Tamoxifen therapy in breast cancer control worldwide.

Reviews/Analyses Tamoxifen therapy in breast cancer control worldwide R.R. Love1 & V. Koroltchouk2 In most developed and many developing countries, breast cancer is the most frequent cancer and the leading cause of cancer death among women. At least 50% of all breast cancer patients worldwide would survive longer, however, if public awareness about and early detection of the condition were increased and greater use were made of efficient treatment of proven value. With early-stage, localized breast cancer, local treatment combined with adjuvant hormonal therapy with tamoxifen, a synthetic estrogen, could save the lives of 6 women out of 100 compared with local treatment alone. Tamoxifen has anti-estrogenic effects not only on breast cancer cells but also on liver metabolism and bone, with concomitant decreases in risk factors for chronic skeletal and vascular system diseases. Long-term tamoxifen treatment causes major adverse clinical effects in <5% of women; menopausal and vaso- motor symptoms occur in the majority of treated women, but their severity lessens over time. Tamoxifen is being considered as a standard therapy and is included in the WHO list of essential drugs for the treatment of breast cancer patients in both developing and developed countries. For the control of breast cancer more successfully worldwide, one challenge is to make tamoxifen therapy available to greater numbers of women. Breast cancer is an increasing global health problem for which a hormonal treatment - tamoxifen - can, in part, provide a solution. The present article briefly reviews the dimensions of the worldwide breast can- cer problem and the available data that support the case for the more widespread application of tamoxi- fen treatment. Finally the challenges in achieving this goal and the new broad research agenda to be addressed are identified. Breast cancer worldwide In 1980 breast cancer was the commonest cancer among women worldwide, with an estimated 572 100 cases, representing 18% of all cancers that affected women (1). Current estimates are that 750 000 new cases will be diagnosed in 1993 (2), 1 Professor, Departments of Human Oncology, Medicine, and Family Medicine and Practice, University of Wisconsin, 7C, 1300 University Avenue, Madison, WI 53706, USA. Requests for reprints should be sent to this address. 2 Scientist, Cancer and Palliative Care unit, World Health Organization, Geneva, Switzerland. Reprint No. 5442 and towards the end of this century over 1 million new cases of breast cancer will be diagnosed annual- ly (3). At present, more than 40% of all breast can- cers are found in developing countries, but the inci- dence pattern in such countries is progressively approaching that of developed countries and it is pre- dicted that it will be more than 50% by the year 2000 (4). More than two-thirds of breast cancers occur among women over 50 years of age, and the inci- dence has been increasing, particularly among post- menopausal women in developed countries. With the aging of populations, the increase in the proportion of elderly women in all countries will lead to a corre- sponding increase in the number of breast cancer cases. Mortality from breast cancer also appears to be increasing, although not in all developed coun- tries (5). The mortality rates in some Western coun- tries remain about a third of incidence rates. In West- ern countries, with the highest incidences at age 40 years, approximately one woman in 1000 develops breast cancer each year, while at age 60 years this rate has increased to one woman in 500. At these incidence levels, 3.3% of healthy 40-year-old women will develop breast cancer before they reach 60 years of age (6). Bulletin of the World Health Organization, 71 (6): 795-803 (1993) © World Health Organization 1993 795 R.R. Love & V. Koroitchouk Dimensions of the challenges of breast cancer control What strategies are there available or are in the pro- cess of being developed for the control of breast can- cer that can be applied worldwide? There are three basic approaches: primary prevention; early diagno- sis and screening; and treatment. Primary prevention may have the greatest potential for the control of breast cancer, but is not applicable at present other than dietary modification, which will probably require to be introduced in adolescence or early in adult life to be fully effective. Although few of the causative factors are easily manipulable, two have recently been receiving more attention and are worthy of comment. First, data are available that im- plicate active and passive cigarette smoking in the development of breast cancer (7, 8). These are of particular concern because of the high and increasing prevalence of smoking in developing countries and provide another cogent reason for directing more vigorous attention to the proliferation of tobacco consumption. Second, more data are becoming avail- able which suggest that long-duration lactation is significantly protective against breast cancer (9). This information should be more widely used in public health programmes worldwide. With these exceptions, practical interventions to prevent breast cancer are not currently available. Approximately half of all breast cancer patients are diagnosed only after the disease has reached an advanced stage, and most of them will not be cured. In its advanced stages, breast