Teasing out the association between oral contraceptives and breast cancer risk

Breast cancer is the most commonly diagnosed cancer among women in the US. Although the breast cancer death rate has substantially declined over the past three decades, it remains one of the top causes of cancer death in women. There are several known modifiable and non-modifiable risk factors for breast cancer. One modifiable factor associated with increased risk is recent use of oral contraceptives, as demonstrated in previous epidemiological studies. However, results from studies have been mixed and whether the use of oral contraceptives is differentially associated with breast cancer molecular subtypes is not yet fully understood. To gain additional insight, researchers from Dr. Christopher Li’s research group in the Public Health Sciences Division conducted a population-based case-case study of premenopausal women. This study was led by Nicole Lorona, graduate student in the Epidemiology Program, and recently published in the journal Hormones and Cancer.

Previous studies were limited by low numbers of participants with the less common subtypes and an inability to assess risk by oral contraceptive type (estrogen and progestin). In addition, some of these studies included pre- and post-menopausal women, which may have contributed to differences in results across some studies. The design of the current study allowed for many of these limitations to be addressed.

The new study included 1,701 premenopausal women diagnosed with breast cancer between June 2004 and June 2015. All women were from either the Seattle-Puget Sound or Albuquerque, NM areas. The authors used medical records to classify breast cancer cases into four different molecular subtypes based on expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2): luminal A (ER+/HER2-), luminal B (ER+/HER2+), H2E (ER-;HER2+), and triple-negative (ER-/PR-/HER2-). The use of oral contraceptives during the five years preceding breast cancer diagnosis was determined through review of medical records or interview-administered questionnaires to the participants. The authors categorized contraceptive use by recency of use into four classes: current (within the year before diagnosis), between one and five years before diagnosis, within the five years before diagnosis, or no use in the five years before diagnosis. For all statistical analyses, the luminal A subtype was used as the reference group and no use of oral contraceptives in the five years before diagnosis was the unexposed group.

Luminal A was the most prevalent subtype among the study participants (~46% of cases), followed by triple-negative (~33%), and H2E and luminal B (each ~11%). Relative to no oral contraceptive use in the five years preceding diagnosis, current use and use within the preceding five years were both associated with a significantly lower odds of H2E breast cancer as compared to luminal A. In addition, increased duration of use was also associated with significantly lower odds of H2E subtype.

In contrast to the H2E group, there were no significant associations between recency or duration of oral contraceptive use and luminal B or triple-negative subtypes. The authors also assessed whether the dose of estrogen or the type of progestin was associated with risk for triple-negative breast cancer but did not find any significant differences.

Lorona summarized the overall results, “Our findings suggest that among young women, oral contraceptive use may have a stronger association with risks of luminal A or triple-negative subtypes of breast cancer than the HER2-overexpressing subtype. These results may be consistent with what is known about the biology of breast cancer. “This contributes to the literature that molecular subtypes of breast cancer differ in etiology, and that some established breast cancer risk factors may be more strongly associated with an increased risk of certain breast cancer subtypes, but not others,” added Lorona. The authors suggested that differences in parity among the groups may in part account for the differential risk, as there is an inverse association between parity and breast cancer risk. Compared to the other groups, women in the H2E group were less likely to be nulliparous.

While the new study adds to our understanding of the association between oral contraceptive use and breast cancer risk, the authors indicated that there is still additional work to do. “Further studies are needed to understand why oral contraceptives may increase risk of only certain subtypes of breast cancer. These elevations in risk could be due to the influence these hormones have on the growth of different types of breast cancer cells or could be correlated with other factors related to breast cancer risk such as pregnancy timing and number,” said Lorona.