Oral Contraceptives and BRCA Cancer: A Balancing Act

Women who inherit a pathogenic mutation in the BRCA1 or BRCA2 gene face high lifetime risks of developing breast and ovarian (or fallopian tube) cancer (1). It is suggested that these women are also predisposed to endometrial cancer (particularly the aggressive serous subtype), albeit with a much lower risk than that of breast or ovarian cancer (2,3). The impact of modifiable exposures on cancer risk, particularly exogenous hormones, including oral contraceptives and hormone replacement therapy (HRT), has been studied extensively. They contain similar hormone profiles, although the doses tend to be much higher in oral contraceptive preparations (4-6).

In women with mutations, an increased risk of breast cancer with use of estrogen plus progesterone (but not estrogen alone) HRT has been described (7); however, the impact of oral contraceptive use is uncertain (8,9). In the general population, oral contraceptives transiently increase the risk of breast cancer, but the absolute risk is small given the typically young age of the user (10). This is not so for BRCA mutation carriers, who typically develop breast cancer at a young age. Complicating matters further, oral contraceptives have a protective effect against ovarian cancer (irrespective of mutation status) and are associated with a 50% (or more) reduction in risk with 5 years of use among BRCA mutation carriers (11-13). Prophylactic bilateral mastectomy is the most effective option to manage breast cancer risk in these high-risk women; however, many (if not most) opt for intensified screening with annual MRI imaging (14). Preventive bilateral salpingo-oophorectomy is recommended (by age 40 years for BRCA1 and by age 45 years for BRCA2 mutation carriers) to prevent ovarian and fallopian cancer, although there is a small residual risk of peritoneal cancer (15). The question of whether a concomitant hysterectomy should be performed is not yet resolved.

With this background information, Schrijver et al. posed the question of whether oral contraceptives affect the lifetime risk of developing cancer in BRCA mutation carriers (16). The authors used a simulation approach to estimate the absolute risks and benefits of combination oral contraceptive use on the cumulative incidence (and mortality) of breast, ovarian, and endometrial cancer combined. They created hypothetical cohorts of 10 000 women with a BRCA1 mutation and 10 000 women with a BRCA2 mutation and calculated the expected impact of oral contraceptives on absolute cancer incidence and mortality. They constructed 18 potential scenarios by altering key variables, including risk-reducing surgery, duration of oral contraceptive use, age at first use, and HRT use after oophorectomy. They assumed that the association between oral contraceptive use and risk of cancer among BRCA mutation carriers is similar to that observed in the general population. They applied BRCA-specific breast and ovarian cancer incidence rates but endometrial cancer rates from the Dutch population at large. Risk reduction with mastectomy and oophorectomy were set to 95% and 80% for breast and ovarian cancer, respectively.

They found that among women with both breasts and ovaries intact, oral contraceptive use was associated with a short-term increase in breast cancer incidence but a long-term decrease in ovarian cancer incidence (and endometrial cancer to a lesser extent) (16). After age 40 years, the net benefit of oral contraceptives exceeded the net risk. This is not surprising given that the increased risk of breast cancer with oral contraceptives is transient, whereas the decrease in ovarian cancer risk is lifelong. Findings were similar for carriers of either gene mutation or when mortality was included as the endpoint.

There are various limitations to using a simulation approach. Notably, the hazard rates associated with the exposures were derived from the general population and not from BRCA mutation carriers specifically; furthermore, they may differ for women with a BRCA1 vs a BRCA2 mutation. There is a need for real data on oral contraceptives in cohorts of mutation carriers to confirm the observations of Schrijver et al. (16)

The question remains: “Do oral contraceptives increase or decrease the lifetime risk of developing cancer in BRCA mutation carriers?” The answer is: “It depends.” For a woman who elects for a preventive bilateral mastectomy, there is no downside to taking an oral contraceptive vis-à-vis her cancer risk. For a woman with both breasts intact and an oophorectomy, there is a net increase in her breast cancer risk—but does the size of the net increase in risk warrant consideration of an alternate form of contraception? Perhaps, but I think we would be better served if we were to rely on real data rather than modeled data, given the assumptions made.

This question is particularly important for a young woman who has just recently discovered she is carrying a BRCA1 or BRCA2 mutation and is considering oral contraceptives for the purpose of contraception. Her decision to take the pill or not is likely made well before she is a candidate for preventive surgery, and it may be unsettling to ask her to consider options for preventive surgeries so well in advance. In some cases, young women may seek genetic testing for the sake of knowing if the pill is safe or is best avoided. Knowledge of the impact of contemporary modes of contraception, including injectable, implants, and intrauterine devices, will be important given their increasing popularity among women of reproductive age.

A second outcome in the study by Schrijver and colleagues (16) is the impact of HRT use after oophorectomy on the incidence of breast cancer. These elevated risks are likely to be real given the well-described impact of progesterone signaling on mammary tumorigenesis as well as on breast cancer risk in women with a BRCA1 mutation (7,17,18). This represents a pressing concern given the need to manage symptoms attributed to early surgical menopause (19-21).

These are all important topics for discussion, and following the publication of the simulation study in this issue, these questions are sure to come up more frequently in sessions with genetic counselors and with other health-care providers.

Funding

None.

Notes

Role of the funder: Not applicable.

Disclosures: The author has no disclosures.

Author contributions: Writing, original draft—JK; writing, editing and revision—JK.

Data Availability

No new data were generated or used for this editorial.