cancer is currently a disease for which palliative therapy is the most prac- tical approach. The reasons for the refractoriness of metastatic breast cancer to significantly life-prolong- ing treatment are multiple: the evolution of the dis- ease over several years, which has preceded this state, results in the development of increasingly malignant clones of tumour cells ; the heterogeneity of the disease, whose recognition and implications have yet to be well defined and understood prevents optimal individual, specific treatment; and most importantly, the absence of adequately effective "curative" therapies distinguishes breast cancer from other malignancies for which remarkably effective drugs have been identified empirically. While our understanding of the biology of breast cancer has grown considerably in recent years, and our abilities to apply the currently available techniques of drug dose density, bone marrow transplantation, and bio- logical response modifiers have increased, practical life-prolonging therapies for populations have yet to be identified. The greatest decrease in breast cancer mortality is likely to result from treating the disease at an earlier stage. In the USA the 5-year relative sur- vival rate for localized breast cancer is approximate- ly 90%; in Russia it is 77%, while the total 5-year survival rates for all stages in these two countries are 75% and 55%, respectively (10, 11). At the Nation- al Cancer Institute, Cairo, the 5-year survival rates were found to be 80%, 52%, 42% and 14% for stage I, II, III and IV, respectively (12, 27). These differ- ences are a function of public awareness of the value to breast cancer patients of early detection and appropriate early treatment. In the absence of public education programmes, and also because of a lack of available resources for either detection or therapy, current statistics show that in developing countries 30% to 80% or more of all breast cancer patients will not be diagnosed until the disease has reached an advanced stage; the large majority will be incurable. Under these conditions, early detection must be the primary means of improving the situation for some time to come. WHO promotes strategies designed to provide patients in all areas of the world with community access to basic facilities for appropriate diagnosis and treatment of breast cancer. The lives of at least 50% of breast cancer patients could be significantly prolonged through the appropriate use of currently available methods of diagnosis and treatment. In developed, particularly Western, countries screening mammography has been the most discussed down- staging strategy. Recent data confirm, however, that mammography is of no benefit to premenopausal women (13) and suggest that when combined with careful breast examination, mammography is only of modest benefit to women aged 50-59 years (14). Therefore, at present, the most practical population approach should include breast self-examination, physical examination by health care workers, mam- mography, or a combination of these, depending on the extent of the breast cancer problem, local resources, and the cultural situation (4). The evidence that mortality from breast cancer can be further reduced through the use of systemic therapies as an adjuvant to surgery or radiotherapy is substantial. A comprehensive meta-analysis of the results from all the randomized trials of adjuvant therapies suggests that use of polychemotherapy for premenopausal women with breast cancer reduces the annual risk of recurrence by 29% and of mortal- ity by 16% (15). Benefits of a similar magnitude result from surgical or radiation oophorectomy in premenopausal women, although this conclusion is based on much smaller numbers of treated women with heterogeneous stages of cancers, and probably should be regarded as hypothesis-generating rather than conclusive evidence for this adjuvant strategy (16). WHO Bulletin OMS. Vol 71 1993796 Tamoxifen therapy in breast cancer control At the population level, the relevance and appli- cation of the polychemotherapy data are somewhat uncertain. First, despite the convincing evidence of benefit from individual trials and the meta-analysis, the evidence for benefits to populations where adju- vant polychemotherapy has been widely applied is weak. Second, the practice of adjuvant polychemo- therapy is highly specialized and expensive and hence its use in developing countries is impractical. In most instances, cancer chemotherapy requires some access to laboratory facilities to monitor white blood cell and platelet counts. Also, the quality of life of some patients will be compromised by adju- vant polychemotherapy (17). Finally, the vast major- ity of the clinical trials of such therapy have involved exclusively Westem women, usually from higher socioeconomic and Caucasian groups, and the relevance of the findings for poorer women from other ethnic groups is unclear. In contrast, as dis- cussed below, the adjuvant tamoxifen may have more widespread application. The current public health challenges of breast cancer control, apart from the more widespread application of tamoxifen adjuvant therapy, are there- fore to encourage patients to present for diagnosis and treatment at an earlier stage of the disease since this offers the most effective approach for a reduc- tion in breast cancer mortality. Tamoxifen as an adjuvant treatment Evidence for efficacy The