Search

Background: Several recent studies suggest that adequate dietary folate may attenuate the risk of breast cancer associated with intake of alcohol. We examined whether the putative benefit extends to women with a family history (FH) of breast cancer using a cohort of 33,552 postmenopausal women aged 55-69 years in 1986., Method: Folate and alcohol intake was estimated from a food frequency questionnaire completed at baseline. Folate was categorized as upper 50th, 31st-50th, 11th-30th, and <10th percentiles. Alcohol use was initially classified into three levels; never drinkers, less than the median and greater than the median. Subsequent models collapsed levels of intake to any versus none. Occurrence of breast cancer was determined through linkage to the Iowa SEER registry. Multivariate-adjusted relative risks (RR) and 95% confidence intervals (CI) were estimated through Cox proportional hazards regression, stratified on FH using non-drinkers with high folate and no FH as the referent group., Results: Through 14 years, 1823 incident cases were identified, 308 among FH+ women. Among FH- women, low folate was not a risk factor among non-drinkers (RR = 0.96, CI = 0.73-1.26), but was among drinkers (RR = 1.40, CI = 1.05-1.86). Drinkers with high folate were not at elevated risk (RR = 1.03, CI = 0.89-1.19). Among FH+ women, low folate was a risk factor among drinkers (RR = 2.21, CI = 1.43-3.41) and non-drinkers (RR = 2.39, CI = 1.36-4.20). Further, drinkers with high folate remained at increased risk (RR = 1.67, CI= 1.30-2.14). However, FH+ women with high folate who did not drink alcohol had no elevated risk., Conclusion: These results suggest that folate may attenuate the risks of postmenopausal breast cancer associated with family history, but only if alcohol use is avoided or minimized.

In a cohort of 27,578 postmenopausal Iowa women, we examined whether the risk with a family history of breast cancer differs by self-reported ethnicity. A total of 1042 breast cancer cases occurred over 10 years of follow-up. Using a phylogenetic tree, ethnicities were combined into five groups: Scandinavian; English, Scottish, Welsh, and Dutch (ESWD); Irish; German; and Other European. The incidence of breast cancer did not differ significantly by ethnicity, although the highest rates were observed among Scandinavian women (488 per 100,000 per year) and the lowest among Irish women (353 per 100,000 per year). The prevalence of a family history of breast cancer was not significantly associated with ethnicity when only first-degree relatives were considered (P = 0.17), but inclusion of data on second-degree relatives increased the statistical significance of the association (P = 0.003). Differences in mean levels of breast cancer risk factors between ethnicities were generally small but statistically significant. Proportional hazards regression was performed to evaluate potential interactions of family history with ethnicity on breast cancer incidence. A family history of breast cancer was associated with increased relative risks among ESWD, Germans, and Other Europeans but not among Irish and Scandinavians. Relative risk estimates were not attenuated upon addition of known breast cancer risk factors to the model, implying that the distribution of these risk factors by ethnicity is unlikely to explain some of the observed ethnic-specific differences between family history and risk of breast cancer. Results of this study could have implications for studies of common genetic polymorphisms and cancer risk.