1735.

[7] OKKAN, S., et al., Radiother. Oncol. 5 4 (1986) 295. [8] UZEL, R., OKKAN, S., КОСА, S., “Results of treatment of cancer of the cervix with a high dose-rate Curietron”, Diagnosis and Treatment of Carcinoma of the Cervix in Developing Areas (Proc. Int. Working Party Meeting Thailand, 1985) (MOULD, R.F., TUNGSUBUTRA, K., Eds), Adam Hilger Ltd, Bristol (1985) 259. [9] ARMITAGE, P., Statistical Methods in Medical Research, Blackwell Scientific Publishers, London (1971) 127,408. [10] MUENZ, L.R., Cancer Invest. 1 5 (1983)455. [11] COMBES, P.F., et al.,Int. J. Radiat. Oncol., Biol. Phys. 11 3 (1985)463. [12] GIRINSKI, T., et al., Int. J. Radiat. Oncol., Biol. Phys. 11 10 (1985) 1783. [13] SUTHERLAND, R., MACFARLANE, W., Br. J. Cancer 37 (1978) 168. [14] URTASUN, R., et al., N. Engl. J. Med. 294 (1976) 1364. [15] MRC WORKING PARTY, Misonidazole for cancer of the cervix, Br. J. Radiol.

51 (1984)491. [16] BATES, T.D., BERRY, R.J., Br. J. Radiol., Special Rep. 17 (1980) Foreword. [17] CIKARIC, S., “The treatment of carcinoma of the uterine cervix by Cathetron at the Institute of Oncology and Radiology in Belgrade — Our 6 years experience”, Gynecological Cancer (Proc. Int. Working Party and Int. Conf. Istanbul, 1980), Z. Kamil Soc., Istanbul (1982) 179.

[18] TAINA, E., Acta Obstet. Gynecol. Scand., Suppl. 1 0 3 (1981) 15. [19] UZEL, R., Br. J. Radiol., Special Rep. 17(1980) 111. [20] BOTERO, D., Am. J. Trop. Med. Hyg. 23 (1974) 1000. [21] LEIMER, R., et al., Acta Tropica 37 (1980) 266.

[22] SKÖLD, M., GNARPE, H., HILLSTRÖM, L., Br. J. Vener. Dis. 5 3 (1977) 44. IAE A-SM-290/76

EXPERIENCE IN THE USE OF HYPOXIC CELL CHEMICAL RADIOSENSITIZERS AND THE POTENTIAL USE OF SOME NEWER RADIOSENSITIZERS

H.K. AWWAD, S. EL BADAWY, M. GHONEIM, M. BARSOUM, M. ZAGHLOUL, H. AKOUSH, M. ABOU ZAID, A. OSMAN, M. EL-NAGGAR Department of Radiotherapy, National Cancer Institute, Cairo, Egypt

Abstract

EXPERIENCE IN THE USE OF HYPOXIC CELL CHEMICAL RADIOSENSITIZERS AND THE POTENTIAL USE OF SOME NEWER RADIOSENSITIZERS. The existence of a significant hypoxia problem in carcinoma in the bilharzial bladder and carcinoma of the cervix uteri is strongly suggested on the basis of studies involving intercapillary distance measurements. Hence studies involving chemical radiosensitizers were limited to these two tumours, in addition to head and neck cancer. Five prospective randomized clinical trials were conducted in the National Cancer Institute, Cairo, investigating the use of misonidazole (MISO). Two of them involved conventional fractionation and patients with T3 bladder cancer and stages IIB and III cancer of the cervix, respectively. A marginal, although insignificant, benefit was observed in the MISO groups. Two other trials involved the use of 3 fractions per day radiotherapy regimens for either head and neck cancer or post-operative irradiation of bladder cancer. No significant therapeutic gain attributable to MISO was observed in either trial. The fifth study involved the use of MISO in combination with concentrated pre-operative irradiation in bladder cancer with no apparent therapeutic gain in the MISO group. The pharmacokinetics of the newer radiosensitizer SR-2508 was studied in patients with bladder cancer. After an IV dose of 2, g/m2, a tumour concentration of 120—140 ßg/g was attained with a very high urinary concentration. The tumour drug concentration was always greater than that of plasma, apparently owing to diffusion from urine. A second drug, Ro-03-8799, has basic properties and hence its concen tration in tumours was shown to be 4-6 times that of blood. The two drugs can be combined since they have different toxicity patterns. A total drug dose of 34 g/m2 of SR-2508 and 21 g/m2 of Roi-03-8799 can be safely administered compared with a maximum permissible dose of 12 g/m2 of MISO.

1. STUDIES ON INTERCAPILLARY DISTANCES AND CAPILLARY DENSITY IN SOME HUMAN SOLID TUMOURS

Solid tumours can outgrow their blood supply with consequent development of regions of vascular inadequacy. According to the oxygen diffusion model of hypoxia [ 1 ], the intercapillary distances (ICDs) can give an indication of tumour

281 282 AYfVfAD et al. cell hypoxia. The ICD measurements were performed in histological preparations in which the capillary endothelial cells were stained, utilizing their alkaline phos phatase activity [2, 3]. It could then be shown that squamous cell bladder cancer suffers more vascular insufficiency than does transitional cell cancer. In tumours of all cell types hypoxia tends to increase with histological grade of differentiation. The associated bladder mucosa apparently suffers less hypoxia. Cervical cancer showed a more efficient microcirculation than squamous cell bladder cancer. However, the pretreatment mean ICD was significantly greater in patients who suffered local recurrence within two years after radiotherapy than in patients who remained without evidence of disease. These studies seem to indicate that a significant hypoxia problem exists in these two tumours and that it is of a degree likely to influence the outcome of radiotherapy. Hence subsequent clinical studies involving the use of the hypoxic cell chemical radiosensitizer misonidazole (MISO) were limited to these two tumours, together with head and neck cancers, since a British study could show a therapeutic benefit from treating in hyperbaric oxygen [4].

2. STUDIES INVOLVING THE USE OF MISONIDAZOLE

Neurotoxicity is a dose limiting factor for the clinical use of MISO. It is generally agreed that the total drug dose should not exceed 12 g/m2. This introduces a serious limitation on the single daily drug dose, particularly when conventional daily fractionation is used. Five prospective clinical trials involving the use of MISO have been completed in the National Cancer Institute and these are briefly reviewed below.

2.1. Radical radiotherapy of bladder cancer: MISO trial

In this study patients with T3 bladder cancer were randomized to radio therapy alone or radiotherapy plus MISO. Radiotherapy comprised three courses, each involving 8 fractions of 250 cGy each, with one and two week interval between the first and second courses, respectively. A booster dose of 4 fractions of 250 cGy each of a reduced volume was then given after a rest interval of one week. MISO was given in a daily dose of 0.5 g/m2 , 3.5 h before irradiation. The two year disease free survival was 52% in the MISO group compared with 44% in the ‘radiotherapy alone’ group; this difference is not statistically significant. Peripheral neuritis was experienced by 63% and recovery was the rule [5].

2.2. Stages IIB and III carcinoma of the cervix: MISO trial

In this study patients with stages IIB and III carcinoma of the cervix uteri were randomized to either radiotherapy alone or radiotherapy plus MISO. IAEA-SM-290/76 283

Radiotherapy comprised external pelvic irradiation to a dose of 4000 cGy, followed by intracavitary radiotherapy. MISO was given in a daily dose of 0.5 g/m2 , 3.5 h before irradiation, and 2 g/m2 were divided into equal doses to be given during intracavitary irradiation. For stage IIB cases the two year survival was apparently greater in the MISO group (70% versus 37% in the control group) but the difference is not statistically significant. For stage III cases the corresponding rates were 44 and 45%, respectively. There was also indirect evidence of a better response in the MISO group, as indicated by a greater degree of reduction of the ICD and more active tumour volume reduction when scored immediately after whole pelvic irradiation. Peripheral neuropathy was observed in 54% of patients and was reversible in all of them.

2.3. MISO in combination with a 3 fraction per day post-operative irradiation regimen of bladder cancer

Patients who after cystectomy proved to have either P3 or P4a bladder cancer were randomized to either no further treatment or post-operative irradiation. The latter patients were then re-randomized to post-operative irradiation alone or irradiation plus MISO. A 3 fraction per day post-operative irradiation regimen was adopted with 125 cGy per fraction and an interval of 3 h between fractions. The total dose amounted to 3750 cGy, divided into 30 fractions over 10 treatment days spread over 12 days. MISO was given in a daily dose of 1.0 g/m2, 2 h before the first fraction. The four year survival amounted to 29% in the cystectomy alone group and to 49% in the cystectomy plus post-operative irradiation groups, irrespective of whether MISO was given or not. Peripheral neuropathy was noted in 85% of patients given MISO with some tendency for late recovery [6].

2.4. MISO in combination with a 3 fraction per day regimen in locally advanced head and neck cancer

Patients with stages III or IV head and neck cancer without evidence of distant métastasés were randomized to either radiotherapy alone or radiotherapy plus MISO. A 3 fraction per day regimen was adopted with a 150 cGy dose per fraction and an interval of 3 h between fractions. A total dose of 4500 cGy was given, divided into 30 fractions over 10 treatment days spread over 12 days. A booster dose.of 2250 cGy divided into 12 fractions over 4 treatment days was considered after a rest interval of 4 weeks. MISO was given in a daily dose of 1 g/m2, 2 h before the first fraction. This trial is still in progress but there seems to be a marginal benefit in favour of MISO. Thirty-eight per cent of the patients developed peripheral neuritis [7]. 284 A W W A D et al.

2.5. MISO in combination with concentrated pre-operative irradiation in bladder cancer

Patients with operable bilharzial bladder cancer were randomized to one of three groups: cystectomy alone, cystectomy preceded by pre-operative irradiation, or cystectomy plus pre-operative irradiation plus MISO. Irradiation comprised five daily fractions of 400 cGy each and cystectomy was donp within 1-3 days. MISO was given in a daily dose of 1.5 g/m2 , 3.5 h before each radiation dose. The two year survival rates were 44, 64 and 64%, respectively, i.e. MISO did not add to the therapeutic benefit of pre-operative irradiation. In this group there were no incidences of peripheral neuropathy.

2.6. MISO neurotoxicity

Peripheral neuropathy was the only toxic manifestation in the five MISO clinical trials. This toxicity was not observed in the concentrated pre-operative radiotherapy trial where the drug dose was 7.5 g/m2. This complication was most frequent in the post-operative 3 fraction per day bladder cancer trial (83%) and least in the head and neck study (38%). A number of factors were shown to influence the risk of peripheral neuro pathy [8]:

— Age: older patients are more susceptible; — Sex: males are generally more susceptible presumably owing to differences in the pharmacokinetics [9]; — Body weight: obesity favours the development of neuropathy; — Tissue exposure to the drug: this was, in our studies, the most significant factor and tissue exposure is readily expressible as the area under the plasma concentration curve; — The degree of upper obstructive uropathy was shown to influence the elimination of the drug with prolongation of the plasma Tm and increase in tissue exposure.

2.7. Appraisal of the MISO studies

It is generally agreed that the maximum total dose of MISO should not exceed 12 g/m2. When this is divided over the course of a daily fractionation conventional schedule, a daily dose of 0.5 g/m2 would be given. The mean blood concentration in the first two studies involving conventional fractionation was about 20 Mg/mL with an expected tumour concentration of the order of 15 ßg/mL. This concentration corresponds to an enhancement ratio for hypoxic cells of about 1.1. Hence it is not surprising that no therapeutic gain resulting from the use of MISO in the first two trials was demonstrated. In the cancer of the cervix IAEA-SM-290/76 285

trial (section 2.2) MISO was administered during intracavitary treatment on the assumption that an oxygen effect still exists during continuous low dose rate irradiation. The drug was given in doses of 0.5 g/m2 every 12 h throughout the application period. The blood drug concentration showed a gradual buildup until the second day when a relatively steady level was reached at a concentration of about 15—20 jug/mL. Attempts were made to increase the daily drug dose by adopting unconventional fractionation schemes. In addition to the radiobiological advantages of accelerated fractionation involving the delivery of multiple fractions per day, such regimens allow the enhancement of a relatively large radiation dose by a single drug dose. The slow elimination of the drug in man favours its use in combination with multiple fractions per day regimens. In the two studies involving the use of 3 fractions per day (sections 2.3 and 2.4) the blood concentration during the three fractions varied within 15% [6, 7]. An average blood concentration of 3 0 -4 0 Mg/mL was obtained with a corresponding enhancement ratio of 1.4— 1.5. Nevertheless, no therapeutic gain could be demonstrated in the post-operative irradiation bladder cancer study with a marginal benefit for head and neck cancer. The failure to demonstrate a therapeutic benefit in the bladder post-operative irradiation study may be due to the absence of serious hypoxia in the small tumour residues left over after cystectomy. The use of concentrated regimens with a few large radiation fractions allows the delivery of a larger daily drug dose. However, the use of such concentrated regimens does not make full use of the spontaneous re-oxygenation that is likely to occur during conventional daily fractionation. This loss may be larger than the possible gain obtained by the use of MISO in suboptimal doses. This may explain the failure to demonstrate any benefit from the use of MISO in the concentrated pre-operative irradiation study (section 2.5). These disappointing results stimulated the search for better chemical hypoxic cell radiosensitizers. To be effective, such agents should be at least as efficient as MISO, but 5 — 10 times less toxic.

2.8. Pharmacokinetics of SR-2508 in patients with bladder and cervical cancer

SR-2508 is a 2-nitroimidazole compound with an electron affinity similar to MISO, but the tumour concentration is about twice that after IV injection of an equimolar quantity of MISO. Based on the estimation of neurotoxicity by rotarod assay in mice and peripheral nerve drug concentration it is estimated that SR-2508 is 8—10 times less neurotoxic than MISO and has a therapeutic index of about 8.0 relative to MISO [10]. The pharmacokinetics of the drug have been studied in patients with carcinoma in the bilharzial bladder and cervical cancer (Table I). In the cases of bladder cancer the tumour concentration varied between 77 and 154 jug/g, which represented 124-244% of the blood level. In contrast, the tumour 286 A W W A D et al.

TABLE I. PHARMACOKINETICS DATA AFTER IV INJECTION OF SR-2508 IN A DOSE OF 2 g/m2 . TUMOUR SAMPLES TAKEN 4 0 -6 0 min AFTER INJECTION. DURATION OF INJECTION 10-15 min

Tumour/plasma Pt No. Tumour Plasma Plasma Ti/2 24 h concentration concentration concentration (beta) urinary excretion ratio (Aig/g) (/ug/mL) (h) (%) (%)

BLADDER CANCER PATIENTS

1 100 55 182 5.4 71 2 124 48 234 6.1 69 3 152 62 244 5.8 75 4 146 87 168 5.9 62 5 154 88 175 5.7 70 6 77 62 124 5.1 56

CANCER OF THE CERVIX PATIENTS

1 62 82 76 5.9 77 2 42 58 72 5.0 68 3 60 99 61 6.1 74 4 106 82 129 6.0 65

concentration in patients with cervical cancer was significantly lower than in the case of bladder cancer and a tumour to plasma concentration ratio of less than 1.0 was noted in 3 out of 4 patients. It is suggested that the high tumour concentration observed in bladder cancer cases is due to diffusion from highly concentrated urine, which would also explain the difference in the tumour concentration between bladder and cervical cancer. The drug concentration in the first hour urine varied between 7000 and 9000 /ug/mL. Such an explanation has still to be verified by studying the diffusion of the drug after intravesical administration. The T ]/2 (beta) of the plasma curve varied between 5.1 and 6.1, which is about one half that of MISO. This relatively rapid elimination reduces the tissue exposure and also the risk of neurotoxicity. Moreover, about 70% of the drug is excreted unchanged in the urine during the first 24 h. This is also a much higher excretion rate than for MISO.

2.9. Pharmacokinetics of Ro-03-8799 in patients with bladder cancer

The pharmacokinetics of Ro-03-8799 have been studied in 4 patients with bladder cancer after IV injection of 1 g/m2 over 10 min. The Tm (beta) of the IAEA-SM-290/76 287 plasma concentration was very close to that of the SR compound and had an average of 5.4 h. Similarly, 70% of the drug is excreted in the urine during the first 24 h. However, unlike with SR-2508, most of the drug is excreted in an altered form. The plasma concentration at 30 min varied between 14 and 18 /ig/mL and the corresponding tumour concentration between 46 and 56 ng/mL, i.e. a tumour to plasma concentration ratio o f 3—4. The basic properties of the drug and the possibility of an acid pH in tumours are believed to account for the high level relative to plasma.

2.10. Combined administration of SR-2508 and Ro-03-8799

The combined administration of SR-2508 and Ro-03-8799 has been studied in two patients after IV infusion of 2 g/m2 of SR-2508 and 1 g/m2 of Ro-03-8799 over 10 min. Both the drug excretion in the urine and the plasma clearance followed the same pattern observed after administration of a single agent. The tumour concentration was also additive.

2.11. Potential use of SR-2508 and Ro-03-8799 in bladder cancer

The two agents have different toxicity patterns. Whereas peripheral neuropathy constitutes the main toxic effect of the SR compound, central nervous toxicity is the main dose limiting toxicity of Ro-03-8799 [11, 12]. SR-2508 seems to offer attractive possibilities and is likely to have a significant advantage over MISO. It has recently been reported by the Stanford group [13] that for a 3-week schedule the lowest cumulative dose producing toxicity was 21.6 g/m2, while no patient was spared neurotoxicity at doses greater than 30 g/m2. In a later report [14] it could be shown that when the duration of drug administration was increased to 5—6 weeks a higher drug dose was tolerated. It was recommended for a long schedule in phase II or III studies to start with a total dose of 34 g/m2, with the possibility of minor escalations during the study. Hence for a curative radiotherapy schedule it should be possible to administer the drug in daily.doses of 1.0—2.0 g/m2 and an enhancement ratio of 1.5 —1.7 should be obtainable. As stated above, the corresponding enhance ment that can be obtained with a daily dose of 0.5 g/m2 of MISO is of the order of 1.1. For such curative radiotherapy courses the individual dose of Ro-03-8799 should not exceed 0.75 g/m2. The tumour concentration attainable after such a dose is of the order of 40 jug/mL and, after adjusting for molecular weights, this does not represent a major advantage over MISO [11]. Two major studies are now planned in the National Cancer Institute on bladder cancer: (a) Concentrated pre-operative irradiation: using 5 daily fractions of 400 cGy each. The drug is given according to one of two schedules: (i) 2.5 g/m2 IV 288 A W W A D et al.

50—60 min before irradiation. This relatively long interval is intended for making use of the possibility of drug diffusion from urine into the tumour. The expected tumour concentration is of the order of 150 jug/g, with a corresponding enhancement ratio of 1.8. This represents a major advantage over the use of MISO in such a study, (ii) An initial dose of 1 g/m2 is administered and an interval of 2 h is allowed to prolong the period of bathing the tumour in the concentrated urine and possibly also to make use of any cytotoxic effect. This is followed by injection of 1.5 g/m2 with irradiation delivered 50—60 min later. (b) Radical radiotherapy of bladder cancer: A combination of the two agents is administered: 1.5 g/m2 of SR-2508 and 0.6 g/m2 of Ro-03-8799, 30—40 min before each radiation dose. In this way it would be possible to give the maximum permissible dose of each agent throughout the entire radical radiotherapy schedule. The same regimen can be applied during radical radiotherapy of cervical cancer.

CONCLUSIONS

( 1 ) MISO proved to be a disappointing chemical radiosensitizer as most trials showed no or only marginal benefit. (2) The use of newer agents should be limited to tumours where evidence of a clinically significant hypoxia problem exists. (3) SR-2508 seems to offer a major advantage over MISO, particularly in a radical radiotherapy setting. (4) A combination of the two newer agents should permit their use during conventional daily fractionation since they have different toxicity patterns.

ACKNOWLEDGEMENTS

This work is supported by IAEA Research Contract No. 3352/RB and by grant No. 842006 within the University Linkage Programme run by the Egyptian Supreme Council of Universities. SR-2508 has been kindly supplied by the National Cancer Institute, Bethesda, USA, while Ro-03-8799 was kindly supplied by Roche Products Ltd, UK (courtesy of Dr. Lenox-Smith).

REFERENCES

[1] THOMLINSON, R.H., GRAY, L.H., Br. J. Cancer 9 (1955) 549. [2] AWWAD, H.K., EL NAGGAR, M., MOKTAR, N.. BARSOUM, M., Int. J. Radiat. Oncol., Biol. Phys. (in press). IAEA-SM-290/76 289

[3] MOKHTAR, N., EL BOLKAINI, N., ABD EL BAKI, H., AWWAD, H.K., Med. J. Cairo Univ. 49(1981) 919.

[4] WATSON, E.R., et al., Br. J. Radiol. 5 1 (1978) 879. [5] AWWAD, H.K., et al., Radiat. Oncol. 2 (1984) 1. [6] ZAGHLOUL, M., et al., Radiother. Oncol 5 (1986) (in press). [7] AWWAD, H.K., et al., Am. J. Clin. Oncol. 6 (1983) 91. [8] AWWAD, H.K., BARSOUM, M., HAMZA, M.R., EL MERZABAN1, M., ABD EL BAKI, H., Egypt. J. Neurol.,Psychiatr., Neurosurg. 21 (1981) 45. [9] DISCHE, S., “Hypoxic cell sensitizers in clinical radiotherapy”, Topical Reviews in Radiotherapy and Oncology (DEELEY, T.J., Ed.), Vol. 1, J. Wright & Sons Ltd, London (1980) 166. [10] BROWN, J.M., Int. J. Radiat. Oncol., Biol. Phys. 10(1984)425. [11] ROBERTS, J.T., BLEEHEN, N.M., WORKMAN, P., WALTON, M.I., Int. J. Radiat. Oncol., Biol. Phys. 10(1984) 1755.

[12] SAUNDERS, M., et al. Int. J. Radiat. Oncol., Biol. Phys. 1 0 (1984) 1759.

[13] COLEMAN, N., et al., Int. J. Radiat. Oncol., Biol. Phys. 1 0 (1984) 1749. [14] COLEMAN, N., et al., Int. J. Radiat. Oncol., Biol. Phys. 12 (in press).

IAEA-SM-290/4

RADIORESPONSIVENESS OF LOCALLY ADVANCED BREAST CANCER Analysis o f factors influencing the response to radiation therapy

J. FODOR, J. TÓTH, G. GYENES National Institute of Oncology, Budapest, Hungary

Abstract

RADIORESPONSIVENESS OF LOCALLY ADVANCED BREAST CANCER. ANALYSIS OF FACTORS INFLUENCING THE RESPONSE TO RADIATION THERAPY. The radioresponsiveness of locally advanced breast cancer was studied in 239 patients treated with primary RT from 1977 to 1985. The response to RT was assessed on clinical and histological bases. For patients with no residual disease or heavily damaged residual disease in the breast and negative axillary lymph nodes after RT, the 5-year probability of local tumour control (LTC) was 90%. Response to RT was mainly affected by the T and N stages. Primary tumour size or RT dose had only a moderate influence on tumour response in this stage of the disease. The 5-year LTC was 37% for N3 and 43% for T4d conditions. Heavy tumour tissue damage was associated with microcapillary lesions, while the micro vasculature around the florid tumour growth was in better condition. Microcapillaries have to be regarded as an important target in RT. The oxygen supply may worsen during fraction ated RT as a result of vascular damage. Chronically hypoxic cells may be killed earlier than euoxic ones. Findings indicate that the radioresponsiveness of locally advanced breast cancer is more likely to be attributable to the biological nature of the disease than to chronically hypoxic cells present during RT.

1. INTRODUCTION

The optimal therapy for patients with locally advanced breast cancer has not been fully established. Despite multimodality treatment — combinations of radia tion therapy (RT), surgery and systemic therapy - both local tumour control (LTC) and survival remain a problem. The administration of radiosensitizers has also been suggested as a future possibility [ 1 ]. The patients involved in this study were treated with primary RT. Their response to RT was assessed on a clinical and a histological basis. The relationship between the grade of response and LTC was examined. Factors supposed to influence tumour response and LTC are: primary tumour size, RT dose, and T and N stages of the disease. The role of microcapillaries in the radioresponsiveness of the tumour and the question of whether hypoxic tumour cells limit the LTC and the response of tumours to RT are also discussed.

291 292 FODOR et al.

TABLE I. T AND N DISTRIBUTION OF ALL THE 239 PATIENTS

T2 T3 T4ABC T4D Total

Ni 0 23 28 9 60 (25%) n 2 11 46 48 26 131 (55%) N3 s 9 23 11 48 (20%) Total 16(7%) 78 (33%) 99 (41%) 46(19% ) 239 (100%)

Note: 5 of the N! conditions were clinically negative (N1A).

2. MATERIAL AND METHODS

The study included 239 women with stage III carcinoma of the breast treated with primary RT from 1977 to 1985. Stage III was defined using the UICC TNM classification [2], but inflammatory carcinoma was included as T4d when the RT was not limited by technical factors. The T and N distributions of all the 239 patients are shown in Table I. Diagnosis of the disease was based on a triple test: physical examination, bilateral mammography and aspiration cytology. Excisional biopsy or limited surgery was not performed before RT. From 1977 to 1981, 100 patients were treated with kilovoltage equipment, while 139 women were treated with a telecobalt unit from 1981 to 1985. The total dose to the primary tumour was estimated to be around 60 Gy and the regional lymph node doses were between 40 and 50 Gy for the X-ray treated patients. The telecobalt base dose was 50 Gy (2 Gy/d, 10 Gy/week) both to the breast and regional lymph nodes. The boost dose to the residual masses was between 10 and 30 Gy. Six weeks after the termination of RT the response of the tumour was assessed on a clinical basis [3]. Patients with an objective response were encouraged to undergo surgery. A significant number of women, however, refused surgery or presented early métastasés. In fact, 54 X-ray and 51 telecobalt treated patients were subjected to surgery (mastectomy and axillary block dissection). The T and N distributions of these patients are presented in Table II. The surgical materials from operated patients were analysed histologically using formalin fixed, paraffin embedded sections, stained with haematoxylin eosin and the PAS technique. The histological findings were classified as follows: In the breast three grades of residual disease were distinguished: (1) No residual disease (NRD) when the primary tumour was completely sterilized on a histological basis. (2) Heavily damaged residual disease (HDRD) when the original histological tumour type was not verifiable because of radiation damage. (3) Typifiable residual disease (TRD) when the histological tumour type could be identified. IAEA-SM-290/4 293

TABLE И. T AND N DISTRIBUTION OF ALL THE 105 PATIENTS OPERATED

T2 T 4ABC ^4D Total

N, 0 16 13 4 33 (31%) n 2 4 27 19 9 59 (56%) N3 1 5 6 1 13 (13%) Total 5 (5%) 48 (46%) 38 (36%) 14(13%) 105 (100%)

Note: 2 of the conditions were clinically negative (N1A).

Two grades of residual disease were distinguished in the axillary lymph nodes (ALN) : (1) ALN are negative after RT (prtNneg). (2) ALN contain tumour tissue after RT (prtNpos). Microcapillary damage was classified as follows: (1) Decrease in microcapillary density. (2) Thickening of capillary walls. (3) Obliteration of vascular lumens.

Post-irradiation and post-surgical therapies: For patients with no response systemic therapy was administered (CMF and/or tamoxifen). Patients with objective responses were treated with hormonal therapy (tamoxifen or castration). The patients were kept under observation for at least 20 months but more than half of them for 60 months with a periodic check-up. The site and time of the first locoregional relapse were recorded. The duration of LTC was calculated from the termination of RT till the appearance of recurrence using Kaplan and Meier’s method [4]. Patients with no response were considered to have persistent local disease.

3. RESULTS

3.1. Tumour response to RT

3.1.1. Change in tumour volume

Complete response (CR) was observed in 56 patients (24%), partial remission (PR) in 156 women (65%) and no response (NR) or progressive disease in 294 F O D O R et al.

TABLE III. RESPONSE TO RADIATION THERAPY BY PRIMARY TUMOUR SIZE

Primary tumour size (mm) Response Total 30-50 50-70 70-90 90 <

CR 9 (26%) 30 (24%) 9 (18%) 8 (27%) 56 (24%) PR 23 (66%) 81 (65%) 37 (74%) 15 (52%) 156 (65%) NR 3 (8%) 14 (11%) 4 (8%) 6(21%) 27 (11%) Total 35 (15%) 125 (52%) 50 (21%) 29 (12%) 239(100%)

TABLE IV. RESPONSE TO RADIATION THERAPY BY PRIMARY TUMOUR DOSE

Dose (Gy) Response Total 50 60 70 80

CR 7 (26%) 12 (32%) 16 (38%) 9 (28%) 44(31%) PR 19 (70%) 22 (58%) 25 (60%) 20 (63%) 86 (62%) NR 1 (4%) 4(10%) 1 (2%) 3 (9%) 9 (7%) Total 27 (20%) 38 (27%) 42 (30%) 32 (23%) 139 (100%)

Note: The telecobalt treated patients (139) are analysed here.

27 patients (11%). The relative response rates for the four groups of primary tumour size are presented in Table III. In this stage of the disease tumour size does not seem to have an essential impact on evoking a better response. The dose and response correlation was analysed in 139 telecobalt treated patients. The rate of CR patients showed a moderate increase up to 70 Gy (Table IV). The relation between the T and N states and the response to RT is shown in Tables V and VI. Patients with a more advanced state showed a weaker response. The relative rate of T4d and N3 states was 52% in the NR group, while in the objective response group it was 15 and 16%, respectively.

3.1.2. Histomorphological change

The response to RT on a histological basis was analysed in all the 105 patients operated. In the breast NRD (10) or HDRD (28) were found in IAEA-SM-290/4 295

TABLE V. T AND N DISTRIBUTION OF ALL THE 212 PATIENTS WITH OBJECTIVE RESPONSE

T2 Ъ T4ABC T4D Total

N, 0 23 28 8 59 (28%) n 2 10 44 45 20 119 (56%) N3 5 9 16 4 34(16%) Total 15 (7%) 76 (36%) 89 (42%) 32 (15%) 212 (100%)

TABLE VI. T AND N DISTRIBUTION OF ALL THE 27 PATIENTS WITH NR

T2 T3 T4ABC ^4D Total

Ni 0 0 0 1 1 (4%) N2 1 2 3 6 12 (44%) N3 0 0 7 7 14(52%) Total 1 (4%) 2 (7%) 10(37%) 14(52%) 27 (100%)

38 cases (36%). The primary tumour remained typifiable in 67 cases (64%). There was no essential difference between the X-ray and telecobalt treated cases. The rate of TRD was 67 and 61%, respectively. The grade of response for the four groups of tumour size is presented in Table VII. The relative rate of NRD or HDRD conditions was the highest in the 30—50 mm group (47%). Correlation between the grade of radiation damage of the primary tumour and the RT dose was analysed in the 51 telecobalt treated patients (Table VIII). On increasing the primary tumour dose from 50 to 70 Gy marked radiation damage of the tumour tissue was observed, but 80 Gy did not enhance this effect. The prtNneg condition was detected in 44 cases (42%) : 18 after X-ray therapy and 26 after telecobalt treatment (18/54 = 33% versus 26/51 = 51%). Parallelism was observed between the volume change and the grade of radiation induced tumour tissue damage. Fifty-two per cent of the CR patients had NRD or HDRD plus prtNneg conditions, while only 12% of the PR women had the same grade of tumour tissue damage (Table IX). Precise quantification of microcapillary injuries was not possible by the method used in this study. A qualitative analysis of the vascular injuries revealed hyaliniza- tion of the capillary walls, obliteration of the vascular lumina and a decrease in the 296 F O D O R et al.

TABLE VII. RESPONSE TO RT ON HISTOLOGICAL BASIS BY PRIMARY TUMOUR SIZE

Grade of Primary tumour size (mm) Total damage 30-50 50-70 70-90 90 <

NRD or 7 (47%) 22 (36%) 7 (37%) 2 (20%) 38 (36%) HDRD TRD 8 (53%) 39 (64%) 12 (63%) 8 (80%) 67 (64%) Total 15 (14%) 61 (58%) 19(18%) 10(10%) 105 (100%)

Note: All the 105 surgical materials are analysed here.

TABLE VIII. GRADE OF RADIATION DAMAGE BY PRIMARY TUMOUR DOSE

50 Gy 60 Gy 70 Gy 80 Gy Total

NRD or 3 (27%) 5 (45%) 7 (50%) 5 (33%) 20 (39%) HDRD

TRD 8 (73%) 6 (55%) 7 (50%) 10 (67%) 31 (61%) Total 11 (22%) 11 (22%) 14(27%) 15 (29%) 51 (100%)

Note: All the 51 surgical materials of the telecobalt treated patients are analysed here.

TABLE IX. RELATIONSHIP BETWEEN VOLUME RESPONSE AND RADIATION DAMAGE

CR PR

NRD or HDRD + prtNneg 11/21 (52%) 10/84(12%) TRD + prtNpos 4/21 (19%) 42/84 (50%) NRD or HDRD 15/21 (71%) 23/84 (27%) TRD 6/21 (29%) 61/84(73%) prtNneg 13/21 (62%) 29/84 (35%) prtNpos 8/21 (38%) 55/84 (65%)

Note: Surgical materials of all the 105 patients operated. IAEA-SM-290/4 297

FIG. 1. Condition of tumour tissue after three months of radiation therapy. Around the heavily damaged microvasculatures typifiable tumour cells cannot be seen.

microcapillary density. NRD or HDRD conditions were always associated with increased vascular damage. The microcapillaries around the morphologically intact tumour tissue were in better condition. The distribution of capillary lesions reflected both primarily vascular and tumour cell damage (Fig. 1).

3.2. Analysis of locoregional recurrence and the duration of LTC

The total number of locoregional failures was 55 during the follow-up period. The 27 NR women were not considered because of persistent local disease. Eighty-two per cent of the relapses appeared in the first two years. The pattern of the failures is presented in Table X. Forty-four per cent of them were associated with macrodeposits which were detectable before treatment, and 56% of them regrew from microdeposits which were undetectable before treatment. Factors relevant to LTC are presented in Table XI and the most characteristic local control curves are plotted in Fig. 2. For patients with NRD or HDRD plus prtNneg the 5-year probability of LTC was 90%, while LTC was only 37% in patients with N3 conditions. The other values were intermediate. 298 F O D O R et al.

TABLE X. SITE OF ALL THE 55 LOCOREGIONAL RECURRENCES

Site of relapse Macrodeposit Microdeposit

Primary tumour 9 _ Skin of the breast - 6 Chest wall - 16 Axillary region 7 1 Supraclavicular region 8 8 Total 24 (44%) 31 (56%)

Macrodeposit = Detectable before treatment. Microdeposit = Undetectable before treatment.

TABLE XI. ACTUARIAL PROBABILITY OF 5-YEAR LOCAL CONTROL

Factor No. of patients Local control (%)

CR 56 72 PR 156 63 NR 27 0

T2 16 63 T3 78 68 ^4 ABC 99 57 T4D 46 43 Ni 60 72 N2 131 58 N3 48 37 NRD or HDRD 38 78 TRD 67 66 prtNneg 44 85 prtNpos 61 60 NRD or HDRD + prtNneg 23 90 TRD + prtNpos 46 58 Whole group 239 59 IAEA-SM-290/4 299

LTC (%)

Months

FIG. 2. Local tumour control according to different factors.

4. DISCUSSION

The results presented emphasize the importance of the volume change and radiation damage of tumour tissue to establish LTC. The primary tumour size had only a moderate influence on the response to RT in locally advanced breast cancer. The dose of RT had a stronger effect on tumour response. ALN infiltrations were more frequently absent after telecobalt therapy than after kilo voltage irradiation (51% versus 33%). On raising the primary tumour dose from 50 to 70 Gy a better response was achieved; however, no futher improvement was observed on increasing the dose from 70 to 80 Gy. The T and N categories proved to be the most important factors in establishing and maintaining LTC. The T4d or N3 stage generally indicated a very bad outcome with 43 and 37% LTC, respectively. T3 and Nj patients did better (68 and 72%, respectively). The analysis of locoregional recurrences revealed that in a significant number of cases even the microdeposits of the tumour were beyond control by RT. Fifty-six per cent of the recurrences regrew from infiltrations undetectable before RT. The findings indicate that the outcome of the disease is influenced by the biological nature of the disease rather than by technical factors. Other reports show the same observations [5—7]. 300 F O D O R et al.