case for more widespread use of adjuvant tamoxifen therapy rests on the quality of the eviden- ce for its efficacy, particularly in reducing breast cancer mortality. This case is extraordinarily strong because of the 40 individual, randomized clinical trials of tamoxifen that have reported consistently on its favourable effects, together with the findings of a rigorous meta-analysis of these trials that has con- firmed their conclusions (16). Here, it is appropriate to refer to the meta-analysis results because they are hypothesis-supporting (and not hypothesis- generating), since the individual trials on which they are based also overwhelmingly draw the same conclusions. In addition, the overview data provide quantitative and qualitative perspectives that are critical for making judgements about the more widespread applicability of tamoxifen. Fig. 1 summarizes the essential results obtained with short-term (usually, a maximum of 2 years) adjuvant tamoxifen therapy. Although modest, the findings indicate that relative to controls, approxi- mately four and six additional women per 100 will be alive 5 years and 10 years, respectively, after this Fig. 1. Ten-year mortality levels with tamoxifen and controls: overall results for 30 000 women. (Reproduced with permission from: Luan, J.M. et al. Risk factors for breast cancer in Shanghai. Cancer research, 48: 1949- 1953 (1988)). uu - 90- 80 - I 60o O 70- >_ co 50- 0 5 No. of years >10 treatment. The proportions of additional women who survive but suffer a recurrence are 8% and 6.5%, respectively, 5 years and 10 years later. Thus, viewed on a population basis, tamoxifen treatment can prolong and save the lives of many women with early-stage breast cancer. Of interest also, is whether further studies on subsets of treated women will provide clarifying information that defines more pre- cisely the most appropriate target groups for this treatment to achieve the greatest benefits. Table 1 provides a summary of some of the important results obtained with tamoxifen. In the trials the same dose appeared to be used and so the lowest widely used dose (20 mg daily) can be taken as standard. The data on the duration of the therapy are incomplete; the majority of trials used treatment periods of 2 years. It is remarkable that despite an apparent cytostatic mechanism of action, tamoxifen produces a persistent benefit many years after cessa- tion of therapy, and that the mortality differences between the treated and untreated groups of women differed by 5-10 years. The overview data suggest that a greater benefit derives from a longer duration of treatment, but at present 2 years must be consid- ered a reasonable standard. While the absolute bene- fits are greater for women who have axillary node metastases and who have tumours with positive or higher measurable levels of estrogen receptors, women with no axillary metastases and low or ab- sent levels of estrogen receptors in their tumours also derive significant benefits from tamoxifen treatment. The effects of adjuvant tamoxifen therapy were qualitatively similar for women who are premeno- WHO Bulletin OMS. Vol 71 1993 -*- Tamoxifen 77.5% -.* Control 73P9<000 5;8.8% 52.6% Pc0.00001 797 R.R. Love & V. Koroitchouk Table 1: Mortality benefits of adjuvant tamoxifen according to dose, duration of treatment, nodal status, and primary tumour estrogen receptor categoriesa Tamoxifen versus no tamoxifen Dose of tamoxifen 20 mg/day versus none 30 mg/day versus none Duration of treatment <1 year versus none >2 years versus none Axillary nodal status Negative Positive Tumour estrogen receptor protein Poor Positive % reduction in annual odds of death from any cause for women aged >50 yearsb 20 (2)c 21 (3) 18 (3) 13 (4) 23 (6) 16 (5) 22 (3) 16 (6) 23 (4) a Adapted from Table 3, ref. 14. b Absolute benefits: If 100 node-positive women have a persis- tent reduction in the annual odds of death of 15% - 6 additional women will be alive after 10 years; if 30%-12 additional women will be alive after 10 years. c Figures in parentheses are standard deviations. pausal or postmenopausal, under or over 50 years of age. In general, younger women appear to benefit less from adjuvant tamoxifen (compare Table 1 and Fig. 1) and in considering strategies for populations based on available data, a case can be made for including other adjuvant therapeutic approaches, such as surgical oophorectomy, although for this there are far fewer data currently available. For older women, among whom the incidence of breast cancer is greater, the benefits of adjuvant tamoxifen are quantitatively and qualitatively similar. Are the available data for tamoxifen different from those for polychemotherapy in premenopausal women in terms of the case they present for applica- tion to populations? It must be stated, in reply, that clear evidence of the benefits to populations from adjuvant tamoxifen therapy is not available at present. If it is accepted that the prevalence of adjuvant tamoxifen use in some Western countries is high, there are several possible explanations for the lack of benefits at the population level, but most of them arise because of confounding caused by an increasing incidence of breast cancer. Tamoxifen therapy is a daily oral treatment and is thus not difficult to administer, but remains expensive in some Western countries. Final- ly, although some of the subjects in the trials were Asian, tamoxifen must be