Oral Contraceptives and the Risk of Breast Cancer

Abstract & Commentary

Synopsis: Among women 35-64 years of age, current or former use of oral contraceptives was not associated with a significantly increased risk of breast cancer.

Source: Marchbanks PA, et al. N Engl J Med. 2002; 346:2025-2032.

Marchbanks and colleagues report the results of a large population-based, case-control study to determine the risk of breast cancer among former and current users of oral contraceptives (OCs). The study population was formed from 4575 women with invasive breast cancer and 4682 matched controls enrolled from Atlanta, Detroit, Philadelphia, Los Angeles, and Seattle. The Centers for Disease Control and Prevention conducted the data coordination. Controls came from the same geographic location. Participants were interviewed using a standardized questionnaire that included photographs about hormonal medications. Sixty-five percent of the women were white and 35% were black. Marchbanks et al state that examination of multiple aspects of OC use revealed little evidence that OCs increase the risk of breast cancer. In particular, neither dose nor duration was associated with an increased risk. Further, among women 45-64 years of age, the risk of breast cancer was not significantly higher among the women currently using OCs containing a low dose of estrogen than among those who had never used OCs. Fully one third of the cases were postmenopausal (n = 1544).

Comment by Sarah L. Berga, MD

This article could be subtitled "when negative results are positive (good)." This is a large study that was properly conducted, and addressed an important and timely topic. The results are reassuring and certainly welcome. But like most good studies, it raises a number of questions that are outside the immediate focus of the study. One of the most compelling questions from my vantage point is why would OC use not raise the risk of breast cancer in older women while HRT use would? This apparent paradox is even more unsettling when one considers that HRT formulations typically contain much smaller amounts of sex steroids and therefore qualify as a lesser exposure in terms of dose. For instance, femHRT contains 5 mg of ethinyl estradiol (EE) while OCs contain between 20-35 mg of EE. There is not a straightforward answer to this question, of course, but there are some candidate explanations that I would like you to consider.

A major difference between HRT and oral contraceptive pills (OCPs) other than dose is that the most commonly used HRT preparation contains a different estrogen type, namely, conjugated equine estrogens (CEE), while all OCP formulations now contain EE. Could it be that the effect of these 2 estrogens, even on a molar basis, upon the breast is not the same? If so, might this be due to the effective half-life of the metabolites? Or is it, as has been hypothesized, due to the high sulfatase of the breast making conjugated estrogens more bioavailable to breast than to other tissues with lesser sulfatase activity? In other words, in giving CEE, are some tissues overdosed relative to others and is this effect not operant for EE? Further, I have worried that to suppress hot flashes, one must give a relatively large dose of CEE because the conjugated moiety cannot transgress the blood-brain barrier. It is only the free fraction that can get into important brain centers. Since giving CEE orally raises SHBG, the free fraction is small. Therefore, to get enough into the brain to quell hot flashes, one must overdose the periphery. The diffusion barrier for EE into the brain might not be the same, although this remains to be shown. Also, since EE is not sulfated in the circulation to the same extent as CEE, the relative exposure of the breast might be lower at a dose that is bioactive for brain or bone.

Another major difference between HRT and OCPs is the progestins used in the most common HRT product and in OCPs. Medroxyprogesterone acetate is a 21-carbon derivative while OCPs contain 19-nortestosterone derivatives. We know little about the differential effects of these progestins upon the breast.

In short, while we try to make sense of an abundance of data on the effects of HRT and OCPs, we must remember not to lump all HRT products together. Perhaps this same argument holds for OCPs, but at least they all contain the same estrogen, even if not in the same dose. The time has come to think critically about how estrogens differ from one another if we are to understand the apparent paradox of the present study. The cartoon that estrogen exposure per se causes breast cancer in a dose- and duration-related manner is outmoded. The present study argues against such a simple conceptualization.

Dr. Berga is Professor and Director, Divison of Reproductive Endocrinology and Infertility, University of Pittsburgh.