Breast cancer and depot-medroxyprogesterone acetate

Bulletin ofthe WorldHealth Organization, 63 (3): 513-519 (1985) © World Health Organization 1985 Breast cancer and depot-medroxyprogesterone acetate WHO COLLABORATIVE STUDY OF NEOPLASIA AND STEROID CONTRACEPTIVES Thepreliminary results ofa study ofthe incidence ofbreast cancer in relation to use of depot-medroxyprogesterone acetate (DMPA) arepresented. Thefindings are based on data from threeparticipating centres in Thailand, and one each in Kenya and Mexico. A relative riskfor breast cancer of 0. 7 was observed in women who had ever usedDMPA; this was not statistically significant. Although no consistent decrease in risk with duration of use was observed, the lowest relative risk (0.5) was observed in women who had used DMPA for three or more years. These findings are based on small numbers and must be considered preliminary. However, they provide no evidence that DMPA increases the risk of breast cancer, and suggest that it may exert a protective effect, particularly in long-term users. Although depot-medroxyprogesterone acetate (DMPA) has been used as a contraceptive in over 80 countries (1), it has not yet been approved for this purpose in the United States of America, in part because of concern that it may increase the risk of breast cancer. This concern is largely based on the development of mammary adenocarcinomas in beagle bitches that received approximately 25 times the usual human dose of DMPA (2). These findings have received detailed scrutiny by individual investi- gators (2) and expert advisory panels to the US Food and Drug Administration (FDA), the US Agency for International Development (AID), and the World Health Organization (1, 3). All have concluded that the beagle is not an appropriate model for humans and that the findings in beagles are of little or no value in assessing the risk of breast cancer in women who use DMPA. The two small but adequately controlled epidemiological studies conducted to date (4, 5) provide no evidence for an increased risk of breast cancer in users of DMPA, although neither study was of sufficient size to detect other than large alterations in risk, and neither assessed risk in long-term users. To assess the influence of DMPA and other steroid contraceptives on risks of mammary, gynaecological, and hepatobiliary malignancies, a collaborative, multinational, hospital-based, case-control study is currently being conducted under the auspices of the ' This report was prepared by David B. Thomas, Liza Noonan, and Anne Whitehead, on behalf of the collaborating investigators listed in Annex 1. All correspondence concerning this report, including requests for reprints, should be addressed to either: Dr David B. Thomas, Fred Hutchinson Cancer Research Center, 1124 Columbia Street, Seattle, WA 98104, USA; or Dr Susan Holck, Special Programme of Research, Development, and Research Training in Human Reproduction, World Health Organization, 1211 Geneva 27, Switzerland. World Health Organization. This report presents preliminary findings of that study on the relationship of DMPA to breast cancer. It is based on data from three participating centres in Thailand, and one each in Kenya and Mexico. At other participating centres, the numbers of women who had used DMPA were insufficient to warrant their inclusion in these analyses. METHODS The background of this study and the methods used have been described previously (6). Briefly, cases are detected in each participating hospital by monitoring new admissions to wards where breast cancer is treated, and by checking outpatient clinics and pathology reports. Cases are restricted to women born after 1930 in all centres except Chiang Mai, where women born up to five years previously are also eligible because DMPA became available earlier there. In all areas, entry into the study is restricted to those who are resident in a defined geographical area served by the hospital and whose neoplasm was not initially detected at a family planning clinic, unless during an initial clinic visit. Controls are selected from among hospitalized women admitted for conditions other than obstetric or gynaecological and thought not to be associated with use of steroid contraceptives. Like the cases, the controls must be residents of a defined area and born after 1925 or 1930, according to the study area. Controls are not matched to individual cases but are selected from specific hospital wards listed in a 4548 -513- WHO COLLABORATIVE STUDY Table 1. Number of breast cancer cases and controls excluded from analysis Cases Controls No. % No. % Total accrued 269 100.0 4501 100.0 No. excluded from