Tamoxifen and contralateral breast cancer in BRCA1 and BRCA2 carriers: An update
Article first published online: 5 DEC 2005
Copyright © 2005 Wiley-Liss, Inc.
International Journal of Cancer
Volume 118, Issue 9, pages 2281–2284, 1 May 2006
How to Cite
Gronwald, J., Tung, N., Foulkes, W. D., Offit, K., Gershoni, R., Daly, M., Kim-Sing, C., Olsson, H., Ainsworth, P., Eisen, A., Saal, H., Friedman, E., Olopade, O., Osborne, M., Weitzel, J., Lynch, H., Ghadirian, P., Lubinski, J., Sun, P., Narod, S. A. and The Hereditary Breast Cancer Clinical Study Group (2006), Tamoxifen and contralateral breast cancer in BRCA1 and BRCA2 carriers: An update. Int. J. Cancer, 118: 2281–2284. doi: 10.1002/ijc.21536
- Issue published online: 21 FEB 2006
- Article first published online: 5 DEC 2005
- Manuscript Accepted: 15 JUN 2005
- Manuscript Received: 24 MAR 2005
- US National Institutes of Health. Grant Numbers: 1R01 CA63682, CA63678, CA57601, CA61231, CA74415
- Canadian Breast Cancer Alliance, The Department of the Army. Grant Numbers: DAMD17-94-J-4299, DAMD 17-94-J-4260, DAMD 17-97-J 4340
- breast cancer;
Women with a mutation in BRCA1 or BRCA2 face a lifetime risk of breast cancer of ∼80%, and following the first diagnosis the10-year risk of contralateral breast cancer is ∼30%. It has been shown that both tamoxifen and oophorectomy prevent contralateral breast cancer, but it is not clear whether there is a benefit in giving tamoxifen to women who have previously undergone an oophorectomy. Furthermore, the relative degree of protection in BRCA1 and BRCA2 carriers has not been well evaluated. We studied 285 women with bilateral breast cancer and a BRCA1 or BRCA2 mutation, and 751 control women with unilateral breast cancer and a BRCA1 or BRCA2 mutation in a matched case-control study. Control women were of similar age and had a similar age of diagnosis of breast cancer and had been followed for as long as the case for a second primary breast cancer. The history of tamoxifen use for treating the first breast cancer was compared between bilateral and unilateral cases. The multivariate odds ratio for contralateral breast cancer associated with tamoxifen use was 0.50 for carriers of BRCA1 mutations (95% CI, 0.30–0.85) and was 0.42 for carriers of BRCA2 mutations (95% CI, 0.17–1.02). The protective effect of tamoxifen was not seen among women who had undergone an oophorectomy (OR = 0.83; 95%CI, 0.24–2.89) but this subgroup was small. In contrast, a strong protective effect of tamoxifen was apparent among women who were premenopausal or who had undergone natural menopause (OR = 0.44; 95% CI, 0.27–0.65). © 2005 Wiley-Liss, Inc.
The lifetime risk of breast cancer in women who carry a deleterious BRCA1 or BRCA2 mutation is ∼80%,1 and following an initial diagnosis of breast cancer the annual risk of contralateral breast cancer is ∼3%.2 Both tamoxifen and oophorectomy have been shown to reduce the risk of contralateral breast cancer in carriers of BRCA1 or BRCA2 mutations.2, 3 In our previous case-control study of 209 bilateral cases and 384 unilateral controls we reported a reduction in contralateral breast cancer risk of 51% associated with tamoxifen use, but the observed protective effect was significant only in BRCA1 carriers. Since this report was published in 2000, we have continued to accrue study subjects to our BRCA carrier registry and we are able to readdress the question of tamoxifen protection with a greater degree of precision and are able to evaluate the magnitude of the protective effect separately for carriers of BRCA1 and BRCA2 mutations. It is also of interest to establish whether or not there is a protective effect of tamoxifen in women who have previously undergone an oophorectomy and to evaluate the protective effect separately for pre and postmenopausal women.
Material and methods
Information on patients with hereditary breast cancer was submitted to the study center by investigators at each of 49 contributing centers in 10 countries. These centers were requested to complete data forms for all known cases of female breast cancer, unilateral or bilateral, carrying a verified BRCA1 or BRCA2 mutation. The majority of these cases were identified through genetic counseling and risk assessment programs were offered to women from high-risk families. The data center received information on a total of 2,972 cases of invasive breast cancer in carriers of pathogenic BRCA1 or BRCA2 mutations.