considered to be signifi- cantly underevaluated in developing countries and among non-Western ethnic groups. With a strong case for a reduction in mortality and an increase in disease-free survival resulting from the use of adjuvant tamoxifen, a more compre- hensive review of its biological effects, other bene- fits, toxicities, and symptomatic sequelae associated with its use is important to place in better context the argument for its more widespread use. Biology of tamoxifen and non-breast cancer effects Tamoxifen is a synthetic estrogen with estrogen agonist and antagonist properties (18). The drug was originally developed as an oral contraceptive, but proved to be ineffective; subsequently, it was found serendipitously to be effective in palliating metasta- tic breast cancer in some women. Over the last 20 years it has been used for this purpose, but also increasingly for adjuvant therapy. The pharmacology of tamoxifen is complex. The maximum concentration in blood occurs 4-7 hours after oral administration, and a steady-state after 4 weeks. The total time required for all detectable serological evidence of tamoxifen and its metabolites to disappear from blood is 6-8 weeks. The teratogen- icity of tamoxifen is not known; this is of importance for premenopausal women who become pregnant while taking the drug, since even if tamoxifen is stopped, the fetus is likely to receive continuous exposure during the entire first trimester. Tamoxifen is metabolized by the liver, excreted in bile, and eliminated from the body in the faeces. In postmeno- pausal women, decreases in gonadotropin levels occur with tamoxifen therapy; levels of estrogens, progesterone, and prolactin do not change. In con- trast, in premenopausal women large increases in total estrogens, estradiol, and progesterone some- times occur. At the cellular level tamoxifen probably acts by combining with nuclear estrogen receptor protein, arresting breast cancer cells at the GI phase. Thus tamoxifen appears to be a cytostatic instead of a cytocidal agent. This is supported by some of the adjuvant trial data, but data for the prolonged effects of the drug suggest that the situation may be more complex. Changes in various biological growth mediators are also produced by tamoxifen: sex- hormone-binding globulin levels increase (which can result in removal of more free estrogen from the circulation); levels of transforming growth factor alpha (a growth stimulatory protein) decrease; and levels of transforming growth factor beta (a growth inhibitory protein) increase. It has been proposed that WHO Bulletin OMS. Vol 71 1993798 Tamoxifen therapy in breast cancer control effects in breast stromal cells mediate tamoxifen's action in neighbouring breast cancer cells. Tamoxifen is therefore a synthetic estrogen that might be expected to have direct estrogenic effects, but also has estrogen-antagonist effects. In addition it causes hormonal and growth factor perturbations whose long-term consequences can be expected to be profound. The most important use of tamoxifen in humans has been as an adjuvant treatment. In these trials, which are discussed above, the focus has been on recurrence and survival from breast cancer. Only recently has greater attention been paid to the pos- sible non-breast organ, tissue, and risk-factor effects of this therapy. Cardiovascular and skeletal effects Hormones exert powerful effects on several organ systems. Among Western women, on whom more studies have been performed, the most profound effects caused by such substances are on the cardio- vascular and skeletal systems; women who survive to undergo menopause are at major risk for chronic diseases of these systems. Almost half of all women in Western societies who pass their 50th birthdays will ultimately die of cardiovascular disease; another large, difficult-to-estimate proportion will suffer from the fracture of osteoporotic bones. These chronic diseases are reflections of marked changes in risk factors that occur following menopause, with associated marked decreases in ovarian hormones. In this context, what is known about the action of tamoxifen? The effects of tamoxifen on risk factors for car- diovascular diseases appear to be generally favour- able and estrogenic. Controlled studies show decreases in the levels of total and low-density lipo- protein cholesterol, fibrinogen and platelets, and an absence of major changes in blood pressure or glucose metabolism in postmenopausal women who are receiving tamoxifen therapy (19) (Fig. 2). These consequences of tamoxifen treatment are clinically important, as indicated by the results of the meta- analysis of tamoxifen adjuvant studies, in which a 25% reduction in vascular deaths was found (16), and by the results of one trial that reported a major reduction in the numbers of postmenopausal women with myocardial infarction (20). While these obser- vations are consistent with risk-factor changes, they should be regarded more as hypothesis-generating than hypothesis-confirming, and in particular their postmenopausal population base should be borne in mind. The effects of tamoxifen on bone mineral den- sity, a major measure of risk for fracture, are also likely to be favourable. The reduction in bone mass associated with the cessation of ovarian estrogen Fig. 2. Mean fasting levels of total cholesterol over time In postmenopausal women receiving tamoxifen or a placebo; bars show the