Regrowth of the tumour within the irradiated region reflects its radioresistance owing to inherent tumour related factors. In this study we tried to find clinical evidence for and against the role of hypoxic cells and microcapillaries in the radioresistance of tumours. According to the original model of Thomlinson and Gray [8], chronically hypoxic cells adjacent to the necrotic area survive the irradia tion. These hypoxic cells - as a result of microcapillary damage — should be killed earlier than euoxic ones. In our cases radiation damage of the tumour tissue was associated with microcapillary lesions but around the florid tumour growth the microvasculature was in better condition. Late damage of the capillary network is preceded by early alterations such as swelling of the endothelian cells and bulging into the lumen, impeding the blood flow [9]. Findings show that the microcirculation deteriorates during and after fractionated irradiation. The radio responsiveness of tumour tissue depends on the inherent radiosensitivity of both the tumour cells and the microcapillaries. The role of the microcapillaries as one of the main possible targets in tumour therapy is also emphasized by Denekamp [10]. Tumours with rapidly proliferating vulnerable microcapillaries, like the exophytic ones, respond well to RT, even if they contain necrotic areas. The regrowth of the tumour from microdeposits also does not support the role of chronically hypoxic cells in radioresistance, since microdeposits may not contain necrotic cells. However, another kind of hypoxia, acute or subacute [11], cannot be excluded either. Oedema of the breast is common in locally advanced disease and reflects the disturbance of lymphatic outflow. Tumour cells under hydrostatic pressure may be hypoxic for a shorter or longer period during irradiation.

REFERENCES

[1] BONADONNA, G., Breast Cancer: Diagnosis and Management, Wiley, New York (1984) 337. [2] UICC, TNM Classification of Malignant Tumours, 2nd edn, UICC, Geneva (1974). [3] HAYWARD, L.J., et al., Eur. J. Cancer 13 (1977) 89. [4] KAPLAN, E.L., MEIER, P., J. Am. Stat. Assoc. 53 (1958) 457. [5] ZUCALI, R., USLENGHI, C., KENDA, R., DONADONNA, G., Cancer 37 (1976) 1422. [6] RUBENS, D.R., ARMITAGE, P., WINTER, J.P., TONG, D„ HAYWARD, L.J., Eur. J. Cancer 13 (1977) 805. [7] TREURNIET-DONKER, D.A., HOP, J.C., DEN HOED-SIJTSEMA, S., Int. J. Radiat. Oncol., Biol. Phys. 6 (1980) 1477. [8] THOMLINSON, P.H., GRAY, L.H., Br. J. Cancer 9 (1955) 539. [9] RUBIN, P., Int. J. Radiat. Oncol., Biol. Phys. 10 (1984) 5. [10] DENEKAMP, J., Acta Radiol. Oncol. 23 (1984) 217. [11] SUTHERLAND, R.M., FRANKO, A.J., Int. J. Radiat. Oncol., Biol. Phys. 6 (1980) 117. IAEA-SM-290/54

PRECLINICAL AND CLINICAL RESULTS OF THE USE OF FLUOSOL-DA (20%) IN RADIOTHERAPY

C.W. SONG, I. LEE, T. HASEGAWA, J.G. RHEE, S.H. LEVITT Department of Therapeutic Radiology, University of Minnesota Medical School, Minneapolis, Minnesota

N. McINTOSH Alpha Therapeutic Corporation, Los Angeles, California United States of America

Abstract PRECLINICAL AND CLINICAL RESULTS OF THE USE OF FLUOSOL-DA (20%) IN RADIOTHERAPY. Perfluorochemicals dissolve large quantities of oxygen in high p 0 2 environments and quickly release oxygen in low p 0 2 environments. Recently, interest in exploiting this unique property of perfluorochemicals to increase tumour oxygenation and thus tumour radiosensitivity has surged. We have been testing the potential usefulness of Fluosol-DA, an emulsion of perfluorochemicals, as a radiosensitizer. When C3H mice bearing RIF-1 tumours in the leg were IV injected with Fluosol-DA (20%) at 12 mL/kg and allowed to breathe carbogen (95% 0 2 + 5% C02) for 1 h before and during a single dose of X-irradiation, the tumoricidal effect of radiation was signifi cantly enhanced. For example, the curability of tumours as estimated from the TCD50 increased by a dose modification factor of 1.47 when the animals were treated with Fluosol-DA (20%) and carbogen breathing. The intratumour p 0 2 in the mice treated with Fluosol-DA and carbogen was measured with the polarographic method employing electrodes 50-60 ßm in diameter. Carbogen breathing alone could cause a moderate increase in p 0 2 but the combination of Fluosol-DA injection and carbogen breathing caused a much greater increase in tumour p 0 2. It was apparent that the increased tumour rèsponse to radiation by carbogen breathing plus Fluosol-DA (20%) injection was due to an improved oxygen supply to the hypoxic cells in the RIF-1 tumour. The present study and reports by other investigators strongly suggest that perfluorochemicals are potentially useful as radiosensitizers of hypoxic tumour cells.

1. INTRODUCTION

A number of experimental data have unequivocally demonstrated that certain fractions of cells in rodent tumours are radiobiologically hypoxic. There is no reason to expect that human tumours, unlike rodent tumours, do not contain hypoxic cells. Although some of the hypoxic cells may undergo reoxygenation during the course of fractionated radiation, it is believed that the presence of radio biologically hypoxic cells in human tumours is the major limiting factor for the

301 302 SONG et al. complete control of tumours by radiotherapy. To eliminate or lessen the hypoxic protection in radiotherapy, a number of different approaches have been tested, such as use of high LET radiations, use of chemical compounds which specifically sensi tize hypoxic cells to radiation or kill hypoxic cells, breathing oxygen at normal or increased pressure, and breathing a mixture of C02 and oxygen (carbogen). Unfortunately, none of these approaches has yet proved to be clinically effective and useful. An interesting development in recent years to solve the hypoxic problem in radiotherapy is the use of perfluorochemicals in combination with breathing carbo gen or pure oxygen [ 1-6]. Perfluorochemicals are hydrocarbon compounds in which the hydrogen atoms are replaced by fluorine. These compounds are known easily to dissolve and release oxygen depending on the environmental oxygen pressure and, as such, they may function as oxygen carriers when injected into the blood circulation [7]. Fluosol-DA is an emulsion of perfluorochemicals initially developed for the treatment of patients suffering from anaemia or ischaemic damage after a stroke or myocardial infarction [8, 9]. During the past several years the potential usefulness of Fluosol-DA to improve tumour oxygenation and thus to increase the radiosensitivity of tumours has been investigated. Based on encouraging results with rodent tumours [ 1-6], phase I clinical trials of Fluosol-DA as a physio logical radiosensitizer have been initiated in several medical centres in the United States of America [10]. This paper summarizes the investigations on Fluosol-DA (20%) at the University of Minnesota.

2. MATERIALS AND METHODS

The efficacy of IV injection of Fluosol-DA (20%) in conjunction with carbogen (95% 0 2 + 5% C02) breathing to increase the response of mouse tumours to X-irradiation was evaluated. The therapeutic gain was estimated from the tumour response and skin damage. The changes in tumour p02 by Fluosol-DA (20%) injec tion and carbogen breathing were measured with the polarographic method.

2.1. Tumour

RIF-1 tumours of female C3Heb/FeJ mice were used. The tumours were induced in the right thigh of mice (25 g) by SC injection of 1 X 10s cells in expo nential growth phase obtained from culture in vitro. Tumours of 7-8 mm in dia meter were used.

2.2. Fluosol-DA (20%) and X-irradiation

The composition and characteristics of Fluosol-DA have been described pre viously [8, 9]. The stock Fluosol-DA emulsion was thawed and mixed with an IAE A-SM-290/54 303 annex solution. The main components of the resultant emulsion were perfluoro- decalin (14%) and perfluorotripropylamine (6%) in Krebs-Ringer’s bicarbonate solution. The emulsion was oxygenated by gassing for 20-30 min with carbogen and injected through the tail veins into the mice at 12 mL/kg. The mice were then placed in a Plexiglas tank and gassed with carbogen [5]. After exposure to the carbogen for 45 min, the mice were anaesthetized by IP injection of pentobarbitol (50 mg/kg). After keeping in the carbogen tank for another 15 min, the mice were mounted in irradiation jigs and the tumours in the legs were irradiated with graded doses of X-rays in a single exposure. The mice breathed carbogen during the irradia tion through specially designed masks placed over the forehead. Some animals were treated with carbogen breathing alone without Fluosol-DA (20%) injection. The physical factors for the X-irradiation were 220 kV, 15 mA and a dose rate of 116.9 rad/min.1 The change in growth rate and also the cure of tumours were scored. Mice without tumours were also treated with Fluosol-DA (20%) plus carbo gen or carbogen alone, the normal legs were exposed to X-irradiation, and the skin reactions were scored to estimate the therapeutic gain.

2.3. Measurement of tumour p02

The changes in tumour oxygenation upon treatment of host mice with Fluosol-DA (20%) and carbogen were investigated by measuring the tumour p 0 2 with the polarographic method. Recessed-tip oxygen microelectrodes with tip diameters of 50-60 ^m were constructed as described elsewhere [10, 11]. Mice were anaesthetized by IP injection of Inactin ((2-methylpropyl)-malonyl-thiourea) and mounted on a board in a Faraday cage, and the electrode was inserted into the tumour with the use of a micromanipulator. A reference electrode (silver-AgCl wire) with a diameter of about 100 /um was also inserted into the periphery of the tumours. The animals were treated with carbogen alone, Fluosol-DA (20%) alone, or Fluosol-DA (20%) plus carbogen, and the tumour p0 2 was obtained from the electrode reading, which was recorded with a chart recorder.

3. RESULTS

3.1. Tumour and skin response

The tumour diameters were measured 2-3 times a week after the treatments. Figure 1 shows, as an example, the changes in the growth rate of tumours after an exposure to 1500 rad [5]. Carbogen breathing alone could significantly enhance the response of tumours to X-irradiation. Much greater enhancement of the tumour response to radiation could be observed when the host animals were injected with

1 100 rad = 1 Gy. 304 SONG et al.

Days After Treatment

FIG. 1. Effects of Fluosol-DA injection (IVj plus carbogen breathing or carbogen breathing alone on the tumoricidal effect of 1500 rad X-irradiation. The changes in the volume of RIF-1 tumours (SC) in the leg of C3H mice are shown as a function of days after the various treatments. Means of 7-8 tumours and SE are shown.

R a d

FIG. 2. Growth delays for the RIF-1 tumour (SC) of C3H mice by various treatments. Means of 7-8 tumours and SE are shown. The best fitting lines are obtained by the least squares method. IAEA-SM-290/S4 305

TABLE I. MODIFICATION OF RADIATION DAMAGE IN RIF-1 TUMOURS AND SKIN BY FLUOSOL-DA (20%) AND CARBOGEN

Treatment Dose modification factors Growth delay TCDso Skin damage of 20 days

Carbogen 1.32 ± 0.07 1.22 ± 0.01 1.13 ± 0.08 Carbogen + Fluosol-DA (20%) 1.96 ± 0.09 1.47 ± 0.03 1.15 ± 0.12

Fluosol-DA (20%) and allowed to breathe carbogen. From the difference in the num ber of days for the control and that for the treated tumours to grow 4 times the initial volume the growth delays owing to the treatments were calculated and are shown in Fig. 2. It can be seen that the X-irradiation induced suppression of tumour growth was markedly enhanced by treating the mice with Fluosol-DA (20%) injec tion and carbogen breathing with a dose modification factor of 1.96 ± 0.09 (SE) (Table I). The dose modification factor for carbogen breathing alone was 1.32 ± 0.07 (SE). The tumours that regressed and did not regrow in 120 days after the treatments were scored as cured and the TCD50 for each treatment protocol was calculated.2 The TCD50 for the tumours treated with X-rays was 5487 ± 564 rad, while that for the tumours of mice treated with carbogen breathing plus Fluosol-DA (20%) was 3741 ± 267 rad. As shown in Table I the ratio of these two TCD50 values (dose modification factor) was 1.47 ± 0.03. Treatment of mice with carbo gen breathing alone reduced the TCD50 to 4503 ± 428 rad with a dose modification factor of 1.22 ± 0.01. The skin damage in the legs was scored every 3 days for 40 days using an arbitrary scale [ 12]. The average skin reaction for each radiation dose was calcu lated and the dose required to induce an average skin reaction of 2.0 was estimated. As shown in Table I, the skin damage from X-irradiation was only slightly increased by carbogen breathing or carbogen breathing plus Fluosol-DA (20%) injection. The therapeutic gain calculated from the dose modification factors for the skin damage and tumour control was 1.28 ± 0.04.

3.2. Tumour p02

Measurements of tumour p 0 2 with microelectrodes demonstrated a small increase in p 0 2 with carbogen breathing alone and a marked increase in p 0 2 with carbogen breathing plus Fluosol-DA (20%) injections. Figure 3 is an original tracing

2 TCDS0 = Radiation dose needed to control 50% of tumours. 306 SONG et al.

!'i| j ¡s ! .j 1 j ■; j j : j- :•] mmHg : ! 1 i . 'i ..! 1 ¡ ! Í: ■ ¡ .... 3 0 0 —п T: ~ " Г " "" i ■' j —j" " t i- ;:;1 i ~ f - г К - " Ч Ч 'М . ; I ! Carbogen ¡ ¡ Carbogén j" ; О wа 20< с о л

С а> loo—H >.О) X О 50J

Time (mjn)

FIG. 3. Actual tracing o f changes in p 0 2 in a RIF-1 tumour under various conditions. Carbogen breathing alone significantly increased tumour p 0 2 and a greater increase in p 0 2 occurred when the host mice were injectedIV with Fluosol-DA (20%) and allowed to breathe carbogen.

Oxygen Tension in Tumour (p02,mmHg)

FIG. 4. Frequency distribution o f рОг in RIF-1 tumours under various conditions. More than 200 readings o f p0 2 values were made in 30-50 tumours for each group. IAEA-SM-290/54 307 of changes in p 0 2 in a tumour. The p02 during air breathing was about 5 mmHg and it significantly increased when the animals breathed carbogen. For unknown reasons the p 0 2 fluctuated considerably during carbogen breathing, in this case ranging from 25 to 70 mmHg. Upon termination of carbogen breathing the tumour p02 declined to about 30 mmHg. When the animal was injected with Fluosol-DA (20%) at 12 mL/kg and allowed to breathe carbogen, the tumour p02 rapidly increased to about 160 mmHg, gradually declined and then stabilized at 120 mmHg. The p 0 2 declined to 30 mmHg after switching from carbogen breathing to air breathing. A more detailed analysis of p 0 2 values in the RIF-1 tumours is shown in Fig. 4. In the control RIF-1 tumours about 90% of the p 0 2 values measured were less than 20 mmHg and none of the p 0 2 values were above 45 mmHg. An injection of Fluosol-DA (20%) to mice breathing air slightly increased the p 0 2. Carbogen breathing without Fluosol-DA (20%) injection significantly increased the p 0 2 and a Fluosol-DA (20%) injection followed by carbogen breathing dramati cally increased the tumour p 0 2 value. In fact, more than 50% of the p 0 2 values measured in the RIF-1 tumours of animals that breathed carbogen after Fluosol-DA (20%) injection were above 45 mmHg.

4. DISCUSSION

The data presented in this paper clearly demonstrate that IV injection of Fluosol-DA (20%), an emulsion of perfluorochemicals, and carbogen breathing can significantly increase the response of tumours to X-rays as measured by tumour growth and tumour cure. The effect of Fluosol-DA in enhancing the response of tumours to X-irradiation was previously attributed to its ability to dissolve large quantities of oxygen in the lung and to rapidly unload the oxygen in low p02 atmosphere in the tumours [1-6]. The results of our measurements of tumour p0 2 with microelectrodes (Fig. 4) are indeed in accordance with the view that tumour p 0 2 is increased by'Fluosol-DA injection and carbogen breathing. In this connection in vivo-in vitro analysis of tumour cell survival after irradiation of the tumours in situ demonstrated a reoxygenation of previously hypoxic cells by Fluosol-DA injection in combination with carbogen breathing [4]. Interestingly, carbogen breathing alone could cause a measurable increase in the response of tumours to radiation in the present study, which is in accordance with the reports by other investigators [13]. It has been suggested that carbogen breathing not only increases arterial p 0 2, but also causes vasodilation which would then increase tumour response to irradiation [13]. Whereas other investigators could not observe any changes in the radio sensitivity of mouse skin by carbogen breathing plus Fluosol-DA (20%) injection, our data show a slight increase in the radiosensitivity of the mouse skin. Con sequently, the therapeutic gain for the cure of RIF-1 tumours, 1.28 ± 0.04, was slightly smaller than the dose modification factor for the tumour response to 308 SONG et al. radiation (Table I). It should be noted that the hypoxic cell fraction in RIF-1 tumours has been reported to be rather small as compared with the hypoxic cell fraction in other rodent tumours. Our unpublished data with SCK tumours, which contain about 40% of hypoxic cells, showed a large therapeutic gain as com pared with that for RIF-1 tumours. The amount of Fluosol-DA (20%) injected into the animals in the present study was 12 mL/kg, which is far less than 40 mL/kg, the maximum dose of Fluosol-DA (20%) allowed by the US Food and Drug Administration for humans. A recent study showed no adverse effect in anaemic patients who received the above men tioned maximum permissible dose of 40 mL/kg [14]. Although no clinical benefit in the anaemic patients was observed in this aforementioned study, partial pressure of arterial 0 2 was significantly increased by Fluosol-DA (20%) injection and about 28% of the oxygen consumed by the body of these patients was that delivered by Fluosol-DA (20%). Phase I clinical trials of Fluosol-DA (20%) for radiotherapy patients are in progress in several medical centres in the USA and preliminary results are encouraging [15]. Further investigation into the use of Fluosol-DA (20%) for the improvement of radiotherapy appears to be warranted.

REFERENCES

[1 ] TE1CHER, B.A., ROSE, C.M., Perfluorochemical emulsion can increase tumor radio sensitivity, Science 223 (1984) 934. [2] TEICHER, B.A., ROSE, C.M., Oxygen-carrying perfluorochemical emulsion as an adjuvant to radiation therapy in mice, Cancer Res. 44 (1984) 4285. [3] FISCHER, J., ROCKWELL, S., MARTIN, D.F., Perfluorochemicals and hyperbaric oxygen in radiation therapy, Int. J. Radiat. Oncol., Biol. Phys. 12 (1986) 95. [4] ROCKWELL, S., Use of a perfluorochemical emulsion to improve oxygenation in a solid tumor, Int. J. Radiat. Oncol., Biol. Phys. 11 (1985) 97. [5] SONG, C.W., ZHANG, W.L., PENCE, D.M., LEE, I., LEVITT, S.H., Increased radio sensitivity by perfluorochemicals and carbogen, Int. J. Radiat. Oncol., Biol. Phys. 11 (1985) 1833. [6] ZHANG, W.L., PENCE, D., PATTEN, M., LEVITT, S.H., SONG, C.W., Enhancement of tumor response to radiation by Fluosol-DA (Abstr.), Int. J. Radiat. Oncol., Biol. Phys. 10 (1984) 172. [7] CLARK, L.C., CLARK, F.W., MOORE, R.E., KINNETT, D.G., INSCHO, E.I., “Room temperature stable biocompatible fluorocarbon emulsions”, Advances in Blood Substitute Research (BOLIN, R.B., GEYER, R.P., NEMO, G.J., Eds), Alan R. Liss, Inc., New York (1983) 169. [8] NAITO, R., YOKOYAMA, J., Perfluorochemical blood substitutes, Fluosol-43, Fluosol-DA 20% and 35%, Technical Information Series No. 5, Osaka, Japan (1978). [9] TREMPER, K.K., LAPIN, R., LEVINE, E„ FRIEDMAN, A., SHOEMAKER, W.C., Hemodynamics and oxygen transport effects of a perfluorochemical blood substitute, Fluosol-DA (20%), Crit. Care Med. 8 (1980) 738. [10] SONG, C.W., LEE, I„ HASEGAWA, K., RHEE, J.G., LEVITT, S.H., Increase in p 0 2 and radiosensitivity of tumors by Fluosol-DA (20%) and carbogen, submitted to Cancer Res., Mar. 1986. IAEA-SM-290/54 309

[11 ] HASEGAWA, T., RHEE, J.G., LEVITT, S.H., SONG, C.W., Increase in tumor p 0 2 by perfluorochemicals and carbogen, submitted to Int. J. Radiat. Oncol., Biol. Phys., Mar. 1986. [12] SCHUMAN, V.L., LEVITT, S.H., SONG, C.W., The radioprotective effect of 5-thio-D- glucose on normal tissues in vivo, Int. J. Radiat. Oncol., Biol. Phys. 8 (1982) 589. [13] KRUUV, J.A., INCH, W.R., McCREDIE, J.A., Blood flow and oxygenation of tumors in mice. I. Effects of breathing gases containing carbon dioxide at atmospheric pressure, Cancer 20(1967) 51. [14] GOULD, S.A., et al., Fluosol-DA as a red-cell substitute in acute anemia, N. Engl. J. Med. 314(1986) 1653. [15] ROSE, C., TEICHER, B., McINTOSH, N., LUSTIG, R., “A clinical trial of Fluosol-DA in advanced head/neck cancer”, Chemical Modifiers of Cancer Treatment (Proc. Conf. Clearwater, 1985) Abstr. No. 4-19.

IAEA-SM-290/74

RADIOSENSITIZATION OF HYPOXIC BACTERIAL CELLS AND ANIMAL TUMOURS BY MEMBRANE ACTIVE DRUGS AND HYPERTHERMIA

K.C. GEORGE, M.A. SHENOY, B.B. SINGH Molecular Biology and Agriculture Division, Bhabha Atomic Research Centre, Trombay, Bombay, India

Abstract

RADIOSENSITIZATION OF HYPOXIC BACTERIAL CELLS AND ANIMAL TUMOURS BY MEMBRANE ACTIVE DRUGS AND HYPERTHERMIA. Among the hypoxic radiosensitizing phenothiazines, chlorpromazine (CPZ), trimeprazine (TMZ) and promethazine (PMZ) protected Swiss mice against whole body irradiation. The protection was more evident at lethal and supralethal doses of X-rays. The protection of the haemopoietic system is indicated by the results of the study. The mechanism of protection is partly attributed to the formation of sulphoxides of the phenothiazine derivatives under the euoxic conditions of irradiation. CPZ and TMZ showed better radioprotection. In view of this and the comparatively high radiosensitizing ability of TMZ under hypoxia, the effects of hyperthermia and anticancer drugs have been studied in a solid mouse fibrosarcoma in the presence of TMZ. The results demonstrate that TMZ enhances the chemotherapeutic effect of melphalan (MEL). Hyperthermia alone increased the efficacy of MEL and cyclophosphamide (CY). When TMZ was present, the combined effect of hyperthermia and the antineoplastic drugs was markedly enhanced. TMZ at clinically acceptable doses caused 60% tumour cure when combined with hyperthermia and MEL while in the absence of TMZ, though no tumour cure was noticed, there was a significant regrowth delay. In contrast, only a regrowth delay was observed with the combined modality of TMZ, hyperthermia and CY without any tumour cure.

1. INTRODUCTION

The basis for searching hypoxic cell radiosensitizers has so far been mainly the electron affinity of compounds among which the nitroimidazoles have received maximum attention [1, 2]. However, in recent years the role of cellular membranes in the lethal effects of radiation has been well established [3—6]. The perturbers of membrane structure may, therefore, prove good modifiers of radiation damage to cells and, on this assumption, various membrane inter acting drugs have been investigated by several workers [7—13]. Phenothiazine derivatives have an advantage over other drugs because of their well known pharmacology, toxicology and pharmacokinetics. The commonly used tranquillizing and antihistaminic drugs such as chlorpromazine (CPZ),

311 312 G E O R G E et al. promethazine (PMZ), prochlorperazine (PCP) and trimeprazine (TMZ) have been found to sensitize several hypoxic single cell systems [9, 14, 15] and both acutely hypoxic and normal mouse tumour cells [16—18] to ionizing radiation. These drugs also exerted cytotoxic effects both in vitro and in vivo [16—20]. A temperature dependent modification of the radiation effect by CPZ was demon strated in bacteria [21 ]. In addition, such phenothiazine derivatives also potentiated the hyperthermic effect in mammalian cells in vitro [20] and in mouse solid tumours in vivo [22, 23]. Among the phenothiazines investigated so far, TMZ proved to be the most efficient modifier of the radiation response of bacteria [14] and also of animal tumours [18]. Furthermore, TMZ proved to be a better potentiator of hyper thermic effects on tumours in vivo [23]. Under oxygenated conditions these phenothiazine derivatives offered radio protection in single cell systems [15], indicating that under a carefully designed clinical protocol they may be beneficially used in radiation therapy causing enhanced damage to hypoxic cells in the tumour and, at the same time, protecting the oxygenated normal tissues surrounding the tumour. This paper presents further results on these drugs, emphasizing the role of membrane active drugs in cancer therapy.

2. RADIOPROTECTION OF MICE BY PHENOTHIAZINES

Preliminary studies in Swiss mice exposed to 250 kV X-ray doses of 6.25, 7.00 and 7.50 Gy, delivered 30 minutes after treatment with single intra- peritoneal doses of CPZ, PMZ and TMZ, demonstrated radioprotection by these drugs. Under similar conditions, however, PCP proved ineffective (Table I). The percentage of dead mice and their mean day of death on irradiation with various X-ray doses for CPZ at 20 mg/kg are shown in detail in Fig. 1. Whereas at lower radiation doses of 6.25 and 6.50 Gy no protection could be noticed either as survival percentage or mean day of death, at the highest doses of 8 or 10 Gy, though 100% lethality occurred, a significant increase in longevity was noticed. At intermediate doses of 6.75, 7.00, 7.25 and 7.50 Gy 100% lethality in untreated animals occurred, while the lethality of drug treated mice remained at 67% only. A radioprotective effect of these drugs has been reported earlier in euoxic bacterial cells and has been ascribed to drug induced fluidity changes in the cellular membrane [15]. Furthermore, under oxygenated conditions the radio- ly tically induced toxic CPZ cation radicals can become oxidized to CPZ sulphoxide [24], which is stable and unreactive and shows no inhibition of cellular enzymes [25]. This may also partly account for the radioprotective effect under oxygen atmosphere of such phenothiazine derivatives. IAEA-SM-290/74 313

TABLE I. EFFECT OF PHENOTHIAZINES ON SWISS MICE GIVEN WHOLE BODY IRRADIATION (7.5 Gy)

Drug Phenothiazine Ratio of Per cent Mean day administered dose dead to total survival of death (mg/kg) mice + SEM

Control Nil 20/20 0.0 10.44 ± 1.28 Prochlorperazine 6.0 10/10 0.0 9.30 ±0.87 Promethazine 4.0 8/9 11.0 12.12 ± 0.87 Trimeprazine 4.0 9/10 10.0 10.66 ± 1.44 8.0 10/11 9.0 10.20 ± 0.81 18.0 7/10 30.0 12.14 ± 0.55 24.0 8/11 27.0 11.89 ± 0.58 Chlorpromazine 5.0 16/17 6.0 12.25 ± 0.35 10.0 12/16 25.0 11.67. ± 0.33 15.0 14/15 7.0 13.57 ±0.64 20.0 12/16 25.0 13.16 ± 1.45

FIG. 1. Histogram showing lethality and mean day of death o f mice given whole body irradiation in the absence and presence o f chlorpromazine. 314 GEORGE et al.

TABLE II. EFFECT OF PHENOTHIAZINE DRUGS AND HYPERTHERMIA ON MOUSE FIBROSARCOMA

Treatment Time (d) to Delay in reach 11 mm diameter tumour growth (mean ± SE) (d ± SE)

Control 2.7 ± 0.2 -

Heat 41°C (tumour core 1 h) 3.8 ± 0.4 1.1 ±0.5 PMZ (2 mg/kg) +41°C 5.2 ±0.5 2.5 ± 0.5 PCP (2 mg/kg)+ 41°C 4.4 ±0.5 1.7 ±0.5 TMZ (2 mg/kg) + 4 1°C 5.5 ± 0.5 2.8 ±0.5

Heat 42° С 4.8 ±0.3 2.1 ±0.4

PMZ (2 mg/kg) + 42°C 9.1 ±0.8 6.4 ±0.8 PCP ( 2 mg/kg) + 42°C 8.9 ±0.5 6.2 ±0.5 TMZ (2 mg/kg) + 42°C 11.0 ± 1.0 8.3 ± 1.0

Heat 43°C 5.5 ±0.3 2.8 ± 0.4 PMZ (2 mg/kg) + 43°C 10.1 ± 0.4 7.4 ±0.5 PCP (2 mg/kg) + 43°C 10.8 + 0.7 8.1 ±0.7 TMZ (0.5 mg/kg) + 43°C 11.8 ±0.6 9.1 ±0.6 TMZ (1 mg/kg)+ 43°C 12.0 ±0.8 9.3 ±0.8 TMZ (2 mg/kg) + 43°C 12.4 ±0.8 9.7 ±0.8

Starting diameter is 8 ± 1 mm. Number of tumours in each group was 7--10.

Furthermore, a decrease in the rectal temperature of 2 to 3°C was noticed on administration of these drugs to the animals. In view of the fact that PCP caused no radioprotection but reduced the animal’s core temperature, the hypothermic effect of these drugs cannot explain their radioprotective property.

3. HYPERTHERMIC RESPONSE OF MOUSE SOLID TUMOURS AS MODIFIED BY PHENOTHIAZINE DRUGS

Hyperthermia alone or in combination with radiation and/or drugs is fast emerging as a useful modality in cancer treatment. Since plasma membrane is the main organelle involved in heat killing of cells [26], phenothiazines, by virtue of their membrane activity, might interact with this phenomenon. IAEA-SM-290/74 315

FIG. 2. Effect of phenothiazine drugs and hyperthermia on mouse fibrosarcoma. Drugs given intravenously 5-10 min before heating. The starting diameter of the tumour was 8 ± 1 mm. Error bars show the SEM. Number o f tumours per group: 7 -1 0 [23]. (o) Heat alone (tumour core 1 h) (л) PMZ (2 mg/kg) + heat (A) PCP (2 mg/kg) + heat (□) TMZ (2 mg/kg) + heat

Indeed, we have earlier shown that CPZ administered before hyperthermia caused substantial delay in tumour growth compared with tumours heated without the drug [22]. On the other hand, no such effect in sarcoma 180 was obtained, presumably because of the differences in the membrane fluidity of these two tumour lines [22]. The effects of the other phenothiazine drugs such as PMZ, PCP and TMZ on the hyperthermic response of a mouse fibrosarcoma are shown in Table II. It can be seen that intravenous administration of any one of these drugs at a dose of 2 mg/kg body weight before heating the tumours to 42 or 43°C significantly potentiated the heat effect, their relative effectiveness being TMZ > PCP = PMZ. At 41°C no significant effect of heat, even in combination with drugs, was noticed. A detailed study employing TMZ demonstrated that, while the drug was ineffective at 0.1 mg/kg, it equally affected the tumour growth at doses of 0.5, 1.0 and 2.0 mg/kg (Table II). 316 GEOR G E et al.

Time of treatment (min)

FIG. 3. Time optimization o f hyperthermic effect in the presence ¿ f trimeprazine in mouse fibrosarcoma. Drug given subcutaneously as a single injection 5-10 min before heating tumours. Error bars show the SEM. (o) Heat 42°С (tumour core 1 h) (A) TMZ (0.9 mg/kg) + 42° С

With a view to getting further insight into the combined effect of these drugs and heat, growth delays versus temperature are plotted in Fig. 2. It can be seen that the slopes of the straight lines between 41 and 42°C are different from those between 42 and 43°C. We computed temperature incremental gain (TIG) as the slopes between these two time intervals which measure the differences between the growth delays on heating the tumours to 41,42 or 43°C. I can then be seen that for heat alone the gain is 1 d between 41 and 42°C, whereas it is only 0.7 d between 42 and 43°C. This may mean that the additional advantage gained on heating to 43 instead of 42° С would be only 70% of that which accrues when tumours are heated to 42 instead of 41 °C. However, when TMZ or PMZ is present, the major benefit is derived between 41 and 42°C (TIG 5.5 and 3.9, respectively) and the additional gains between 42 and 43°C are only about 25% of this value (TIG 1.4 and 1.0, respectively). Similarly, in the presence of PCP the gain between 42 and 43°C (TIG 1.9) is approximately 40% of that between 41 and 42°C (TIG 4.5). It can thus be surmised that, while in tumours treated with heat alone significant advantage may be achieved by increasing the temperature to 43°C, in combination treatment with TMZ or PMZ the major benefit is realized at precisely 42°C; treating at 43°C is therefore of little additional advantage. Considering the fact that these and other membrane active drugs interact with cellular membranes and modify their micro-environment [27] as well as their response to heat [28], it is likely that they might shift the temperature profiles for hyperthermic killing as seen in the present study, thus necessitating suitable modifications in treatment protocols. IAEA-SM-290/74 317

E _£ О) ш «Е ТЗ

Days after treatment

FIG. 4. Effect o f trimeprazine and melphalan on mouse fibrosarcoma. Drugs given sub cutané ousiy daily for 5 days. Starting diameter was 6 ± 1 mm. Error bars show the SEM. Number of tumours per group: 8. Control O—------o MEL ( 0.05 mg/kg) X------X MEL (0.5 mg/kg) ▲— ▲ MEL (1 mg/kg) ------■ TMZ (0.9 mg/kgj + MEL (0.05 mg/kg) a —— ■ TMZ (20 mg/kg) + MEL (0.05 mg/kg) □—------□ TMZ (0.9 mg/kg) + MEL (0.5 mg/kg)

In view of the modifications in the membrane caused by these drugs we further investigated time optimization of hyperthermic effects in the presence of TMZ. Figure 3 shows that maximum advantage is obtained when the heat treatment is given for 60 minutes, irrespective of the presence of the drug.

4. CHEMOSENSITIZATION BY TRIMEPRAZINE TARTRATE

The anticancer action of some commonly used chemotherapeutic agents such as MEL and CY was enhanced when tumour bearing animals were pretreated with TMZ. Figure 4 shows that the tumoricidal effect of MEL was concentration dependent, resulting in 75 and 100% tumour cure at a daily subcutaneous dose of 1 and 4 mg/kg, respectively, for 5 days. At doses less than 1 mg/kg, MEL only caused a delay in tumour growth without any cure. TMZ alone was ineffective up to a dose of 20 mg/kg while, at a much lower dose of 0.9 mg/kg, it significantly enhanced the tumoricidal effect of MEL at doses of 0.05 and 0.5 mg/kg. 318 G E O R G E et al.

Days after treatment

FIG. 5. Effect of trimeprazine, melphalan and hyperthermia on mouse fibrosarcoma. Drugs given subcutaneously daily for 5 days. On the 5th day, within 5-10 min following injection, tumours were locally heated. Starting diameter was 6 ± 1. Error bars show the SEM. Number of tumours per group: 7-15. •— —Ф Control о----— о Heat 42° С (tumour core 1 h) X— —X TMZ (0.9 mg/kg)+ 42°С ▲ ----A MEL (0.05 mg/kg) + 4 2° С Л ----Л MEL (0.5 mg/kg)+ 42°С ■— TMZ (0.9 mg/kg) + MEL (0.05 mg/kg) + 42° С □— ----□ TMZ (0.9 mg/kg) + MEL (0.5 mg/kg) + 42° С

The chemosensitizing effect of TMZ on CY was less pronounced. TMZ, however, was equally effective when given at doses of 0.9, 10 or 20 mg/kg in combination with MEL and CY. It is significant to note that the acceptable clinical dose of TMZ is in the range of 0.6 to 0.9 mg/kg.

