A190

Introduction: Mammographic density and circulating sex hormones are two well confirmed predictors of breast cancer risk. While mammographic density was hypothesized to reflect cumulative exposure to estrogens, recent studies have challenged this premise. This prospective study directly examined the independence of circulating hormone levels and mammographic density as predictors of breast cancer risk. Methods: We conducted a nested case-control study within the Nurses' Health Study cohort. Blood samples were collected during 1989 and 1990. The 253 breast cancer cases and 520 matched controls included in the analysis were postmenopausal and not taking postmenopausal hormones at the time of blood collection and mammography. Endogenous hormone levels of estradiol (E2) and testosterone (T) were measured in plasma. Mammographic density was assessed using computer-assisted software. Results: Circulating E2 and T, and mammographic density were all significantly associated with breast cancer risk. In general, the risk of breast cancer associated with mammographic density and circulating sex steroids was unchanged when each was mutually adjusted for the other. The relative risk of breast cancer associated with being in the highest quartile compared to the lowest of mammographic density went from 3.8 to 3.9 (95% CI, 2.2-6.9; P for trend<0.0001) when additionally adjusted for circulating E2. Circulating levels of E2 (OR=2.4, 95% CI 1.4-4.0) and T (OR=2.0, 95% CI 1.2-3.1) were both still associated with breast cancer risk after adjustment for mammographic density. Women in the top tertile of both mammographic density and plasma hormone levels were at a substantially increased risk of breast cancer relative to those in the lowest tertile of both (E2 and density: OR=4.1 95% CI 1.7-9.8; T and density: OR=6.0, 95%CI 2.6-14.0). Although circulating hormone levels did not significantly modify the effect of mammographic density on breast cancer risk, these results suggest that women with high sex steroid levels and high breast density are at a high risk of developing breast cancer. In addition, mammographic density was positively associated with both ER+/PR+ (top vs bottom quartile OR=2.0, 95%CI 1.1-3.7; P-trend 0.08) and ER-/PR- tumors (comparable OR=2.3, 95%CI 0.7-7.7; P-trend 0.12; P-heterogeneity=0.56). Conclusions: Circulating sex steroid levels and mammographic density are strong independent predictors of postmenopausal breast cancer risk. It is plausible that the relationship between premenopausal hormone levels and breast cancer is mediated by premenopausal breast density. Future studies should examine this relation among premenopausal women in which levels of both are higher, and should consider alternative mechanisms by which mammographic density increases the risk of breast cancer.

[Fifth AACR International Conference on Frontiers in Cancer Prevention Research, Nov 12-15, 2006]