analysis: Not interviewed 1 9 7.1 249 5.5 Prior history of breast cancer 2 0.7 2 0.04 Used other injectable contraceptive 2 0.7 88 2.0 No. included in the analysis 246 91.4 4162 92.5 specified order. At the beginning of each week, all English version. In Kenya, interviews are conducted women who were admitted within the previous 48 in the local language, but the answers are recorded hours to a particular ward and who meet the eligibility directly on to the English questionnaire. A calendar is criteria are asked to participate. The same procedure incorporated into the questionnaire to aid in the is repeated on succeeding wards until two controls per establishment of dates of use of steroid contra- case have been selected for the week. This procedure ceptives. is continued on the following week beginning with the Attempts are made to validate selected items on the ward listed after the last one visited. Because this questionnaire, including use of DMPA, by review of procedure initially resulted in the cases tending to be hospital and clinic records. In Mexico, where older than the controls, during the second and subse- injectable contraceptives other than DMPA are also quent years of the study, on alternate weeks only used by appreciable numbers of women, users of an women over 40 years of age were selected. Since this unknown type of injectable contraceptive were study considers other neoplasms in addition to those assumed to have used DMPA if injections had been of the breast, there are more than two controls per received every three months. case of breast cancer. All cases are provisionally diagnosed by a local A standardized questionnaire is used to obtain pathologist. Slides from all participating centres are information on the known and suspected risk factors also sent to a single reference pathologist for review for the neoplasms under study, as well as a complete and classification according to the International obstetric and contraceptive history. The question- Histological Classification of Tumours (7). naire was initially written in English. Spanish and All data are coded and sent to a coordinating centre Thai translations are used in Mexico and Thailand, in Seattle where they are monitored for quality, and the results are subsequently transcribed onto the processed, and analysed. Table 2. Total numbers of cases and controls in each centre and numbers who had ever used DMPA Cases Controls DMPA-users DMPA-users Total Total Centre No. % No. % Kenya 25 1 4.0 301 24 8.0 Mexico 47 0 0.0 695 23 3.3 Siriraj 74 4 5.4 1228 66 5.4 Chulalongkorn 39 0 0.0 1009 75 7.4 Chiang Mai 61 10 16.4 929 193 20.8 Total 246 15 6.0 4162 381 9.2 514 BREAST CANCER AND DMPA Table 3. Relative risks of breast cancer in relation to potentially confounding variables No. of No. of Relative % of controls who Variable Level cases controls risk' had used DMPA Age at menopause Premenopausal 200 3511 1.0 9.7 (years) < 45 18 315 0.7 6.3 > 45 28 336 1.0 6.3 Menopausal status" Premenopausal 200 3506 1.0 9.7 Artificial menopause 9 143 0.7 2.8 Natural menopause 37 508 0.8 7.3 Age at first live birth' < 20 40 1148 1.0 12.5 (years) 20-24 81 1339 1.9 11.5 25-29 34 464 2.4 9.9 > 30 29 197 4.7 12.2 Nulliparous" 62 1011 3.8 1.3 History of benign No 228 4082 1.0 9.2 breast disease Yes 18 80 3.2 7.5 Family history of No 242 4145 1.0 9.1 breast cancer Yes 4 17 5.8 11.8 Age at menarchee < 13 56 990 1.0 6.5 (years) 14-1 5 93 1782 0.9 8.1 > 15 96 1370 1.1 12.5 a Adjusted for age and centre. b Excluding 3 controls who had never menstruated and 2 with unknown menopausal status. ` Excluding 3 controls with unknown age at first live birth. d Including 2 cases and 1 7 controls with previous pregnancy but no live birth. e Excluding 1 case and 17 controls with unknown age at menarche and 3 controls who had never menstruated. For this report, the unconditional logistic regres- sion model for large strata (8) was used to calculate estimates of relative risks adjusted for various poten- tially confounding variables. All potentially con- founding variables were entered into the regression models as stratified variables. RESULTS This report is based on data on controls and those cases considered by the reference pathologist to have carcinoma of the breast, for whom complete data were received at the coordinating centre by 29 December 1983. Table 1 shows the cases excluded from analysis because: (a) they had not been inter- viewed; (b) they had a prior history of breast cancer; or (c) they had used other injectable contraceptives. The two cases and 86 of the 88 controls who had used other injectable contraceptives were from Mexico. The proportions not