5. COMBINED EFFECT OF HYPERTHERMIA AND ANTICANCER DRUGS MODIFIED BY TRIMEPRAZINE

The effect of TMZ on the combined treatment of tumours with MEL and hyperthermia is shown in Fig. 5. As TMZ appeared fully effective at 42°C, hyperthermia was given at this temperature for 1 hour (see Section 3). Heat alone, at 42°C, delayed the tumour growth significantly. When combined with MEL, hyperthermia considerably increased its effectiveness. It can IAEA-SM-290/74 319 further be seen that TMZ, at 0.9 mg/kg, also enhanced the heat effect. A combination of TMZ (0.9 mg/kg), MEL (0.05 mg/kg) and heat produced a growth delay of 8.2 ± 0.9 d, which was greater than the sum of the delays of the three treatments given individually (3.4 ± 1.0 d). Increasing the MEL dose to 0.5 mg/kg in this modality resulted in 60% tumour cure as compared with the mere growth delay noticed when only MEL and heat were applied (Fig. 5). In the remaining 40% of animals the tumours showed a much longer delay in growth than those treated without TMZ. Although TMZ also increased the combined effect of heat and CY, no tumour cure was noticed, indicating the superiority of MEL over CY in this combination treatment. The mechanism by which TMZ may potentiate the cytocidal effects of MEL, CY and hyperthermia is still not understood. A small but significant reduction of 0.8 ± 0.1 °C in the rectal temperature of mice when TMZ was combined with MEL or CY may partly affect the drug pharmacokinetics, resulting in increased plasma levels of the cytotoxic drugs. In addition, pheno thiazines may reduce the drugs’ efflux from cells owing to their calcium influx blocking action [29, 30], thereby enhancing the cytotoxicity of MEL and CY. A similar mechanism of potentiation of cytotoxicity of adriamycin and vincristine by CPZ and trifluoperazine has been reported earlier [31, 32]. Trifluoperazine has also been reported to interact with bleomycin and hyper thermia in mouse tumour cells [33]. In addition, TMZ being a membrane interacting agent [27] also alters membrane permeability and thereby enhances cellular uptake of the cytotoxic drugs. These phenomena may be augmented at elevated temperatures.

6. CONCLUSION

Our results thus indicate that TMZ, in addition to being a hypoxic cell radiosensitizer and euoxic radioprotector, may also significantly improve tumour therapy if used in combination with antineoplastic drugs such as MEL or CY and hyperthermia.

REFERENCES

[1] ADAMS, G.E., DEWY, D.L., Biochem. Biophys. Res. Commun. 12 (1963) 473. [2] FOWLER, J.F., “Chemosensitization and radiosensitization by nitroimidazoles”, New Chemo- and Radio-Sensitizing Drugs (BRECCIA, A., FOWLER, J.F., Eds), Lo Scarabeo, Bologna (1985) 61. [3] SHENOY, M.A., SINGH, B.B., GOPAL-AYENGAR, A.R., Sciences (N.Y.) 160 (1968) 999. [4] MYERS, D.K., Adv. Biol. Med. Phys. 13 (1970) 219. 320 G E O R G E et al.

[5] SHENOY, M.A., JOSHI, D.S., SINGH, B.B., GOPAL-AYENGAR, A.R., Adv. Biol. Med. Phys. 13 (1970) 255. [6] ALPER, T., “Cell death and its modification: The roles of primary lesions in membranes and DNA”, Biophysical Aspects of Radiation Quality (Proc. Symp. Lucas Heights, 1971), IAEA, Vienna (1971) 171. [7] SHENOY, M.A., SINGH, B.B., GOPAL-AYENGAR, A.R., Nature (London) 248 (1974) 415. [8] GEORGE, K.C., SHENOY, M.A., JOSHI, D.S., BHATT, B.Y., SINGH, B.B., GOPAL- AYENGAR, A.R., Br. J. Radiol. 48 ( 1975) 611. [9] SHENOY, M.A., GEORGE, K.C., SINGH, B.B., GOPAL-AYENGAR, A.R., Int. J. Radiat. Biol. 28 (1975) 519. [10] DJORDJEVIC, B., Radiology 131 (1979) 515. [11] YONEI, S., Int. J. Radiat. Biol. 36(1979) 547. [12] YATVIN, M.B., SCHMITZ, B.J., DENNIS, W.H., Int. J. Radiat. Biol. 37 (1980) 513. [13] YAU, T.M., KIM, S.C., Br. J. Radiol. 51 (1980) 551. [14] MANIAR, H.S., SINGH, B.B., Int. J. Radiat. Biol. 44 (1983) 399. [15] MANIAR, H.S., NAIR, C.K.K., SINGH, B.B., Radiat. Environ. Biophys. 23 (1984) 279. [16] GEORGE, K.C., SRINIVASAN, V.T., SINGH, B.B., Int. J. Radiat. Biol. 38 (1980) 661. [17] SHENOY, M.A., SINGH, B.B., Indian J. Exp. Biol. 18 (1980) 791. [18] GEORGE, K.C., SINGH, B.B., Indian J. Exp. Biol. 22(1984) 305. [19] SHENOY, M.A., SINGH, B.B., Int. J. Radiat. Biol. 34 (1978) 595. [20] SHENOY, M.A., BIAGLOW, J.E., VARNES, M.E., DANKEL, J.W., Int. J. Radiat. Oncol., Biol. Phys. 8 (1982) 725. [21] SHENOY, M.A., SINGH, B.B., Radiat. Environ. Biophys. 24 (1985) 1 13. [22] GEORGE, K.C., SINGH, B.B., Br. J. Cancer 45 (1982) 309. [23] GEORGE, K.C., SINGH, B.B., Br. J. Cancer 51 (1985) 737. [24] FELMEISTER, A., DISCHER, C.A., J. Pharm. Sei. 53 (1964) 756. [25] SAMUELS, A.M., CAREY, M.E., Gastroenterology 74 (1978) 1186. [26] HAR-KEDAR, I., BLEEHEN, N.M., Adv. Radiat. Biol. 6 (1976) 229. [27] SEEMAN, P.P., Pharmacol. Rev. 24 (1972) 583. [28] YATVIN, M.B., Int. J. Radiat. Biol. 32(1977) 513. [29] LANDRY, Y., AMELLAL, M., RUCKSTUHL, M., Biochem. Pharmacol. 30 ( 1981) 2031. [30] WEISS, B., PROZIALECK, W.C., WALLACE, T.L., Biochem. Pharmacol. 31 (1982) 2217. [31] TSURUO, T., LIDA, H., NOJINE, M., TSUKA GOSHI, S., SAKURAI, Y., Cancer Res. 43(1983) 2905. [32] KRISHNAN, A., SAUERTEIG, A., WELLHAM, L.L., Cancer Res. 45 (1985) 1046. [33] MIRCHEVA, J., SMITH, P.J., BLEEHEN, N.M., Br. J. Cancer 53 (1986) 99. DISCUSSION

(Summary of discussion held on Papers IAEA-SM-290/43, 15,40,4, 54 and 74)

In response to a question, S. Krishnamurthi (Paper IAEA-SM-290/43) indicated that patients who had reached hospital had mostly cancer but no leukoplakia, and that in the early days leukoplakia had been treated with implant or external radiotherapy but that now local leukoplakia was treated surgically and extended cases with radiotherapy. Replying to questions, A. Hidayatalla (Paper IAEA-SM-290/15) pointed out that to reduce tumour volume before irradiation he had tried multi-drug treatment with bleomycin and methotrexate and had obtained good results but no controlled treatment had been done, and that owing to the large bilateral masses of lymph nodes it had been impossible to shield the spinal cord and parotid gland in most cases, but he had not met radiation myelitis. In response to a question, S. Okkan (Paper IAEA-SM-290/40) indicated that in his previous single-dose study the enhancement ratio (ER) had been found to b e 1.35—1.48 in vivo with a dose of 0.1—0.5 mg/g for ornidazole. Ahúm an skin erythema study with a single dose had also shown an ER of 1.24 with a dose of 2.5 g/m2. In view of this he had not expected an ER of more than 1.2. Replying to a comment that the tumour response depended not only on factors such as hypoxia or euoxia but also on others (share of stroma, differentiation of tumour cells, etc.), and that in a study like the one discussed these factors should be taken into account S. Okkan pointed out that he had given cases according to the prognostic factors such as haemoglobin level, histopathology and age, but the number of patients was too small at that time to make any conclusions about the significance of each factor in the trial. A speaker commented that in her country over 66 to 70% of breast cancers were being seen at a late stage which did not come within the scope of surgery and radiotherapy alone could successfully treat the tumour only in 26% of patients with stage P3, P4, and that after combined radiotherapy and chemotherapy the rate of cure of breast tumours and even remote métastasés reached 56%. In his response, J. Fodor (Paper IAEA-SM-290/4) pointed out that according to his experience based on observation of patients with T4_4d and N3, the role of chemotherapy has not been established. Another speaker stated that the paper presented by J. Fodor constituted a model which could be used to develop data on a larger scale in Europe (in Scandinavian countries, the United Kingdom, the German Democratic Republic, the Federal Republic of Germany, etc.). Such data could serve as a guide for optimal treatment protocols to be largely used by radiotherapists in developing countries. J. Fodor mentioned that the answer was contained in that statement.

321 322 DISCUSSION

In response to questions, C.W. Song (Paper IAEA-SM-290/54) pointed out that toxicity of Fluosol-DA (20%) had been minimal and that, generally, one out of ten patients did not tolerate injection of Fluosol-DA (20%). An increase in alkaline phosphatase activity had been noticed in 50% of patients treated but it subsequently subsided and kidney function remained intact. The study on animals had shown no late effects. A conventional fraction schedule of 250 cGy a day for five days a week to a total dose of 6000 cGy had been used and the Fluosol-DA (20%) intravenous injections with 0 2 breathing had been done once a week. Another speaker commented that in Japan perfluorochemicals had been widely used as artificial blood without serious toxic effects, and therefore Fluosol-DA (20%) could be regarded as a potentially useful sensitizer. In response to a question, K.C. George (Paper IAEA-SM-290/74) pointed out that the mechanism by which these phenothiazine drugs acted as radio sensitizers or radioprotectors in animals was still not understood. However, studies on bacterial cells had shown that these drugs exhibited radiosensitizing properties under hypoxic conditions, while under oxygenated conditions they acted as radioprotectors. Whereas the radiosensitizing effect under hypoxia was attributed to the radiation short lived radicals of the drug which could be toxic to cells, the radioprotective effect in euoxic cells was ascribed to the fluidization of the cellular membrane on treatment with such drugs, which would facilitate diffusion of the non-protein sulphydrides within the membrane thus allowing chemical restitution of damaged sites. Further, with the case of chlorpromazine under oxygenated conditions, the radiolytically induced chlorpromazine cation radicals could be oxidized to chlorpromazine sulphoxide which is non-toxic and this might also partly account for the radioprotective effect under euoxic conditions. A short answer to the question is that these drugs would sensitize hypoxic cells to ionizing radiation and protect oxygenated ones against radiation damage irrespective of whether they are malignant or not. In response to another question, K.C. George mentioned that he had not yet started clinical trials with these drugs, that radiosensitizing or radioprotective effects of phenothiozine derivatives and, particularly, chlorpromazine, produced by doses in which they are usually applied as anti-emetics were still being investigated, that the doses of the drug needed to sensitize hypoxic cells in a clinical situation might be high, but the potentiation of the hyperthermia effect in vivo by these drugs had been achieved at clinically acceptable doses of 0.5-5.0 mg/kg body weight. DOSIMETRY AND TECHNOLOGY

IAEA-SM-290/53

CAESIUM MANUAL AFTERLOADING INTRACAVITARY TREATMENT OF CARCINOMA CERVIX: A SIMPLE METHOD OF DOSE CALCULATION

K.A. EL-GHAMRAWI, M.M. MAHFOUZ*, R. MOULD, O. ZAKI Kasr El-Einy Centre of Radiation Oncology and Nuclear Medicine (NEMROCK), Faculty of Medicine, Cairo University, Cairo, Egypt

Abstract

CAESIUM MANUAL AFTERLOADING INTRACAVITARY TREATMENT OF CARCINOMA CERVIX: A SIMPLE METHOD OF DOSE CALCULATION. For the proper use of the computer calculated dose rate tables of Mould and Hobday for the Amersham 137Cs manual afterloading system with ±10% accuracy, a standard application should be employed. Using anteroposterior and lateral X-ray films of the pelvis, certain guidelines have been developed to describe such a standard application. Seventy-eight patients with carcinoma cervix were treated with the Amersham 137Cs manual afterloading system. In 9 instances there was disagreement between the ready-made dose rate tables and the two-dimensional detailed calculations. In 5 of these 9 cases the disagreement was due to non-standard application. In the remaining 4 cases the reasons were not clear. In all stage I cases (10 patients) no single local recurrence could be detected during a mean observation time of 20 months. Grade III or IV rectal or bladder reactions were not observed. Thus, this simple method of dose calcu lation could be used in developing countries for standard applications and when a trained medical physicist is not available.

1. INTRODUCTION

Carcinoma cervix is a common neoplasm in developing countries. Early stages are well curable by intracavitary brachytherapy. A major problem in the use of this method is the lack of trained medical physicists. Mould and Hobday [1] have published dose rate tables for the Amersham caesium-137 manual after loading system. Hospital experience with this system is described by Bateman et al. [2]. However, the indiscriminate use of such tables may result in serious under- or overdosage at the tumour site and/or critical organs. The aim of this study is to define the standard application in which the ready-made tables could be used with an error not exceeding ±10% and to describe NEMROCK experience in treating carcinoma cervix with this simple method.

* Professor Emeritus of Radiation Oncology and Nuclear Medicine, Faculty of Medicine, Cairo University, Cairo, Egypt.

325 326 EL-GHAMRAWI et al.

TABLE I. CLINICAL STAGING OF CARCINOMA CERVIX (FIGO STAGING SYSTEM)

Stage No. of patients Percentage

I 10 12.8 11a 3 3.8 lib 37 47.5 Щ 23 29.5 IV 2 2.6 Recurrent 3 3.8

Total 78 100.0

TABLE II. RADIATION TREATMENT PROTOCOL FOR CARCINOMA CERVIX

Clinical stage Treatment recommended

la Two intracavitary insertions, 70 Gy to Point A Ib & 11 a Two intracavitary insertions, 70 Gy to Point A plus uni- or bilateral parametrial external irradiation bringing dose to pelvic wall to 60 Gy lib & III Whole pelvic external irradiation, 45 Gy in 22 fractions, followed by one intracavitary insertion giving 35-40 Gy to Point A IV Palliative single intracavitary insertion

2. MATERIAL AND METHODS

Dose rates to point A and isodose distributions, within the pelvis have been studied with computer models [1 ]. Models have been made for an ideal applica tion and for various clinical situations in which the application is not ideal. The various geometrical variants have been studied and guidelines developed to define a standard application as seen in anteroposterior (AP) and lateral X-ray films of the pelvis. IAEA-SM-290/S 3 327

TABLE III. FREQUENCY OF THE USE OF INTRAUTERINE 137Cs SOURCES

Type of source No. of times used Percentage

Long 68 77.3 Medium 13 14.8 Short 3 3.4 Extra long 4 4.5

TABLE IV. FREQUENCY OF THE USE OF VAGINAL 137Cs SOURCES

Type of source No. of times used Percentage

Large 9 10.2 Medium 53 60.2 Small 23 26.2 Tandem 3 3.4

Seventy-eight patients with carcinoma cervix in various clinical stages were treated exclusively or in part with the Amersham 137Cs manual afterloading system. The dose rates to point A as derived from the ready-made tables were compared with the two-dimensional dose calculations. The differences were analysed in view of the derived guidelines for standard application.

3. RESULTS AND DISCUSSION

Between 1 September 1982 and 30 April 1986, 78 patients with carcinoma cervix were treated with the Amersham 137Cs manual afterloading system. Their mean age was 51.6 years, ranging between 30 and 70 years. The clinical stages and treatment protocols adopted are shown in Tables I and II, respectively. The mean value of the length of the intrauterine canal was 62 mm. This explains why in 77% of the applications the long intrauterine source train was used (Table III). The medium sized vaginal sources were the commonest sources used (60%; Table IV). This information could be very useful for developing countries when ordering intracavitary brachytherapy equipment. It is also important to note that the washer system was used in the majority of cases (87%; Table V). 328 EL-GHAMRAWI et al.

TABLE V. FREQUENCY OF THE USE OF WASHERS AND SPACERS

Type Number Percentage

Washer 74 87.1 Spacer 11 12.9

FIG. 1. Anteroposterior view of the Amersham applicator showing various derived geometri cal variants.

The washer applicator gives more stability and a constant relationship between the two ovoids. The following measurements are determined from an anteroposterior X-ray film of the pelvis (Fig. 1): (1) Beta angle: the angle between the axis of the intrauterine tube and the vaginal axis. (2) J-distance: the distance between the lower surface of the locking flange and the front face of the most advanced vaginal ovoid. This measures the degree of separation between the intrauterine train and the vaginal sources. The wide separation could be the result of the presence of a cauliflower tumour or the use of inappropriate vaginal ovoid. IAEA-SM-290/S3 329

FIG. 2. Lateral view o f the Amersham applicator showing alpha angle and R-distance.

TABLE VI. CRITERIA FOR STANDARD APPLICATION

(1) Beta angle less than 20° (2) J-distance <15 mm (3) K-distance < 1 0 mm (4) Alpha angle 90-180' (5) R-distance > 1 0 mm

(3) К-distance: the distance between the two horizontal lines passing the mid points of each vaginal source. This distance measures the amount of displace ment of one ovoid relative to the other. The following measurements are made from the lateral X-ray film: (1) Alpha angle: the angle between the vaginal axis and the axis of the intra uterine tube (Fig. 2). (2) R-distance: the vertical distance between the most posteriorly placed vaginal source and the axis of the rectal catheter. Table VI shows the derived guidelines for standard application in which the dose rates at point A right and left differ by more than ±10%. Variations of anatomical angulation between uterine and vaginal axes are shown in Tables VII and VIII. In 9 instances the table values differed from the individually calculated values by more than 10%. In 5 of these 9 cases the disagreement was due to non-standard application. In the remaining 4 cases no clear explanation could be found. 330 EL-GHAMRAWI et al.

TABLE VII. VARIATION OF ANATOMICAL ANGULATION BETWEEN UTERINE AND VAGINAL AXES (BETA ANGLE) IN AP X-RAY FILM

Beta angle No. of cases Percentage

0 -5 ° 24 30.8 о 1 Os О 22 28.3 11-15° 17 21.8 16-20° 7 8.9 2 1 -2 5 ° 4 5.1 2 6 -3 0 ° 4 5.1

TABLE VIII. VARIATION OF ANATOMICAL ANGULATION BETWEEN UTERINE AND VAGINAL AXES (ALPHA ANGLE) IN LATERAL X-RAY FILM

Alpha angle No. of cases Percentage

90-100° - - 101-110° 6 7.7 111-120° 6 7.7 121-130° 9 11.5 131-140° 12 15.4 141-150° 16 20.5 151-160° 20 25.6 161-170° 6 7.7 171-180° 1 1.3 > 180° 2 2.6

The above mentioned data indicate that these derived guidelines are valid to discriminate between standard and non-standard cases in 95% of applications (84/88 applications). It is also noted that this system was used easily by junior staff without serious complications. The mean rectal dose rate was 28.6 ± 16.3 rad/h.1 The short term results in stage I cases treated only with brachytherapy are satisfac tory (Table IX).

1 1 r a d = 1.00 X 10“2 Gy. IAEA-SM-290/5 3 331

TABLE IX. TREATMENT RESULTS OF STAGE I CASES WITH BRACHY THERAPY

Total number of cases treated 10 Locally free of disease 9 Lost to follow-up 1 Liver metastasis 1

Note: Mean follow-up period 20 months, Duration of follow-up 3-42 months.

4. CONCLUSION

In developing countries, where financial resources and medical staff are limited, the earlier detection of carcinoma cervix, its treatment with brachy therapy at peripheral hospitals and the use of a very simple method of dose calculation are very encouraging and safe.

REFERENCES

[1] MOULD, R.F., HOBDAY, P.A., Radiation Dosimetry for Amersham Caesium-137 Manual Afterloading System for Gynaecological Brachytherapy, Amersham (1984). [2] BATEMAN, T.J., DAY, T.J., SKEGGS, D.B.L., 5 years hospital experience with Amersham 137Cs manual afterloading system, Br. J. Radiol. 56 (1983) 401.

IAEA-SM-290/32

Invited Paper

CODE OF PRACTICE FOR ABSORBED DOSE DETERMINATION IN PHOTON AND ELECTRON BEAMS

H. SVENSSON*, P. ANDREO**, J. CUNNINGHAM***, K. HOHLFELD+

* Radiation Physics Department, University of Urnea, Umeâ, Sweden

**Seccion de Fisica, Departamento de Radiología, Hospital Clínico Universitario, Zaragoza, Spain ***Physics Department, Princess Margaret Hospital, Toronto, Ontario, Canada + Physikalisch Technische Bundesanstalt, Braunschweig, Federal Republic of Germany

Abstract

CODE OF PRACTICE FOR ABSORBED DOSE DETERMINATION IN PHOTON AND ELECTRON BEAMS. An advisory group was set up by the IAEA to suggest measures to be taken for the production of a dosimetry protocol. The authors of the paper were chosen to be authors. The Agency was of the opinion that such a protocol would be of great value not only to the network of Secondary Standard Dosimetry Laboratories (SSDL) but also to hospitals providing radiation treatment for cancer patients. The report includes recommendations on the procedure for determining the absorbed dose at low and medium energy X-rays, and high energy photon and electron radiation. Advice on equipment, measurement geometry and quality assurance is given. It was decided that the symbols and formalism should follow the ICRU recommen dations. The numerical data on interaction coefficients follow the recommendations of the standards laboratories (i.e. CCEMRI). Correction factors (i.e. katt and km) to be applied for about 40 types of commercial ionization chambers were computed as it was considered that it would be difficult to restrict the use to a few types of chambers, as in the NACP protocol, or advise the users on how to carry out complicated computations, as in the AAPM protocol. A part of the report is devoted to conventional X-rays. In this case a very general type of formalism is suggested. It was found that there is a lack of information on the correction factors to be applied for different types of chambers. Furthermore, it was found that conven tional dosimetry procedures, often used in determining the absorbed dose at the medium

333 334 SVENSSON et al. energy range of X-rays, underestimate the absorbed dose by several per cent. More work is needed in this field. An independent evaluation of the dosimetry resulting from the applica tion of this protocol has been carried out for high energy photon and electron radiation using the FeS04 dosimeter as a reference. The agreement in absorbed dose values was generally within fractions of one per cent. The conclusion is, therefore, that use of this report can give an absorbed dose determination sufficiently accurate for practical applications in radia tion therapy.

1. INTRODUCTION

Modern day practices of radiation therapy emphasize the need for high doses of radiation to be delivered with increased accuracy. With the availability of improved anatomical information from sophisticated diagnostic imaging procedures, the information required to achieve better accuracy in the determination of the absorbed dose delivered to the patient is available. It has been demonstrated that the success or failure of radiation treatment depends on the absorbed dose delivered to the tumour and that this generally should not vary by more than a few per cent from the prescribed values [1-3]. This accuracy is necessary, at least for some types of tumours, in order to secure a large probability of eradication of the tumour without unacceptable complications. An essential component of the total uncertainty in the delivery of an absorbed dose to the tumour is the uncertainty in the calibration of the radiation beam. It has been shown [4] that the three components, namely, determination of the absorbed dose in a water phantom, dose planning, and irradiation of the patient, contribute almost equally to the overall uncertainty in the absorbed dose delivery. In this analysis it was assumed that the very best methods were used in the com plete procedure. An overall uncertainty of about 8% (corresponding to one standard deviation in uncertainty) was derived for the absorbed dose delivered to a point of interest in the tumour. The radiation treatment was assumed to be carried out with high energy photon or electron radiation. A large number of absorbed dose intercomparisons have been carried out by different departments that show much larger variations in the stated dose values than would be expected if a good procedure is followed [5-7]. The uncertainty in the absorbed dose determination of a reference point in a water phantom is thus often much larger than would be expected for the complete therapy procedure (including dose planning, patient set-up, etc.). It is also often not possible for the physician to adjust the dose level in a proper way following local clinical experience as errors in dosimetry may be of different sizes for different beam qualities within the department. The aim of the present code of practice in dosimetry is to improve the accuracy in dose determination. Local codes have been published by many national organizations [8-14]. In general, however, they are somewhat too specific in that they serve the requirements of the facilities in the countries where the documents originated. The authors of this code have been involved in the IAEA-SM-290/32 335 development of national procedures in their own countries. Further, an advisory group consisting of persons from several different countries should, it is hoped, ensure a very broad applicability of this code of practice.

2. MEASUREMENT CHAIN

The complete measurement chain is shown in Fig. 1. It starts with the absolute determination of exposure or air kerma at the Primary Standards Dosimetry Laboratory (PSDL), then the calibration is transferred to the Secondary Standards Dosimetry Laboratory (SSDL), where the hospital’s (user’s) chamber is calibrated and, finally, the absorbed dose is determined using this chamber in the user’s beam. The procedure for high energy photon or electron radiation is shown in the diagram, where the calibration at PSDL and SSDL is carried out in a 60Co gamma beam. A similar procedure is used for the determination of absorbed dose at low (10-100 kV) and medium (100-300 kV) energy X-ray qualities. In the complete calibration chain a large number of interaction coefficients (e.g. sw> ад, Sgj-ajj., W/e, , see below for explanation of symbols) and correction factors (e.g. km, katt and pu) are introduced. It is important to have a coherent system of values for these coefficients and correction factors in order to minimize systematic errors. The numerical data on stopping powers, W/e (= 33.97 J-C"1 ), g, and energy absorption coefficients follow the recommen dations of the CCEMRI(I) [15]. The complete calibration chain from the PSDL to the user is therefore considered in this report.

3. ABSORBED DOSE DETERMINATION

3.1. High energy photon and electron radiation

The absorbed dose to water is to be determined using an ionization chamber applying the Bragg-Gray equation. The final equation including interaction and correction factors has the form:

Dw,u(Peff) = Dair(sw, air Ju'Pu'Peel (1) where Dw u (Peff) is the absorbed dose to water at the position of the effective point of measurement of the ionization chamber, Dair is the mean absorbed dose to air in the ionization chamber air cavity, (sw air)u is the ratio of mean stopping powers water to air for the radiation quality in use at the effective point of measurement, and pu and pce] are correction factors discussed below. The problem faced by the physicist is first to determine D jj from the ionization chamber measurement in the radiation beam of interest and then to choose the appropriate values for (sw>air)u and pu, pcel. 336 SVENSSON et al.

й rt Ф гН Х> 3 i X» В О a» bo >> rt г-i 2; 'S» О й -р •С rt (Ö -нс -p о •н о о 11 и >> 1—1 ф • CO rt С тз O й >> 3 о г. E _Й -И О ■р о* •*н rt Р rt +-> •гЧ -р Й 3 О гН С rt ■H rt rt о N S*á rt 3 •н •н S « о 4 -р С С rt О II О С й •н Й •г4 о Х> о O -Р •н •н Ф S -p О ■р гН г-Н \ o Ф rt rt Х> o rt и й о В «M и Х> •н Й О •н -р х: •H Й О i—i rt -р rt ф ад >» rt W x> со CJ x) й -р ТЗ о rt rt í о rt rt -и § rt vo й X > rt -р «и > о cd rt ф о •H CJ о со М X* ■р Ф 1 й Й . й С й й •н Ф •H ■н +* ф 2 •н rt H rt •<н со rt rt о О ф o o ■н ft Ф со -p S *н <м X о й &D Ф <м Ф Й ф rt й й Ф й ф w й rt ° 3 -<н О •Н С ft o (Ц со <4 О О rt “гН xJ а * со rt X « rt Ф Ч С С Ф В é3* Е о О О g X i rt й •н Ф •н Ф -p ф i -р С -р Л5 ад Х> X о •н rt 1 с rt й Й O й Х> •H (О Ф Ф rt Д й й й -Р -Р г-Н O •и ф ф С Ф rt C •rH rt со С •н 'Ö о •H -p Р 3 ф

Й Ф ft -p 3 G Ф 1 О E n Й •h o o г-t Ф V Й •H •H -P rt -p •p 3 Ф ы o rt ¿3 E •H Ф M o tj В C rH x: й Й о О Ф C -p -p rt -р •H O •H o + ‘H Ф 5 H Ф -P Й ф Й Ф ■H Ф rH Ф x: x> x> x> + e « § •H cl ft Л rt ¿¡■s Éi o o Calibration Calibration chain utilized for the determination of absorbed dose gamma Co in 60 radiation, high energy X-rays or electron radiation.

О О

IQSd 10SS U3Sn F/G. 7. IAEA-SM-290/32 337

Determination of D aj.. The mean absorbed dose to air in the cavity is obtained from Dair = Nd Mu . where ND is the absorbed dose to air chamber factor. This factor is determined from the air kerma calibration in the 60Co gamma beam. This is possible as it is assumed that the same meter reading (M) is obtained for a given value of Dair whether the irradiation is carried out at the users’ beam quality (index u) or at the calibration quality (index c). The relation between air kerma K^jj. c at the calibration quality and D ^ is

Dair = Kair; c ( 1 ( 2 ) where g = 0.003 (for definition see Ref. [16]), is a factor correcting for the non-air equivalence of the ionization chamber, and kgtt is à factor correcting for the absorption and scattering in the ionization chamber wall and buildup material. The effect of the central electrode is not included (see below). Values of the product km-katt for almost 40 different types of chambers are given in the protocol. The intent is that the hospital physicist should not need to calculate the values. Stopping power ratio (sw ац.)и. The Spencer-Attix cavity theory was used for the determination of stopping power ratios (sw ajr)u . The cut-off energy, Д, used for the restricted stopping powers was chosen as 10 keV. For electron radiation the stopping power data are taken from Monte Carlo calculations of electron spectra [13]. Input parameters for the choice of(sw ац.)и are the mean energy at the phantom surface, E0, and the depth to the effective point of mea surement, z. Eo is to be determined at the depth of the 50% depth dose [3]. For high energy photon radiation the stopping power data were taken from Ref. [17]. Monte Carlo calculations [18] were used for generating electron energy spectra and determining average stopping power values. The parameter for speci fying beam quality is TPRio, i.e. the tissue phantom ratio for 20 and 10 cm depth. The field size is 10 cm X 10 cm. (Corresponding values of D20/D 10 are also given, i.e. the ratio of absorbed dose at 20 and 10 cm depth using SSD = 100 cm.) Perturbation correction factor, pu. The factor pu corrects for the non-water equivalence of the chamber (i.e. chamber wall material and air cavity) when mea surements are carried out in the users’ beam. The corrected meter reading should be representative for a point in the phantom where the effective point of measure ment is situated. For electron radiation Рц corrects for the differences in scattering of primary electrons in the water and the air cavity [11, 19, 20]. The material of the chamber wall can be disregarded if this is of low atomic number. For photon radiation the pu factors differ from those given in other proto cols as they are valid only if an effective point of measurement is used. This definition is used as the pu factors then correct only for the non-water equiva lence of the chamber wall. The replacement of water by the air cavity is thus accounted for by the choice of ?eff- The AAPM [13] reports a procedure to calcu late pu (based on Ref. [21 ]). In this protocol pu values are given as a function of the photon beam quality with the wall material as a parameter (see Fig. 2). It is assumed that the wall thickness is 0.5 mm. The uncertainty in using these 338 SVENSSON et al.

FIG. 2. Perturbation correction factor pu as a function o f the quality of photon beams for different materials o f the chamber walls. It is assumed that the chamber is of thimble type having a wall thickness o f 0.5 mm [36]. factors also for other wall thicknesses should be small as the Рц correction is always fairly small for plastic or graphite chamber walls. Correction factor for perturbation of the central electrode, pcei. None of the national protocols correct for the perturbation effect caused by the central electrode. The correction can be neglected if the electrode material is of graphite or plastic. However, in commercial ionization chambers the central electrode is frequently made of aluminium and a correction ought to be carried out [22-24]. The correc tion should in principle be applied for both the determination of D^/Mç (Eq. (2)) at the calibration quality and for the measurement (Eq. (1)) in the users’ beam. This correction factor is relatively small and, furthermore, often partly cancels out in the two steps: calibration and dose measurements. In this protocol a simplified procedure is used. Thus a global correction factor, ptej, is introduced in Eq. (1). This factor is given for different beam qualities in Table I for a cylindrical ioni zation chamber (Farmer type). It is seen that a maximum error of about 2% can be introduced for some types of chambers and beam qualities by disregarding this correction.

3.2. Low energy and medium energy X-rays

For low energy X-rays (i.e. 10 to 100 kV) it is recommended that either a plane parallel chamber together with a phantom is calibrated directly in absorbed IAEA-SM-290/32 339

TABLE I. CORRECTION FACTOR pcel TO BE USED IF THE CHAMBER HAS A CENTRAL ELECTRODE OF ALUMINIUM (DATA FOR A FARMER-TYPE CHAMBER [23, 24, 31])

Electrode radius Electrons Photons 60Co and photons (mm) (h^)max>25 MeV № y)max<25 MeV

0.5 1.008 1.004 1.000 1 1.015 1.008 1.000 1.5 1.020 1.010 1.000 2.5 1.032 1.016 1.000 dose to water per unit scale division or the chamber is calibrated in air kerma in free air and the absorbed dose to water is determined from the relationship

Dw — Мц Nj( B(¿len/P)w, air (3) where NK is the air kerma calibration factor, В is the backscatter factor, and (7xen/p)w air is the ratio of the mean mass energy absorption coefficient water to air. Values of В and (psn/p)w ^ are given in the IAEA protocol. For medium energy X-rays it is recommended that the absorbed dose be determined at 5 cm depth. This is in agreement with recommendations of the ICRU [ 1 ] but differs from the procedure used by many hospitals where measure ments are made free in air and Eq. (2) is used for absorbed dose calculations. The ionization chamber is, in this protocol, to be calibrated in air kerma per unit scale division in a beam of as similar a radiation quality as possible to that used at the hospital. In the user’s beam the chamber centre, which is considered to be the effective point of measurement, is placed at 5 cm depth. The following relationship holds

Dw — Мц Njçku (Меп/^-^w, air Pu (4) where Dw, Mu, NK and (j¡Ten/p)w ^ have the conventional meanings (see above), ku is a correction factor that is introduced since the air kerma calibration factor may be quality dependent. This factor corrects for the difference between radia tion quality for the calibration, free in air, and for the measurements (by the user) in the water phantom. For the chambers recommended in this protocol it has been assumed that ku = 1.00. pu is the perturbation correction factor that corrects for the replacement of water by an air cavity equal to the external size of the ioni zation chamber. Without this factor Eq. (1) would give the absorbed dose in a small mass of water at the centre of the air cavity. New values for (Men/P)w,air and pu have been recommended [25]. The absorbed dose determined will therefore differ by several per cent from previous values. 3 4 0 SVENSSON et al.

BS64 different l-values ( PSDls BS64) Protocols BSBO в- correction BSB2

it's a weak link in the dosimetric chain

FIG. 3. Schematic drawing showing that the stopping power data used by the PSDLs and the hospital physicists (in the different national dosimetry protocols) are based on different compu tations by Berger and Seltzer (the year of the references is given in the figure). This will intro duce errors in dosimetry at the hospitals. A change in the data at the PSDLs was carried out in 1986 but a corresponding change was not made by the hospitals. The IAEA protocol uses the same set o f dosimetry data as are applied at the PSDLs.

