Dietary antioxidant intake and risk of breast cancer.

A151

Background: Laboratory and animal studies suggest that dietary antioxidants may reduce breast cancer risk but there are limited epidemiological studies. Methods: We computed the dietary antioxidant intake from an antioxidant food table (based on the Ferric Ion Reducing Antioxidant Power assay) and from dietary data collected by food frequency questionnaires in 1991, 1995 and 1999 from 93,865 women in the Nurses Health Study II. Using multivariate relative risks (RR) and 95% confidence intervals (95% CI), we evaluated the association of antioxidant intakes with breast cancer risk in women aged between 26 and 46 years at baseline in 1991. We also examined the association with levels of smoking status. Results: During 10 years of follow-up, we documented 1,369 cases of invasive breast cancer. The multivariate RR (95% CI) comparing highest to lowest quintiles of cumulative average antioxidant intakes was 0.95 (0.79, 1.15; p-value for test of trend = 0.70). Within levels of smoking status, the multivariate RR (95% CI) comparing highest to lowest quintiles of cumulative average antioxidant intakes among current smokers was 1.62 (1.04, 2.54; p-value for test of trend = 0.02); among past smokers, 0.54 (0.30, 0.99; p-value for test of trend = 0.30) and among never smokers 0.91 (0.72, 1.14; p-value for test of trend = 0.47). The p-value for the test of interaction between antioxidant intakes and smoking status was 0.53. Conclusion: We found no overall association between dietary intakes of antioxidants measured by a total antioxidant assay and risk of breast cancer. However there was a positive association among current smokers and the suggestion of an inverse association among past smokers that requires further study.