Does breast size modify the association between marnmographic density and breast cancer risk?

Background: Both the absolute and the percent of mammographic density are strong and independent risk factors for breast cancer. Previously, we showed that the association between mammographic density and breast cancer risk tended to be weaker in African American than in White U.S. women. Because African American women have a larger breast size, we assessed whether the association between mammographic density and breast cancer was less apparent in large than in small breasts. Methods: We assessed mammographic density on mammograms from 348 African American and 507 White women, 479 breast cancer patients and 376 control subjects, from a case-control study conducted in Los Angeles County. We estimated odds ratios (OR) for breast cancer with increasing mammographic density, and the analyses were stratified by mammographic breast area. Results: Median breast size was 168.4 cm(2) in African American women and 121.7 cm(2) in White women (P for difference <0.001). For absolute density, adjusted ORs (95% confidence intervals) per increase of 10 cm(2) were 1.32 (1.13-1.54),1.14 (1.03-1.26), and 1.02 (0.98-1.07) in the first, second, and third tertiles of breast area, respectively (P for effect modification by breast area = 0.005). The results for percent density were similar although weaker; adjusted ORs per 10% increase (absolute value) in percent density were 1.22 (1.05-1.40), 1.22 (1.06-1.41), and 1.03 (0.90-1.18 P for effect modification by breast area = 0.34). Conclusion: Our results indicate that the association between mammographic density and breast cancer may be weaker in women with larger breasts.