4. EVALUATION OF THE CODE OF PRACTICE

Uniformity in dosimetry using this protocol has been excellent (within a few tenths of one per cent) among the standard laboratories, as shown in several intercomparisons [26-28]. However, the same values on interaction coefficients have generally been used by the different laboratories. The intercomparisons have thus shown consistency, but accuracy in absorbed dose determination has not been proven. Recently, for instance, revised stopping power data were introduced that directly influence the determined value of the absorbed dose. There is generally reasonable uniformity in measurement procedures at the hospitals within a country. However, a weak link in the dosimetry chain is that stopping powers of different values are used by the PSDLs and in the different national protocols. The situation is schematically described in Fig. 3. In the IAEA protocol the aim has been to produce a consistent calibration chain from the PSDL through the SSDL to the hospital. For high energy photon and electron radiation the IAEA Code of Practice was evaluated by Mattsson and Svensson [29]. As a reference ferrous sulphate dosimetry was used (emG = 352 X 10~6m 2 -kg“1- Gy-1 was applied). As can be seen in Table II, the differences in absorbed dose values for the two methods IAEA-SM-290/32 341

TABLE II. ABSORBED DOSE DETERMINATION BASED ON THE ‘IAEA IONIZATION CHAMBER METHOD’ COMPARED WITH RESULTS FROM FERROUS SULPHATE DOSIMETER MEASUREMENTS [29]. THE LARGEST DIFFERENCE IS FOR AN EXPERIMENTALLY USED VERY ‘CLEAN’ ELECTRON BEAM (10 *MeV)

Difference Quality Djk i ) FeSO , (%)

Gamma 60Co 2.017 2.018 -0 .0 5 X-ray 4 MV 2.017 2.017 0 X-ray 16 MV 2.002 1.996 +0.3 Electron 18.8 MeV 1.998 2.002 -0 .2 Electron 10 *MeV 2.005 1.964 +2.1 * Electron 10 MeV 2.003 2.008 -0 .2 5 Electron 6 MeV 2.006 2.010 -0 .2

were generally a few parts of one per cent. Only in electron beams generally not used for radiotherapy the difference was as large as 2%. (This may, however, at least partly depend on the uncertainty in using Rs0 to determine E0 and thus also be applied as an imput parameter for sw air.) Corresponding evaluations have been made by Mattsson [30] for several national protocols. The difference was here somewhat larger, between -1.5 and + 2.3%. In this study a graphite ionization chamber was used. The variation would increase to a maximum of 1 or 2% if chambers of other wall materials were used [31, 32]. This relatively good uniformity is due to the fact that some errors in the dosimetry cancel out along the calibration chain. This difference in determined dose values is probably of minor importance as compared with other errors in the complete procedure used for the treatment of patients. However, the lack of consistency in the calibration procedure may be a far larger problem as it may confuse the hospital physicist. The currently used dosimetry for medium energy X-rays is based on ICRU Reports 23 and 24 [1, 33]. Recently, Kubo [34] and Mattsson [31] determined absorbed dose to water values, using a water calorimeter, that were several per cent higher than conventional values. Cunningham et al. [35] showed that the values of (^en/p)water; gleite are very different at small field sizes at the surface of the phantom and at greater depths for large field sizes with a beam quality of 250 kV X-rays. Lack of this information previously has almost certainly led to errors of as much as 10% in dose determinations. For medium energy X-rays the IAEA protocol uses experimental determina tions of pu carried out at the Physikalisch-Technische Bundesanstalt (Braunschweig, Federal Republic of Germany). The dosimetry for medium beam qualities at the 342 SVENSSON et al. reference condition (depth 5 cm, field size 10 cm X 10 cm) gives about 5 to 10% higher values than what has conventionally been determined. This implies that at least some of the RBE values determined using these X-ray qualities as a reference must be in error by this percentage. The difference in RBE between 60Co gamma radiation and medium energy X-rays would thus be less than currently believed.

5. CONCLUSION

The IAEA dosimetry protocol should make it possible to improve consist ency and accuracy in dosimetry. An evaluation of this procedure shows that the agreement with other reference methods (i.e. dosimetry based on the ferrous sulphate dosimeter or the water calorimeter) is excellent. It is hoped that the extensive amount of tabulated data provided should make it simple for physi cists to use this protocol. The changes in dose values for high energy photon and electron radiation would be 2 or 3% at the most compared with dosimetry based on national protocols. For medium energy X-rays, on the other hand, the diffe rence may be as large as 10%.

REFERENCES

[ 1 ] INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASUREMENTS, Determination of Absorbed Dose in a Patient Irradiated by Beams of X or Gamma Rays in Radiotherapy Procedures, ICRU Publications, Bethesda, ICRU Rep. 24 (1976). [2] STEWART, J.G., JACKSON, A.W., The steepness of the dose response curve both for tumour cure and normal tissue injury, Laryngoscope 85 (1975) 1 107. [3] BRAHME, A., Dosimetric precision requirements in radiation therapy, Acta Radiol. Oncol. 23 (1984) 379. [4] SVENSSON, H., Quality assurance in radiation therapy: physical aspects, Int. J. Radiat. Oncol., Biol. Phys. 10, Suppl. 1 (1984) 59. [5] SVENSSON, H., Dosimetric measurements at the Nordic medical accelerators. II. Absorbed dose measurements, Acta Radiol., Ther., Phys., Biol 10 (1971) 1102. [6] EISENLOHR, H.H., JAYARAMAN, S., IAEA-WHO cobalt-60 teletherapy dosimetry service using mailed LiF dosemeters. A survey of results obtained during 1970-75, Phys. Med. Biol. 22(1977) 18. [7] JOHANSSON, K.-A., MATTSSON, L.O., SVENSSON, H., Dosimetric intercomparison at the Scandinavian radiation therapy centres. I. Absorbed dose intercomparisons, Acta Radiol. Oncol. 21 (1982) 1. [8] HOSPITAL PHYSICISTS ASSOCIATION, A Practical Guide to Electron Dosimetry 5-35 MeV, HPA Rep. Ser. No. 4, HPA, London (1971). [9] HOSPITAL PHYSICISTS ASSOCIATION, A Practical Guide to Electron Dosimetry below 5 MeV for Radiotherapy Purposes, HPA Rep. Ser. No. 13, HPA, London (1975). [10] HOSPITAL PHYSICISTS ASSOCIATION, Revised code of practice for the dosimetry of 2 to 35 MV X-rays, and of caesium-137 and cobalt-60 gamma-ray beams, Phys. Med. Biol. 28(1983) 1097. IAEA-SM-290/32 343

NORDIC ASSOCIATION OF CLINICAL PHYSICS, Procedures in external radiation therapy dosimetry with electron and photon beams with maximum energies between 1 and 50 MeV, Acta Radiol. Oncol. 19 (1980) 55. NORDIC ASSOCIATION OF CLINICAL PHYSICS, Electron beams with mean energies at the phantom surface below 15 MeV, Acta Radiol. Oncol. 20(1981) 403. AMERICAN ASSOCIATION OF PHYSICISTS IN MEDICINE, A protocol for the deter mination of absorbed dose from high-energy photon and electron beams, Med. Phys. 10 (1983) 741. SOCIEDAD ESPAÑOLA DE FISICA MEDICA, Procedimientos Recommendados para la Dosimetría de Fotones y Electrones de Energías comprendidas entre 1 MeV y 50 MeV en Radioterapia de Haces Externos, Comité de Dosimetría en Radiotherapia, SEFM Publ. No. 1/1984, Madrid (1984). COMITE CONSULTATIF POUR LES ETALONS DE MESURE DES RAYONNEMENTS IONISANTS (Section 1), Bureau International des Poids et Mesures, Sèvres, 8th Meeting (1985). INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASUREMENTS, Radiation Quantities and Units, ICRU Publications, Bethesda, ICRU Rep. 33 (1980). BERGER, M.J., SELTZER, S.M., Stopping Powers and Ranges of Electrons and Positrons, National Bureau of Standards, Gaithersburg, Rep. NBSIR 82-2550 (1982). ANDREO, P., BRAHME, A., Stopping power data for high energy photon beams, Phys. Med. Biol, (in press). INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASUREMENTS, Radiation Dosimetry: Electron Beams with Energies between 1 and 50 MeV, ICRU Publications, Bethesda, ICRU Rep. 35 (1984). JOHANSSON, K.-A., MATTSSON, L.O., LINDBORG, L., SVENSSON, H., “Absorbed dose determination with ionization chambers in electron and photon beams having energies between 1 and 50 MeV”, National and International Standardization of Radia tion Dosimetry (Proc. Symp. Atlanta, 1977), Vol. 2, IAEA, Vienna (1978) 243. ALMOND, P.R., SVENSSON, H., Ionisation chamber dosimetry for photon and electron beams. Theoretical considerations, Acta Radiol., Ther., Phys., Biol. 16 (1977) 177. KRISTENSEN, M., Measured influence of the central electrode diameter and material on the response of a graphite ionization chamber to cobalt-60 gamma rays, Phys. Med. Biol. 28(1983) 1269. MATTSSON, L.O., Application of the water calorimeter, Fricke dosimeter and ioniza tion chamber in clinical dosimetry. An evaluation of correction factors and interaction coefficients, Thesis, University of Gothenburg, Gothenburg, Sweden (1984). ROGERS, D.W.O., BIELAJEW, A.F., NAHUM, A.E., Ion chamber response and Awa]] correction factors in a Co-60 beam by Monte Carlo simulation, Phys. Med. Biol. 30 (1 9 8 5 )4 2 9 . SCHNEIDER, U., Bestimmung der Wasser-Energiedosis im Wasserphantom mit frei in Luft kalibrierten Ionisationskammern. Jahresbericht 1985, Physikalisch-Technische Bundesanstalt, ISSN 0340-4366, Braunschweig (1986). NIATEL, M.-T., LOFTUS, T.P., OETZMANN, W., Comparison of exposure standards for 60Co gamma rays, Metrología 11 (1975) 17. ZSDANSZKY, K., “Primary and secondary standards of dosimetry. Calibration methods in Hungary”, National and International Standardization of Radiation Dosimetry (Proc. Symp. Atlanta, 1977), Vol. 1, IAEA, Vienna (1978) 107. HOFMEESTER, G.H., “Calorimetric determination of absorbed dose in water for 1-25 MeV X-rays”, Biomedical Dosimetry: Physical Aspects, Instrumentation, Calibration (Proc. Symp. Paris, 1980), IAEA, Vienna (1981) 235. 3 4 4 SVENSSON et al.

[29] MATTSSON, L.O., SVENSSON, H., (1987), Comparison of absorbed dose determination using ferrous sulphate dosimeter and the ionization chamber method, to be published in Acta Radiol. Oncol. [30] MATTSSON, L.O., Comparison of water calorimetry and ionisation chamber dosimetry in 100 and 200 kV X-ray beams, CCEMRI(I)/85-l5, BIPM Sèvres (1985). [31] MATTSSON, L.O., JOHANSSON, K.-A., Experimentally Determined Wall Correction Factors km and katt for Cylindrical Ionization Chambers used in High Energy Photon and Electron Beam Dosimetry, Radiation Physics Department, University of Göteborg, Internal Report (1984). [32] MIJNHEER, B.J., WITTKAMPER, F.W., Comparison of recent codes of practice for high-energy photon dosimetry, Phys. Med. Biol. 31 (1986) 407. [33] INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASUREMENTS, Measurement of Absorbed Dose in a Phantom Irradiated by a Single Beam of X or Gamma Rays, ICRU Publications, Bethesda, ICRU Rep. 23 (1973). [34] KUBO, H., Water calorimetric determination of absorbed dose by 280 kVp orthovoltage X-rays, Radiother. Oncol. 4(1985) 275. [35] CUNNINGHAM, J.R., WOO, M., ROGERS, D.W.O., BIELAJEW, A., The dependence of mass energy absorption coefficient ratios on beam size and depth in a phantom, Med. Phys. 13(1986) 496. [36] ANDREO, P., NAHUM, A.E., BRAHME, A., Ionization chamber dependent wall correc tion factors in dosimetry, Phys. Med. Biol, (in press). IAEA-SM-290/61

CLINICAL DOSIMETRY IN CZECHOSLOVAKIA

V. LAGINOVÁ Institute of Clinical Oncology, Bratislava J. NOVOTNŸ Institute of Radiation Dosimetry, Prague Czechoslovakia

Abstract

CLINICAL DOSIMETRY IN CZECHOSLOVAKIA. The treatment of cancer patients is a multidisciplinary effort involving the synergistic interaction of various medical branches. It is the duty of the radio-oncology team to make sure that the quality of the patients’ care and the accuracy of the dose delivered during the treatment is the best achievable under the current state of the art. Since 1984 a new ‘Protocol for clinical dosimetry of therapeutic photon and electron beams in the energy range of 10 keV—50 MeV’ has been in use in all Czechoslovak radiotherapy departments. The approach of this Protocol is similar to that used in NACP or AAPM protocols, i.e. (a) the response of an ionization chamber is characterized by a mass calibration factor; (b) water or plastic phantoms are recommended for primary dosimetry; (c) all chamber dependent and radiation dependent parameters remain explicit in the dose calculation. The first experience since the introduction of the Protocol in radiotherapy practice can be summarized as follows: (1) uniformity in clinical dosimetry throughout the whole country and the intercomparison of national standards can determine the consistency of clinical dosimetry between different countries; (2) improvement in the precision of clinical dosimetry procedures in radiotherapy; (3) improvement of quality assurance in radiotherapy in all radiotherapy departments; (4) possibility of objective comparisons of radiotherapy treatment results between different institutions.

1. INTRODUCTION

The treatment of a cancer patient is a multidisciplinary effort involving the synergistic interaction of clinical oncology, pathology, radiography or ultrasono graphy, surgery, chemotherapy, haematology, immunology and radiation therapy. A treatment protocol is decided upon following an examination and evaluation of the patient by the oncology team. This protocol may be based on one or more adjuvant modalities, the most common being surgery, chemotherapy and radio therapy. Radiotherapy usually involves delivery of a certain quantity of radiation in several fractions to be administered over a certain period of time. It is the duty of the radio-oncology team to make sure that the quality of the patient’s care and the accuracy of the dose delivered during the treatment is the best achievable under the current state of the art.

345 6 4 3

Quality assurance procedures AIOÁAD NOVOTNY AND LAGINOVÁ

FIG. 1. Flow chart of Protocol procedures. IAEA-SM-290/61 347

Precision in radiotherapy is required at different stages which can be summarized as follows: (a) knowledge of the prescribed dose, (b) precise determination of the target volume, (c) treatment unit parameters, (d) estimation of the absorbed dose, (e) patient set-up, (f) dose application. The responsibility for the first two steps lies clearly with the radiotherapist and the rest mainly with the physicists. Different steps are involved in the accurate delivery of a treatment. To achieve the desired accuracy, quality assurance is required for all these steps but the nature of the quality assurance is very different for each one. Each procedure must be carried out with great care and skill in order to secure high quality treatment. In addition, the equipment must fulfil certain specifications both when first taken into use and during all its future operation during radiation treatment. Basic recommenda tions for different quality assurance procedures have been published by interna tional [1-4] and national [5—7] organizations. Since 1984 a new ‘Protocol for clinical dosimetry of therapeutic photon and electron beams in the energy range of 10 keV— 50 MeV’ has been used in all Czechoslovak radiotherapy departments. The main features and the first experience since the introduction of the Protocol are discussed.

2. GENERAL CONSIDERATIONS

The purpose of this Protocol is to provide radiophysicists with an accurate method of determining the dose to water from the photon and high energy electron beams used for radiation therapy, i.e. X-rays from 10 kV up to 50 MV, 137Cs and 60Co gamma rays, and electrons with energies from 5 to 50 MeV. The calibrated cavity method, which has provided medical physics and radiotherapy with a relatively simple and accurate means of dosimetry for almost two decades, is used in the Protocol. The approach of the Protocol is similar to that used in NACP [5] or AAPM [6] protocols, i.e. (a) the response of an ionization chamber is characterized by a parameter Ng that is a function of a variety of chamber dependent parameters in addition to the 60Co exposure calibration factor; (b) polystyrene and acrylic plastics or water phantoms are recommended for primary dosimetry; (c) all chamber dependent and radiation dependent parameters remain explicit in the dose calculations. The flow diagram in Fig. 1 shows each component of the Protocol. The major components are listed in the second line, and the physical parameters or other dosimetric data related to each of the components are listed in the third line.

(a) A 60Co calibration factor for the user’s ionization reference ionization chamber is a prerequisite for the application of the Protocol. Although the 60Co calibration factor is not a major component of the Protocol, it is assigned a prominent position in the flow diagram so as to assure the user that this national 348

Co beam Electron beam AIOÁAD NOVOTNY AND LAGINOVÁ

FIG. 2. Diagram for calculating the absorbed dose to water from phantom measurement in electron beams. IAEA-SM-290/61 349 reference standard plays an essential but less explicit role in radiation therapy dosimetry. Because there is no central Secondary Standard Dosimetry Laboratory in Czechoslovakia, calibrations with 60Co gamma rays are performed in a primary laboratory situated at the Institute of Radiation Dosimetry (IRD) of the Czechoslovak Academy of Sciences in Prague and in an accredited radiotherapy centre, the Institute of Clinical Oncology in Bratislava, which serves as a secondary standardization laboratory for radiotherapy centres. X-ray calibration of special ionization chambers is performed at the Institute of Hygiene in Prague. The consistency of calibration sources is regularly checked by international inter comparison or by postal TLD intercomparison organized by the IAEA and WHO.

(b) The cavity gas calibration factor, Ng, is the dose to the gas (usually air) in the chamber per unit electrometer reading. It is a constant for all radiation qualities for which the average energy expended in the production of one ion pair (W) is the same as that of 60Co gamma rays (W/e = 33.85 ± 0.15 JC-1 for dry air). This is the case for radiation covered by the Protocol. The cavity gas calibration factor can be obtained from the primary standardization laboratory or from the secondary laboratory. The cavity gas calibration factor is calculated using the equations given in the Protocol or in the Code of Practice which is available to standardization laboratories. The calculation of Ng is made only once, and subsequently only if the 60Co exposure calibration factors are to be changed.

(c) The ratio of the average stopping powers and the ratio of the average mass energy absorption coefficients, which are necessary to calculate the dose to the phantom material, are functions of the spectrum of electrons and photons at the point of measurement, and this spectrum, in turn, is a function of the incident energies of X-rays, gamma rays or electrons. The Protocol recommends that beam quality parameters be determined from in-phantom depth ionization measurements.

(d) The user may choose any of three materials for a dosimetry phantom: water or polystyrene and acrylic plastics. As water is the reference material for dose calibration, the additional step of transferring dose to plastic to dose to water is avoided by using a water phantom. Recommended calibration depths for water phantoms are given in the Protocol for each radiation quality. It is recommended that the user choose a dosimetry phantom that, based upon the data provided in the Protocol and his own special requirements, will yield the most accurate and reproducible dosimetry for his radiation beams.

(e) When ionization chamber measurements are made in the phantom, the product of Ng and the electrometer reading is the dose to gas in the chamber from the radiation fluence at the measurement point. The dose to the phantom material 35 0 LAGINOVA AND NOVOTNY

CO X *

<0 00 О ffi 3 ЗЙ ^ о . E ~ -

<£ тз oo О) > * о со

X с4 FIG. 3. of of error and uncertainty magnitudes Sources connected with radiotherapy procedures. IAEA-SM-290/61 351 that replaces the chamber when it is removed from the phantom is given by the product of the dose to gas, the ratio of the stopping power of the phantom material to air or the ratio of mass energy absorption coefficients of air to material (in the case of low energy X-rays), and the factors to account for phantom replacement (total perturbation factor). Figure 2 shows the flow diagram for the calculation of absorbed dose from phantom ionization measurements for electron beams. Similar diagrams are given in the Protocol for all radiation qualities.

(f) When plastic dosimetry phantoms are used, the theory and data enable the user to determine the dose to plastic. Having followed the irradiation procedures in the Protocol, the phantom fluence at the measured point in the plastic phantom is the same as would be obtained at the calibration depth in a water phantom. When photon fluences are equal, the dose to water is related to the dose to plastic by the ratio of their average mass energy absorption coefficients. After correcting the dosimeter reading for lower electron fluence in polystyrene and acrylic plastics, the dose to water is given by the product of the dose to plastic and the ratio of their unrestricted mass stopping powers and is called the transfer factor.

Each step in the measurement chain, from the standard dosimetry laboratory to the radiation treatment of patients, contributes to the uncertainty of the absorbed dose delivered to the irradiated tissue. Individual errors of each step were analysed [8] and individual uncertainties for each step as well as cumulated uncertainties were established. Figure 3 shows the main sources of error and the individual and cumulated uncertainties at the 95% confidence level. On the basis of this analysis, the permissible limits for total uncertainty of absorbed dose estimation in a reference point in radiation therapy were established in the Protocol as U95 = ±5% for gamma rays, high energy X-rays and electron beams and U95 = ±7% for low energy X-rays (below 500 kV). The dosimetry data on which the irradiations are based should be checked regularly for consistency. The number of tests may depend to some extent on the behaviour of the particular therapy unit and its intended use. A quality asssurance programme for X-rays, 60Co gamma rays and accelerator beams is suggested in the Protocol, where the frequencies and number of basic tests are given for each step in the dosimetric chain. All necessary energy dependent and chamber dependent data are tabulated in the Protocol and semi-empirical formulas are given in the appendix for quick calculation of the calibration factor Ng, the absorbed dose and other factors on small computers or pocket calculators. Standard programs in Basic and for the HP-41 calculator are available. 35 2 LAGINOVÁ AND NOVOTNŸ

FIG. 4. Diagram showing the general features of most recommendations used in radiation therapy.

3. COMPARISON WITH OTHER PROTOCOLS

Since 1980 a number of national and international protocols have been published [1—7]. These protocols are concerned with the calibration procedures for photon and electron beam absorbed dose determination by ionization chamber measurements. The ionization chambers are calibrated, at single photon energy, against a national standard which would, therefore, be the same for all users in a particular country and would be constant with time. In addition, intercomparison of national standards determines the consistency of the machine calibrations between countries. If the assumption is made that national standards agree within a known value (generally better than ±1%), then any differences in machine calibrations will be due to the protocols. Although all protocols appear to be rather similar, there are some funda mental differences in the approaches used. The general outline for all protocols is shown in Fig. 4. The absorbed dose to water can be determined using equations IAEA-SM-290/61 353 in the protocols for both photon and electron beams. For each protocol it can be seen that the ratio of the absorbed dose to water divided by the product of the 60Co exposure factor and electrometer reading is a constant. Using a value of parameters given in each protocol, for a given set of conditions, this ratio has been calculated. A graphite walled ionization chamber, which is allowed in all protocols, has been used for comparison. The calculations have been performed both for photon and for electron beams and for different energies. Figure 5 shows an example for three energies. The average values obtained from all protocols are compared with the values obtained using procedures given in the Czechoslovak Protocol. The values differed by no more than ±1.5% from these values. The differences between protocols can be explained by the different values for the parameters, including stopping power values, mass energy absorption coefficient values, cut-off energies, wall attenuation factor, etc.

4. INTERCOMPARISON OF 60Co UNITS

It is expected that the introduction of the Protocol into radiotherapy practice will bring: (1) uniformity in clinical dosimetry throughout the country; in addition, intercomparison of national standards can determine the consistency of clinical dosimetry between different countries; (2) improvement of the quality assurance of clinical dosimetry procedures in radiotherapy centres; (3) the possibility of objective comparisons of radiotherapy treatment results between different centres. To test the impact of the Protocol on dosimetry in radiotherapy centres a postal TLD intercomparison of absorbed doses has been carried out by the Institute of Radiation Dosimetry. The method has been described in detail elsewhere [9]. Briefly, alumino- phosphate glass discs are prepared in the IRD and are distributed to radio therapy centres by post. The recipients are requested to irradiate the TLD dosimeters with a dose of 2 Gy at a depth of 5 cm in a water phantom. The dosimeters are returned to the IRD and the absorbed dose is determined by comparing them with TLD discs irradiated with 60Co gamma rays of the primary standard. The total uncertainty of this method has been evaluated to be ±3% at the 95% confidence level. Intercomparisons of the IRD primary standard with other primary or secondary standards have shown a maximum deviation of 0.3%. The results are expressed as the percentage deviation between the dose quoted by the radiotherapy centre and the dose measured by the IRD. A positive deviation means that the dose measured by the IRD is higher than the quoted dose. Figure 6 shows the percentage deviation plotted against the percentage 3 5 4 LAGINOVÁ AND NOVOTNŸ

0.96

X-RAYS: 25 MV

0.94

ä 0.92

0.90

DIN NACP NCRP H PA AAPM CS ORGANIZATION M Reading electrometer N x Calibration coefficient for exposure X CP Ref. - P AC N D IN - D IN No. 6809 Klinische Dosimetrie, Therapeutische A nw endung gebündelter endung nw A [6] Ref. Therapeutische - Dosimetrie, M protocols: P Klinische A A 6809 different No. to IN D according - established doses absorbed f o IN D Comparison 5. FIG.

NCRP - NCRP Rep. No. 69: D osim etry o f X-ray and Gamma R a y Beams fo r Radiation r fo Beams y a R Gamma and X-ray f o etry osim D 69: No. Rep. NCRP - NCRP NUMBER OF INSTITUTIONS (%) N AC P = Nordic Association o f Clinical Physics Clinical f o e orm N Association Industrie Nordic = eutsche D = P AC N IN D A AP M = American Association o f Physicists in Medicine. in f Physicists o Association American = M AP A radiotherapy, in etry dosim beam RP C N electron r fo Practice f o Code and [7] Ref. - HPA HPA = Hospital Physicists Association Physicists Hospital = HPA 10 20 30 I. . eut psa necmaio of o hrp uis n Czechoslovakia. in units therapy Co f o intercomparison postal f o Results 6. FIG. = = 1 - 1 -9 -7 -5 -3 - 3 7 1 1 15 13 11 9 7 5 3 1 -1 3 - 5 - 7 - 9 - -11 3 -1 -15 National Council o f Radiation Protection Radiation f o Council National hs Md Biol. Med. Phys. Elektronenstrahlung und Gamma Röntgen, Therapy in the Energy Range 10 k e V to 50 MeV. 50 to V e k 10 Range Energy the in Therapy [5] 30, 30, 18) 1169. (1985) IAEA-SM-290/61 VI ON (%) N IO T IA EV D

-i ------1 ------1 ------■ 1 ■ ■ ■ - --- ■ i 1------1 i ■—

355

35 6 LAGINOVÁ AND NOVOTNY

deviation for all results (25 centres took part in the intercomparison). The dashed curve in Fig. 6 shows the results of the intercomparison carried out in 1975 using a ferrous sulphate dosimeter. An acceptable deviation for such simple dose measurements in radiotherapy is ± 5%. All the measurements were found to be within this limit; 60% of the measurements were even within ±3%. Comparing this result with the inter comparison carried out in 1975, one can see an important improvement in the dosage. This is mainly caused by the introduction of the Protocol, improved instrumentation and improved quality assurance of the clinical dosimetry in radiotherapy centres.

5. CONCLUSION

The goal of radiotherapy is to eradicate a tumour without causing severe damage to healthy tissues. An overall precision of about ±10% (at the 95% confidence level) in the absorbed dose at any point of the patient is required to meet this goal. The overall uncertainty is a combination of several components. An improvement, therefore, necessitates more work in several fields from fundamental dosimetry to practical problems such as fixation of the patient. There are many ways to improve this situation, for instance, to apply quality assurance programmes to all equipment in use (dosimeter instruments, dose planning systems, simulators, accelerators, etc.) and to carry out dose measurements on each patient. The Protocol solves only one part of the whole problem. Regulations and recommenda tions from authorities or hospital physicists organizations are of great importance in order to give minimum requirements on equipment, quality assurance programmes and competence of personnel.

REFERENCES

[ 1 ] INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASUREMENTS, Radiation Dosimetry: X-Rays and Gamma Rays with Maximum Photon Energies Between 0.6 and 50 MeV, ICRU Rep. No. 14, IÇRU Publications, Washington, DC (1969). [2] INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASUREMENTS, Measurement of Absorbed Dose in a Phantom Irradiated by a Single Beam of X or Gamma Rays, ICRU Rep. No. 23, ICRU Publications, Washington, DC (1973). [3] INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASUREMENTS, Determination of Absorbed Dose in a Patient Irradiated by Beams of X or Gamma Rays in Radiotherapy Procedures, ICRU Rep. No. 24, ICRU Publications, Washington, DC (1976). [4] INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASUREMENTS, Radiation Dosimetry: Electron Beams with Energies Between 1 and 50 MeV, ICRU Rep. No. 35, ICRU Publications, Bethesda, MD (1984). IAEA-SM-290/61 357

NORDIC ASSOCIATION OF CLINICAL PHYSICS, Procedures in external radiation therapy dosimetry with electron and photon beams with maximum energies between 1 and 50 MeV, Acta Radiol. Oncol. 19 (1980) 55. AMERICAN ASSOCIATION OF PHYSICISTS IN MEDICINE, A protocol for the determination of absorbed dose from high energy photon and electron beams, Med. Phys. 10(1983) 741. HOSPITAL PHYSICISTS ASSOCIATION, Revised Code of Practice for dosimetry of 2 to 25 MV X-ray, caesium-137 and cobalt-60 gamma ray beams, Phys. Med. Biol. 28 (1983) 1097. NOVOTNY, J., KOVÁR, Z., Cesk. Radiol, (in print). ^ WAGNER, R., NOVOTNY, J., JIROUSEK, P., KOVAR, I., HORÁKOVÁ, I., Res. Rep. ÚDZ CSAV 185/86 (1986).

IAEA-SM-290/27

TOTAL GAMMA DEPTH DOSE DISTRIBUTION FROM A CALIFORNIUM -252 SOURCE IN A TISSUE EQUIVALENT PHANTOM

F.A. EL-BAKKOUSH Medical Application Division, Radiation and Radioisotopes Application Sector, Tajura Nuclear Research Centre T.S. AKKI, R.M. MEGAHID Neutron Physics Division, Physics and Materials Science Sector, Tajura Nuclear Research Centre Tripoli, Libyan Arab Jamahiriya

Abstract

TOTAL GAMMA DEPTH DOSE DISTRIBUTION FROM A CALIFORNIUM-252 SOURCE IN A TISSUE EQUIVALENT PHANTOM. Measurements were made of the depth dose distribution of total gamma rays produced when collimated beams of neutrons and gamma rays emitted from a californium-252 source are passed through a water phantom. The phantom was exposed to direct and boron-carbide filtered beams of 3, 5 and 10 cm diameter. The total gamma absorbed dose was measured by means of 7LiF thermoluminescence dosimeters. The depth dose distribution of total gamma rays was studied by measuring the gamma dose at different positions along the beam axis and in planes parallel and vertical to the beam direction. The absorbed dose values obtained were used to plot groups of attenuation curves in planes vertical to the beam axis. These attenuation curves were used to construct groups of isodose charts when direct and boron carbide filtered beams are used. In addition, the data obtained were integrated in planes vertical to the beam axis to obtain the attenuation of total gamma dose when an infinite plane monodirectional source is used.

1. INTRODUCTION

Californium-252 has a potential use in radiotherapy as a replacement of pure gamma emitters because of its lower oxygen enhancement ratio (OER) [1-4]. At lower dose rates (— 10 rad/h)1 the OER of the mixed neutron and gamma ray emission from 252Cf is found to be approximately 1.5, which is indistinguishable from that resulting from the fission spectrum alone. This is explained by the fact that at very low dose rates, the gamma ray component has a negligible biolo gical effect [5-7]. As the dose rate is increased to approximately 100 rad/h, the

1 1 rad = 1.00 X 10~2 Gy.

359 3 6 0 EL-BAKKOUSH et al. relative biological effectiveness (RBE) of the gamma rays can no longer be ignored. In addition, when comparing acute exposures of high LET radiation such as neutrons with a low LET radiation such as gamma rays, the RBE varies with the dose [8-10]. Therefore, the values of LET, RBE and OER vary as the ratio of neutron to gamma dose changes with the tissue depth in the case of 2S2Cf because the radiation emitted is a mixture of neutrons and gamma rays. The irradiation of the human body with neutrons from discrete sources or from broad beams generated by 14 MeV neutron generators is becoming increasin gly important in medical diagnosis and cancer therapy [11,12]. In diagnostic applications the entire body or parts of the body may be exposed to neutrons [13]. However, in therapy a small part of the body is exposed to a collimated neutron beam or to neutrons from an implanted 252Cf source. Therefore, it becomes of interest to determine the spatial dose distribution from neutrons and gamma rays at selected sites for both applications. The spatial dose distribution in a phantom irradiated by mixed radiation beams of neutrons and gamma rays can be obtained by calculation or by measur ing the depth dose at various locations in a tissue equivalent phantom [14-18]. The data obtained by the most accurate computational methods become less accurate when the irradiated phantom contains irregularities such as bone or air cavities of different geometries. Therefore, it becomes necessary to measure the dose distribution in a tissue equivalent phantom to obtain the data required for the treatment planning programme [19]. This paper presents the data obtained by measuring the depth dose distri bution of total gamma rays at various positions in a phantom irradiated by colli mated beams of neutrons and gamma rays from a 252Cf source. The data can be used to derive information such as total gamma dose attenuation within and out side the collimated beam of penetrating radiations. They can also be used to provide an indication of the dependence of the direct delivered and scattered depth dose on the beam diameter.

2. EXPERIMENTAL DETAILS

Measurements of the total gamma depth dose distribution resulting from the penetration of 252Cf radiation in a tissue equivalent phantom were performed in a water filled polyethylene tank. The tank had outer dimensions of 50 X 50 X 50 cm. The water tank was placed in front of the horizontal channel of the irradiation cell which contains a 50 pg 252Cf neutron source. The 252Cf is in the form of a palladium cermet pellet doubly encapsulated in a welded stainless steel capsule 8 mm in diame ter and 10 mm in length. The source is stored at the bottom of the central tube of the irradiation cell and is moved to the desired position during measurements by a remote method. The thickness of the paraffin wax and the lead shielding surround ing the irradiation cell are designed to accommodate up to 500 ßg of 252Cf source. A schematic diagram of the experimental layout is shown in Fig. 1. The gamma dose distribution was measured by means of thermoluminescence detectors based on lithium fluoride made into solid discs by incorporation with IAEA-SM-290/27 361

1 9

FIG. 1. Schematic diagram of the experimental layout. 1 - Irradiation cell; 2 - Polyethylene tank; 3 - 252С /source; 4 - Radiation collimator; 5 - Sample holder; 6 - Sample guide; 7 - Boric acid; 8 — Lead shield; 9 - Paraffin wax.

Teflon. This detector has proved to be particularly useful for measuring the absorbed dose distribution in tissue equivalent phantoms [20-22]. The use of •this detector for such applications has grown rapidly in the last decade because it possesses the following characteristics: — Doses can be measured over a very wide range of integral dose — Energy dependence is very small — Quick reusability and acceptable accuracy — Small physical size of the dosimeter — Ability to store radiation exposure information over a period of many months. The available TLD used is7 LiF discs in a Teflon matrix 12.7 mm in diameter and 0.4 mm in thickness. The TL response of the dosimeter was measured by a Toledo 654 TL reader. The TL detectors were calibrated with the aid of a 60Co source of known emission rate and the TL response for a certain emission rate was used to convert the measured TL response to absolute gamma dose. The dosimeters were annealed before each measurement in a preheated oven at 300° С for 3 hours, followed by rapid decrease to 80°C where they were kept for 16 hours. The gamma doses were measured at different positions along the beam axis and in planes parallel (Z-direction) and vertical (R-direction) to the beam axis. Each set of measurements was repeated at least three times to reduce the experi mental error. However, the experimental and statistical errors were calculated and it was found that the gamma doses measured have errors of about ±6%, while at positions of low dose intensity the error was within ± 15%. 362 EL-BAKKOUSH et al. IAEA-SM-290/27 363

3. RESULTS

The total gamma dose distribution as a function of tissue depth was deter mined in a water filled phantom. Water can be considered to be dosimetrically equivalent to soft tissue since the absorbed dose for the same radiation field is the same for both media. The data reported by Lauridsen and Jensen [23] indicate that at large depths (>50 cm) the difference between the target dose rate for water and total soft tissue as body liquid is 17% for 137Cs and 11% for 60Co. However, at small depths (below 20 cm) the difference is negligible. The distributions of total gamma ray doses in planes vertical to the beam axis and located at various depths are shown in Fig. 2. This figure presents the gamma dose measured at different positions in the water phantom irradiated with direct and B4C filtered collimated beams of 3, 5 and 10 cm diameter. The curves are plotted as gamma dose rate per hour versus tissue depth. These curves show that the intensity of the total gamma dose decreases with increasing tissue depth parallel and vertical to the beam direction. It can be seen that the dose distri bution in the phantom is influenced significantly by beam diameter. The dose rate measured at different positions within and outside the beam vicinity increa ses with increasing beam diameter. When the attenuation curves are compared it can be observed that at tissue depths within the beam vicinity the attenuation curve has an initial shoulder, while at tissue depths greater than the beam diameter the curve is an exponential function of depth. The attenuation curves also show that the removal of neutrons with energies below 10 keV by the boron carbide filter causes a noticeable decrease in the dose rate at small tissue depths. This can be attributed to the fact that the contribution of secondary gamma rays produced from the attenuation and the absorption of neutrons with energies below 10 keV is effective at small depths. The depth dose distributions in planes parallel and vertical to the beam axis were used to plot the isodose charts for planes crossed by the beam axis. These isodose charts are a prerequisite for calculating the prescribed absorbed dose to the target volume in the patient. These isodose charts are illustrated in Fig. 3. The curves mapped on the charts are for positions of equal absorbed dose expressed as dose rates (mrem/h). It can be seen that the observed dose distribution in the phantom largely depends on the beam diameter, the tissue depth and the energy range of the neutrons passing through the phantom. Therefore, it becomes of prime importance in radiotherapy to use isodose charts exactly specified for the particular radiation source, beam geometry and phantom shape. The attenuation function of a plane infinite monodirectional source is very important for accurate calculation of the radiation attenuation in a medium with no geometrical attenuation from the source-target distance. This function can be easily derived from the attenuation function of a point monodirectional source. In practice, it is difficult to obtain a powerful point monodirectional source. For this reason, a strong source is usually used, i.e. a disc monodirectional source. The attenuation function of a disc monodirectional source depends on the diameter of the beam and the information obtained can be used to derive the attenuation function of a plane infinite monodirectional source. 3 6 4 EL-BAKKOUSH et al.

a 25 N" 50 mrem/h n-lncm

100 4. \ N \ s ' / "■ *■ A' A'' 10

1000 mremjÇb \ ( i l i V /Л i i I 1 1 ХЧ M 1 U \\ 21 18 15 12 9 6 3 3 6 9 12 15 18 21 25 — 10 mrem/h D = 5 C m

/ z' 0 \ s ' 50 mrem/h N, x.