included for other reasons were similar in all five centres. Table 2 shows the numbers of cases and controls from each centre and the number who had ever used DMPA. Because both the number of controls per case and the proportion of controls that had used DMPA varied among the centres, all estimates of relative risks based on data from all five centres combined were adjusted for centre. They were also adjusted for age because cases tended to be older than controls in all centres, and DMPA use varied with age (data not shown). The distribution of other known risk factors for breast cancer, which could be confounding variables, is shown in Table 3. Risk of breast cancer was found to be reduced in women with an early menopause and, after adjusting for age, to be lower in women with either a natural or artificial menopause than in premenopausal women. Risk increased with age at first live birth, and was elevated in nulliparous women with prior benign breast lesions and a family history of breast cancer. The expected decline in risk with age at menarche was not found. With this exception, these observations are consistent with those of other 515. WHO COLLABORATIVE STUDY Table 4. Relative risks of breast cancer in relation to use of DMPA Relative risks' adjusted for: Category of use of DMPA No. of cases" No. of controls' Age and centre Four variables' Never used 230 3754 1.0 1.0 Some use 15 378 0.6 (0.3-1.0) 0.7 (0.4-1.2) Months of use: 1-12 7 149 0.8 (0.4-1.7) 0.8 (0.4-1.8) 13-36 5 107 0.7 (0.3-1.8) 0.9 (0.4-2.2) > 36 3 103 0.4 (0.1-1.2) 0.5 (0.2-1.5) Unknown 0 19 - Age at first use: < 30 years 7 179 0.8 (0.3-1.7) 1.0 (0.4-2.3) > 30 years 8 199 0.5 (0.2-1.1) 0.5 (0.3-1.1) U Figures in parentheses give 95% confidence limits. b One case and 30 controls excluded because of unknown values for one or more confounding variables. Adjusted for age, centre, nulliparity, and age at birth of first child. investigators and provide reassuring evidence for the validity of the data. As shown in the last column of Table 3, the proportion of controls that had used DMPA increased with age at menarche, and was particularly low in nulliparous women. The relative risk of breast cancer in users of DMPA was adjusted for age, centre, use of oral contra- ceptives, calendar year of menarche, and for each of the variables in Table 3. The year of menarche was included because use of DMPA was found to vary by year of menarche in both cases and controls. Only adjustment for age at birth of first child and nulli- parity (as a combined variable) resulted in an esti- mated value appreciably different from that obtained by controlling only for age and centre. As shown in Table 4, the age- and centre-adjusted relative risk in women who had ever used DMPA was 0.6, and of borderline statistical significance (the upper limit of the 95%o confidence interval was 1.0). However, adjustment also for nulliparity and age at birth of first child gave a relative risk of 0.7 with a 95% confidence interval that included 1.0. The relative risk of breast cancer was lower in women who had used DMPA for more than 3 years than in shorter-term users although, after adjustment for nulliparity and age at birth of first child, there was no clear trend of a decrease in risk with increasing duration (Table 4). The relative risks shown are based on small numbers of users and their 95% confidence intervals all include 1.0. The last two rows of Table 4 show that the reduc- tion in risk of breast cancer in users of DMPA, if any, is largely confined to women with a first exposure after 30 years of age. Relative risks in DMPA users did not vary significantly by age, year of menarche, menopausal status, family history of breast cancer, age at menarche, age at first live birth (including nulliparity), history of benign breast disease, or use of oral contraceptives. However, because of the small numbers of cases that had used DMPA, the power of this study to detect such interactions is low. Too few women had used DMPA before the birth of their first child to allow investigation of the influence of such use on risk of breast cancer. DMPA had been used by only 13 of 1011 nulliparous controls, and by none of the 62 nulliparous cases; it was known to have been used before a first live birth by only 8 of 3148 parous controls and by 1 of 184 parous cases. Small numbers of early users similarly precluded an investigation of the influence of DMPA when first used at a relatively young age. DISCUSSION The preliminary findings presented in this paper clearly provide no evidence that DMPA increases the risk of breast cancer in humans. Results from one small case-control study of 30 