standard errors. (Reproduced with permission from: Love, R.R. et al. Effects of tamoxifen on cardiovascular risk factors in postmenopausal women. Annals of internal medicine, 115: 860-864 (1991)). I 0) E al) CD) a) 0 C) cr 0 CD Cl, CD 0 3 0 1- ,3 6 12 18 Study time point (months) production involves mainly the more metabolically active trabecular bones in the spine and to a lesser degree in the hips. In postmenopausal women, tamoxifen has a clear bone-density-preserving effect on the lumber spine (Fig. 3) (14). This observation is consistent with data from animal studies which sug- gest that tamoxifen is an antiresorptive agent similar to estrogen and thus may be associated over time with decreased rates of bone fracture. Fig. 3. Change in mean lumbar spine bone mineral den- sity (BMD) In postmenopausal women receiving tamoxi- fen or a placebo; bars show the standard errors. (Repro- duced with permission from: Love, R.R. et al. Effects of tamoxifen on bone mineral density in postmenopausal women with breast cancer. New England journal of medi- cine, 326: 852-856 (1992)). (D Cl)coQC', .0 WHO Bulletin OMS. Vol 71 1993 799 R.R. Love & V. Koroltchouk Other significant and undefined effects Because the major sources of data about the effects of tamoxifen have been adjuvant trials, which were designed to evaluate cancer endpoints, our under- standing about the consequences of this therapy is incomplete. These trials and the findings of the meta- analysis discussed above will, nevertheless, continue to be a major source of data. While the long list of incompletely defined effects shown in Table 2 should prompt caution and further evaluation and research, these uncertainties must be placed in context. With adjuvant tamoxifen treatment, recur- rence of breast cancer and death are unquestionably averted for postmenopausal women, and probably also in premenopausal women (Table 1 and Fig. 1). The meta-analysis determined that the vast majority of deaths were from breast cancer; thus, while there may be morbidity caused by tamoxifen treatment that has not yet been fully identified, as well as pos- sible long-term mortality effects, over a 5-10-year period after diagnosis of breast cancer the deaths from cancer and from all causes reduced. As discus- sed above for the major chronic diseases of Western women aged over 50 years, tamoxifen appears, if anything, to be protective. The breadth of data that have generated the pos- sible (hypothetical) effects shown in Table 2 will not be reviewed here, but the reasons for the particular entries are commented on below. A useful review of Table 2: Effects of tamoxifen therapy that are incom- pletely defined In women of all ages Uterine endometrium: carcinogenic Uterine myometrium: carcinogenic Liver: carcinogenic Colorectum: carcinogenic Haematopoietic system: carcinogenic/protective Central nervous system: mood-altering Coagulation: thrombophlebitis Cardiovascular: lipoprotein, blood pressure, glucose levels Eye: macular, retinal, lens effects Hepatobiliary: cholelithiasis Immune system: functional effects Drug interactions Skeletal: bone mineral density changes at all sites Breast cancer: treatment of recurrence In premenopausal women Ovary: carcinogenic Breast: carcinogenic Hormonal: patterns, levels and frequency of changes Cardiovascular: lipid, fibrinogen, platelet changes Gynaecological: symptomatic, infectious Vasomotor: symptomatic Pregnancy: teratogenic, contraception this topic has been carried out by Nayfield et al. (22). The reported hormonal effects on uterine tissues, the liver, and most recently, on the colorectum can promote tumour growth. There are very limited data at present to corroborate that these effects are a sig- nificant if at all a real concem. The growth-factor- altering effects of tamoxifen may protect against haematopoietic malignancies. Depression occurs with tamoxifen, but its frequency is poorly described. Thrombophlebitis appears to occur at excess rates of 1 per 800 women-treatment years; risk factors for this complication have not been fully characterized. Although some good quality lipid, fibrinogen, and platelet data are available for post- menopausal women, data on lipoprotein, blood pres- sure, and glucose levels are needed since minor changes in these parameters could be critical for determining the risk for cardiovascular disease and are probably controllable. Whether the prevalence of various ocular conditions is occasionally increased by tamoxifen therapy is unknown, but some data indicate that this should be more carefully studied (23). Since estrogen therapy is associated with increased rates of cholelithiasis, and tamoxifen has similar effects on lipids and lipoproteins, the inci- dence of cholethiasis among women who are receiv- ing tamoxifen should be evaluated. The effects of hormones such as tamoxifen on the immune system are of increasing interest; in one Swedish study, hos- pitalization rates for immune system disorders were lower among tamoxifen-treated women. Bone miner- al density changes at sites other than the lumbar spine also need to be assessed. Since tamoxifen is metabolized in the liver and binds to different body tissues and proteins, its pos- sible interaction with other drugs warrants evalua- tion. Finally, tamoxifen is not a curative adjuvant therapy. Many women experience recurrences of breast cancer (metastatic disease), while taking tamoxifen. The mechanisms of