Among the 2,972 cases, there were 611 cases of bilateral breast cancer (20.6%). Bilateral cases were excluded from the current study if the first cancer was diagnosed prior to January 1, 1970 (i.e. before tamoxifen was in use (n = 180)), if the contralateral cancer occurred within 1 year of the diagnosis of the initial breast cancer (n = 47) or if the case was diagnosed with ovarian cancer at any time prior to the contralateral breast cancer (n = 28). Synchronous bilateral cases were excluded because there would be no opportunity for the patient to take tamoxifen before the onset of the contralateral cancer. A total of 356 eligible cases of bilateral breast cancer was identified.
Women with unilateral breast cancer in the registry database were eligible to serve as controls. Controls were born within 3 years of the birth date of the case, and were diagnosed with breast cancer at an age within 2 years of the age of the first diagnosis of breast cancer of the case. Cases and controls were carriers of mutations in the same gene (BRCA1 or BRCA2). Cases and controls were also matched for oophorectomy (yes/no—1 or more years prior to the age of second cancer diagnosis in the bilateral case). Each control was followed for a period at least as long as the follow-up period of the matched case. Women were ineligible to serve as controls if they were diagnosed prior to 1970, if they had a contralateral mastectomy or if they had a diagnosis of ovarian cancer prior to, or during, the follow-up period. For each bilateral case, we attempted to identify one or more unilateral control patients. No woman received tamoxifen prior to the diagnosis of the initial breast cancer. All of the exposures in cases and controls are defined for the time-period equivalent to the period before the contralateral cancer of the matched case.
Because of the requirement for genetic testing of the study subjects (which is available to living women) these represent prevalent cases; the median year of diagnosis of the first breast cancer was 1992 (range, 1971–2000). Our study was restricted to living cases, because it is only possible to perform mutation analysis on living women, and because risk factor information was obtained by questionnaire. The questionnaires were completed between the years 1980 and 2004, on average 7.5 years after the diagnosis. Mutation analysis was performed using several established detection techniques, and all mutations were confirmed by direct sequencing of DNA samples.
The case and control women completed a questionnaire that asked about the medical and surgical treatment of the initial breast cancer. In some centers the questionnaire was administered by telephone interview. The year of diagnosis of the initial and contralateral breast cancer was recorded. Women were asked if they had received tamoxifen as a treatment for the first breast cancer, and if so, the dose, dates and duration of tamoxifen treatment. In this study, tamoxifen use was defined as that given as treatment for the initial breast cancer, and not for the treatment of recurrent cancer. Women were also asked to report on the type of surgical treatment they received (lumpectomy, unilateral mastectomy or bilateral mastectomy). Women were asked if they had undergone bilateral oophorectomy, either before cancer diagnosis or thereafter. Women were asked about chemotherapy (yes/no) and radiotherapy (yes/no) as treatment of the initial breast cancer. Additional variables of interest included smoking, reproductive history and ethnicity.
A matched case-control analysis was performed. The frequency of tamoxifen use was compared between the bilateral cases and unilateral matched controls. The univariate odds ratios (and p-values) for contralateral breast cancer associated with tamoxifen use were calculated using conditional logistic regression. The odds ratio for contralateral breast cancer associated with tamoxifen use were adjusted for the other covariates, including other treatments received (radiotherapy and chemotherapy), smoking (ever/never), parity and oral contraceptive use (ever/never). To estimate the protective effect of tamoxifen separately for BRCA1 and BRCA2 carriers, and for women with and without oophorectomy, estimates were generated for these subgroups using the matched subsets. All calculations were done using the SAS Statistical Package.
The characteristics of the 285 cases of bilateral breast cancer and 751 unilateral controls were similar (Table I). The average time duration from the first cancer to the diagnosis of contralateral cancer was 5.7 years for the cases. The average period of follow-up was 7.4 years for the controls, and each control was followed for a period at least as long as that of the matched case.
|Bilateral cases||Unilateral controls|
|Year of birth (mean)||1948.0||1948.0|
|Year of diagnosis (mean ± SD)||1987.6 ± 6.2||1988.7 ± 6.2|
|Year of interview (mean ± SD)||1999.5 ± 2.5||1999.3 ± 2.3|
|Age of 1st breast cancer (mean ± SD)||40.1 ± 7.5||41.0 ± 7.5|
|Age of 2nd breast cancer (mean ± SD)||45.8 ± 8.0|
|Place of residence, N|
|Canada||59 (21%)||199 (27%)|
|United States||119 (42%)||249 (33%)|
|Europe||95 (33%)||263 (35%)|
|Israel||12 (4%)||40 (5%)|
|Oral contraceptives (ever)||60.7%||62.4%|
Tamoxifen use was reported by 12.3% of the bilateral cases and by 24.6% of the unilateral controls (p < 0.0001). The univariate odds ratio for tamoxifen use and contralateral breast cancer was 0.45 (95% CI, 0.29–0.70) (Table II). The results of the multivariate analysis, controlling for other treatments, ethnic group, oral contraceptive use, parity and smoking were similar (Table 2).