cases (4), and one 516 BREAST CANCER AND DMPA prospective study with 7 cases in 5003 users of DMPA (5), provided similar reassurance, with estimated values of relative risks in women who had ever used DMPA of 1.0 and 0.69, respectively. Published results from human studies have so far provided no evidence to suggest an increased risk of breast cancer with DMPA use. The finding of a relative risk of 0.7 in users of DMPA in this study, although not statistically signifi- cant, is compatible with the value of 0.69 observed in the above-mentioned prospective study. Further support for a protective effect is provided by two studies of progestogen administration in conjunction with estrogen therapy at menopause. A prospective study found rates of breast cancer to be lower in women who received estrogens plus cyclic proges- togen (some of which was medroxyprogesterone ac- etate) than in women treated with estrogens only (9). In a small randomized trial, women who received oral medroxyprogesterone acetate plus an estrogen had lower rates of breast cancer than women who received a placebo (10). The finding in the present study that the reduction in risk was greatest in women who had used DMPA for the longest period of time provides additional evidence that DMPA may be protective against breast cancer. This study, although larger than any other con- ducted to date, has not yet accumulated sufficient cases to provide a stable estimate of the relative risk in women who have ever used DMPA, or to allow more detailed analyses. Given the prevalence of use of DMPA in the controls, with the current number of 246 cases studied, there is a 90% chance of detecting actual relative risks of less than 0.4 or greater than 1.9 in women who have ever used DMPA. To provide sufficient data to detect smaller alterations in risk, and to allow for more refined analyses of greater power, data collection is continuing in all col- laborating centres represented in this report. RU RESUME CANCER DU SEIN ET ACtTATE DE MEDROXYPROGESTERONE-RETARD Ce rapport presente les resultats preliminaires d'une etude cas-temoins collective multinationale effectuee en milieu hospitalier sur les relations entre l'acetate de medroxy- progesterone-retard (DMPA) et le cancer du sein. Il est base sur l'analyse de donnees provenant de cinq centres partici- pants, trois en Thaflande, un au Kenya et un au Mexique. Dans chaque hopital participant, on depiste les cas en surveillant les nouvelles admissions dans les salles reservees au traitement du cancer du sein et en procedant A des enquetes sur les consultations externes et les rapports d'anatomo-pathologie. Dans tous les centres, les cas sont limites aux femmes nees apres 1930, sauf A Chiang Mai ou les femmes nees jusqu'A 5 ans auparavant sont egalement etudiees du fait que le DMPA y a et disponible plus t6t. Partout, les cas retenus en vue de cette etude se limitent aux personnes residant dans une zone geographique precise desservie par l'h6pital et dont le neoplasme n'a pas et depiste A l'origine dans un dispensaire de planification familiale, A moins que ce ne soit au cours de la premiere visite. Pour chaque cas on choisit environ deux temoins parmi les femmes hospitalisees pour des raisons autres qu'obste- triques ou gynecologiques et normalement sans rapport avec l'utilisation de contraceptifs steroidiens. Un questionnaire type permet de recueillir des renseigne- ments sur les facteurs de risque, connus et presumes, associes aux neoplasmes etudies ainsi que sur tous les antecedents en matiere d'obstetrique et de contraception. Les frottis de tous les centres sont envoyes A un meme anatomo-pathologiste centralisateur pour examen standard et classification selon la Classification histologique inter- nationale des tumeurs. Toutes ces donnees sont codees puis envoyees A un centre coordinateur, a Seattle, ou, apres un contr6le de qualite, elles sont traitees et analysees. Dans ce rapport, on a utilise le modele de regression logistique inconditionnelle applicable aux strates d'effectif eleve pour calculer les risques relatifs estimatifs (corriges pour differentes variables parasites eventuelles). Toutes ces variables ont et incorporees au modele de regression ou elles servent de base a la stratification. Le present rapport repose sur l'interrogatoire de 4162 temoins et de 246 femmes dont l'anatomo-pathologiste centralisateur a considere qu'elles etaient porteuses d'un cancer du sein. Quinze cas (6%) et 381 temoins (9,2%) avaient precedemment utilise du DMPA; on a estime que le risque relatif de cancer du sein pour les femmes ayant pris du DMPA A une periode quelconque de leur vie etait de 0,7 (intervalle