100 mrem/h \\ '\ \\ 7 ) \| 1*К i I V И i i 1 i i i v i 4 i i in 21 18 15 12 9 6 3 3 6 9 12 15 18 21 25_ D = 3 c m ——J 0 mrem/h

S N. / /■ \ \ \ \ / s ' 1 - 50 mrem/h \ \

1 iV i 1 ( ^ \ í NI i "1 i 1 ' i i 21 18 15 12 9 6 3 3 6 9 12 15 18 21 Water depth along vertical direction (cm)

FIG■ 3. Isodose charts o f total gamma rays in water phantom irradiated with collimated beams from a 252C f source. ------Direct beam; ------Boron carbide filtered beam.

The attenuation function for total gamma quanta in a water phantom irra diated by an infinite plane monodirectional californium source was derived from the measured depth dose distributions following a method used by Megahid et al. [24]. The relations between the integrated value of depth dose obtained for an infinite plane monodirectional source are presented with similar relations for disc collimated beams in Fig. 4. From the attenuation relations shown in this figure, the values of the total attenuation coefficients (ß) have been determined by least squares fit. The values obtained were used to calculate the average value of the relaxation length (X) for total gamma rays in water when the phantom is irradiated with disc collimated beams. These are: IAEA-SM-290/27 365

Water depth (cm)

FIG. 4. Attenuation of total gamma doses in water phantom irradiated by disc collimated (D.) and infinite plane (In.) monodirectional beams from a 2s2Cf source. ------Direct beam; ------Boron carbide filtered beam.

5.87 ± 0.45 cm for direct beams 5.83 ± 0.39 cm for B4C filtered beams In the case of an infinite plane monodirectional source the average values of X are: 6.71 ± 0.30 cm for direct beams 6.34 ± 0.31 cm for B4C filtered beams It is evident that the value of X derived from the attenuation relations given for a disc monodirectional source is less than the value given for an infinite plane monodirectional source. This can be attributed to the fact that the side scattered radiations have more effect at large tissue depths in the case of an infinité plane source than for disc collimated beams. It can also be seen that the values of X derived for B4 С filtered beams are slightly smaller than those derived for direct beams. This can be attributed to the fact that the contribution of secondary gamma rays resulting from the absorption of neutrons with energies below 10 keV is more pronounced at small depths than at large tissue depths. 366 EL-BAKKOUSH et al.

4. CONCLUSIONS

The results obtained from this investigation have shown that the dose rate measured at various positions within and outside the beam vicinity increases with increasing beam diameter. In addition, the accumulated gamma dose at any posi tion largely depends on the tissue depth. The attenuation curves and isodose charts show that the exclusion of neutrons with energies below 10 keV causes a decrease in the total gamma dose at small tissue depths, especially for beams of large diameters. Therefore, in teletherapy, it becomes essential for the correct application of collimated beams of a 2s2Cf source, to know not only the dose delivered to the malignant tumour, but also the accumulated dose at different locations within the phantom representing more closely that part of the human body to be exposed to californium neutrons.

ACKNOWLEDGEMENTS

The authors would like to express their gratitude to Dr. S. Adalah, Head of the Radiation and Radioisotopes Application Sector, and to Dr. M. Khalaf, Head of the Physics and Materials Science Sector, for their interest, help, encourage ment and enthusiasm throughout this study and to the technical staff of the workshop for manufacturing all the experimental facilities.

REFERENCES

[1] HALL, G.S., ROSSI, H., The potential of californium-252 in radiotherapy preclinical measurements in physics and radiotherapy, Br. J. Radiol. 48 (1975) 777. [2] MARUYAMA, Y., Rapid clearance of advanced pelvic carcinomas by low dose rate 2S3Cf neutron therapy, Radiology 133 (1979) 473. [3] “Californium-252 radiobiology and'medical applications”, Californium Utilization (Proc. Int. Symp. Brussels) AIKEN, S.C., Savannah River Lab. 1 (1980). [4] MIRCEA, N.S., LAWRENCE, H.L., MARTIN, R., Californium-252 dosimetry in phantoms of various dimensions, Radiology 137 (1980) 789. [5] HALL, E.J., “The oxygen effect of californium-252”, Some Physical, Dosimetry and Biomedical Aspects of Californium-252 (Proc. Educational Seminar Karlsruhe, 1975), IAEA, Vienna (1976) 163. [6] DJORDJEVIC, B., ANDERSON, L.L., KIM, S.H., Oxygen enhancement ratios in Hela cells irradiated with californium and radium sources, Radiology 107 (1973) 429. [7] DREW, R.M., FAIRCHILD, R.G., ATKINS, H.L., The oxygen enhancement ratio measured with Hela cells and protracted irradiation from 2S2Cf and 137Cs, Radiology 104 (1972) 409. [8] ATKINS, H.L., FAIRCHILD, R.G., ROBERTSON, J.S., Comparison of irradiation by Cf and radium on skin of swine, Radiology 96 (1970) 161. [9] WITHERS, H.R., OLIVER, G.D., GLENN, D.W., Response of mouse jejunal crypt cells to low dose-rate irradiation with californium neutrons or radium gamma rays, Radiat. Res. 48(1971)484. IAEA-SM-290/27 367

[10] BRENNER, D.J., SMITH, F.A., Dose and LET distributions due to neutrons and phantoms emitted from stopped negative pions, Phys. Med. Biol. 22 (1977) 451. [11] BODDY, K., HOLLOWAY, I., ELLIOT, A., Phys. Med. Biol. 19 (1974) 379. [12] ING, I., CROSS, W.G., Spectra and dosimetry related to neutron irradiations of the human body, Phys. Med. Biol. 20 16 (1975) 906. [13] BEWLEY, D.K., Curr. Top. Radiat. Res. 6 (1970) 294. [14] KRISHNASWAMY, V., Phys. Med. Biol. 17 (1972) 56. [15] KRISHNASWAMY, V., Phys. Med. Biol. 19 (1974) 886. [16] JONES, T.D., Health Phys. 27 (1979) 87. [17] BEWLEY, D.K., McCULLOUGH, E.C., PAGE, B.C., SAKATA, S., Phys. Med. Biol. 19 (1974) 831. [18] GREENE, D., MAJOR, D., Phys. Med. Biol. 19 (1974) 448. [19] OLDENBURG, U., BOOZ, J., Microdosimetry (Proc. 3rd Symp. Stresa, 1971), Euratom Rep. EUR-4810 d.f.e.p. (1971)511. [20] BJARNGARD, E.E., JANES, D., “Thermoluminescent dosimeters of LiF and CaF2 : Mn incorporated in Teflon”, Solid State and Chemical Radiation Dosimetry in Medicine and Biology (Proc. Symp. Vienna, 1966), IAEA, Vienna (1967) 99. [21] LANGMEAD, W.A., WALL, B.F., A TLD System Based on Lithium Borate for the Measurement of Doses to Patients Undergoing Medical Irradiation, National Radiolo gical Protection Board, Harwell (1975). [22] EL-BAKKOUSH, F.A., AKKI, T.S., EL-GHAWI, U.M., MEGAHID, R.M., “Dose attenua tion and build-up factor for gamma-rays from 132 Cs and 60Co in tissue equivalent mate rials”, Radiation Physics (Proc. 3rd Int. Symp. Ferrara, 1985), to be published. [23] LAURIDSEN, B., JENSEN, P.H., “Experimental verification of internal dosimetry calculations construction of a heterogeneous phantom based on human organs”, Radia tion Physics (Proc. 3rd Int. Symp. Ferrara, 1985), to be published. [24] MEGAHID, R.M., MAKARIOUS, A.S., EL-KOLALY, M.A., Attenuation of reactor thermal neutrons in a bulk shield of ordinary concrete, Ann. Nucl. Energy 8 (1981) 79.

IAEA-SM-290/31

SIMPLE DOSE CALCULATION METHOD

FOR BRACHYTHERAPY W ITH 192 Ir G. HORGAS, V. LOKNER, B. POKRAJAC, S. SPAVENTI Nuclear Medicine and Oncology Clinic, Clinical Hospital “Dr. M. Stojanovic”, Zagreb, Yugoslavia

Abstract

SIMPLE DOSE CALCULATION METHOD FOR BRACHYTHERAPY WITH 192Ir. The application of prepared tables for dose determination in interstitial radiotherapy with 192Ir wires (limb separation 12 mm) implanted according to the Paris geometry rules in the treatment of cancers of the lower lip, anterior tongue and floor of the mouth is described. As the number of possible radioactive source settings is limited by the small dimensions of the tumours, the most characteristic distributions were calculated and tabulated. Even in the case of imperfect implant settings or other chosen limb separations the calculation of the individual therapy dose is possible. Using the manual afterloading technique and simple tabulated dosimetry based on the Paris geometry rules, a small radiotherapy unit in a developing country can treat its patients without desirable but costly and often unavailable remote after loading devices and computers for dose calculations. The advantages and limitations of using these tables in practice are discussed.

1. INTRODUCTION

The role of 192 Ir wires and pins in short distance low dose rate radiotherapy is well established. If the application geometry is properly combined with well chosen dose prescriptions, the results are excellent both for high cure rate and good tolerance by normal tissues. In the majority of dosimetryisystems, such as the Manchester system [ 1 ], the dose delivered to the tumour tissue is specified at a given distance outside the treated volume. A new concept of source geometry and dosimetry known as the Paris system was introduced in 1966 with the aim of facilitating the work of radiotherapists dealing with interstitial therapy [2]. At that time the system was limited to coplanar patterns only but it was later extended to implants with several planes using the concept of elementary cells such as ‘equilateral triangles’ or ‘squares’ and their summits [3-5]. The difference between the systems is that the dose delivered in the Paris system is specified through an isodose surface strictly defined in relation to the minimum dose rate between a pair or a group of sources. Such isodose surfaces should closely surround the tumour or target volume. The setting of the sources must at the same time ensure homogeneity of the dose inside the implanted volume. Although dosimetry relies on computer calculations [6], manual calculations remain possible with the ‘point technique’ [7].

369 3 7 0 HORGAS et al.

This paper presents our experience with 192 Ir wires in interstitial radio therapy using the manual afterloading technique proposed by Paine [8] for treating cancers of the lower lip, anterior tongue and the floor of the mouth. Tumours at these locations usually cover only a few cubic centimetres, so only a small number of sources is required. For dosimetry we use computer prepared tables for typical wire settings. The tables are used for the reference and basal dose rate readings in the central plane assuming perfect source settings and for a number of imperfections still considered to be in conformity with the Paris geometry rules. We are aware of the fact that an ideal implant dosimetry could not be generally used for the source settings actually achieved. However, our experience is that with carefully set steel guide needles in a great number of cases the ideal or near ideal geometry could be maintained.

2. BASIC PRINCIPLES OF THE PARIS SYSTEM

For a better understanding of our tables it is necessary to give a brief summary of the essential principles of the Paris system, which are easily achieved in practice and well suited for 192Ir wires. • The arrangement of the radioactive sources must be rectilinear, parallel, equidistant and such that their centres are in the same plane perpendicular to the direction of the sources. The nominal linear activity of the sources must be uniform and identical for all sources. Dosimetric calculations are done in a central plane which is the plane at right angles to the long axes of the sources. Inside the implanted volume within the central plane there are some geometrically defined basal points representing the lowest dose rates. The target volume must be enclosed with the reference isodose surface defined as 85% of the basal dose rate [3]. The radioactive source spacing can be selected freely. The proper prediction of the treatment volume as well as the appropriate selection of active length and setting of the sources are important. This has been discussed previously [4, 5]. Clinical practice has revealed that it is difficult to ensure good parallelism for smaller spacings and, at the same time, slight errors in parallelism are less important for larger spacings. Too large a space between the sources could cause necrosis as a consequence of overdose sleeves around the sources at points which receive double or more the reference dose. Therefore we use standard thick 192 Ir wires (Amersham, 0.6 mm diameter, platinum clad wires) in the form of Pierquin’s ‘hairpins’ with 12 mm separation between the limbs as well as single pins. IAEA-SM-290/31 371

щ ш в

How good is the setting of v ire s ? (e g separation is reasonably constant) I i* u ifliimiinfi:»iimununmimuuniim...... good

Is the setting listed in the Tables ? Make a central plane (CP) Ц11И'Ж1ИН1П11!1!!1!1!1И1П1И111111Ш‘И11!Н1!1ЦШИ1!111111ЧИ!111Ж111!!'!П!1Ш11П11Ш drawing of your setting yes and determine points for ( ШШМШШШМШШМШШШМШММШшШМтшШК basal dose (BD) calculations Standard separation of wires( 12 mm) ? A...... " I l l » " " ......

ш1ШШ!111<1!1!и!11ШШШ1Ш1йШМ!И11ИМ11ШШ111Ш!М11111111111111!11111!1Ш111И1111ШШшг Ifllll!lllllllllllllllllllllll,l!lllllllllllllllllllllllllllllll.lllllllllllllllllllll!lllllllll итпншго ПТШПШХ Determine effective length of Determine effective length of wiresires j I wires and using Curves 3 add 1 идишшшшиищщшш/ contributions of all the w ires to BD point(s). If there is more thar one BD point calculate BDUbies lll!l!!llllllllll!lllll!ll!lll!lllllllllllllll!lllfllll!!lllll!llllll!llli!l!llll!flllll1líí!ri!ílIÍ^ Using effective length for a given setting find uncorrected reference dose ( RDTlb|ej )

Calculate RDtlble, I ...... "i...... i....i... What is the nominal linear activity of vires ? Determine correction factor f. la nominal linear activity (ud/ mm) 120 Ai Ci/ mm

How old ( in weeks ) are the wires? Find the correction factor f (Curve 1)

iiiiiiiiiiiiiiiiiiiiuiiiuuniiiiiHiiiiiwimiiHiiiiiiiMMiMiii.iiiii.iii'i'iHfTiiiiuBiiiiiiiiiiiiwifwwffwi Correct RD Convertí in days and if needed use Curve 2 tables in order to determine for how long (hours) the RD-RD, *f *f therapy should be prolonged due to the table« t I disintegration of Ir-192 during therapy time ...... ШЧ1|11М1111ПЧЧ11111иУ ЦЦ1ЦП1И111М111П1!П11111И11И1МПИ11ИШ1ЧЧ1111ИНИ1ЧШЧ11Н1ЧПП11П111М111111111>'111Н111И111И11Н1111ШШ/

Determine therapy time Total dose /да For a given setting of wires RD therapy time is determined

FIG. 1. The flow chart as a basic orientation for radiotherapists in dose rate calculations. 372 HORGAS et al.

Disintegration corrections 1.0 0.9 (C): 0.8 0.7 0.6 Curve 1 ; 0.5 0.4 0.3 0.2 0 1 2345678 9 1011 12 13 1415 161718 1920 Time (weeks) Duration of therapy corrections 12

S 10 о £ 8

Curve 2

'(d)- 3 4 5 6 7 8 10 Time (d) FIG. 2. (a) Source position in the floor o f the mouth controlled by lateral roentgenogram; (bj Central plane section where RD encloses target volume, A and В are BD points. Table for RD and BD dose rate reading; (cj Source decay in weeks; (d) Time to be added in hours to the total time o f implant. IAEA-SM-290/31 373

FIG. 3. Iridium ‘hairpins’ are not in ideal position according to the Paris system and individual dose calculation is necessary.

3. USE OF THE TABLES

The tables are based on dose rate calculations for 192 Ir thick wires with 4.44 MBq/mm (120 /zCi/mm) nominal linear activity, but the values could easily be adjusted for different linear activities. It was assumed that the sources are parallel with homogeneous activity distributions. Both filtration through the platinum cladding and absorption in the tissue were taken into account. For different effective wire lengths the absorbed doses in Gy/h were calculated and tabulated for various source settings. Various methods can be used to check the actual distribution of the radioactive sources of the implant in the central plane. Conventional radiography • using orthogonal views at right angles to one another or routine sagittal or frontal tomograms are the simplest methods of choice [7]. To ensure parallelism of the two adjacent ‘hairpin’ limbs steel guide needles are used. Parallelism between pairs of such needle holders as well as between single pin holders is also required if the implant is of complex form. This could easily be achieved. With the help of a simple flow chart (Fig.l) radiotherapists can use the tables for cases where ideal or near ideal settings are detected with radiographic 3 7 4 HORGAS et al.

(a) CURVE 3 .1

Effective length (cm)

(b) CURVE 3.2

Effective length (cm)

FIG. 4. BD points calculated on the basis of (a) curve 3.1 and (b) curve 3.2.

control (Fig. 2). If the distribution of the sources in the central plane section is not tabulated and a satisfactory dose distribution is still expected (Fig. 3), radiotherapists can perform individual dose calculations by the reconstruction of the central plane source distribution. Single pin dose rate curves (Fig. 4) for basal and reference doses may easily be applied for this purpose. IAEA-SM-290/31 375

It is necessary to avoid constructions of bad implants, such as obtuse triangles or rectangles, where even the most sophisticated calculation cannot bring about a satisfactory dose distribution and therefore good treatment results. The tables contain decay correction curves as well as the treatment time correction curve to permit appropriate adjustment of the dose rates resulting from disintegration of 192Ir before and during radiotherapy.

4. CONCLUSION

Brachytherapy has become a method of choice in oncology because of its well known advantages in the treatment of some tumours. The Paris system of geometry and dosimetry is a simple method utilizing an individual treatment approach and the advantages of sources like 192 Ir. Modern remote afterloading machines and computers offer complete protection of the medical staff and permit perfect source settings and dose calculations. Not being in the position to exploit the advantages of these high technology devices, we use the manual afterloading technique and our tables for simple dose calculations based on the Paris rules. This approach enables brachytherapy to be used even by a small radiotherapy unit not equipped with high technology machines and experts for dose calculations. The programme of dose calculation enables direct dose readings to be made when the control of source position shows a perfect implant in the central plane section. In cases where the central plane source setting is not perfect, the individual dose can be calculated with the help of supplementary curves. The main limitation of these tables is that the basal and reference doses can only be calculated for the central plane section of the implant.

ACKNOWLEDGEMENT

The technical assistance by Mr. G. Vranic is gratefully acknowledged.

REFERENCES

[1] QUIMBY, E.H., Am. J. Roentgenol. 57 (1947) 622. [2] PIERQUIN, B., DUTREIX, A., Br. J. Radiol. 40 (1967) 184. [3] PIERQUIN, B., DUTREIX, A., PAINE, C.H., CHASSAGNE, D., MARINELLO, G., ASH, D., Acta Radiol. Oncol. 17 (1978) 33. [4] MARINELLO, G„ DUTREIX, A., PIERQUIN, B„ CHASSAGNE, D., J. Radiol. Electrol. 59 (1978) 621. [5] DUTREIX, A., MARINELLO, G., PIERQUIN, B., CHASSAGNE, D., HOULARD, J.P., J. Radiol. (Paris) 60 (1979) 21. [6] BOISSEERIE, G., MARINELLO, G., J. Radiol. (Paris) 60 (1979) 327. [7] PIERQUIN, B„ FAYOS, J.V., Am. J. Roentgenol-. 87 (1962) 585. [8] PAINE, C.H., Clin. Radiol. 23 (1972) 263.

DISCUSSION

(Summary of discussion held on Papers IAEA-SM-290/53, 32, 27 and 31)

In response to a speaker, K.A. El-Ghamrawi (Paper IAEA-SM-290/53) pointed out that the aim of brachytherapy in stage IV carcinoma of the cervix had been haemostatic only if radical surgery had been impossible. In response to questions from several speakers, H. Svensson (Paper IAEA-SM-290/32) pointed out that the Code of Practice followed all international documents on the subject and was the most up-to-date protocol; that it gave the procedure for determining absorbed dose at one point in the water phantom and the next step was to go to the patient; that the new procedure would reduce the difference in RBE factors between gamma rays and X-rays; and that the Code of Practice did not touch upon the problem of neutron dosimetry, which never theless could be done quite accurately. Replying to questions, R.M. Megahid (Paper IAEA-SM-290/27) recalled the physical characteristics of 2s2Cf and indicated that the goal of his investigation was to model external beam therapy using 252Cf as a source of collimated radiation. According to R.M. Megahid’s opinion, the collimated beam of 252Cf radiation could be used for irradiation of skin tumours and small-depth tumours. The information obtained is also useful for the determination of RBE and OER, which depend on the share of neutrons and gamma rays which differ with depth. In the discussion, doubt was expressed of the possibility of using 2S2Cf as a source for external beam therapy as the neutron energy and the dose rate are too low. Answering questions, G. Horgas (Paper IAEA-SM-290/31 ) pointed out that the dose distribution calculation had been made by a radiotherapist or a medical physicist using tables according to the recommendations of Prof. Pierquin, in 1979; that no corrections in dose distribution had been made for bones or air cavities as they had been behind the target volume; that in the case of non-ideal setting up of 192Ir sources when, for example, the distance between the two wires was more than 12 mm, individual calculations had been made, and this could certainly be done in a hospital where medical physicists are available. A discussion was held on the question of cost effectiveness of interstitial brachytherapy using 192Ir. Different points of view were expressed by representatives of developing countries. According to G. Horgas’ opinion, the treatment was acceptable for his hospital in spite of having to purchase the 192Ir from abroad; however, according to another speaker, in spite of 192Ir being available in his country the technique was very expensive and should be applied only in limited cases. According to a third opinion, it was very difficult to make comparisons and final conclusions could not be made without consideration of all the expenses involved in the methods, including cost of premises, training of staff, equipment, and, of course, including the benefit to the patients.

377

ORGANIZATION OF RADIATION THERAPY IN DEVELOPING COUNTRIES

( P a n e l ) Chairman: M.M. Mahfouz (Egypt) Members: N.M. Bleehen (United Kingdom) R .J. Morton (United States of America) M. Nofal (IAEA) N.T. Racoveanu (WHO) T. Rebolledo (Venezuela) M. Snelling (United Kingdom) R. Walstam (Sweden) Scientific Secretary: Y. Skoropad (IAEA) IAEA-SM-290/79

RADIOTHERAPY IN DEVELOPING COUNTRIES -

CONSTRAINTS AND POSSÎBLE SOLUTIONS

N.T. RACOVEANU Consultant, World Health Organization, Geneva

Abstract

RADIOTHERAPY IN DEVELOPING COUNTRIES - CONSTRAINTS AND POSSIBLE SOLUTIONS. The major constraints of radiotherapy in developing countries are the results of the following conditions: (1) Cancer is not recognized as a health problem by the health authorities because of the high burden constituted by communicable diseases, malnutrition, etc. (2) Radiotherapy facilities and specialized personnel are not available in developing countries (40%) or are insufficient (60%) with regard to the needs. (3) Adequately designed com prehensive cancer programmes are non-existent, rendering the efforts of the existent cancer control means inefficient. The lack of cancer prevention, early detection, diagnosis, and so on, lead to the referral of late stages of various tumours for which mostly palliative treatment is possible. For improving this situation, measures to correct the above causes are necessary such as: (i) Initiation of appropriate comprehensive cancer programmes in all countries; (ii) Systematic development of radiotherapy facilities within the above programmes. Radio therapy is a cost effective cancer treatment modality in developing countries. Approximately 1400-1900 teletherapy machines and the appropriate number of specialists (radiotherapists, medical physicists, technicians) will be necessary for an acceptable coverage till the year 2000; (iii) Manual afterloading brachytherapy on a larger scale could be considered as an alternative of reducing the financial effort necessary to improve radiotherapy by at least 30%.

1. INTRODUCTION

Health care in developing countries is confronted with a number of problems for which no easy solution is available. Amongst these major problems are the control of particular parasitic diseases (malaria, schistosomiasis, onchocerciasis, etc.), the birth rate, infant mortality, malnutrition, etc. To these specific public health problems of the developing world are added those which are common in the industrialized world such as cardiovascular diseases and cancer, which are increasing continuously in a number of countries. The progress realized in the control of the most common communicable diseases, such as smallpox, diarrhoeal diseases, and all those included in the Expanded Programme of Immunization, etc., has resulted in a change in the mortality patterns in large areas of the developing world (People’s Republic of

381 382 RACOVEANU

00 -н co (N IO vq -H O со 00 ON o Ó o O n ЧО 00 \T í \ n \£> СО ro co CO 40 tu

3 a о o«

& о со l> — es O n co ON о pN 00 ш Z. 00 co _h‘ co J (N —» — Г4 н z ¡э О и о Z Г- m l> i/*> 00 Е r-~ r~ r— r- r-~ Г- r- o \ ON ON ON ON O n ON о _l ”■ P-J ТЭ 1 1 1 ! 1 1 1 W о O n CO CO m CO co co CO Tf- > NO Г~- t"- r- Г- r- r- r- Г- Г-» ON O n O n ON ON ON O n O n O n ON оы ö- О ей щ «

D Z < к Z о С! о и •а Л Xi и а и ш < сга z §• 0 & i X! w I С/3 а. 1 ™ i Q ь •* I Ö ’S с 60 о з JD с В Z 5 о О 3 о и О X Ú¿ Q¿ О W i l ¡5и ? tu Ä .N < "3 U ■§к Щ а bQ СО ^ о< a О, Щ Ъо с »4о о1н о < о £ В 3 3 3

н U < < < < < < < ш ш ш Europe Yugoslavia Slovenia — 1973-1976 242.0 173.1 IAEA-SM-290/79 383

China, some areas in Latin America, the Middle East and South East Asia). Cardio vascular diseases and cancer are in the forefront of the major causes of mortality or hospital bed occupancy for the whole or a limited segment of the population in the above mentioned regions. This trend will probably continue and expand to most or all of the developing countries in the near future, thereby rendering cancer an important public health problem for which the health services are not yet prepared. The reasons for this unpreparedness are multiple, but some of the more important ones could be identified as follows: ( 1 ) The heavy burden represented by the specific public health problems of developing countries, which at present absorb a large part of the human and financial resources available for health care; (2) The lack of recognition of cancer as a public health problem because of the unavailability of data on incidence, bed occupancy by cancer patients, or on general health expenditure for this group of diseases; (3) The non-existence of a well conceived and easily adaptable model for comprehensive cancer control - something not yet fully understood - developed and applied in most of the industrialized countries where cancer has constituted a major public health problem for a long time; (4) The heavy initial investment represented by modern radiotherapy, which is one of the major means of cancer therapy — an investment beyond the actual financial possibilities of a very large number of developing countries. This list, which is by no means comprehensive, explains why at present cancer control is included in the health programmes of only a very limited number of developing countries and radiotherapy facilities exist in only approximately 60% of the developing countries, large areas of Africa, South East Asia, etc., having no access to such facilities. In presenting this problem here, it is hoped that fruitful discussions held during the Symposium will lead to useful recommendations on how to improve the existing situation, which has grown worse during the last three to four decades in a number of areas, particularly on the African Continent.

2. CANCER AS A HEALTH PROBLEM IN THE DEVELOPING WORLD

Reliable information on cancer incidence is available only for restricted areas in the developing world. Table I shows some of the available data on cancer incidence gleaned from various countries which have offered it for compilation by IARC. Table I demonstrates the large variation in the incidence of cancer in the developing world from 75 to approximately 320 new cases per year per 100 000 population — a result of a number of factors such as the age structure 3 8 4 RACOVEANU of the population, the various environmental and socio-cultural conditions, the reliability of cancer detection, diagnosis and registration. Not only does the cancer incidence vary widely throughout the developing world but also the distribution of major cancer sites, as demonstrated in Tables II—IV. These tables demonstrate a very important fact for cancer control — a limited number of cancer sites constitute the major number of cancers prevalent in both the developing and the industrialized world. A large number of these cancers are at present controllable by either primary or secondary prevention and, therefore, a comprehensive programme of cancer control including this aspect of cancer prevention could lead to an important decrease in the number of cancer cases in the developing world if such measures are efficiently applied. Unfortunately, practical experience of such preventive measures applied on a large scale is not yet available and, therefore, it is not expected that developing countries will be able to avoid the cancer epidemics at present seen in the industrialized world. The increase in cancer incidence already observed in the developing countries and the change in the frequency of major.cancer sites as shown in Table V constitute factors which should be considered in the elaboration of an appropriate cancer control programme. Comparing frequency of cancer sites during the two periods shown in Table V, particularly in males, an increase of bronchus, larynx, pharynx and oesophagus cancer and a decrease of oral cavity, skin, gastrointestinal and bone cancer are to be seen. Most of the types of cancer which have increased are linked with smoking habits, which must have increased in the population long before the change in cancer trends became evident. Such changes are well known in other areas of the world, stressing the fact that the health services of developing countries are becoming aware of the pressing necessity of introducing com prehensive cancer programmes. Using the data on cancer incidence from Table I and those of major site distribution (Tables II—IV), it is possible to extrapolate the probable number of cancer cases occurring each year in the developing world. The result of this extrapolation is presented in Table VI. Table VI permits an evaluation of the number of new cancers in the developing world, which comes to more than 3.5 million per year, of which cancers of the cervix uteri and breast account for more than 300 000 cases each and cancers of the liver, skin and the upper part of the GI tract account for more than 200 000 cases each. A first conclusion is that most of the above types of cancers are considered as preventable today by either: ( 1 ) Primary prevention — which means preventing their occurrence by reducing the exposure to conditions which produce those cancers — smoking, solar radiation, carcinogens in food, betel and other chewing substances, etc. (2) Secondary prevention — which means preventing precancerous lesions from evolving to full malignancy by early detection of precancerous lesions as is usually the case with carcinoma cervix uteri, oral cavity, etc. IAEA-SM-290/79 385

TABLE II. MAJOR CANCER SITES (% OF ALL CANCERS) IN LATIN AMERICA AND AFRICA

Brazil Colombia Males Females Males Females

Skin 16.9 Breast 20.5 Skin 20.2 Cervix 21.8 Stomach 13.2 Skin 15.9 Stomach 18.6 Skin 16.9 Bronchus 9.3 Cervix 14.1 Prostate 7.8 Breast 12.9 GO OO Unspecified Unspecified 7.9 Bronchus 7.6 Stom ach 9.5 Prostate 5.7 Stom ach 6.2 Unspecified 5.7 Unspecified 6.6 Larynx S.O Corpus 4.1 Bladder 3.9 Ovary 4.2 Oesophagus 4.4 Colon 3.2 Lymphomas 3.4 Thyroid 2.7 Bladder 4.4 Ovary 3.0 Larynx 2.4 Gall bladder 2.5 Colon 3.4 Rectum 2.4 Myeloid 2.3 Bronchus 1.9 leukaemia

Brain 2.5 Bronchus 2.1 Brain 2.2 Colon 1.8

Cuba Senegal Males Females Males Females

Bronchus 23.3 Breast 17.5 Liver 38.8 Cervix 21.3 Prostate 11.3 Cervix 10.7 Skin 12.2 Breast 14.8 Skin 9.1 Bronchus 10.4 Lymphomas 10.6 Liver 13.5 Stomach 6.6 Skin 7.8 Unspecified 5.5 Skin 10.3 Bladder 4.8 Colon 5.9 Stom ach 4.3 Ovary 6.1 Larynx 4.6 Stom ach 4.2 Connective Unspecified 3.5 tissue 3.8 Unspecified 4.0 Corpus 3.7 Prostate 3.5 Lymphom as 2.7 Colon 3.8 Ovary 2.7 Bladder 3.5 Stom ach 2.2 Oesophagus 2.8 Rectum 2.6 Brain 3.3 Connective 2.2 tissue Pancreas 2.6 Unspecified 7.8 Hodgkins 2.2 Brain 2.1

Another conclusion is that with the exception of cancers of the stomach and liver, the most common cancers seen in developing countries, such as bronchus, cervix uteri, breast, skin, oral cavity, oesophagus, nasopharynx, larynx, prostate and bladder could be efficiently treated with radiotherapy. 386 RACOVEANU

TABLE III. MAJOR CANCER SITES (% OF ALL CANCERS) IN ASIA

China India Males Females Males Females

Stomach 23.1 Cervix 14.5 Oesophagus 9.3 Cervix 20.3 Bronchus 20.1 Stom ach 13.2 Bronchus 9.3 Breast 17.2 Liver 14.6 Breast 12.9 Unspecified 8.5 Unspecified 8.2 Oesophagus 9.7 Bronchus 11.1 Larynx 8.4 Oesophagus 7.5 Rectum 3.5 Liver 5.7 Tongue 7.4 Ovary 5.8 Bladder '2.9 Oesophagus 5.0 Stom ach 6.0 M outh 4.4 Colon 2.9 Thyroid 4.4 Hypopharynx 5.9 Stom ach 3.5 Nasopharynx 2.6 Colon 3.7 M outh 4.8 Tongue 2.7 Unspecified 2.6 Rectum 3.6 O ropharynx 3.6 Bronchus 2.7 Pancreas 1.7 Unspecified 2.8 Rectum 2.4 Colon 2.1

Pakistan Thailand Males Females Males Females

Bronchus 9.7 Breast 25.1 Oral cavity 13.6 Cervix 33.2 Oral cavity 8.6 Oral cavity 10.1 Liver 10.0 Breast 13.4 Hypopharynx 6.3 Cervix 8.5 Bronchus 7.5 Oral cavity 9.6 Skin 5.7 Ovary 4.7 Oesophagus 5.9 Skin 4.3 Larynx 5.6 Oesophagus 4.5 Nasopharynx 5.8 Leukaemia 3.3 Lymph nodes 5.4 Skin 3.4 Skin 5.6 Bronchus 2.4 GI tract 5.1 Hypopharynx 3.2 Leukaemia 5.5 Liver" 2.3 Oesophagus 4.8 Haem ato poietic 2.7 Larynx 5.0 Nasopharynx 1.7 H aem ato poietic 4.7 Unspecified 2.7 _ _ _ _

Bone 4.2 GI tract 2.6 — — — IAEA-SM-290/79 387

TABLE IV. MAJOR CANCER SITES (% OF ALL CANCERS) IN EUROPE

Hungary Poland Males Females Males Females

Bronchus 19.8 Skin 15.4 Bronchus 25.6 Breast 16.2 Stom ach 17.8 Breast 15.0 Stom ach 19.2 Cervix 14.0 Skin 11.6 Stom ach 10.4 Skin 5.9 Skin 6.1 Prostate 7.3 Cervix 7.8 Larynx 5.8 Corpus 5.6 Lip 5.1 Corpus 5.9 Rectum 4.3 Ovary 5.4 Rectum 4.5 Colon 5.2 Colon 4.0 Colon 4.9 Larynx 3.9 Bronchus 5.1 Prostate 3.4 Liver 3.8 Colon 3.3 Rectum 4.8 Unspecified 3.1 Rectum 3.7 Unspecified 3.1 Ovary 4.8 Pancreas 3.1 Bronchus 3.7 Bladder 3.0 Gall bladder 4.2 Liver 3.0 Stom ach 11.0