|Case (N = 285)||Control (N = 751)||Univariate analysis||Multivariate analysis|
|OR (95% CI)||p-value||OR (95% CI)||p-value|
|Ever||35||185||0.45 (0.29–0.70)||0.0004||0.47 (0.30–0.74)||0.001|
|Ever||24||137||0.48 (0.29–0.79)||0.004||0.50 (0.30–0.85)||0.01|
|Ever||11||48||0.39 (0.16–0.94)||0.03||0.42 (0.17–1.02)||0.05|
There were 848 subjects with BRCA1 mutations (228 bilateral cases and 620 unilateral controls) and there were 188 subjects with BRCA2 mutations (57 bilateral cases and 131 unilateral controls). On average, the BRCA1 carriers were diagnosed with their first cancer at 40.3 years and the BRCA2 carriers were diagnosed at 44.6 years. Tamoxifen use was reported for 19% of the BRCA1 carriers and for 31% of the BRCA2 carriers. The protective effect of tamoxifen was similar in both BRCA1 and BRCA2 carriers; among the BRCA1 cases and matched controls, the univariate odds ratio was 0.48 (95% CI, 0.29–0.79) and among the BRCA2 carriers the odds ratio was 0.39 (95% CI, 0.16–0.94).
Cases and controls were matched for oophorectomy status. This was done to evaluate the protective effect of tamoxifen separately for women with and without intact ovaries. The results suggest that tamoxifen is protective against contralateral breast cancer in women with intact ovaries (Table III) but the subgoup of oophorectomized women was small (26 cases and 123 controls). A protective effect was not observed among women who had undergone oophorectomy (OR = 0.83). It is also of interest to establish whether tamoxifen is protective after natural menopause (Table IV). The observed protective effects of tamoxifen were similar for women who were diagnosed with their first breast cancer before menopause (OR = 0.54; 95% CI, 0.27–1.05) and those initially diagnosed after menopause (OR = 0.33; 95%CI, 0.11–1.01). There did not appear to be any residual protection offered by tamoxifen beyond 10 years of the first breast cancer diagnosis (Table V).
|Status||Odds ratio||(95% CI)||p-value|
|Oophorectomy (n = 26)||0.83||(0.24–2.89)||0.77|
|No oophorectomy (n = 259)||0.44||(0.27–0.65)||0.0009|
|N||Odds ratio||95% CI||p-value|
|Both cancers premenopausal||86||0.31||(0.11–0.82)||0.02|
|First cancer premenopausal, second cancer postmenopausal||114||0.54||(0.27–1.05)||0.07|
|Both cancers postmenopausal||37||0.33||(0.11–1.01)||0.05|
|Natural menopause only|
|First cancer premenopausal, second cancer postmenopausal||22||0.13||(0.02–1.10)||0.06|
|Both cancers postmenopausal||13||0.45||(0.09–2.37)||0.34|
|Years since diagnosis of first primary breast cancer|
|1–5 years (N = 168 sets)||5–10 years (N = 68 sets)||>10 years (N = 49 sets)|
|Tamoxifen, any use, all carriers|
|Ever||0.46 [0.2–0.79] 0.005||0.42 [0.16–1.10] 0.08||0.99 [0.13–7.61] 0.99|
We observed that the risk of contralateral breast cancer was reduced by more than 50% in carriers of BRCA1 and BRCA2 mutations when tamoxifen was given as treatment for the initial breast cancer. In our earlier smaller study3 a significant effect was seen only for the BRCA1-positive subgroup; now the effect is statistically significant for BRCA2 carriers as well. The current study extends our previous study by the addition of 15 new centers and 2 new countries (Israel and Austria). The number of study subjects is approximately double that of our previous report (∼40% of the subjects in the current study were included in that report).