de confiance a 95%: 0,4-1,2) une fois e1imin6e l'influence globale de l'age, du centre, de l'age A la naissance du premier enfant et de la nulliparite. Une correction effectuee pour d'autres facteurs de risque de cancer du sein n'a pas modifie cette estimation de facon appreciable. Le risque avait sa plus faible valeur chez les femmes ayant utilise du DMPA pendant au moins 3 ans (risque relatif=0,5, intervalle de confiance A 95%: 0,2-1,5) ben qu'on n'ait pas pu mettre en evidence de nette tendance A la baisse avec la duree d'utilisation. La diminution apparente du risque se limitait aux femmes qui avaient pris du DMPA pour la premiere fois apres l'age de 30 ans. Le nombre de cas rassembles dans cette etude est encore insuffisant pour qu'on ait une estimation stable du risque relatif chez les utilisatrices de DMPA A une periode quel- 517 518 WHO COLLABORATIVE STUDY conque de leur vie et pour qu'on puisse proceder a des analyses plus fines, de puissance statistique convenable: par consequent, la collecte des donnees se poursuit dans tous les centres dont on a parle dans ce rapport. Cependant, il est certain que ces resultats preliminaires ne temoignent nullement d'une augmentation du risque de cancer du sein chez les femmes par le DMPA, mais donneraient plutot a penser qu'il a un r6le protecteur. REFERENCES 1. Facts about injectable contraceptives: Memorandum from a WHO meeting. Bulletin of the World Health Organization, 60: 199-210 (1982). 2. FRASER, 1. S. & WEISBERG, E. A comprehensive review of injectable contraception with special emphasis on depot-medroxyprogesterone acetate. Medical journal of Australia, I (Suppl.): 1-9 (1981). 3. ROSENFIELD, A. ET AL. The Food and Drug Admini- stration and medroxyprogesterone acetate. What are the issues? Journal of the American Medical Association, 249: 2922-2928 (1983). 4. GREENSPAN, A. R. ET AL. The association of depot- medroxyprogesterone acetate and breast cancer. Contraception, 21: 563-569 (1980). 5. LIANG, A. P. ET AL. Risk of breast, uterine corpus, and ovarian cancer in women receiving medroxypro- gesterone injections. Journal of the American Medical Association, 249: 2909-2912 (1983). 6. WHO Collaborative Study of Neoplasia and Steroid Contraceptives. Invasive cervical cancer and combined oral contraceptives. British medical journal, 290: 961-965 (1985). 7. SOBIN, L. H. ET AL., ed. A coded compendium of the International Histological Classification of Tumours. Geneva, World Health Organization, 1978. 8. BRESLOW, N. E. & DAY, N. E. Statistical methods in cancer research. Vol. 1. The analysis of case-control studies. Lyon, International Agency for Research on Cancer, 1980 (IARC Scientific Publication, No. 32). 9. GAMBRELL, R. D. JR. ET AL. Decreased incidence of breast cancer in postmenopausal estrogen-progestogen users. Obstetrics and gynecology, 62: 435-443 (1983). 10. NACHTIGALL, L. E. ET AL. Estrogen replacement therapy. II: A prospective study on the relationship to carcinoma and cardiovascular and metabolic problems. Obstetrics and gynecology, 54: 74-79 (1979). Annex I PARTICIPATING CENTRES AND INVESTIGATORS The data collection centres and investigators were as follows: University of Nairobi, Nairobi Centre for Research in Reproduction, Nairobi, Kenya J. K. G. Mati, Patrick R. Kenya, Alfred Kungu,' D. Gatei.' Mexico General Hospital, Mexico City, Mexico Hector Rodriguez Cuevas,a Socorro Benavides Salazar,b Jorges Albores Saavedra,c Patricia Ontiveros." Principal investigator. " Co-investigator. Pathologist. Chiang Mai University, Faculty ofMedicine, Chiang Mai, Thailand Suporn Silpisornkosol,a Tieng Pardthaisong, b Nimit Martin,b Choti Theetranont.c Chulalongkorn University, Faculty of Medicine, Department ofObstetrics and Gynaecology, WHO Collaborating Centre for Research in Human Reproduction, Bangkok, Thailand Banpot Boonsiri,a Pramuan Virutamasen, b Chansuda Wongsrichanalai,/ Prasarn Jimakorn.c Mahidol University, Faculty of Medicine, Siriraj Hospital, Department of Obstetrics and Gynae- cology, Family Planning Research Unit, Bangkok, Thailand Suporn Koetsawang," Daungdao Rachawat,b Nivat Chantarakul.' BREAST CANCER AND DMPA Coordinating Centre Fred Hutchinson Cancer Research Center, Division of Public Health Sciences, Seattle, WA, USA David B. Thomas (Study Coordinator), Diane Roseman (Epidemiologist), Anne Whitehead (Statistician), Liza Noonan (Statistician). World Health Organization Susan Holck, Special Programme of Research, Development and Research Training in Human Reproduction, World Health Organization, Geneva, Switzerland. Reference pathologist Helge Stalsberg, University of Tromso, Institute of Medical Biology, Tromso, Norway. 519