this tamoxifen resis- tance and the optimal management of these patients are only beginning to receive attention (24). It is, however, notable that the incidence of second pri- mary, i.e., contralateral, breast cancer is significantly lower for women receiving tamoxifen treatment; the meta-analysis reported above found a 39% reduction in the incidence of this event (16). There are additional effects of particular concern for younger premenopausal women, in the main because fewer women in this category have partici- pated in the adjuvant trials of the drug. Of greatest importance is whether in younger women, the direct and indirect hormonal effects of tamoxifen may be carcinogenic to the breast and ovaries. Available data WHO Bulletin OMS. Vol 71 1993800 Tamoxifen therapy in breast cancer control indicate that tamoxifen markedly increases the levels of estrogens in some women at frequencies that have not yet been completely determined, and one trial has reported that this is associated with an increase in contralateral breast cancer (18). Epithelial ovarian cancer is affected by factors that influence ovulation or epithelial disruption, which tamoxifen causes. Specific cardiovascular effects have not been evalu- ated in premenopausal women, and the gynaecologi- cal effects of tamoxifen are poorly described. The details of and risk factors for the vasomotor symp- toms produced by tamoxifen are also poorly described. Finally, how to achieve optimal contra- ception with tamoxifen and its risks of teratogenesis are not known, but are important issues for premeno- pausal women. The above matters deserve particular research attention and monitoring, with the more widespread use of adjuvant tamoxifen being justified by the sig- nificant mortality benefits, particularly for postmeno- pausal women. Because it is likely that most of the serious consequences resulting from use of tamoxi- fen are rare, any associations will only be observed through population monitoring. Contraindications In practical terms, there are few women, particularly postmenopausal women, who have histories that should prevent them from taking tamoxifen. Table 3 summarizes the suggested absolute and relative contraindications. While certainty about retinal macular changes following tamoxifen therapy is lacking, they are a possible side-effect and their seriousness warrants particular prudence (23). The other absolute contraindications are based on concems about hormonal carcinogenesis, teratogen- esis, and inefficacy of tamoxifen therapy for breast cancer. Table 3: Contraindications to adjuvant tamoxifen Absolute Retinal macular oedema or degeneration History of benign or malignant liver tumour secondary to oral contraceptives Pregnancy Other hormonal therapy (estrogens, oral contraceptives) Relative History of thrombophlebitis, particularly hormone-related History of depression, particularly hormone-related Cataract Drugs: chlorpromazine, chloroquine, thioridazine, amiodarone and other antiarrhythmics Severe vasomotor symptoms Polycystic ovaries The relative contraindications for tamoxifen therapy shown in Table 3 are conditions that may be important risk factors for major complications. The drugs listed in Table 3 can exhibit retinal toxicity and could act synergistically with tamoxifen to produce ocular conditions. Tamoxifen invariably increases the number and intensity of vasomotor symptoms, and up to a third of postmenopausal women receiving the drug may develop annoying gynaecological symptoms (26). For some women who experience such symptoms before beginning tamoxifen therapy and who are prescribed estrogen therapy because of their breast cancer, the resultant intensification of these symptoms may be intoler- able. Widespread use of adjuvant tamoxifen: benefits and challenges By the year 2000 if 1 million women per annum are diagnosed to have breast cancer, the large majority at the earlier stages of the disease, available data sug- gest that the deaths of 60 000 of them could be aver- ted by adjuvant tamoxifen treatment and that many thousands more will face a recurrence of breast cancer, and death from the condition will be delayed. Most breast cancer therapy can be administered at the community level and does not require the ser- vices of highly specialized doctors. Tamoxifen treat- ment can be given by surgeons, radiotherapists, or family doctors. Two-thirds of breast cancers occur among women aged more than 50 years, and this group is best served by the use of adjuvant tamoxi- fen, which is easily administered by medical person- nel with minimal training and experience, and causes few side-effects. Tamoxifen can be taken for long periods without monitoring or frequent follow-ups. For most developing countries, where an ideal medi- cal infrastructure will not be available for some decades to come, tamoxifen should receive priority as a breast cancer treatment altemative; it can save medical resources as well as lives. Laboratories that are not equipped to compete in the search for a cancer cure could instead carry out research on ways to deliver simpler therapies with minimal requirements and costs, in order to provide as much coverage as possible for breast cancer patients in both developed and developing countries. Such target-oriented research is needed to identify the optimal methods for delivering chemotherapeutic drugs and tamoxifen. It is unrealistic to expect tamoxifen compliance