Romania Yugoslavia Males Females Males Females

Stomach 21.6 Breast 17.6 Bronchus 23.2 Breast 18.5 Bronchus 17.7 Cervix 16.5 Stom ach 17.6 Skin 11.5 Skin 8.0 Stom ach 14.1 Skin 8.2 Stom ach 11.4 Liver 5.8 Skin 8.1 Prostate 6.8 Cervix 8.9 Lip 5.7 Liver 4.9 Rectum 4.6 Ovary 5.7 Prostate 5.0 Bronchus 3.8 Unspecified 4.5 Corpus 5.6 Bladder 4.0 Corpus 3.7 Colon 3.2 Unspecified 5.0 Rectum 3.4 Ovary 3.6 Bladder 3.0 Rectum 4.9 Larynx 3.3 Rectum 3.5 Larynx 3.0 Colon 4.2 Unspecified 2.9 Unspecified 2.6 Oesophagus 2.4 Bronchus 3.8 388 RACOVEANU TABLE VI. CANCER INCIDENCE AND MAJOR SITE DISTRIBUTION IN DEVELOPING COUNTRIES (EXTRAPOLATED FIGURES) Ü4 Z Z, < о О 2 О *® 2 о 0 с 00 W .£ S ON о 00 о о СЛ £ и о .s § u о О чо о VO о о о о о £ <с Oí —н . ». с £; ° ° X I OS о о о о о о (N о 2 >» с I 1

о О <УЭ ° 2 ° я о g I ° (N О i/-> О О О о о о 1 г- о С/Э CQ s В .s о о о о о го 2 л 1-1 о о о С л л С а I S 1 6 6 о о о « Он о ЧО ю о щ Л < о V« о 1н V 1 ,

IAEA-SM-290/79 OJ 4t 1Л щ М о я О Р2 g (N Я X о ° -С (N I о CS о о о о о о о о о CQ -н 00 I о •g Д2 8 ° 2 о, cd ьо 3 в .*_>Св О о tí 3 о E и £ £ и I

I/“)о о - *-<00 — о — Г-о о о ° « я О о '> о i á 00 S > ° S S

12 с/5 о ^ Щ дз 40 к о « о ^ о м сч 00 о о «Д 4) О о 5 О го о о о о СЛ CQ Си ~ J3 / Ш с/з 2 е £ С 4) g g 4) w с С в з ев ч • 1ч о m £ со 3 « ~ « е сЗ о с ев О 2 2 I О ° I I I о- 3 CL* £ о и I

te ч> ,

о о со о о о o' xi со г- ю го о о о о СО Л ТО ш S М ■*-* о >» В а СЛ . ^ . J-н с С/5 о £ св 4) ев о 3 i-. 1 1 ты С"» О О rs \п о о о О wo гм г- о o' ю "О •ТЗ сл О« ев ею 3 св V а> О 1н 1 1 сч 1/1о о «JO о о о и ‘> ■<и X ев >ч с X Ul 1-. 1 1 —' О (N О » 3 ^ с I 1 2 2 1 ° ° 1 о I

о о о л 2 я) о I

J ГЧÇJ о о о о о •гом о « Jd О о ^ о --3 с о о о со н *Л) о ^ чо ° ° о о .d о о о о CQ(Л I л .. .. ^ s ° & О й о о 2 о 1м о с АЛ 3 е -4-1 Л о «Í о § § 1 ° 1 о I го I 1 о _> (N Üi ев й i

Oesophagus — 210 ООО 389 390 RACOVEANU

TABLE VII. NEW CASE LOAD PER YEAR IN SOME RADIOTHERAPY CENTRES

Country - City Number of cases treated per year in successive years

1 2 3 4 5 6 7

Burma — Rangoon 1909 1933 1659 1909 2237 - - Sudan - Khartoum 701 662 688 741 805 870 913 Thailand - Bangkok 2900 3054 3077 3159 3032 — -

3. CONSTRAINTS OF RADIOTHERAPY IN DEVELOPING COUNTRIES

Radiotherapy in developing countries at present faces a number of major constraints which could be identified as: (1) the absence of comprehensive cancer programmes which leads to: (i) inadequate cancer prevention; (ii) late detection/diagnosis of cancer cases; (iii) the major number of cancer cases diagnosed are susceptible only to palliative treatment; (2) the scarcity of radiotherapy facilities, which results in poor access to such treatment modality because of: (i) distance; (ii) lack of cancer specialists and cancer awareness of medical personnel; (iii) poor referral of patients; (3) inadequate curative results of the radiotherapy performed because of: (i) the high percentage of advanced stages amongst patients treated; (ii) overcrowdedness of existing radiotherapeutic services — alternating with underutilized services as a result of distance, poor referral, inability to pay, etc. (iii) inappropriate quality assurance of radiotherapy; (iv) inadequate records and patient follow-up. The absence of comprehensive cancer programmes is common to a large number of countries throughout the world; such programmes were started around 1975 in a limited number of industrialized countries and are extending slowly. Historically, cancer was a disease treated individually by a number of medical specialists starting with the general practitioner and ending with a surgeon, radiotherapist or, more recently, chemotherapist. The idea of a cancer team and a cancer centre or institute, where comprehensive treatment of the malignant disease could be performed, was adopted after 1950 but only with great caution. IAEA-SM-290/79 391

TABLE VIII. STAGE WHEN THE TREATMENT IS STARTED IN CARCINOMA CERVIX UTERI IN VARIOUS DEVELOPING COUNTRIES (FIGO Report 1976-78)

Per cent by stage Country I II III IV

Algeria3 18.6 63.5 12.4 4.5 Brazil 1 1 .9 -1 6 .9 2 3 .0 -3 4 .5 4 6 .6 -5 4 .3 2 .2 -1 0 .8 India 9.8 34.3 55.9 0 Mexico 26.4 49.6 12.8 11.2 Poland 1 3 .1 -2 0 .6 5 1 .5 -6 1 .8 2 0 .5 -2 5 .6 1.6—2.3 Romania 10.4 39.7 46.1 3.8 Tunisia3 17.5 27.5 38.0 17.0 Yugoslavia 1 7 .3 -4 3 .5 2 5 .5 -4 4 .2 2 4 .3 -3 7 .4 2 .6 -8 .8

3 Diagnostic and Treatment of Carcinoma of the Cervix in Developing Areas, Adam Hilger, Ltd, Bristol (1985) 354 pp.

During the last 10— 15 years many developing countries have created cancer centres or institutes which concentrate a number of possibilities for cancer diagnosis and treatment in a single institution. Not all such centres are conceived as integrated cancer treatment institutions, many being just a centralized place where radio therapy, and possibly chemotherapy, are administered. However, in a few specific cases the centres dispose not only of all diagnostic and treatment modalities but also some additional activities such as early detection, cancer registration, patient follow-up, and health education. This case is nevertheless exceptional in developing countries, where most radiotherapy departments remain isolated, very often treating only late stage or recurrent cases after surgery and other treatment modalities. Table VII demonstrates why such radiotherapy centres are at present unable to cope with the continuous increase in case load and have gradually reached a plateau in the number of new cases treated per year. Such a situation is very common in most developing countries. As a result of insufficient cancer detection/diagnosis in developing countries, a high percentage of the cases which are referred for treatment are in an advanced stage. Table VIII, which is based on figures obtained from the 1976— 78 report of the International Federation of Gynaecology and Obstetrics (FIGO), demonstrates this fact for carcinoma of the cervix uteri — a type of malignancy 392 RACOVEANU

TABLE IX. NUMBER OF PATIENTS WITH CARCINOMA CERVIX UTERI SURVIVING 5 YEARS AFTER TREATMENT FROM THE INITIAL 1000 PATIENTS AT ALL STAGES (FIGURES EXTRAPOLATED ON THE BASIS OF TABLE VIII AND SURVIVAL RATE RECORDED IN THAILAND

No. of cases surviving 5 years Country I II III IV Total

Algeria 140 311 31 3 485 India 72 167 141 0 380 Mexico 193 24 32 8 474 Romania 76 193 116 2 387 Sweden 365 248 38 4 655 Tunisia 128 134 96 11 369

which can be easily detected in the precancerous (in situ) stage and can therefore be efficiently treated by simple conization, a procedure which is cheap and has practically no complications. The lack of early detection results in late stages being referred for treatment, the cost of which increases exponentially, whilst the complications and probability of recurrence are directly proportional to the stage. For comparison, it can be noted that in Sweden the distribution of stages by per cent is: I — 41%; II — 42%, III — 11% and IV — 6%. Even with the best treatment modalities the end results are poor when the tumour stage is advanced. Therefore, the five year survival rate is low in countries where 40—60% of cases are treated in stages III and IV. The absence of reliable follow-up in most developing countries does not allow a quantified presentation of this observation. Assuming the survival rates given by P. Tepmongkol (Thailand) as valid for other developing countries, an extrapolation is possible for some countries as presented in Table VIII. Table IX was prepared with such extrapolated data adding comparable data published by some cancer institutions in the industrialized world. In preparing Table IX, the following hypothesis was used: 1000 cases of carcinoma cervix uteri are treated in each place, the stage distribution is the one given in Table VIII and the 5 year survival rate is similar to that in Thailand (I - 73.2%; II — 48.6%; III - 25.5% and IV - 6.8%) while for the control, the cancer stage distribution and survival rates are those obtained at Radiumhemmet, Stockholm. IAEA-SM-290/79 393

Table IX demonstrates that the chances of survival are much higher, 40% or more, when carcinoma of the cervix is detected at an earlier stage, assuming that the quality of the radiotherapy treatment is equal. Apart from the constraints mentioned above concerning the overcrowding of radiotherapy services and the late stage of presentation, the quality of the radiotherapy is also very important, as is well demonstrated by the Patterns of Care Study undertaken by the American College of Radiology. It is almost impossible today, when objective data on the quality of radio therapy such as rates of disease free survival, complications, recurrences, are not available, for the developing world to demonstrate the good or poor quality of the radiotherapy. I am obliged, therefore, to use an indirect indicator for this purpose — the deviation in the measurement of the dose of teletherapy machines resulting from the IAEA-WHO TLD Postal Dose Intercomparison. Results of such measurements for the period 1969— 82 are presented in Table X. This table points out a number of facts: (1) roughly only 60% of the approximately 600 radiotherapy departments compared are measuring the output of their 60Co teletherapy machines with acceptable accuracy (±5%); (2) 20% of the rest have errors of less than ±10% and the others produce errors of greater magnitude; (3) the accuracy of dose measurement is better in some areas such as Europe, whilst other regions of the world present greater discrepancies in the measured values. The Programme of Quality Assurance in Radiotherapy, which is now being developed by WHO with support from the IAEA and non-governmental organiza tions (IFOMP, ICRU, ISR, etc.), aims at improving the quality of the radiotherapy and at an adequate evaluation of the results of this treatment modality on the outcome of cancer.

4. SUGGESTIONS FOR THE IMPROVEMENT OF RADIOTHERAPY IN DEVELOPING COUNTRIES

The constraints on radiotherapy as identified in the present paper could be used in defining measures to improve the actual situation. Such measures have, of course, to be realistic and adapted to the economic, social, health and other characteristics of the developing world, otherwise the suggestions constitute an intellectual exercise. Radiotherapy must be considered as an integral part of a comprehensive cancer programme, otherwise the success of this therapeutic modality remains limited in the control of cancer. An outline of a comprehensive cancer programme and the method of its implementation in a country or an area is beyond the scope of this paper. Recommendations for such purposes are available from WHO, UICC and other sources [12— 14]. Any country which intends to introduce radiotherapy or improve the existing radiotherapy must start by defining a cancer programme with 3 9 4 RACOVEANU

cn oo ON и I VO Г- On

uo r*** Os

G\ 40 On

(N 00 On

U Ш Û I o ЧО Os

oÛ S w CU Ш X H o г 1—4 pá D Q Q tu Z < aaH O C/3 Й ¡3 СЛ td Q£¡

>< W J «c H IAEA-SM-290/79 395 all components of cancer prevention, detection, diagnosis, treatment, rehabilitation, health education, etc., and consider within this broad programme the amount of financial, manpower, educational and other resources which can be expended in the field of radiotherapy. Such an approach will avoid the errors made so often in the past when radiotherapy started in isolation was regarded just as an expensive method of palliating late stage malignancies.

4.1. Cost of radiotherapy

Radiotherapy is costly, particularly for the developing countries which have a per capita health expenditure of US $5— 10 or 100 at most. Table XI gives some figures for the cost of various teletherapy machines, as reviewed by WHO TRS 644. The table does not include brachytherapy, which is an additional treatment modality for a given number of tumours and is much less expensive, especially when manual afterloading devices are used. The average cost of a treatment using 137Cs sources and a manual afterloading system as applied in the IAEA-WHO project in Egypt could be estimated to be approximately US $30-40. The cost of radiotherapy was compared with that of an alternative treatment - chemotherapy — for which some estimates are given in Table XII. From this table it is obvious that, despite the high initial investment of teletherapy, the cost per patient is much more advantageous if 60Co sources are used, which is from a third to a quarter of the cost of chemotherapy as indicated in Table XII. Manual afterloading brachytherapy, a method which unfortunately at present is very little used in the developing world, seems to be the most cost effective radiotherapy modality for those types of malignancies where such modalities can be used efficiently. It is interesting to note that during the present Symposium a number of other sites - oral, oesophagus, etc. - where brachytherapy could be successfully used were discussed, enlarging the perspectives of this method, which represents the most rational option for improving radio therapy in developing countries.

4.2. Use of radiotherapy

Table XIII, taken from WHO TRS 644, shows the frequency of the use of radiotherapy, surgery and other treatment modalities for a number of common cancer sites. By comparing Tables VI and XIII it is possible to conclude that radiotherapy and surgery, adequately applied, could control a large proportion of the most prevalent cancers in the developing world. The liver represents the only important type which mostly needs chemotherapy — although the results are still far below expectations and surgical alternatives are available. 3 9 6 RACOVEANU

CU < Oí И X H S < W 03 < 2 oà Ш H X w Qà O tU H z w s H-*O h D O' w tUo сл W Он ¡X H OO D O 2 < >

O h O tu 00 w H < g H oo W Ш H СЛ D J O < O > o t> o

fee

Cn 1 Ci = 3.70 X lO 10 Bq. t? IAEA-SM-290/79 397

TABLE XII. COST OF CHEMOTHERAPY (US$ 1979 VALUE)

Chlormetine + vincristine + prednisone + procarbazine 6 cycles 777

Bleomycin + chlormetine + vincristine + procarbazine 6 cycles 716 Doxrubicin + bleomycin + vinblastine + procarbazine 6 cycles 2119 Vinblastine weekly/1 year 1030

The above arguments support the idea that radiotherapy should take its proper place in cancer control in the developing world as it is a method which can be applied in the most common cancers and which, from the health economic viewpoint, has a favourable cost/benefit ratio in comparison with chemotherapy. There is a lack of information on the cost of surgery, which explains why a comparison of surgery versus radiotherapy is not considered here.

4.3. Future planning for radiotherapy

In a report published in 1982 a description of the worldwide situation of radiotherapy in 1979 is given and the very poor coverage which existed in Africa, Asia, the Pacific area and part of Latin America is demonstrated. The situation has not changed essentially since then. Few radiotherapy departments have been established in the developing world since 1979; moreover, some have ceased operations because of equipment breakdown, emigration of specialists, etc. The statement that radiotherapy is unavailable for a large proportion of cancer patients in the developing world and is applied too late to the majority of those able to reach such a facility remains valid at the end of 1986. Drastic action to change this situation should be undertaken by the health authorities of developing countries, supported by the UN Agencies such as WHO, IAEA and ANDP, donor agencies and non-governmental organizations. Because of the heavy initial investment for radiotherapy, such an undertaking should be carefully planned to avoid any wastage of resources, very often seen in such cases. Table XIV presents the actual coverage of the population in developing countries with teletherapy machines, the optimal coverage and the coverage which would be acceptable to improve the present situation by the year 2000. All figures represent population in million per one machine with the exception of the last row. 398 RACOVEANU

TABLE XIII. FREQUENCY OF RADIOTHERAPY AND SURGERY

Treatm ent Cancer distribution Radio Surgery Surgery Other None therapy plus alone (%) (%) alone radio (%) (%) therapy (%)

Mouth, pharynx 59 14 23 0.4 4 Sinus, larynx 68 16 7 0.6 8 Skin, connective tissue 52 5 41 0.3 2 Breast 16 42 27 7 8 Cervix uteri 59 13 13 0.2 5 Other female genital organs 11 31 35 5 18 Bladder, kidney 38 8 46 2 18 Lymphoma, leukaemia 26 6 4 40 24 Lung, bronchi 35 4 19 22 29 Digestive tract 1 0.7 53 4 41 All cancers 21 12 33 7 26

Table XIV gives an estimate of the number of radiotherapy machines necessary to ensure an acceptable coverage by the year 2000. The total number is between 1400 and 1900 machines. With the current actual facilities for construction of 60Co teletherapy machines, and in particular production of 60Co sources, it will be impossible to provide the number of machines required. It must also be mentioned that the cost of such equipment is of the order of US $350— 475 million, to which the cost of the premises, training of personnel, dosimetric and other auxiliary equipment should be added — at least doubling the above figures. The US $700— 950 million represent a yearly investment of more than US $60— 70 million per year for the period until the year 2000, which is not feasible with the existing health budgets of the developing countries. A solution would be to alternate teletherapy with a much larger use of brachytherapy, as already started in Egypt by IAEA/WHO. One brachytherapy set with manual afterloading equipment costs approximately US $ 10 000— 12 000. It is known that brachytherapy cannot replace external beam radiotherapy, but it can take a good part of the work-load of teletherapy machines in a well planned development of radiotherapy services. IAEA-SM-290/79 3 99

4.4. Radiotherapy personnel Radiotherapy personnel constitute another major constraint on the improvement of radiotherapy in the developing countries. To train an experienced radiotherapist, able to work alone, a minimum of 6-8 years is necessary; a medical physicist needs 2-3 years of training and experience, and a radiotherapy technician at least 2 years. At the same time it must be mentioned that most of the industrialized countries have a shortage of radiotherapists, which will increase the difficulty of training the number of radiotherapists, medical physicists and other personnel required by developing countries. A solution to this difficulty is to establish training programmes for radiotherapy personnel in the developing countries, as has already been done by WHO in Colombo (Sri Lanka). It is advisable to identify as soon as possible the centres where such training programmes could be established, to prepare adequate training curricula, adapted to the specific cancer problems of the countries concerned, and start such training as soon as possible. Financial resources are necessary for such activities and efforts should be made at inter national and national levels to provide funds for these training programmes.

4.5. Quality assurance in radiotherapy Quality assurance in radiotherapy constitutes another activity which should be developed to improve radiotherapy as was considered by WHO. A questionnaire distributed by WHO in 1984 has shown that quality assurance in radiotherapy is recognized as a necessary activity by most specialists in developing countries, being applied mostly on a voluntary basis. It is difficult to find out how detailed the controls made on dosimetric pro cedures, treatment planning and machine set-up are because records of these are not usually maintained. Furthermore, there seems to be a limitation of QA to the physical aspects of treatment and not to the clinical ones or the patient record, follow-up, and results evaluation. All these aspects are considered in the WHO publication which was prepared after the Workshop on Quality Assurance in Radiotherapy, at Reisenburg, December 1984. Another Training Workshop on Quality Assurance in Radiotherapy, held in Passau in mid-October 1986, will prepare a technical manual for quality assurance in radiotherapy which, it is hoped, will help in promoting and standardizing QA practices in developing countries.

5. CONCLUSIONS

A review of the present constraints on radiotherapy in developing countries presents a number of conclusions and suggestions for improvement. ( 1 ) Developing countries are unprepared for the change taking place in the health status of their populations — the decrease in communicable diseases and the increase in cancer, cardiovascular and other chronic diseases; 400 RACOVEANU

O— v~i^ О H z o Щ Q Z I—I w Ыz K 3 u oD. < p- % Оч < o Pi I w o co я wH -I w H ВЙ Он z o H c/5 -I< w D zu ÜH W 2 Q Dz

kJ u CQ z

(2) Comprehensive cancer programmes have been elaborated and adopted only in a very small number of developing countries and, in general, health authorities in the developing world place cancer very low on their lists of priorities; (3) Radiotherapy in most developing countries is either non-existent or offers a very low level of coverage. At the same time, because of the lack of cancer prevention and early detection, radiotherapy is applied mostly to advanced tumours where only palliative results are possible; (4) To change the actual situation it is necessary that developing countries, with the support of UN Agencies, non-governmental organizations, donor agencies and other funding organizations, adopt an urgent and well planned programme aimed at: (i) introducing comprehensive cancer programmes in all developing countries which have decreased the morbidity for most common transmissible diseases to a given level; (ii) improving the coverage of the population with cancer services, as demanded by the magnitude of cancer incidence trends; (iii) establishing a coherent plan for developing radiotherapy services which includes training radiotherapy personnel at centres organized in the developing countries; purchase and installation of tele- and brachytherapy equipment on a large scale; introduction of a coherent quality assurance programme in radiotherapy.

BIBLIOGRAPHY

Diagnosis and Treatment of Carcinoma of the Cervix in Developing Areas (Proc. Int. Working Party Meeting, Thailand, 1985) (MOULD, R.F., TUNGSABUTRA, K., Eds), Adam Hilger Ltd, Bristol and Boston (1985) 354 pp.

FIGO - Annual Report on the Results of Treatment of Gynecological Cancer, Vol. 19, Statements of results obtained from 1976 to 1978 inclusive, Radiohemmet, Stockholm (1985) 285 pp.

INTERNATIONAL AGENCY FOR RESEARCH ON CANCER, Cancer Incidence in Five Continents, Vol. 4 (WATERHOUSE, J., MUIR., C., SHANMUGARATNAM, K., POWEL, J., Eds), I ARC, Lyon (1982) 812 pp.

Malignant Tumours, Report of a Multicentre Study, Pakistan Medical Research Council, Karachi (Aug. 1982) 12 pp., 26 Append.

Proceedings of the Sixth Meeting of the International Working Party on the Use of Radionuclides and Afterloading Techniques in the Treatment of Cancer of the Uterus in Developing Areas and International Conference on Gynaecological Cancer, Istanbul, 25-29 Aug. 1980, 347 pp.

Quality Assurance in Radiation Therapy - Clinical and Physical Aspects, Quality Assessment in Radiation Oncology (Proc. 1st Int. Symp. Washington, 1983), Int. J. Radiat. Oncol., Biol. Phys. 10, Suppl. 1 (1984) 202 pp. 402 RACOVEANU

RACOVEANU, N.T., “IAEA/WHO TLD dose intercomparison and the need a quality assurance programme in radiotherapy”, Computers in Radiotherapy and Oncology (MOULD, R.F., Ed.), Adam Hilger, Bristol and Boston (1984) 215. Radiotherapy Department, General Hospital, Rangoon, Cancer Registry Report 1978 (mimeo graphed) 48 pp. STJENSWARD, J., et al., Cancer control: Strategies and priorities, World Health Forum 6 (1985) 160. STJENSWARD, J., et al., National cancer control programmes and setting priorities, Cancer Detect. Prev. 9 (1986) 113. Tumour Registry, Cancer Institute of the Siriraj Hospital, Statistical Report 1975, Faculty of Medicine, Siriraj Hospital, Mahidol University (1975) 21 pp. WORLD HEALTH ORGANIZATION, Control of oral cancer in developing countries — A WHO meeting, Bull. World Health Organ. 62(b) (1984) 817. WORLD HEALTH ORGANIZATION, Optimization of Radiotherapy, Report of a WHO Meeting of Investigators, World Health Organ., Tech. Rep. Ser. 644, WHO, Geneva (1980) 89 pp. WORLD HEALTH ORGANIZATION, Quality Assurance in Radiotherapy, WHO, Geneva (in press). IAEA-SM-290/80

Invited Paper

EARLIER DETECTION OF CARCINOMA OF THE CERVIX UTERI IN SLOWLY DEVELOPING COUNTRIES

M.M. MAHFOUZ*, S. EL-HADDAD, K.A. EL-GHAMRAWI, O. ZAKI, F. HAGGAG Kasr El-Einy Centre of Radiation Oncology and Nuclear Medicine (NEMROCK), Faculty of Medicine, Cairo University, Cairo, Egypt

Abstract EARLIER DETECTION OF CARCINOMA OF THE CERVIX UTERI IN SLOWLY DEVELOP ING COUNTRIES. The concept of earlier detection of carcinoma of the cervix uteri is presented and compared with early detection and diagnosis. The experience with brachytherapy using Amersham l31Cs sources and the manual afterloading applicator by teams composed of gynaecologists, radio therapists, radiodiagnosticians and physicists is presented. The preliminary results of the intro duction of the system in 3 teaching hospitals and 12 district and general hospitals covering 16 million Egyptians, i.e. 38% of the population, in 10 out of 26 governorates in Egypt are discussed.

Carcinoma of the cervix uteri (CCU) is one of the common neoplasms in some of the slowly developing countries (SDCs). In such countries oncological health services coverage is always inadequate. Cases usually present in very advanced stages. During the past five decades the management of CCU by surgical and radiotherapeutic procedures has become established. Minor surgery, e.g. cervical coni zation, has been curative in cases of carcinoma in situ (CIS), while extended surgery is needed for the late clinical stages II, III and IV. Furthermore, teletherapy and brachytherapy (BRT) are commonly used for cure and palliation of the disease. It was a surprise to learn that in more than 50 SDCs there are no radiation therapy services, and that in 85 other countries where oncology services do exist the linkage between such services and other health care levels is extremely poor.

* Professor Emeritus of Radiation Oncology and Nuclear Medicine, Faculty of Medicine, Cairo University, Cairo, Egypt.

403 40 4 MAHFOUZ et al.

Such facts were aired by different members of the International Working Party (IWP) for the control o f CCU during their meeting in Cairo in March 1978. During that meeting, the experience in CCU management of the Cairo University Kasr El-Einy Centre of Radiation Oncology and Nuclear Medicine (NEMROCK) was presented:

(a) Since the start of radiological services in Cairo in 1923, BRThas been utilized in the treatment of CCU. The manual and automatic low dose rate after loading techniques have been in use since 1976. (b) More than 85% of cases were late in presentation — stages IIB, III and IV. All cases were referred from gynaecology out-patient clinics all over Egypt. (c) The neoplasm is the second most common tumour among Egyptian females suffering from malignant disease. (d) Manual low dose rate afterloading brachytherapy (MAL-BRT), is an effective and dose programmable technique.

The first author raised the possibility of using vaginal smears on patients of the gynaecology out-patient (GOP) clinics only, in order to increase the detection rate of early cases among patients attending district and general hospitals, instead of their use as a mass population programme. At the following IWP meeting in Mexico City in 1982, Dr. C.B.C. Taylor of the IAEA explored the possibility of extending the modern treatment of CCU to early curable cases in SDCs. Professor M. Nofal, Division of Life Sciences, IAEA, co-ordinated the efforts of crystallizing the concept with Dr. N.T. Racoveanu, Chief Medical Officer of Radiation Medicine of the WHO. A joint project was worked out and agreed upon between the Egyptian Government represented by NEMROCK and the IAEA, for the earlier detection and afterloading BRT for CCU. The project was supported by the Italian Govern ment (March 1983) and implemented by the IAEA. The basic idea of the project is to make a curative radiotherapy procedure BRT available to health care facilities as near as possible to the CCU patients. This idea is believed to represent an economically valid solution for the initiation and develop ment of some radiotherapy facilities in developing countries that will not be able to have fully established radiotherapy services in the near future. Such a strategy does not solve the management problems of late stages of that disease. However, it should allow the cure of some of the early cases and prevent their evolution into incurable cases. Furthermore, trained gynaecologists and radiologists could, thus, intercept early cases of CCU at the district hospital level; thus, reducing the sufferings of some patients who, if left untreated, would be incurable. The above described concept was discussed and endorsed by specialists attending the IWP meetings held in Cairo in 1978 and in Mexico City in 1982. It has been approved and implemented by the Government of Egypt and the IAEA. The Government of Italy offered the financial support, while WHO offered technical advice. A Technical Project Advisory Committee IAEA-SM-290/80 405

DAMIETTA

i- 0

- 100 F A Y O U M G H 2 D H

- 200

— 3 0 0 km

10 governorates out of 26

7 DH = District hospital 7 GH = General hospital 2 UH = University hospital _1_Ref. centre = NEMROCK 17

HEALTH COVERAGE = 11 X 106

FIG . 1. Cancer registry, brachytherapy and cytology netw ork in Egypt (IAEA/W H O Project). 40 6 MAHFOUZ et al.

TABLE I. COMPARISON BETWEEN EARLY (EYD) AND EARLIER DETECTION (ERD) OF CCU CONCEPTS

Item Early detection (EYD) Earlier detection (ERD)

Population scope of the Whole population Females attending GOP scieening activity clinics Scope of motivation for Whole population Members of the health cancer control and health team education Health care infrastructure Large number Small number requirements Type of cancer risk High and low cancer risk High cancer risk patients patients examined patients only Collection of vaginal lry, 2ry, and 3ry health 2ry and 3ry health care smears at care levels levels Cancer registry Nation wide registry Hospital or departmental registries Health organization and Elaborate Rather simple administration Cost per case detected High Expected to be low

(PACT) was set up to give advice and guidelines on the treatment protocol, the training syllabus and the evaluation of the project. It is chaired by Dr. M. Snelling and its members are from Italy, the United Kingdom, France, Egypt, the IAEA and WHO. The project is carried out by the staff of NEMROCK, Kasr El-Einy Centre of Radiation Oncology and Nuclear Medicine o f Cairo University. The role of a limited vaginal smear service and of a departmental cancer registry was found to be mandatory to achieve early detection of the disease among patients of the gynaecology out-patient clinics of the district and general hospitals of an area or a geographical zone. Thus, early cases of CCU and dysplasias when detected could be managed. Furthermore, the vaginal smear (VS) test has to be applied to all patients above 35 years of age attending the gynaecological out-patient clinics as an integral part of the routine clinical examination. Such a procedure depends on the successful motivation of the health team and not on the population at large. The concept of earlier detection (ERD) was, thus, developed. It addresses itself to the detection, treatment and the follow-up of dysplasia, CIS and the invasive stages of CCU among a limited group of high cancer risk females attending the gynaecology out-patient departments of district (D), general (G) and university hospitals (U) (Fig. 1). IAEA-SM-290/80 407

Peripheral activity

District (DH) 2ry activity HCL • 80-100 beds • Gynaecology and radiodiagnosis ■ Collection of vaginal smears Peripheral activity ' Peripheral cancer registry Peripheral activity ------1------• 4 0 0 0 -1 0 0 0 0 smears/a ------| I______" ______I

General (GH) NEMROCK (RC & UH> University (UH)

3ry activity HCL 3ry activity HCL 3rv activity HCL • 300-450 beds ■ Gynaecology and radiodiagnosis • 400-1700 beds • Gynaecology and • Radiotherapy .• Gynaecology and radiodiagnosis • Radiophysics radiodiagnosis • Collection of • Pathology department • Radiotherapy vaginal smears • Central cytology department • Radiophysics • Cytopathology • Central cancer registry • Pathology - cytology • Cancer registry • Training centre • Cancer surgery • 10000-15000 smears/a - Evaluation • Cancer registry * 15000-25000 smears/a

I CENTRAL ACTIVITY |

F IG .2. Institution and departm ental activities in the hospital netw ork fo r the earlier manage m ent o f CCU in Egypt (IAEA/W H O Project).

2 r y H C L 3 rV H C L

DH GH UH RC (NEMROCK)

X^X'X'XvXvX’X vX vX ’XvX’X'X'X^X'X-X'XvXvXvXvXXi Vaginal smear collection

х*х*Х‘Х ‘’***,-'-х -Х ‘Х * х * х о х * Х ‘У ‘‘*'*'-х*х*,*‘” Х*'*Х” Х ” У - ,-*— *‘**,*ч Registration ____■ ■■■■■...... ■ ...... : . ■ j

...... i ...... ; ...... •lannea v • •

FIG. 3. Earlier detection and management o f CCU in Egypt (IAEA/W HO Project). 408 MAHFOUZ et al.

CCL Central cytology laboratory MOH Ministry of health CR Central registry PCCP Peripheral cytology collecting post FU Follow-up PCL Peripheral cytology laboratory GOP Gynaecology out-patient clinics pR Peripheral registry

FIG . 4. Brachytherapy project fo r CCU vaginal sm ear and cancer registry flow in Egypt

(IAEA/W H O Project).

The concept of early detection (EYD), on the other hand, addresses itself to the detection and diagnosis of early cases of CCU in the whole population of females in the country. A comparison of the two concepts is shown in Table I. The Earlier Detection (ERD) concept depends on: — The motivation of the members of the health tèam in a hospital — An efficient cytology service network among the affiliated hospitals — An efficient linkage between the D and G hospitals’ gynaecological services with the oncological services — The availability of treatment services for early cases of CCU in the G and D hospitals — An efficient hospital or departmental cancer registry network to control the follow-up (FUP) of suspected and latent cases. IAEA-SM-290/80 409

Early detection (EYD) requires:

— A cytological services network involving all levels of health care in the country. Such a service usually requires a formidable degree of organization and administration — Public awareness of CCU amongst all females of the country. — A highly organized national cancer registry system.

Three main challenges posed by the concept of ERD of CCU are:

(1) Detection and treatment of as many as possible of the early cases before they become incurable. (2) Extension of effective low cost diagnostic and treatment services for early cases of CCU in district (D) and general (G) hospitals (Figs 2, 3). (3) Training of non-radiotherapists to conduct BRT efficiently and safely.

The Egyptian project of earlier detection of CCU addresses itself to a pilot project. It depends on the existing health care facilities of the country. In Egypt the health care delivery (HCD) system is composed of about 3250 rural health centres distributed among 26 governorates covering a population of 48 million of which 51.2% are females. There are 11 teaching hospitals (univer sity hospitals (UH)), 70 general hospitals (GH) and 168 district hospitals (DH). All these hospitals have gynaecological and radiodiagnostic services. However, radiotherapy, radiophysics and radiation oncology facilities are only available in 6 UH oncology centres. The project comprises five major activities:

(1) Registration of all female cases (above 35 years) presenting to the gynaeco logy out-patient departments of the 17 affiliated hospitals (Fig. 1). A central registry for these hospitals in NEMROCK will also include the CCU cases detected and the dysplasia cases (Fig. 4). (2) Vaginal smear activities to be performed at three levels (Fig. 4): peripheral cytology collecting posts (PCCPs), situated in the district hospitals; peri pheral cytology laboratories (PCLs), situated in the general and university hospitals; and the central cytology laboratory (CCL) of NEMROCK. (3) Training of teams from the project affiliated hospitals, composed of gynae cologists, radiologists, medical physicists and radiotherapists. Radiotherapy and medical physics services are to be extended to hospitals which have only gynaecologists and radiologists (PACT (I), IAEA/RL/99). The course offers 108 hours of theoretical and practical training over a period of three weeks. Three training courses have been completed with a total of 69 trainees: three from the Sudan, two from Kenya, two from Tanzania and two observers (one from Italy and one from India), the rest being Egyptians. 41 0 MAHFOUZ et al.

TABLE II. STANDARD 137Cs SOURCE SET OF BRT FOR CCU FOR THE AFFILIATED HOSPITALS (PERIPHERAL LEVEL)

Type of source train3 Unit cost Standard set (103 US$) Quantity Cost ( 103 US$)

Uterine Short 1.1 1 1.1 Medium 1.5 1 1.5 Long 1.9 1 1.9 Extra-long 2.3 - -

Ovoid Small 1.1 - - Medium 1.1 2 2.2 Large 1.1 - -

Tandem Medium 2.7 1 2.7 Long 3.1 - -

Total 15.9 9.4

“Standard Amersham 137Cs source trains.