We found tamoxifen to be almost equally effective in BRCA1 and BRCA2 carriers. We did not have the estrogen-receptor status of the breast cancers in this study; however, it is expected that the majority of BRCA1-associated breast cancers will be ER-negative and that the majority of BRCA2-associated breast cancers will be ER-positive.4, 5, 6, 7, 8, 9 These and previous data suggest that tamoxifen is effective in preventing ER-negative second primary cancers. The estrogen-receptor status of the first breast cancer does not influence the actuarial risk of contralateral breast cancer in BRCA1 carriers.2 There is a greater than the expected degree of concordance in the ER-status in bilateral breast cancers in BRCA1 carriers; nevertheless, we previously found that 54% of second primary breast cancers, which arise in BRCA1 carriers following an ER-positive breast cancer, are ER-negative.10 It is therefore quite unlikely that our observed overall reduction of 50% could be due entirely to the prevention of ER-positive breast cancers. Furthermore, we have shown that prior tamoxifen use is not associated with an increased proportion of ER-negative breast cancers, as would be the case if tamoxifen preferentially prevented ER-positive breast cancers.2, 10 In studies on contralateral breast cancer in noncarriers, tamoxifen treatment of the first cancer has been shown to be predictive of the ER-status of the contralateral cancer; in particular, women who had used tamoxifen were more likely to be diagnosed with ER-negative cancer than women who did not.11 These data suggest that in the noncarrier breast cancer population, tamoxifen selectively prevents ER-positive second primary breast cancers. This was also the conclusion of the NSABP P1 prevention study regarding first primary cancers.12 The biological basis for this difference is not yet known.
Premenopausal oophorectomy in BRCA1 carriers also significantly reduces the risk of first primary and contralateral breast cancer.13, 14, 15 By matching our cases and controls on oophorectomy status we were able to address the possibility of a joint protective effect of oophorectomy and tamoxifen on contralateral breast cancer risk. We did not observe a significant protective effect on contralateral breast cancer risk with tamoxifen among the subgroup of women who had previously undergone an oophorectomy. However, this subgroup was small; there were only 26 women who developed contralateral breast cancer following oophorectomy in the study population, attesting to the strong protective effect of oophorectomy. The lack of an additive effect of oophorectomy and tamoxifen in the present study contrasts with the results of the historical cohort study of Metcalfe, et al.2 They reported an odds ratio of 0.09 for the risk of contralateral breast cancer in young BRCA carrier women who had undergone oophorectomy and had taken tamoxifen as well, compared to women having neither treatment. The reason for the discrepancy in the 2 studies is not known.
Our study does not permit us to recommend the optimum duration of tamoxifen for chemoprevention, because of our study design, and the small size of our subgroups. Because this is a case-control study we consider only tamoxifen use in the controls during the period prior to the time diagnosis in the case and therefore these values do not accurately reflect the entire course of tamoxifen use by the patient.
The results of the present study confirm our earlier report. There are more cases and controls in the present series and our matching criteria are now more stringent. In addition, in the present study, cases and controls were matched on oophorectomy status. Since 2000 there has been only one other report which attempted to evaluate tamoxifen use in BRCA carriers. King, et al. observed no protective effect of tamoxifen in BRCA1 carriers identified with breast cancer within the NSABP prevention trial.16 However, the number of study subjects with BRCA1 mutations in that study was extremely small (N = 8) and no conclusion could be drawn. In the present study, which includes 848 BRCA1 carriers, a clear protective effect was observed. Given the large number of centers participating in the current study it is unlikely that a study of similar magnitude will report on this topic in the near future. No randomized trials are currently underway of tamoxifen use in unaffected BRCA carriers.
There are several limitations to our study. The interview took place on average 7 years after the diagnosis of the first cancer. It is quite unlikely that the women would not accurately recall if they had ever used tamoxifen, but it is possible that women who survived for this amount of time may not be completely representative of the breast cancer population at large.
On the basis of these data, we believe that it is reasonable to offer tamoxifen to young women with intact ovaries and a BRCA1 or BRCA2 mutation for the prevention of contralateral breast cancer. Many authors recommend oophorectomy to women with breast cancer and a BRCA1 mutation (primarily to prevent second primary ovarian cancer),17 and it is not clear if these women would also benefit from tamoxifen. However, tamoxifen does appear to be effective as chemoprevention in women who have undergone natural menopause.
It is compelling to speculate that tamoxifen given prophylactically to carrier women with no history of breast cancer will reduce the incidence of first primary breast cancer in premenopausal women. In these women, oophorectomy will reduce the breast cancer risk by about one-half.15 Our current data do not permit us to recommend tamoxifen to unaffected carriers who have undergone a prophylactic oophorectomy, and more research in this area is needed. It is possible that other antiestrogenic regimens, such as an aromatase inhibitor, will be effective in oophorectomized women and that maximal protection will be afforded by combination therapy.
- 15Oophorectomy and breast cancer in BRCA carriers. J Clin Oncol, in press., , , , , , , , , , , , et al.