in symptom-free women in many developing countries; therefore, a depot form of the drug, should be sought. Currently, the cost of a daily dose of oral tamoxifen is still too high for WHO Bulletin OMS. Vol 71 1993 801 R.R. Love & V. Koroltchouk most developing countries; however, it should be possible to decrease the price by increasing its use and by producing generic versions. The major challenges to increasing tamoxifen use are summarized in Table 4. Clinical research is needed to address four issues to help define those public health efforts that are likely to be the most beneficial. While the postulated benefits of the thera- py are based on a 2-year course of tamoxifen, some data suggest that treatment for a longer period pro- vides greater benefits. As further data on the precise benefits associated with different durations of treat- ment become available, a cost-benefit curve can be drawn to facilitate definition of a rational public health approach. At present, insufficient data on pre- menopausal women are available to support use of adjuvant tamoxifen as a priority therapy for this group. It is important to investigate the benefits of tamoxifen treatment in developing countries, where breast cancer affects younger women more, especial- ly those aged 40-50 years. In developing countries, where pill-taking would present problems, a large clinical trial of a depot form of tamoxifen should be evaluated as part of a comprehensive breast cancer programme in which delay in presenting for diagno- sis is studied and a careful assessment of medical resource use is undertaken. Because health care financial resources are severely limited in developing countries, agreements with major pharmaceutical companies must be made to provide large amounts of tamoxifen at less than current Western prices. In addition, in individual countries precise quantifica- tion of the population-wide benefits to be gained will enable use of tamoxifen therapy to be placed ration- ally in the list of priorities developed under compre- hensive national cancer control plans to provide opti- mal care and improve the quality of life for breast cancer patients and to ensure that scarce resources Table 4: Major challenges to increasing adjuvant tamox- ifen use for women with breast cancer worldwide Clinical research Definition of optimal duration of treatment Increased definition of known adverse effects and their inci- dence Further data on breast cancer benefits in premenopausal women Data on breast cancer benefits in non-Western populations Applications Decreasing the cost of treatment Development of injectable depot treatment Quantitation of benefits for use in national cancer control plan- ning Public health strategies that combine down-staging and adju- vant tamoxifen are directed towards providing the maximum benefit. Finally, since the benefits of adjuvant tamoxifen are likely to be lower for stage-III or regionally advanced breast cancer, which is often the most commonly presented form in developing countries, programmes should make greater educational efforts or increase the frequency of breast examinations to downstage the disease at diagnosis, as well as make tamoxifen more readily available. Resume La place du tamoxifene dans la lutte mondiale contre le cancer du sein Le cancer du sein est un probleme de sante mon- dial de plus en plus preoccupant auquel un traite- ment hormonal (par le tamoxifene) peut apporter une solution partielle. Cette forme de cancer est la plus frequente chez les femmes et l'on prevoit un million de cas nouveaux par an en l'an 2000. Actuellement, il n'existe pas de moyens pratiques de pr6vention, mais le tabagisme pourrait jouer un r6le plus important qu'on ne le pensait jusqu'ici dans le d6veloppement de ce cancer, tandis que l'allaitement prolong6 semble avoir un effet pro- tecteur notable. II est important de diagnostiquer le cancer du sein a un stade aussi pr6coce que possible pour limiter la mortalite et la morbidite. De nombreux essais cliniques randomises et une meta-analyse portant sur 40 essais et 30 000 femmes ont montr6 qu'un traitement par le tamoxi- fene, venant en complement des mesures locales et r6gionales, pr6sentait des avantages significa- tifs. Ce traitement ameliorerait le taux de survie a dix ans d'environ 6 pour 100 et dans de nom- breux cas, la recidive serait retardee. La plupart des donnees relatives au tamoxifene concernent les femmes menopaus6es; avant la menopause, les avantages sont moins nets. Qualitativement toutefois, on ne peut d6finir aucun groupe de femmes pour lesquelles le tamoxifene ne presen- terait aucun interet. Le tamoxifene a des effets estrogeniques et anti-estrogeniques. Heureusement, ses effets sur les facteurs de risque de maladie cardio-vasculaire et d'ost6oporose chez les femmes menopausees semblent favorables. Les concentrations de cho- lest6rol total et de cholesterol LDL sont abais- s6es, tandis que la densit6 osseuse se maintient. Ces effets donnent a penser que le tamoxifene pourrait reduire le risque de maladies cardio-vas- culaires et de fractures consecutives a l'osteopo- rose, principales causes de mortalite et de morbi- dit6 chez les femmes agees dans les pays WHO Bulletin OMS. Vol 71 1993802 Tamoxifen therapy in breast cancer control developpes. Bien que le tamoxifene ait des consequences globalement b6n6fiques sur la mor- talite, certains de ses effets n'ont pas encore e parfaitement