FIG. 5. HCLs and management of the various clinical stages of CCU in Egypt (IAEA/WHO project. IAEA-SM-290/80 411

Details of the syllabus and an evaluation of these training programmes were presented at PACT (II) (IAEA/RL/113) and PACT (III) (IAEA/RL/121). (4) Equipment acquisition

- For brachytherapy, a standard set composed of intrauterine source trains — short, medium and long —, vaginal ovoid sources — small, medium and large — and a medium tandem, with a total cost of approximately US$ 9400 (Table II). The package also includes disposable applicators, a rectal dose dosimeter, and radiation protection equipment. — For cancer registry a data processing unit with the necessary formating facilities and programs for registration and follow-up are under procurement (Figs 2 and 4).

(5) The treatment of early CCU cases and their follow-up as well as the follow-up of the dysplasia cases (Figs 3, 5).

At present the cytology service is being set up and the equipment is being delivered to the affiliated hospitals. The clinical experience in brachytherapy in NEMROCK with this system is presented at this meeting by Professor Kamal El-Ghamrawi (Paper IAEA-SM-290/53). We are looking forward to expanding the project to other developing countries having high prevalence of CCU.

IAEA-SM-290/82

THE ESTABLISHMENT OF RADIOTHERAPY IN A DEVELOPING COUNTRY Experience from a project in Nairobi, Kenya

R. WALSTAM, J. EINHORN Department of Radiation Physics and Radiumhemmet, The Karolinska Institute, Stockholm, Sweden

Abstract

THE ESTABLISHMENT OF RADIOTHERAPY IN A DEVELOPING COUNTRY. EXPERIENCE FROM A PROJECT IN NAIROBI, KENYA, A nucleus for radiotherapy was established in Nairobi, Kenya, in 1968. The project was initiated and for some years supported by Sweden. Extensive financial support was, however, also given voluntarily by Kenyan organizations and governmental authorities. The department developed according to the needs perceived in the practical work and as expressed by the local authorities. The aim was to introduce radiotherapy, investigate its usefulness and train local staff to take over as soon as possible. The experience of the first ten years is summarized and conclusions are drawn with respect to clinical findings, educational requirements, research aspects and general problems involved in a similar project.

1. INTRODUCTION

The project was initiated as a result of an increasing interest in Burkitt’s lymphoma in the 1960s. An air-bridge had been established between Nairobi and the Department for Tumour Biology at the Karolinska Institute in Stockholm. The ENT surgeon P. Clifford in Nairobi supplied on a weekly basis deep frozen tumour specimens to Prof. G. Klein for scientific investigations. This scientific co-operation was presented at a meeting in Stockholm in 1966. Among other topics the complete lack of radiotherapy facilities in central Africa was discussed at this meeting. Through negotiations with Swedish research funds we were able to obtain financial support for a temporary project: a two year period to investi gate the possibilities of treating Burkitt’s lymphoma with orthovoltage X-rays. This proposal was presented to hospital and health authorities at an inter national scientific conference on Cancer in Africa, held in Nairobi in January 1967 [1 ]. During our discussions it was firmly stated by the health authorities in Kenya that they would prefer us to establish a nucleus for future permanent radiotherapy in the country. We referred to the recommendations of the IAEA and WHO and suggested that at least a 60 Co telegamma unit and a proper training programme would be required. Our limited funds could not cover such an advanced task.

413 41 4 WALSTAM and EINHORN

However, the discussion continued by correspondence with the local society, the Kenya Cancer Council (KCC), and the Ministry of Health (MoH) and, a few months later, an agreement was signed with the Karolinska Institute to support the establishment of a radiotherapy department at the Kenyatta National Hospital in Nairobi (KNH).

2. THE PROJECT

The aims of the project were changed to:

— introduce radiotherapy in this part of Africa — evaluate the need for radiotherapy in the East African population — educate and train the various categories of staff in a radiotherapy department — evaluate the results and give advice on possible future development.

With these aims in mind funds were made available in Sweden to supply a new 60 Co apparatus with source and the necessary auxiliary equipment and to provide the salaries for the supporting staff for two to three years. The MoH and KCC promised to support the construction of a building for the 60 Co unit and a small out-patient department and to select local staff and counterparts for training. Through a national campaign in Kenya significant contributions were offered, for instance by President Mzee Yomo Kenyatta himself and by the East African Women’s League and Rotary. Many enthusiastic colleagues at KNH and in the Medical School also supported the project in many ways. The plans for a small department to be established within the hospital premises was sent to Nairobi in the summer of 1967. The building was ready for installation of the 60 Co unit in 1968. The inauguration ceremony on 11 December 1968, the 5th anniversary of Kenya’s independence, was conducted by President Kenyatta. The agreement provided that local staff should be selected immediately and that their education and training should be carried out locally as much as posssible. It was estimated that the establishment of radiation therapy as a recognized speciality in the country would require at least 20 years.

3. EXPERIENCE

The development of the department and the experience in various aspects are discussed under separate subheadings. After an initial period of six months - when the two authors of this paper worked together at the department - many Scandinavian consultants such as radiotherapists, radiation physicists and radiographers took over the responsibilities at the department. They were usually IAEA-SM-290/82 415 employed on a yearly basis. Under the agreement with the Karolinska Institute the two authors were asked to take over the overall responsibility for the development of the department which necessitated a visit to the site once a year on average. During the initial period only those patients who, according to medical judgement, could benefit most from the therapy were accepted. Treatment was given free of charge and, occasionally, patients from various neighbouring countries were admitted.

3.1. Equipment and buildings

The main unforeseen developments in the department were based on local experience. Soon after the start of the department we discovered a need for superficial X-ray radiotherapy for skin cancer and Kaposi’s sarcoma. We also noted the existence at KNH of some nuclear medicine equipment which could easily be repaired and brought into operation. Thus an extension of the department was requested which was finalized in 1971. A 100 kV X-ray machine was purchased through KCC and, in 1971, a second hand, simple photographic isotope scanner — a gift from a hospital in Sweden — was installed and used for a preliminary country-wide thyroid study. A relatively high incidence of cancer of the cervix — also amongst fairly young women — was known to exist in Kenya. Dr. Constance W ood, in London, had previously discussed with the East African Women’s League the importance of radiotherapy for this disease and suggested the establishment of proper facilities. After an initial period when pre-operative external beam therapy was given it was decided to purchase equipment for brachytherapy and to establish a small department for these patients. The application theatre and 5 ward beds were arranged in an existing building where the necessary reinforcement of the structural shielding could be applied. In accordance with the training programme for the staff, the brachytherapy method introduced was the Stockholm method. For handling sources and applicators special equipment for 137 Cs irradiators, developed at the Royal Marsden Hospital in London, was purchased. This separate section of the radiotherapy department was ready for operation in 1974. An increasing number of patients and improvements in technology, together with the wearing out of existing equipment, made further development in recent years necessary. A separate isotope department was established with new equipment supplied largely by the IAEA and an extended radiotherapy department with a new 60 Co apparatus and a modem X-ray simulator was taken into operation in 1983. The original radiotherapy department and its irradiation facilities will, according to reports, be used for dosimetry standardization and calibration by the radiation protection authority established at the MoH. During a visit to Nairobi in early 1984 we were also informed about plans for the establishment of radiotherapy facilities in Kisumu and Mombasa in order to cover other parts of the country. 416 WALSTAM and EINHORN

3.2. Staffing

During the initial period the department was mainly staffed by consultants recruited in Scandinavia. Radiotherapy consultants were in charge of the department for almost ten years, radiation physicists for about four years, radiotherapy radio graphers for more than two years and a scientific secretary for some ten years. There was unfortunately a delay of two years in the selection of counterparts for radiotherapy and radiation physics. The recruitment of radiographers was less troublesome but, on the other hand, their employment in the radiotherapy department was generally of fairly short duration. According to the agreement education and training should be given locally as much as possible. The first radiotherapists and radiation physicists had extensive local training culminating in approximately one year of advanced training at similar departments in Sweden. As mentioned earlier, the initial research funds from Sweden could only support Scandinavian staff for two to three years. Since the prolongation of the project was mainly caused by the delay in recruitment of counterparts and by the continuous development of new, medically motivated activities of the depart ment, it was possible to obtain further support through a bilateral agreement with the Swedish International Development Authority (SIDA). In recent years, not covered by this paper, the recruitment of radiotherapists, radiation physicists and radiographers has continued. They have been exposed to different training schemes in various countries and their experience would therefore probably be of great value to the IAEA and WHO. It might also be of interest to mention in this connection that the first two physicists recruited are now serving their country in other positions: one heading the Kenyan Radiation Protection Authority, the other being senior lecturer in the Radiology Department at the Medical School.

3.3. Clinical experience

A clinical registry was established when the radiotherapy department was started. This registry has been utilized for repeated statistical studies, for instance on the incidence and age distribution for various malignancies. Interesting findings have been reported regarding the frequency of tumours seen at the department and the age distribution of patients treated. There seem to be differences, possibly of epidemiologic interest, between patients of African, Asian or European heredity. Another interesting observation is the stage distribution of patients with various carcinomas. Particularly in carcinoma of the cervix a great difference was noted as compared with the situation in Sweden [2]. Similar differences have also been reported between other developing and industrialized countries IAEA-SM-290/82 417 and this is of great importance for the choice of irradiation technique - and therefore also of the equipment for external beam and/or brachytherapy.

3.4. Research activities

The close links between several research departments in Nairobi and in Stockholm provided opportunities for carrying out a great number of research projects, both experimental and clinical. Thus the studies on Burkitt’s lymphoma could be intensified and extended to radiation response. This very fast growing tumour did not respond well to ordinary fractionation but showed dramatic response to superfractionation [3,4]. The presence of Epstein-Barr virus (EBV) was studied not only in Burkitt’s lymphoma cells, but also in others, for instance nasopharyngeal tumours, which have a high incidence in some tribes [5]. Another example is a randomized trial on the value of pre-operative radiotherapy in the early stages of carcinoma of the cervix, a study which could only be conducted as long as no brachytherapy facilities were available [6]. A great number of scientific papers were published by local and Scandinavian authors during the period, until 1980, when the project was run under the above cited agreement [3-8]. We hope that the links established will be available for future research.

4. DISCUSSION

The development from a research proposal - limited in time and resources — to a department continuously expanding over a decade was not anticipated when the project started. We were, of course, faced with many administrative and technical problems as well as with a completely new clinical situation. We had to improvise and find unconventional solutions to unexpected difficulties of many different kinds. Thanks to the never failing support from Kenyan organizations, the hospital staff and the Medical School as well as to the financial support from SIDA and Swedish research funds, it was possible to complete our task. We think that the link established between the university institutions in Kenya and Sweden and the common research interests were of considerable importance.

5. CONCLUSION

The establishment of a nucleus for radiotherapy in a developing country requires relatively long term support from both the developing country and the industrialized world. At least 10 to 20 years are usually required to build up a 418 WALSTAM and EINHORN recognized radiotherapy service. Medical students, who will be involved in the referral of patients in 30 to 50 years, should, if possible, become acquainted with this treatment modality and its advantages and restrictions. It is therefore useful to introduce this speciality at university clinics even when only a very limited number of patients can be admitted. Our experience fully supports the statements made in a WHO Report on the optimization of radiotherapy particularly for developing countries [2]:

— Cancer is an increasing health problem in all developing countries — Treatment resources are generally insufficient both in large cities and in rural areas — The frequency of young cancer patients is often higher than in industrialized countries — Different cancer forms predominate in different countries — The majority of the patients come in for treatment with very advanced tumours — Cancer patients make demands on hospital resources irrespective of whether specialized treatment resources are available or not.

REFERENCES

[1 ] Cancer in Africa, A selection of papers presented at the East African Medical Research Council Scientific Conference in Nairobi, January 1967 (CLIFFORD, P., LINSELL, C.A., TIMMS, G.L., Eds), East African Publishing House, Nairobi (1968). [2] WORLD HEALTH ORGANIZATION, Optimization of Radiotherapy, WHO Technical Report Series No. 644, WHO, Geneva (1980). [3] NORIN, T., Radiation therapy in Burkitt’s lymphoma. Long term results, Acta Radiol. 16 4(1977) 289. [4] NORIN, T., ONYANGO, J., Radiotherapy in Burkitt’s lymphoma. Conventional or super fractionated regime — Early results, Int. J. Radiat. Oncol., Biol. Phys. 2 (1977) 399. [5] LARSSON, L.-G., et al., Radiation therapy of nasopharyngeal carcinoma in East Africa, Acta Radiol. 15 (1976) 305. [6] EINHORN, J., ONYANGO, J., WALSTAM, R., “Combined therapy versus surgery in the treatment of carcinoma of the cervix uteri in East Africa”, Carcinoma of the cervix in Developing Areas, Adam Hilger Ltd, Bristol and Boston (1985). [7] EDSMYR, F., et al., Clinical efficiency of Bleomycin in oesophageal and skin carcinoma in East Africa, The East African Med. J. (Aug. 1973). [8] EDSMYR, F., LARSSON, L.-G., ONYANGO, J., WANGURU, S., WOOD, M., Radiation therapy in the treatment of keloids in East Africa, Acta Radiol. (Ther.) 13 (1974) 102. IAEA-SM-290/85

TRAINING OF RADIATION THERAPY TECHNOLOGISTS

R.J. MORTON Radiotherapy Development Branch, Radiation Research Program, National Cancer Institute, Bethesda, Maryland, United States of America

Abstract TRAINING OF RADIATION THERAPY TECHNOLOGISTS. People are the key to a quality radiotherapy programme. The most expensive and most sophisticated equipment still requires skilled and dedicated personnel to operate it safely and effectively. Since the radiation therapy technologist interacts with the patient on a daily basis and is charged with administering the radiation to the prescribed volume of the patient in the prescribed amount, his or her training is paramount to the outcome of the therapy. While one could make a case that the greater the training, the better the technologist, there are certain minimum standards below which the person should not treat patients. This minimum level is discussed. More than an opinion, the paper is a proposal to increase the number and quality of radiotherapy technologists in developing countries. The most efficient and least expensive method of providing this training is to establish a training programme within the country of need using personnel well educated in these topics. The fastest way to establish such programmes would be for the IAEA/WHO to convene a working group of experts to design the curriculum and develop packaged training and reference materials. The IAEA/WHO should consider establishing a committee to accredit these training programmes to ensure uniform minimum standards of education.

People are the key to a quality radiotherapy programme. The most expensive and most sophisticated equipment still requires skilled and dedicated personnel to operate it safely and effectively. The training and education o f all personnel providing cancer treatment are the most important factors in the many facets of patient care such as diagnosis, prescription, treatment planning, treatment delivery, and follow-up. Most often the physician and physicist are well trained and experienced. But, frequently, treatment is delivered by personnel, whom I shall refer to as ‘equipment operators’, who are undertrained for their job. The incentives to invest in this' person’s training and education are often non-existent. They are the lowest paid members of the professional staff and, therefore, are often felt to require less education. This is a false economy. The equipment operator may know little more than to align the light field with the treatment port outline and how to set the treatment equipment controls and turn it on. Since this person interacts with

419 420 MORTON the patient on a daily basis and is charged with administering the radiation to the prescribed volume of the patient in the prescribed amount, his or her training is paramount to the outcome of the. therapy. Further, he or she is the one person who has daily intimate knowledge of the performance of the treatment equipment and can most quickly recognize potential problems. A radiation therapy technologist, by my definition, possesses broader knowledge of the many facets of radiation therapy by virtue of the education he or she receives. In the United States of America, radiation therapy technologists attend a 24 month programme which includes classroom and clinical aspects. But what works in the developed countries after a long history of programme evolution, from on-the-job training to a formal accredited educational programme, is not suitable for a beginning programme in a developing country. It is more correct to look at the process that evolved in the developed countries. Early on diagnostic X-ray technologists were hired to treat patients and their education was supplemented by in-house short courses tailored to the needs of that particular clinic. As the number of qualified people increased, the technologists started taking jobs in other clinics. There they often had to be retrained to perform at the standard expected in the new clinic. This led to the need to standardize education and training so that an entry level radiation therapy technologist hired by any clinic could be expected to have a certain minimal background. The standards continued to improve and eventually the educational programmes became accredited. Accreditation ensured that all schools received periodic review, that the instructors met specified minimal standards, and that the curriculum was standardized for all schools. This is the progression I would like to suggest for developing countries, but I would encourage that we build on this experience and that the time-scale be accelerated. There must be certain minimal standards for employment as an equipment operator, and higher standards for entry level radiation therapy technologists. As experience and numbers of technicians increase in any country the standards can be raised to meet the higher quality performance expected. Eventually, accreditation of the programmes is important to establish the same standards for all programmes. This will give confidence to those hiring graduates from any programme that a student has received the minimal education felt necessary by an expert accreditation committee. How does one start to improve the standards in a particular country? First, each country needs to tailor its programme to its culture. Economic and social factors have to be taken into consideration as well as the educational level of the trainees available to enter the radiation therapy technologist education programme. Many countries have imported technologists educated in other countries; others have sent their countrymen abroad for education. These methods may sometimes be successful, but frequently problems arise. Since the cultures and economic structures are different, medical practices and standards differ. This often leads to dissatisfaction on the part of both the technologist and clinic personnel when IAEA-SM-290/85 421 a foreign trained technologist is employed. The best approach is to develop and improve educational programmes within the country. There is one other aspect that is often ignored and later reduces the quality and effectiveness of the graduate technologist. These new graduates will require good role models to put into practice what they have learned. The rest of the staff must be well qualified and motivated. Good clinicians, physicists and technologists are necessary to instil the quality of performance expected. Some one is administering radiation to your patients now and they are doing the best they can, given the training they have received. Their level of education must be increased before new students can be trained in that facility. These people may provide some of the instruction in the programme and that requires them to be knowledgeable and experienced. So, your present technologists or equipment operators are your first students. One could make a case that the greater the education, in quality and quantity, the better the technologist. While I agree, there are minimum standards of education below which the technologist should not administer radiation to patients. This is both for his or her own safety and the safety of the patient, as well as for the quality and effectiveness of treatment provided to the patient. One can think of the two major tasks of effectiveness and safety as each having a group of topics that pertain, with some overlap occurring. I propose the following topics be taught to prepare the technologist for the effective use of radiation for therapy: — Anatomy — Oncology terminology — Patient interaction and observation — Patient record maintenance — Equipment operation The safety aspect requires some additional topics: — Photon radiation biology — Photon principles and interactions — Personal radiation safety — Equipment terminology — Equipment safety — Quality assurance — Emergency procedures Notice that I refer to photon radiation biology and photon principles and interactions. This is to reduce confusion that may be introduced by discussions on alpha and beta particles, let alone neutrons. I propose that the content specifics be directed to the needs of the trainee. This keeps it relevant and shortens the training period. For the content of each of these topics I propose that the International Atomic Energy Agency (IAEA) and World Health Organization (WHO) convene a committee of experts to design the curriculum, develop teaching material and design examinations to test the students. Training 422 MORTON aids such as slides and videotapes could also be developed. Thus, a country that wanted to start a programme for radiation therapy technologists would have all the guidance material available and all countries would provide the same, or at least similar, education. As examples of material that could be developed, I offer two. First, ‘A Primer on Theory and Operation of Linear Accelerators in Radiation Therapy’, which was co-authored by Dr. C.J. Karzmark and myself for the United States Center for Devices and Radiological Health (CDRH) [1]. The Primer contains the basic terminology needed to converse on the subject and a very basic description of the operation of each component. Further, the Primer is supplemented by a three part videotape that gives graphic representation of components and shows a linear accelerator in use. The second example is ‘Quality Assurance in Radiation Therapy; A Manual for Technologists’, edited by M. Wizenberg, and published by the American College of Radiology with funding from CDRH [2]. This manual was written by a task force and panel of physicians, physicists and technologists in radiation therapy. Each chapter gives a graphic example of the component receiving quality assurance and contains sections on: — Graphic representation of satisfactory and unsatisfactory — Purpose and importance — Sources of malfunction — Spot check frequency — Spot check objective Materials and methodology — Materials and tools — Principles of spot check — Techniques of spot check — Sources of error — Records of results and actions taken — Action guidelines — Summary of learning outcome — References This manual is further supported by an Instructor’s supplement [3]. The supplement includes notes for the instructor, questions to ask the students and discussion topics. The Bibliography to this paper lists the sources for these and other useful material for programme development. Finally, I propose that the IAEA/WHO establish an international accreditation committee to set standards for educational programmes in radiation therapy technology and to review the programmes to ensure their quality. This would further reduce the burden for each country. Within these programmes the students could then be awarded diplomas which carry an IAEA/WHO seal of accreditation and would be recognized in many countries. This would allow less developed countries to raise the quality of radiation therapy by hiring these graduates and, eventually, to have their own educational programmes. IAEA-SM-290/85 423

Let us not forget that the cancer patient’s care is improved when treatment personnel qualifications are improved. Quality care requires quality people!

REFERENCES

[1] KARZMARK, C.J., MORTON, R.J., A Primer on Theory and Operation of Linear Accelerators in Radiation Therapy, United States Department of Health and Human Services, Center for Devices and Radiological Health, Rockville, MD 20857, USA (1981). [2] WIZENBERG, M.J. (Ed.), Quality Assurance in Radiation Therapy; A Manual for Technologists, American College of Radiology, Reston, VA 22091, USA (1982). [3] WIZENBERG, M.J. (Ed.), Quality Assurance in Radiation Therapy; A Manual for Technologists — Instructor’s Supplement, American College of Radiology, Reston, VA 22091, USA (1982).

BIBLIOGRAPHY

Curriculum Guide for Radiation Therapy Technology, American Society of Radiologic Technologists Educational Foundation, Inc., Albuquerque, NM 87123, USA (Latest Edition). Essentials and Guidelines of an Accredited Educational Program for the Radiation Therapy Technologist — Adopted by the American College of Radiology, American Medical Association and American Society of Radiologic Technologists (Joint Review Committee on Education in Radiologic Technology), Chicago, IL 60606, USA (Latest Edition).

IAEA-SM-290/6S

ESTADO ACTUAL DE LA RADIOTERAPIA EN VENEZUELA

N. URDANETA, R. MILLAN, A. RODRIGUEZ, A. COLINA, E. ALVAREZ, A. VERA, M. BITTAR, J. MATTOUT, M. AGUILERA, L. RUAN, R. PERDOMO, C. NIEVES, R. VERA Cátedra de Radioterapia, Facultad de Medicina, Universidad Central de Venezuela, Caracas, Venezuela

Memoria presentada por T. Rebolledo

Abstract-Resumen

PRESENT STATUS OF RADIOTHERAPY IN VENEZUELA. The purpose of the work was to discover the conditions obtaining in the practice of radio therapy in Venezuela; 15 public radiotherapy centres were visited, evaluated and information obtained on physical environment, treatment units, intracavitary radiotherapy groups, medical and para-medical personnel, recording of clinical data and number of patients. In the groups it was observed that: only one of 17 voltage stabilizing units was operating; the four linear accelerators can only be operated irregularly and there are 17 60Co units operating, but the majority of those are antiquated and worn out. One centre has a simulator and another has a computer. In intracavitary radiotherapy, approximately 50% of the centres do not carry out utero-vaginal implants in proper conditions. There are sufficient radiotherapists, but their working conditions cannot be compared with those in developed countries. There is a lack of auxiliary personnel, radiation physicists and dosimetrists. There is no training for radiotherapy technicians and there are not enough of them. In 50% of the centres, no treatment plans are made; one-third do not have any devices for immobilization. Four hospitals do not use radiotherapy histories or treatment graphics, and in four institutions no tumour record is used. The conclusion is that oncological radiotherapy is not adequate in most of the centres and it is recommended that: the damaged equipment should be repaired; the 60Co units with inadequate source-axis distances should be replaced; new sources should be acquired for the equipment with source-axis distances of 80 cm; maintenance should be constant; simulators should be provided; damaged intracavitary radiotherapy equipment and sources should be replaced; the numbers and training of technical personnel should be increased; physicists and dosimetrists should be trained; a national radiation physics centre should be set up, and the recording of clinical information should be improved.

ESTADO ACTUAL DE LA RADIOTERAPIA EN VENEZUELA. Este trabajo se realizó para conocer las condiciones actuales de la práctica de la radioterapia en Venezuela. Con este fin se visitaron 15 centros públicos de radioterapia, evaluándose y obteniendo información sobre: ambiente físico, unidades de tratamiento, grupos de radioterapia intracavitaria, personal médico y paramédico, registro de datos clínicos y número de pacientes. En los grupos se observó que: sólo una de 13 unidades de ortovoltaje está en funcionamiento; la operabiUdad de los 4 aceleradores lineales es irregular; existen 17 unidades de 60Co en funcionamiento, pero la mayoría anticuadas y deterioradas. Un centro posee simulador y otro

425 426 URDANETA et al. un computador. En radioterapia intracavitaria, alrededor del 50% de los centros no realiza el implante útero vaginal en condiciones adecuadas. Existe un número apropiado de radioterapeutas, pero sus condiciones de trabajo no pueden compararse con las de países desarrollados. Hay falta de personal auxiliar, de físicos de radiaciones y de dosimetristas. No existe formación de técnicos de radioterapia y su número es insuficiente. En el 50% de los centros no se realizan planes de tratamiento; en un tercio no se hacen radiografías de localización, y sólo dos centros fabrican dispositivos para inmovilización. Cuatro hospitales no emplean historia de radioterapia ni gráficos de tratamiento; en 4 instituciones no se emplea registro de tumores. En conclusión, la radioterapia oncológica no se practica de manera adecuada en la mayoría de los centros, recomendándose: la reparación de equipos dañados; la sustitución de las unidades de cobalto 60 con distancias fuente-eje inadecuadas; la adquisición de nuevas fuentes para los equipos con distancias fuente-eje de 80 cm; un mantenimiento constante; la dotación de simuladores; el reemplazo de equipos y fuentes de radioterapia intracavitaria deteriorados; el incremento y entrenamientos del personal técnico; la formación de físicos y dosimetristas; la creación de un centro nacional de física de radiaciones; y el mejoramiento del registro de información clínica.

1. INTRODUCCION

Para el año 1986 deben presentarse en Venezuela alrededor de 18 000 nuevos casos de cáncer [1]. Según el anuario de estadística vital del año 1982 [2], la mortalidad por cáncer fue de 7600 casos. Esta enfermedad constituye la tercera causa de muerte y la segunda por enfermedades. Según cifras de países desarrollados, un 50-60% de cancerosos requiere irradiación en algún momento de su evolución [3]. En Venezuela, los cánceres del cuello uterino, pulmón, próstata y mama son muy frecuentes y muchos se presentan en etapa avanzada, de allí que la necesidad de radioterapia curativa o paliativa en Venezuela es igual si no superior a la de países desarrollados. Un mejor conocimiento de la historia natural de la enfermedad, así como el trabajo multidisciplinario, los avances técnicos, la disposición de equipos modernos y la preparación del personal médico y paramédico en radioterapia han permitido que las radiaciones ionizantes hayan contribuido a mejorar las cifras de curación de muchos tumores.

2. MATERIAL Y METODOS

Este trabajo se realizó por medio de visitas a los servicios de radioterapia de 15 hospitales (Fig. 1 ), intercambiándose ideas con los radioterapeutas y consta tándose el estado del ambiente físico, unidades de tratamiento y aplicadores intracavitarios. Igualmente se obtuvo información sobre personal médico y paramédico, registro de datos clínicos (historia, registro de tumores, seguimiento) y número de pacientes tratados. IAEA-SM-290/65 427

К ONCOLOGICO PADRE MACHADO H. ONCOLOGICO LUIS RAZETTI H. VARGAS DE CA R A CA S H.CENTRAL DE LAS FF.AA.

FIG. 1. Centros de radioterapia visitados.

2.1. Ambiente físico

El Cuadro I muestra los datos de las condiciones del ambiente físico. La evaluación de los locales se basó en lo siguiente: acceso al público, área, venti lación e iluminación de ambientes de espera, consulta y tratamiento, oficinas médicas y salas de reuniones; igualmente, se tuvo en cuenta la existencia de espacios para el revelador y la aplicación de anestesia para niños. Se clasificaron los ambientes como adecuados, inadecuados o inexistentes. 428 URDANETA et al.

CUADRO I. EVALUACION DEL AMBIENTE FISICO

Espacio Adecuado Inadecuado No existente

Sala de espera 10 5 Consulta 11 4 Ambiente, equipos de radioterapia 14 1 Espacio para oficinas 12 2 1 Espacio para reuniones 9 6 Revelador 3 1 11 Aplicación de anestesia 5 10

CUADRO II. EVALUACION DE LAS UNIDADES DE RADIOTERAPIA

Tipo de unidad № de equipos Funcionamiento Sí No

Ortovoltaje 13 1 12 Cobalto 60* 17 17 Acelerador lineal 4 3 1 Simulador 1 1 Computador 1 1

* 1 centro tiene una fuente de cesio.

En la mayoría de las instituciones, el ambiente físico posee los requerimientos mínimos para prestar buen servicio. Alrededor del 50% de los servicios no posee espacio para reuniones. Cuatro hospitales poseen espacio para revelador propio y cinco para anestesia, lo que dificulta la práctica de ciertas actividades clínico- docentes.

2.2. Unidades de radioterapia (Cuadro II)

Existen 13 unidades de ortovoltaje, de las cuales solo funciona una. Hay 4 aceleradores lineales (4 a 6 MeV) cuya operabilidad es irregular; 17 unidades de 60Co están en funcionamiento, pero cuando se analiza su estado (Cuadro III), IAEA-SM-290/6S 429

CUADRO III. ESTADO DE LAS UNIDADES DE COBALTOTERAPIA

Factor evaluado Adecuado Inadecuado

Años en funcionamiento 3 14 (<10 años) Rendimiento 1 16 « 1 0 0 R/min) Distancia fuente-piel 5 12 (<80 cm) Tamaño del campo máximo en isocentro 5 12 (= >25 cm) (<25 cm) Capacidad de rotación 13 4 (Sí - No) Mesa de tratamiento 11 6 Mantenimiento 7 10 (Sí - No) Calibración del equipo 8 9 (<1 año)

CUADRO IV. EVALUACION DE LOS IMPLANTES UTERO-VAGIN A LES

Condiciones del implante Adecuado Inadecuado

Anestesia general o raquídea 13 2 (Sí - No) Rx de localización 8 7 (Sí - No) Tiempo de adquisición de las fuentes 3 12 (%10 años) Revisión de las cargas de cesio 137 5 10 (Sí - No) Estado de los equipos de Henschke 6 9 430 URDANETA et al. teniendo en cuenta los patrones recomendados [4], la mayoría no reúne las condiciones mínimas aceptables para practicar radioterapia curativa —casi todas están anticuadas, deterioradas y con insuficiente mantenimiento o calibraciones periódicas. Una institución posee un simulador y otra un computador para planes de tratamiento y dosimetría (Cuadro II).

2.3. Implantes utero-vaginales

Desde 1973 existe en Venezuela un Plan Nacional de tratamiento del cáncer del cervix (cobaltoterapia externa y cesioterapia intracavitaria con aplicadores de Henschke) [4]. Por la importancia de esta patología en la mujer venezolana, interesa analizar la calidad de implantes utero-vaginales y condiciones de equipos y fuentes radiactivas. El Cuadro IV muestra que en 2 centros no realizan implantes con anestesia adecuada y en 7 no realizan placas de localización para verificar la colocación correcta del equipo. En la mayoría de las instituciones el estado de los aplicadores y de las cargas de cesio 137 no es el más apropiado.

2.4. Personal médico

Se presenta una relación entre el número de radioterapeutas y el número aproximado de pacientes nuevos vistos anualmente. Vale destacar que existen 3 centros en Caracas donde se tratan más de 600 pacientes anualmente (Cuadro V); en el interior del país se cuenta con 3 servicios que tratan anualmente 400 pacientes (Cuadro VI). Existen 4 hospitales donde sólo se cuenta con un especialista. Al analizar la relación médicos-paciente por año, parece haber una distribución acorde, y hasta por encima de los requerimientos sugeridos por expertos [5], quienes recomiendan un radioterapeuta por cada 200-250 pacientes. Sin embargo, en Venezuela esta relación no es válida por completo porque el personal médico no trabaja a tiempo completo; en la mayoría de los servicios no se dispone de personal paramédico suficiente y bien entrenado.

2.5. Personal paramédico

Para analizar los recursos humanos disponibles, se establecieron dos grupos, uno referente a los técnicos radiólogos y enfermeras y otro al resto del personal (Cuadros VII y VIII). En cuanto al primero, el número de técnicos radiólogos en más del 50% de los servicios es insuficiente. En Venezuela no existe una escuela para formación de técnicos en radioterapia, por tanto, este personal integrado por técnicos en IAEA-SM-290/6S 431

CUADRO V. RELACION ENTRE RADIOTERAPEUTAS Y NUMERO DE PACIENTES

o o N" de radioterapeutas N~ de pacientes Institución oncólogos nuevos por año (medio tiempo) (aproximadamente)

Caracas: Instituto de Oncología Luis Razetti 6 800 Hospital Oncológico Padre Machado 6 800 Hospital Vargas 3 360 Hospital Central de las Fuerzas Armadas 3 220 Hospital Universitario de Caracas 6 700

CUADRO VI. RELACION ENTRE PERSONAL MEDICO Y NUMERO DE PACIENTES

N2 de radioterapeutas N° de pacientes Lugar oncólogos nuevos por año (medio tiempo) (aproximadamente)

Caracas 24 2880 Maracay 1 160 Valencia 2 500 Barquisimeto 3 400 Coro 1 150 Maracaibo 2 400 Mérida 2 180 San Cristóbal 3 170 Barcelona 1 240 Cumana 1 100 Ciudad Bolívar 2 300 TOTAL 42 5480 432 URDANETA et al.

CUADRO VII. PERSONAL PARAMEDICO EN RELACION AL NUMERO DE PACIENTES

Relación Personal Adecuada Inadecuada

Técnicos radiólogos3 6 9 Enfermeras11 13 2

a Relación adecuada: 1 X 100 pacientes nuevos por año. b Relación adecuada: 1 X 200 pacientes nuevos por año.

CUADRO VIII. EVALUACION DEL PERSONAL PARAMEDICO

Personal Existente No existente

Físico de radiaciones 4 11 Dosimetristas 6 9 Auxiliar técnico 3 12 Trabajadoras sociales 10 5

CUADRO IX. EVALUACION DE LA CALIDAD DEL TRATAMIENTO

Condición del tratamiento Presente Ausente

Planes de tratamiento y dosimetría 8 7 Radiografías de localización 10 5 Dispositivos individuales y moldes 2 13

CUADRO X. ANALISIS DEL REGISTRO DE INFORMACION CLINICA

Existente Inexistente

Historia de radioterapia 10 5 Gráficos 11 4 Registro de tumores 11 4 IAEA-SM-290/6S 433 radiodiagnóstico no está bien capacitado ni puede ofrecerle al médico toda la ayuda necesaria. En cambio, el número de enfermeras si resultó adecuado. En lo referente al personal restante, existe carencia de físicos, dosimetristas y auxiliares técnicos en la mayoría de los centros. Afortunadamente, en el campo del servicio social existe una buena cobertura en la mayoría de las instituciones.

2.6. Calidad del tratamiento radiante

Uno de los aspectos más importantes del presente estudio fué la evaluación de la práctica de la radioterapia. Para esto se tomaron en cuenta tres parámetros (véase el Cuadro IX). Se constató que en aproximadamente la mitad de los servicios no se realizan planes de tratamientos, que las radiografías de localización no se practican en un tercio, y que sólo dos de los centros fabrican dispositivos y moldes para inmovilización.

2.7. Registro de información clínica

Para evaluar el funcionamiento de todo servicio hospitalario se recabaron datos acerca de: historia propia de radioterapia, utilización de gráficos y existencia de un registro de tumores en el servicio o en el hospital (Cuadro X). Existen servicios de radioterapia donde no se realiza la historia clínica separadamente, ni se emplean gráficos. Cuatro hospitales no poseen registro de tumores, ni en la institución ni en el servicio de radioterapia.

2.8. Unidades de larga estancia

De los 15 centros, 3 disponen de un albergue para alojar a los pacientes de áreas remotas.

3. CONCLUSIONES Y RECOMENDACIONES

Al analizar los datos podemos concluir que la radioterapia no se practica de una manera adecuada en Venezuela por falta de recursos humanos y materiales y debido a que no se ha dado a esta especialidad toda la importancia que merece. Creemos que para cambiar este panorama poco halagador, deben tomarse a la brevedad varias medidas para mejorar la calidad del ejercicio de esta especialidad y extender esta terapéutica a toda la población que la requiera.

