Alcohol intake and invasive breast cancer risk by molecular subtype and race in the Carolina Breast Cancer Study

Lindsay A. Williams, Andrew F. Olshan, Chui Kit Tse, Mary Elizabeth Bell, Melissa A. Troester

Alcohol is an established breast cancer risk factor, but there is little evidence on whether the association differs between African Americans and whites. Invasive breast cancers (n = 1,795; 1,014 white, 781 African American) and age- and race-matched controls (n = 1,558; 844 white, 714 African American) from the Carolina Breast Cancer Study (Phases I-II) were used to estimate odds ratios (ORs) and 95 % confidence interval (CI) for pre-diagnosis drinks per week and breast cancer risk. African American controls reported lower alcohol intake than white controls across all age groups. Light drinking (0 to ≤2 per week) was more prevalent among African American controls. Moderate-to-heavy drinking was more prevalent in white controls. African Americans who reported drinking >7 drinks per week had an elevated risk compared to light drinkers [adjusted OR, 95% CI 1.62 (1.03-2.54)]. A weaker association was observed among whites [adjusted OR, 95% CI 1.20 (0.87-1.67)]. The association of >7 drinks per week with estrogen receptor-negative [adjusted OR, 95% CI 2.17 (1.25-3.75)] and triple-negative [adjusted OR, 95% CI 2.12 (1.12-4.04)] breast cancers was significant for African American, but not white women. We observed significantly elevated ORs for heavy intake at ages <25 and >50 years of age for African American women only. We found no evidence of statistical interaction between alcohol intake and oral contraceptive use or smoking. Drinking more than seven alcoholic beverages per week increased invasive breast cancer risk among white and African American women, with significant increases only among African American women. Genetic or environmental factors that differ by race may mediate the alcohol-breast cancer risk association.