caracterises. L'absence de donn6es detaill6es sur ses effets avant la m6nopause est particulierement pr6occupante. Le traitement par le tamoxifene n'est contre-indiqu6 que dans de rares cas: oedeme maculaire ou degenerescence de la retine, autre traitement hormonal et grosses- se. Les r6sultats des essais cliniques constituent des arguments solides en faveur de la g6n6ralisa- tion de l'emploi du tamoxifene comme adjuvant, mais les recherches se poursuivent pour 6tablir la duree optimale du traitement, reconnaitre ses effets indesirables et evaluer ses avantages avant la menopause et pour les populations non occi- dentales. Pour mettre le tamoxifene a la disposi- tion des populations et se rapprocher du but recherche, qui est de reduire de 60 000 le nombre mondial de deces annuels dus au cancer du sein, il faudrait lancer des programmes de sant6 publi- que pour etudier la possibilit6 d'administrer le pro- duit sous forme d'injections retard, quantifier soi- gneusement les avantages de ce type d'interven- tion pour la population et favoriser un diagnostic aussi pr6coce que possible. Les essais cliniques pourraient etre l'occasion de faire participer acti- vement les agents de sante a la d6finition des nouvelles strat6gies visant a rendre le traitement par le tamoxifene accessible a davantage de femmes atteintes d'un cancer du sein. References 1. Parker, D.M. et al. Estimates of the worldwide fre- quency of sixteen major cancers in 1980. Interna- tional journal of cancer, 41: 184-197 (1988). 2. Parkin, D.M. et al. Estimates of the worldwide inci- dence of eighteen major cancers in 1985. Interna- tional journal of cancer, 54: 594-606 (1993). 3. Miller, A.B. The role of screening in the fight against breast cancer. World health forum, 13: 227-285 (1992). 4. Koroltchouk, V. et al. The control of breast cancer: a World Health Organization perspective. Cancer, 65: 2803-2810 (1990). 5. Stjernsward, J. et al. Progress against cancer? New England journal of medicine, 315: 965 (1988). 6. Love, R.R. The risk of breast cancer in American women. Journal of the American Medical Associa- tion, 257: 1470 (1987). 7. Hirayama, T. Health effects of active and passive smoking. In: Aoki, M. et al., ed. Smoking and health 1987. Amsterdam, Elsevier, 1988, pp. 75-86. 8. Palmer, J.R. et al. Breast cancer and cigarette smoking: a hypothesis. American journal of epidemi- ology, 134: 1-13 (1991). 9. Luan, J.M. et al. Risk factors for breast cancer in Shanghai. Cancer research, 48: 1949-1953 (1988). 10. 1987 Annual cancer statistics review including can- cer trends, 1950-1985. National Cancer Institute, Div- ision of Cancer Prevention and Control, 1987, p. 78. 11. Berezkin, D.P. et al. [Breast cancer patients sur- vival (based on the data of the All-Union Centre for Evaluation of Treatment Efficacy)]. Voprosiy On- cologij, 32(9): 17-24 (1986) (in Russian). 12. Omar, S. et al. End results of breast cancer thera- py. I. A prospective study: breast cancer. Cancer prevention and therapy, 1-13, Feb. 1989. 13. Miller, A.B. et al. Canadian National Breast Screen- ing Study: I. Breast cancer detection and death rates among women aged 40 to 49 years. Canadian Medical Association journal, 147(10): 1459-1476 (1992). 14. Miller, A.B. et al. Canadian National Breast Screen- ing Study: II. Breast cancer detection and death rates among women aged 50 to 59 years. Canadian Medical Association journal, 147(10): 1477-1488 (1992). 15. Early Breast Cancer Trialists' Collaborative Group. Systemic treatment of early breast cancer by hormonal, cytotoxic or immune therapy. Lancet, 339: 72-85 (1992). 16. Early Breast Cancer Trialists' Collaborative Group. Systemic treatment of early breast cancer by hormonal, cytotoxic or immune therapy. Lancet, 339: 1-15 (1992). 17. Essential drugs for cancer chemotherapy: Memoran- dum for a WHO meeting. Bulletin of the World Health Organization, 63: 999-1002 (1985). 18. Love, R.R. Tamoxifen therapy in primary breast cancer: biology, efficacy and side-effects. Journal of clinical oncology, 7: 803-815 (1989). 19. Love, R.R. et al. Effects of tamoxifen on cardiovas- cular risk factors in postmenopausal women. Annals of internal medicine, 115: 860-864 (1991). 20. McDonald, C.C. & Stewart, H.S. Fatal myocardial infarction in the Scottish adjuvant tamoxifen trial. British medical journal, 303: 435-437 (1991). 21. Love, R.R. et al. Effects of tamoxifen on bone min- eral density in postmenopausal women with breast cancer. New England journal of medicine, 326: 852-856 (1992). 22. Nayfield, J.G. et al. Potential role of tamoxifen in prevention of breast cancer. Journal of the National Cancer Institute, 83: 1450-1459 (1991). 23. Pavildis, N.A. et al. Clear evidence that long-term low-dose tamoxifen treatment can induce ocular tox- icity. Cancer, 69: 2961-2964 (1992). 24. Touchette, M. Tamoxifen resistance in breast can- cer. Journal of NIH research, 4: 67-72 (1992). 25. Houghton, J. et al. The NATO and CRC trials of adjuvant tamoxifen therapy. In: Jordan, V.C., ed. Long-term tamoxifen treatment for breast cancer. Madison, WI, University of Wisconsin Press (in press). 26. Love, R.R. et al. Symptoms associated with tamoxi- fen treatment in postmenopausal women. Archives of internal medicine, 151: 1482-1487 (1991). 27. Omar, S. et al. Staging of breast cancer and end results of therapy. In: Omar, S. & Contesso, G., ed. Breast cancer. London, Korba, 1988, pp. 33-34. WHO Bulletin OMS. Vol 71 1993 803