3.1. Equipos de radioterapia

Las recomendaciones que se sugieren en este terreno son las siguientes:

1 ) La reparación de los aparatos de radioterapia convencional y la adquisición de este tipo de equipos para los centros que no los poseen. 43 4 URDANETA et al.

2) Creemos que el equipo ideal para Venezuela sigue siendo la bomba de cobalto 60, siempre y cuando ésta reúna las condiciones señaladas. Por tanto: a) Deben sustituirse las unidades que posean una distancia fuente-eje menor de 80 cm , o que no tengan capacidad de rotación. No creemos conveniente cambiar solo la fuente de estos equipos, pues esto sería una solución parcial del problema. De acuerdo a expertos de países desarrollados, los servicios que posean solo unidades de cobalto 60 que no reúnan las características mencionadas deben practicar radioterapia paliativa o, de lo contrario, deben ser clausurados sino mejoran sus equipos [6]; b) A los aparatos de cobalto 60 con una distancia fuente-eje de 80 cm, con capacidad de rotación, debe instalárseles una nueva fuente radiactiva que sobrepase las 100 cGy/min. Esta última consideración debe ser tomada en cuenta para los equipos a adquirir. Todas estas unidades deben disponer de los accesorios mínimos, como bandejas, plomos protectores, filtros en cuña, curvas de isodosis, etc. ; c) Seria recomendable la compra de estas unidades a un solo proveedor de reconocido prestigio y confiabilidad, el cual debería asegurar un lapso de garantía, una existencia de repuestos y un servicio de mantenimiento preventivo; d) Deben realizarse visitas y calibraciones periódicas por un físico para constatar el funcionamiento de los equipos. 3) Asegurar a los aceleradores lineales existentes un mantenimiento constante que garantice su óptima operabilidad. 4) Dotar de simuladores a todos los serivicios. 5) Sustituir los aplicadores de Henschke que estén en malas condiciones. Actual mente se construyen en Venezuela equipos similares, menos costosos y de buena calidad [7]. Las cargas de cesío [7] deben renovarse para mejorar la calidad del implante en cuanto al tiempo de duración, y ser sometidas a revisiones periódicas. 6) Desarrollar la radioterapia intersticial limitándola a los centros de mayor afluencia de pacientes pues, debido al costo del material radiactivo (iridio 192, cesio 137) y a la mayor complejidad de esta terapéutica, no creemos posible realizarla en todos los servicios.

3.2. Recursos humanos

En este campo se sugiere lo siguiente:

1 ) Estimular al personal médico existente en los diversos servicios de radio terapia, mediante cursos periódicos de actualización y pasantías en las instituciones donde por el volumen de trabajo exista mayor experiencia en el manejo de determinada neoplasia. Igualmente, visitar estos centros para IAEA-SM-290/65 435

intercambiar opiniones, definir conductas terapéuticas en casos difíciles y consolidar el trabajo en equipo. Es fundamental que en los servicios donde solo exista un radioterapeuta se creen nuevos cargos para disponer de al menos dos especialistas, facilitando así las labores asistenciales y fomentando los hábitos de superación científica. Debe mantenerse la formación de especialistas en esta materia, la cual se realiza en la Universidad Central de Venezuela, así como estimular la formación de otros especialistas en oncología, tales como oncólogos médicos, cirujanos oncólogos, etc., lo cual traerá como consecuencia el mejor manejo del paciente por parte de equipos multidisciplinarios. 2) Prestar al personal técnico adecuada atención, mejorando su formación mediante cursos de especialización en técnica de radioterapia a nivel superior. Por otra parte, un buen número de servicios no posee un número adecuado de este personal, situación que debe corregirse a la brevedad. 3) En vista de que más del 50% de los centros no cuenta con dosimetristas, es necesaria su formación, para lo cual podrían recibir un entrenamiento combinado con la carrera de técnicos en radioterapia, y desempeñar ambas funciones en los hospitales que no tengan un número elevado de pacientes. 4) Es conveniente contar con auxiliares técnicos para la realización de moldes y dispositivos de inmovilización individuales. Sólo 3 hospitales disponen de este personal. 5) Es indispensable contar con físicos de radiaciones en los servicios que traten alrededor de 400 pacientes por año [5]. Para los otros hospitales debe disponerse de este personal en calidad de consultante. Así, por medio de visitas periódicas, puede asegurarse que la calibración de los equipos es correcta, así como la supervisión y orientación de los dosimetristas. 6) Se plantea la creación de un centro de física de radiaciones, que debe estar dotado de un computador para planes de tratamiento, de equipos de calibración y dosimetría y de personal apropiado, adscrito a algún hospital oncológico para prestar servicios a todas las unidades de radioterapia y procesar todos los planes complejos. Para esto, los radioterapeutas y el dosimetrista de cada servicio enviarían por correo rápido el contorno anatómico del área a tratar y los datos físicos necesarios para la elaboración del plan, el cual sería devuelto por el mismo medio al sitio de origen.

3.3. Registro de información clínica

Para hacer una evaluación del funcionamiento de cualquier servicio clínico y conocer los resultados obtenidos con determinada terapéutica, es fundamental contar con un buen sistema de historias clínicas que contenga toda la información pertinente y gráficos descriptivos de las lesiones y de los campos de irradiación. Además, es indispensable contar con un mecanismo de seguimiento eficaz del 436 URDANETA et al. paciente, manejado por el servicio social y coordinado por el radioterapeuta. A todo esto debe agregársele un buen registro de tumores, para conocer la frecuencia de los diferentes tipos de neoplasias y planificar una conducta coherente destinada a mejorar la atención médica y los resultados terapéuticos. Aunque en las dos terceras partes de los servicios se cumplen los requisitos de historia de radioterapia, existen algunos centros que ni siquiera llenan estas condiciones, ni poseen un registro de tumores. Consideramos indispensable que se implementen estos sistemas de recolección de información clínica, para lo cual se requiere un esfuerzo conjunto del radioterapeuta del centro y las autoridades competentes.

3.4. Ambiente físico

A pesar de que buena parte de los centros visitados poseen ambientes de trabajo adecuados, la dotación en cuanto a material clínico, examen físico, reveladores, etc., debe mejorarse.

3.5. Unidades de larga estancia

Un problema confrontado en Venezuela es la falta de alojamiento apropiado para los pacientes que provienen de áreas remotas. En ocasiones, el enfermo no asiste regularmente al tratamiento o lo abandona por esta causa. En otras, ocupa innecesariamente una cama de hospitalización por el tiempo que dura la irradiación. Por estas razones han surgido los albergues, que han cumplido una labor importante en las 3 instituciones donde funcionan. Creemos que deben extenderse a la gran mayoría de los centros para asegurar a los pacientes un ambiente humano y agradable durante el tratamiento radiante.

3.6. Necesidades de radioterapia en Venezuela

Según cifras del Comité de Estudios de Radioterapia Oncológica de los Estados Unidos [5], una unidad de supervoltaje debe tratar al menos 300 pacientes nuevos por año; esta cifra podría llevarse hasta 400 pacientes si se utiliza el equipo a tiempo completo. En Venezuela, según cifras de la División de Docencia de la Dirección de Oncología, la tasa de incidencia estimada es de 98 cánceres por 100 000 habitantes [1 ]. Ahora bien, de acuerdo con los datos de la Oficina Central de Estadística e Informática, el censo de población para 1985 es de 17 316 741 habitantes [8]. Teniendo en cuenta la tasa de incidencia, obtendríamos un estimado de 16 970 casos en 1985, cifra comparable a la estimada por Capote [1 ]. Se calcula que la radioterapia debe aplicarse al 60% de todos los pacientes con cáncer; por tanto, 10 182 pacientes requerirían radioterapia anualmente. En los centros de IÂEA-SM-290/65 437 radioterapia se tratan 5480 pacientes al año; existe un déficit de 4700 pacientes que no reciben radioterapia curativa o paliativa. Con la rotación de recursos humanos y equipos adecuados y modernos en los hospitales puede brindarse a la población venezolana afectada de cáncer una atención más apropiada, en cuanto a radioterapia. Este análisis del estado actual de la radioterapia en Venezuela es básico para concientizar al grupo humano que labora en este campo y a las autoridades del área sanitaria con el fin de elevar la calidad de la radioterapia oncológica. De otra manera, todas estas consideraciones se quedarán, como en muchas otras ocasiones, limitadas al papel.

REFERENCIAS

[1] CAPOTE, L., comunicación personal, agosto de 1985. [2] MINISTERIO DE SANIDAD Y ASISTENCIA SOCIAL, Anuario de Epidemiología y Estadística Vital, Año 1982, Caracas (1984). [3] BRADY, L., et al., Radiation oncology programs for the present and future, Cancer 55 (1985)2037-2050. [4] COMITTEE FOR RADIATION ONCOLOGY STUDIES, Criteria for Radiation Oncology in Multidisciplinary Cancer Management, Report to the Director of the National Cancer Institute, National Institute of Health (Feb. 1981). [5] VERA, R., Tratamiento radiante del cáncer de cuello uterino, Plan Nacional, Acta Oncol. Venez. (1978) 51-86. [6] HANKS, G., DIAMOND, J., KRAMER, S., The need for complex technology in radiation oncology. Correlation of facility characteristics and structure with outcome, Cancer 55 (1985) 2198-2201. [7] RIVAS, de L., comunicación personal, agosto 1985. [8] SILVA, de N., Oficina central de Estadística e Informática, comunicación personal, 1985.

PANEL DISCUSSION

(Summary of discussion held on Papers IAEA-SM-290/79, 80 and 65)

Opening the discussion on the Panel Session, M.M. Mahfouz expressed his expectation that most participants and observers at the Symposium, both from industrialized and developing countries, would co-operate as has been exercised in the past, to look into, analyse and propose actions for the development of radiotherapy in the less developed areas of the world. Paper IAEA-SM-290/79 raised several important problems of implementing radiotherapy in developing countries which were discussed and commented. It was stressed that the concept of oncology was multidisciplinary and radiotherapy as a treatment modality of cancer should be regarded as part of the concept together with surgery and chemotherapy; that it was very useful to realize from the presentation the needs of developing countries up to the year 2000 regarding personnel, equipment, and the cost of the package of the equip ment; and that the main problem was to decide on how so many 60Co units could be procured and proper training be done; that close co-operation of specialists from industrialized and developing countries as well as assistance from international bodies, national voluntary organizations and industrialized countries might help developing countries to obtain more information about their problems and to assist in solving them. Different points of view were expressed in the discussion regarding the needs of developing countries in types of facilities for external radiotherapy. A speaker, referring to the higher dose rate and better characteristics of photon beams of linear accelerators compared with 60Co units, argued for supplying accelerators to large centres in developing countries, where they should be the main instruments for external high energy radiotherapy. Another participant, referring to at least 12 years’ experience in the operation of 60Co units and linear accelerators in a developing country, pointed out that because 60Co units needed less maintenance and service they should be ideal sources of high energy radiation in developing countries, which faced serious difficulty in providing maintenance and service. N.T. Racoveanu noted in his comment that supply of linear accelerators to developing countries should be well argumented and personnel should be trained to operate and maintain them properly as had already been done with such sophisticated diagnostic machines as CT-scanners, ultrasound machines, etc. Replying to a question, M. Nofal pointed out that the IAEA paid great attention to the problem of training, that training courses and workshops were held in Japan, Malaysia and Egypt and, in addition, on-the-job training was provided through fellowships and by supplying experts to many institutions in the developing world, and that the Agency was willing to co-operate with any specialized scientific body within the regulations and resources available.

439 440 PANEL DISCUSSION

In response to a question, M.M. Mahfouz (Paper IAEA-SM-290/80) stressed that the Egyptian project aimed at salvaging early cases of carcinoma of the cervix and preventing them from progressing to incurable stages by means of early detection and applying simple radiotherapy techniques in district hospitals as near as possible to the patients’ homes, this being the corner-stone of the philosophy of the project. In response to a comment, M.M. Mahfouz pointed out that the Egyptian project had been designed for carcinoma of the cervix of stage Is and IA, where the anatomy of the uterus was not yet distorted by the extension of the disease; therefore a standard application was justified. Dose calculations and direct measurements in the rectum had confirmed that. An anaesthetist was involved in the examination of the patients required for clinical staging as well as for the dilation of the cervix and the curettage biopsy specimens. If the gynaecologist wanted to do a panhysterectomy with or without lymph adenectomy and had appropriate facilities including those needed for blood transfusion, he might be welcome to do that remembering that such extended surgery had its own, relatively higher mortality than low dose rate brachytherapy as well as that the patient stayed in hospital 14—21 days. A speaker pointed out that there was different representation of carcinoma of the cervix of stage I in NEMROCK (about 1% before the project started) and National Cancer Institute (NCI) of Egypt (about 17%), which suggested that the NEMROCK figure was not representative. Answering the comments, M.M. Mahfouz replied that the NCI clinical material having a high percentage of stage I cases proved the philosophy of the IAEA/WHO project, as cases had been referred to the NCI from many hospitals (general and district) after a very active motivation programme launched by the NCI while the NEMROCK clinical material represented the state of affairs without any motivation exercise. In response to a comment on the important role of public awareness of the cancer problem in the successful promotion of cancer care, M.M. Mahfouz pointed out that he supported the idea of awareness within the health team, but not public awareness because that was a form of political pressure on the decision maker, which was not always acceptable. Answering a comment, M.M. Mahfouz mentioned that high dose rate afterloading equipment was rather expensive compared with low dose rate equipment, that high dose rate brachytherapy could only be performed in big radiotherapy centres where there were appropriate facilities and experienced staff, while low dose rate brachytherapy could be done in district hospitals. A participant commented that the implementation of the Egyptian project was very useful for the Sudan too because it permitted a treatment team to be trained at the Cairo training course, low dose rate brachytherapy equipment to be obtained and treatment of the patients to be started, but the training did seem rather long for qualified radiotherapists and medical physicists. PANEL DISCUSSION 441

In reply to the comment, M.M. Mahfouz agreed with the above opinion and informed the audience that a special programme would be set up for these two types of specialists in future training courses. Commenting on the Egyptian project, a participant mentioned that at first she had been rather sceptical about allowing radioactive sources outside the radiotherapy department, something different from what one had been taught before in training. However, understanding that it meant that cases of carcinoma of the cervix would be discovered early and treated more successfully and also that there would be a reduction in advanced cases flooding radiotherapy depart ments, she believed the project was an excellent idea. However, when the treat ment was repeated it should be always done under the supervision of a large, well equipped and staffed centre like NEMROCK, which should remain in touch with patients and trainees. Commenting on Paper IAEA-SM-290/65, M.M. Mahfouz pointed out that it showed a very good example of how to conduct surveillance on the state of the art in a particular speciality; that it would be a very good exercise and example for every developing country because it was very important to evaluate an existing situation before looking at what might be required in the future and, in the meantime, to undertake the necessary actions.

SUMMARY OF PANEL DISCUSSION

The organization of radiotherapy in developing countries should be regarded as an integral part of a multidisciplinary approach to cancer control within the framework of oncology and closely linked with health services. Radiotherapy always called for a lot of resources (money for equipment and training and organization), but requests for those were always tacked on to the end of the health requirements of any hospital or any health care service. Therefore, developing countries would have to call for external assistance, which might or might not be provided, or would have to generate their own philosophy of educating national health care decision makers that radiotherapy was a very important tool in the eradication of cancer. The intention to assist developing countries in solving that problem was on the agenda of decision makers in the IAEA and WHO. That was a good example of the active attitude of the two international organizations towards fulfilling the needs of developing countries, which, it was hoped, would be continued. Representatives of developing countries also hoped that UICC would co-operate with the IAEA and WHO. To get together, as at this Symposium, was the only way for radiotherapists to draw the attention of the international community to the various problems that exist and to their possible solution. The Symposium turned out to be the best forum for participants from industrialized and developing countries as well as for manufacturers of radiotherapy equipment, to interact for the benefit of cancer patients in the developing areas of the world.

CHAIRMEN OF SESSIONS

Session I M.M. MAHFOUZ Egypt Session II N.M. BLEEHEN United Kingdom Session III S. KRISHNAMURTHY India Session IV M. SNELLING United Kingdom Session V N.T. RACOVEANU WHO Session VI R. UZEL Turkey Session VII R.J. MORTON USA Session Vin N.T. RACOVEANU WHO Session IX (Panel) M.M. MAHFOUZ Egypt

SECRETARIAT OF THE SYMPOSIUM

Scientific Y. SKOROPAD Division of Life Sciences, Secretaries: IAEA, Vienna

M. GUSTAFSSON Division of Life Sciences, IAEA, Vienna

Administrative H. BAKHOUM Division of External Relations, Secretary: IAEA, Vienna

Editor: B. KAUFMANN Division of Publications, IAEA, Vienna

443

LIST OF PARTICIPANTS

Abdin-Bey, M.R. College of Veterinary Medicine, King Faisal University, P.O. Box 1757, Al-Hassa 31982, Saudi Arabia

Akki, T.S. Tajura Nuclear Research Centre, P.O. Box 397, Tripoli, Libyan Arab Jamahiriya

Altemark, U.M.F. Buchler GmbH, Harxbütteler Strasse 3, D-3300 Braunschweig, Federal Republic of Germany

Altmann, H. Institut für Biologie, Österreichisches Forschungszentrum Seibersdorf Ges.m.b.H., A-2444 Seibersdorf, Austria

Arcangeli, G. Istituto Medico e di Ricerca Scientifica S.p.A., Via Santo Stefano Rotondo 6, 1-00184 Rome, Italy

Arias, C.F. Comisión Nacional de Energía Atómica, Avenida del Libertador 8250, 1429 Buenos Aires, Argentina

Awwad, H.K. Department of Radiotherapy, National Cancer Institute, Cairo, Egypt

Badib, A.O. Faculty of Medicine, Alexandria University, Alexandria, Egypt

Ben Attia, A. Institut Salah-Azaiz, Boulevard du 9 Avril, Bab-Saadoun, Tunis, Tunesia

Binder, W. Strahlentherapeutische Klinik, Universität Wien, Alserstrasse 4, A-1090 Vienna, Austria

Bleehen, N.M. Department of Clinical Oncology and Radiotherapeutics, Addenbrookes Hospital, Hills Road, Cambridge CB2 2QQ, United Kingdom

445 44 6 LIST OF PARTICIPANTS

Boecker, W. Institut für Medizin, Kernforschungsanlage Jülich GmbH, Postfach 1913, D-5170 Jülich, Federal Republic of Germany

Clubb, B.S. Department of Oncology, King Faisal Specialist Hospital and Research Centre, P.O. Box 3354, Riyadh 11211, Saudi Arabia

Deneufbourg, J.-M. Service de radiothérapie, Hôpital universitaire, Boulevard de la Constitution 66, B-4020 Liège, Belgium

Djakaria, M. Department of Radiotherapy, Medical School, University of Indonesia, Jl. Salemba 6, Jakarta, Indonesia

Durosinmi-Etti, F.A. Lagos University College of Medicine, Lagos, Nigeria

El-Bakkousch, F.A. Tajura Nuclear Research Centre, P.O. Box 397, Tripoli, Libyan Arab Jamahiriya

El-Ghamrawi, K.A. Radiotherapy Department, Kasr El-Einy Centre of Radiation Oncology and Nuclear Medicine (NEMROCK), Faculty of Medicine, Cairo University, Cairo, Egypt

Fletcher, K.E. Amersham International, White Lion Road, Amersham, Bucks HP7 9LL, United Kingdom

Fodor, J. National Institute of Oncology, XII Rath Gy. u. 7/9, H-1525 Budapest, Hungary

Fox, J.S. Atomic Energy of Canada Limited, 413 March Road, P.O. Box 13500, Kanata, Ont. K2K 1X8, Canada

Gallo, M. Via Gramsci, 37, 1-27010 Guissago Stazione (PV), Italy

George, K.C. Molecular Biology and Agriculture Division, Bhabha Atomic Research Centre, Trombay, Bombay 400 085, India LIST OF PARTICIPANTS 447

Glaser, F.H. Klinik und Poliklinik für Radiologie der Medizinischen Akademie Erfurt, Nordhäuser Strasse 74, DDR-5010 Erfurt, German Democratic Republic

González-Miranda, С. Servicio de Radioterapia, Hospital Militar, Holanda No. 050 Providencia, Santiago, Chile

Gupta, B.D. Institute of Medical Education and Research, Chandigarh, India

Hidayatalla, A. Radiation Oncology Academic Unit, Faculty of Medicine, University of Khartoum, P.O. Box 677, Khartoum, Sudan

Höfer, R. Abteilung für Nuklearmedizin, II. Medizinische Universitätsklinik Wien, Gamisongasse 13, A-1090 Vienna, Austria

Hone, C.P. Nuclear Energy Board, 20-22 Lower Hatch Street, Dublin 2, Ireland

Horgas, G. Nuclear Medicine and Oncology Clinic, Clinical Hospital “Dr. M. Stojanovié”, Vinogradska cesta 29, YU-41000 Zagreb, Yugoslavia

Hradil, M. Czechoslovak Atomic Energy Commission, Slezská 9, CS-120 29 Prague 2, Czechoslovakia

Jullien, D. Commissariat à l’énergie atomique, Institut de protection et de sûreté nucléaire, B.P. 6, F-92265 Fontenay-aux-Roses, France

Kato, H. Japan Atomic Energy Research Institute, Tokai-mura, Naka-gun, Ibaraki-ken, Japan

Kirkpatrick, W.W. Atomic Energy of Canada Limited, 413 March Road, P.O. Box 13500, Kanata, Ont. K2K 1X8, Canada

Konoplyannikov, A. Research Institute of Medical Radiology of the Academy of Medical Sciences of the USSR, 249020 Obninsk, Kaluga region, Union of Soviet Socialist Republics

Krishnamurthi, S. Cancer Institute, Adyar, Madras 600 020, India 448 LIST OF PARTICIPANTS

Krompholz, K. Swiss Federal Institute for Reactor Research, CH-5303 Würenlingen, Switzerland

Kuipers, T. Rotterdam Radio-Therapeutic Institute, Dr. Daniel den Hoed Cancer Centre, Groene Hilledijk 301, NL-3075 EA Rotterdam, Netherlands

Laginová, V. Institute of Clinical Oncology, Heydukova 10, CS-812 50 Bratislava, Czechoslovakia

Lanché, T. Instituto Nacional de Pediatría, Insurgentes Sur 3, 700-С, Delegación Coyoacan, 04530 México 22 D.F., Mexico

Luande, J. Tanzania Tumor Centre, Ocean Road Hospital, P.O. Box 3592, Dar es Salaam, Tanzania

Mahfouz, M.M. Radiology Department, Faculty of Medicine, Cairo University, Cairo, Egypt

Mayer, A. E. Weil Hospital and Polyclinic, Centre of Radiation Therapy, Uzasoki u. 29, H-l 145 Budapest, Hungary

Megahid, R.M. Tajura Nuclear Research Centre, P.O. Box 397, Tripoli, Libyan Arab Jamahiriya

Merkle, K.H. Zentralinstitut für Krebsforschung, Robert Rössle Institut, Lindenberger Weg 80, DDR-1115 Berlin-Buch, German Democratic Republic

Morton, R.J. National Cancer Institute, Radiation Research Program, 9000 Rockville Pike, Bethesda, MD 20892, United States of America

Mouelé Soné, A. Centre Alexis Vautrin, F-54511 Vandoeuvre-les-Nancy, France

Moussa-Mahmoud, L. Faculty of Medicine, Alexandria University, Alexandria, Egypt

Okkan, S. Department of Radiotherapy, Cerrahpaça Faculty of Medicine, University of Istanbul, Aksaray-Istanbul, Turkey LIST OF PARTICIPANTS 449

Onoyama, Y. Department of Radiology, Osaka City University Medical School, 1 Asahi-machi, Abeno-ku, Osaka 545, Japan

Onyango, J.N. Department of Radiotherapy, Kenyatta National Hospital, P.O. Box 20834, Nairobi, Kenya

Pagliero, K.M. Oncology Centre, Royal Devon & Exeter Hospital, Barrack Road, Exeter EX2 5DW, United Kingdom

Patricio, M.B. Instituto Portugués de Oncología de Francisco Gentil, Palhavá, Avenida Lima Bastos, Lisbon, Portugal

Pokrajac, B. Nuclear Medicine and Oncology Clinic, Clinical Hospital “Dr. M. Stojanovié”, Vinogradska cesta 29, YU-41000 Zagreb, Yugoslavia

Powel-Smith, C.J. Department of Radiotherapy, Elliot Hospital, Manchester, NH 03103, United States of America

Puribhat, S. Division of Radiation Oncology, National Cancer Institute, Rama VI Road, Phyathai, Bangkok 10400, Thailand

Racoveanu, N.T. World Health Organization, (WHO) Avenue Appia, CH-1211 Geneva 27, Switzerland

Rebolledo, T. Cátedra de Radioterapia, Universidad Central de Venezuela, Caracas, Venezuela

Scheifer, G. Clinique Saint Louis, B.P. 73, L-9001 Ettelbruck, Luxembourg

Shanta, V. Cancer Institute, Adyar, Madras 600 020, India

Singh, P. Institute of Radiotherapy, Oncology and Nuclear Medicine, The General Hospital, 50586 Kuala Lumpur, Malaysia

Snelling, M. 7, Elm Place, London SW7 3QH, United Kingdom

Song, C.W. Department of Therapeutic Radiology, University of Minnesota Medical School, 420 Delaware Street S.E., P.O. Box 494 Mayo, Minneapolis, MN 55455, United States of America 45 0 LIST OF PARTICIPANTS

Sorbe, B.G. Gynaecology Department, Örebro Medical Centre, S-701 85 Örebro, Sweden

Stumpf, J. E. Weil Hospital and Polyclinic, Oncoradiological Centre, Uzasoki u. 29, H-1145 Budapest, Hungary

Sugahara, T. Health Research Foundation, Matsuo Building 4F, Kawaramachi-Marutamachi sagaru, Kamigyo-ku, Kyoto 602, Japan

Svensson, H. Radiation Physics Department, University of Umeâ, S-901 85 Umeâ, Sweden

Department of Radiotherapy and Oncology, St. Mary’s Hospital, Milton, Portsmouth, Hants, United Kingdom

Tavares, M.A.C. Instituto Portugués de Oncología de Francisco Gentil, Palhavá, Avenida Lima Bastos; Lisbon, Portugal

Tisljar-Lentulis, G.M. Institut für Medizin, Kemforschungsanlage Jülich GmbH, Postfach 1913, D-5170 Jülich, Federal Republic of Germany

Uzel, R. Department of Radiotherapy, Cerrahpasa Faculty of Medicine, University of Istanbul, Aksaray-Istanbul, Turkey

Van der Linden, P.M. Radiotherapy Department, Academic Hospital Utrecht, Catharynesingel 101, NL-3511 GV Utrecht, Netherlands

Vázquez, T. Facilidad de Medicina, Hospital de Clínicas, Departamento de Oncología, Avenida Italia s/n, Montevideo, Uruguay

Vilhena, M. Instituto Portugués de Oncología de Francisco Gentil, Palhavá, Avenida Lima Bastos, Lisbon, Portugal

Walstam, R. Department of Radiation Physics and Radiumhemmet, The Karolinska Institute, P.O. Box 60204, S-10401 Stockholm, Sweden LIST OF PARTICIPANTS

Wolber, G. Institut für Nuklearmedizin, Deutsches Krebsforschungszentrum, Im Neuenheimer Feld 280, D-6900 Heidelberg, Federal Republic of Germany

Zaránd, P. E. Weil Hospital and Polyclinic, Oncolradiological Centre, Uzasoki u. 29, H-l 145 Budapest, Hungary

AUTHOR INDEX

Num erals refer to the first page ofpaper(s) by the author concerned

Abdel Rahman, E.: 261 El-Senoussi, M.A.: 139 Abou Zaid, M. .- 281 Ertem, A.; 125 Aguilera, M. : 425 Feinendegen, L.E.: 173 Aissi, A.; 155 Feola, J.M.î 155 Akçasu, A.: 271 Fodor, J.j 291 Akki, T.S.s 359 George, K.C.; 311 Akoush, H . : 281 Ghoneim, М. .• 281 Altemark, U.M.F.: 47 Glaser, F.H.; 13 Altmann, H .. 61 González-Miranda, C.: 97 Alvarez, E.i 425 Gopalan, N.: 249 Amer, M.H.s 139 Gouda, M.Y.: 207 Andreo, P . 333 Gyenes, G.: 291 Arcangeli, G..* 211 Haggag, F.* 403 Awwad, H.K.; 281 Hamad, H.M.A.: 261 Babu, A.M.: 121 Hasegawa, T. .• 301 Badib, A.O.; 207 Hidayatalla, A.: 261 Badinez, L.: 97 Höfer, R.; 163 Barsoum, M..* 281 Hohlfeld, K.: 333 Benassi, M.: 211 Horgas, G..« 369 Bittar, M.: 425 Howard, G.C.VJ. г 187 Bleehen, N.M.: 187 Jullien, D.; 79 Boecker, W..- 173 Kasdorf, P.: 199 Booz, J.: 173 Khalil, S.A.; 207 Borrego, R.: 87 Kida, A.: 229 Cabral, R.: 105 Коса, S.: 125 Campana, F..* 7 9 Konoplyannikov, A . : 221 Carpino, S.; 211 Krishnamurthi, S.: 249 Chotigavanich, C. 33 Kuipers, T.j 27 Clubb, B.S.s 139, 155 Kun, E..* 61 Colina, A.: 425 Laginová, V.: 345 Cunningham, J.: 333 Lanché, Т.г 87 Cyb, A.: 221 Lee, I.s 301 Deneufbourg, J.-M.; 111 Levitt, S.H.í 301 Einhorn, J.г 413 Lokner, V.: 369 El Badawy, S.: 281 Luande, J.: 135 El-Akkad, S.M..- 139 Mahboubi, E . г 139 El-Bakkoush, F.A.: 359 Mahfouz, M.M.r 325, 403 El-Ghamrawi, K.A.; 325, 403 Martínez, G.: 87 El-Haddad, S.: 403 Mattout, J.; 425 El-Naggar, M.; 281 Mauro, F.: 211

453 454 AUTHOR INDEX

McIntosh, N.: 301 Schultz, H.P..- 139 Megahid, R.M.: 359 Shanta, V.: 249 Millan, R..- 425 Shenoy, M.A.: 311 Morton, R.J.: 419 Singh, B.B.: 311 Mould, R. .• 325 Snelling. M.: 3 Nakajima, T.: 229 Sombooncharoen, S.: 33 Németh, G.; 41 Sondhaus, C.A. .- 173 Neves, M.: 105 Song, C.W.: 301 Newman, H.F.V.; 187 Spaventi , S . .• 369 Nieves, C.г 425 Stumpf, J.: 41 Novotny, J..* 345 Sugahara, T.; 193 Okkan, S.: 125, 271 Svensson, H. : 333 Onoyama, Y.: 229 Svoboda, V.H.J.: 147 Onyango, J.N.: 121 Tanaka, M. : 229 Osman, A.; 281 Tangkaratt, S.; 33 Pagliero, K.M.: 71 Tisljar-Lentulis, G.: Patricio, M.B.í 105 Tole, N .: 121 Perdomo, R.: 425 Tóth, J.: 291 Pokrajac, B.; 369 Tsumura, M.; 229 Pomplun, E . : 173 Turan, N. ; 271 Ponte, A. de.* 105 Turkan, S.: 125, 271 Puntumchinda, P.: 33 Tuschl, H.: 61 Puribhat, S.: 33 Urdaneta, N.: 425 Quandt, S.M.: 47 Uzel, R.: 125, 271 Quick, C.A..- 139 Vadon, G.: 41 Racoveanu, N.T.: 381 Vasantan, A.i 249 Rajevic, J.: 97 Vazquez, T.: 199 Reddy, N.M.S.: 249 Vera, A.: 425 Rhee, J.G.: 301 Vera, R.: 425 Rivera, R.: 87 Vilcoq, J.-R..- 79 Rodríguez, A.: 425 Vilhena, M.: 105 Rowland, C.G.: 71 Walstam, R .: 413 Ruan, L.: 425 Yazici, Z.: 271 Salama, A.M.: 207 Zaghloul, M.: 281 Schaden, W..- 173 Zaki, O.: 325, 403 Schneeweiss, F.H.A.: 173 INDEX OF PAPERS BY NUMBER

Page IAEA--SM-290/ Page

291 53 325 71 54 301 139 56 135 155 59 41 105 60 207 87 61 345 14 7 65 425 199 67 121 27 69 187 261 70 229 97 73 211 47 74 311 359 76 281 33 77 173 369 78 163 333 79 381 111 80 403 125 81 3 271 82 413 249 83 13 193 84 61 79 85 419 221

455 HOW TO ORDER IAEA PUBLICATIONS

An exclusive sales agent for IAEA publications, to whom all orders and inquiries should be addressed, has been appointed in the following country:

UNITED■ STATES OF AM ERICA BERNAN - UNIPUB, 4611-F Assembly Drive, Lanham, MD 20706-4391

In the following countries IAEA publications may be purchased from the sales agents or booksellers listed or through major local booksellers. Payment can be made in local currency or with UNESCO coupons.

ARGENTINA Comisión Nacional de Energía Atómica, Avenida del Libertador 8250, RA-1429 Buenos Aires AUSTRALIA Hunter Publications, 58 A Gipps Street, Collingwood, Victoria 3066 BELGIUM Service Courrier UNESCO, 202, Avenue du Roi,B-1060 Brussels CHILE Comisión Chilena de Energía Nuclear, Venta de Publicaciones, Amunategui 95, Casilla 188-D, Santiago CHINA IAEA Publications in Chinese: China Nuclear Energy Industry Corporation,Translation Section, P.O. Box 2103, Beijing IA E A Publications other than in Chinese: China National Publications Import & Export Corporation, Deutsche Abteilung, P.O. Box 88, Beijing CZECHOSLOVAKIA S.N.T.L., Mikulandska 4, CS-11686 Prague 1 Alfa, Publishers, Hurbanovo námestie 3,CS-815 89 Bratislava FRANCE Office International de Documentation et Librairie,48, rue Gay-Lussac, F-75240 Paris Cedex 05 HUNGARY Kultura, Hungarian Foreign Trading Company, P.O. Box 149, H-1389 Budapest 62 INDIA Oxford Book and Stationery Co., 17, Park Street, Calcutta-700 016 Oxford Book and Stationery Co.,Scindia House, New Delhi-110 001 ISRAEL Heiliger and Co., Ltd, Scientific and Medical Books, 3, Nathan Strauss Street, Jerusalem 94227 ITALY Librería Scientifica, Dott. Lucio de Biasio "aeiou". Via Meravigli 16, 1-20123 Milan JAPAN Maruzen Company, Ltd, P.O. Box 5050,100-31 Tokyo International PAKISTAN Mirza Book Agency, 65, Shahrah Quaid-e-Azam, P.O. Box 729, Lahore 3 POLAND Ars Polona*Ruch, Céntrala Handlu Zagranicznego, Krakowskie Przedmiescie 7, PL-00-068 Warsaw R O M A N IA llexim, P O. Box 136-137, Bucharest SOUTH AFRICA Van Schaik Bookstore (Pty) Ltd, P.O. Box 724, Pretoria 0001 SPAIN Di'az de Santos, Lagasca 95, E-28006 Madrid Di'az de Santos, Balmes 417, E-08022 Barcelona SWEDEN AB Fritzes Kungl. Hovbokhandel, Fredsgatan 2, P.O. Box 16356, S-10327 Stockholm UNITED KINGDOM Her Majesty's Stationery Office, Publications Centre, Agency Section, 51 Nine Elms Lane, London SW8 5DR USSR Mezhdunarodnaya Kniga,Smolenskaya-Sennaya 32-34, Moscow G-200 YUGOSLAVIA Jugoslovenska Knjiga,Terazije 27,P.O. Box 36, YU-11001 Belgrade

Orders from countries where sales agents have not yet been appointed and requests for information should be addressed directly to: Division of Publications dap ÿ International Atomic Energy Agency Wagramerstrasse 5, P.O. Box 100, A-1400 Vienna, Austria

INTERNATIONAL SUBJECT GROUP: I ATOMIC ENERGY AGENCY Life Sciences/Radiation Biology V IEN N A, 1987 PRICE: Austrian Schillings 9 4 0 ,-