Supplemental antioxidant intake and plasma protein carbonyl concentrations in the Long Island Breast Cancer Study Project

B56

Plasma protein carbonyl is a marker of oxidative stress associated with cancer risk and the aging process. Among cases and controls in the population-based Long Island Breast Cancer Study Project, we analyzed associations between intake of specific antioxidant supplements, at low and high doses, and plasma protein carbonyl concentrations, taking effect modification by breast cancer treatment into account. Cases were 1508 women diagnosed with in-situ or invasive first primary breast cancer during 1996-97 who were recruited within a few months after diagnosis. Controls were 1556 frequency age-matched women without a history of breast cancer identified through random digit dialing. All participated in an in-person interview to obtain information on prediagnostic exposures and behaviors; 98% of cases and 98% of controls completed a food frequency questionnaire that assessed the use of dietary supplements over the previous 10-15 years, and 73% of both cases and controls provided blood samples. Of the cases who provided blood samples, 39% did so after initiating chemotherapy or radiation therapy. We categorized supplemental intake of vitamin C, vitamin E, zinc and beta-carotene as none, low or high. High intake was defined as greater than twice the amount supplied by a One-A-Day type multivitamin (120 mg vitamin C, 40.2 mcg vitamin E as alpha-tocopherol, 15 mg zinc, and 1200 IU beta-carotene). We created an antioxidant index to measure the combined intake of supplemental antioxidants. Blood plasma samples were analyzed for protein carbonyl concentrations. Among controls, linear regression was used to examine the association of supplemental antioxidant intake on plasma protein carbonyl concentrations. Among cases, linear regression assessed whether treatment modified the association of antioxidant intake on plasma protein carbonyl concentrations. Likelihood ratio tests (LRT) assessed interaction using a significance level of 0.10 due to limited power. Contrary to our hypotheses, among control women, users of high doses of supplemental vitamin C (β-coef 1.16, 95% CI 0.20 - 2.12), vitamin E (β-coef 1.06, 95% CI 0.03 - 2.08), and the antioxidant index (β-coef 1.00, 95% CI -0.001 - 1.99) had higher plasma protein carbonyl concentrations, than non-users. Users of low doses of antioxidant supplements did not differ from non-users. Among women with breast cancer, neither chemotherapy nor radiation therapy affected plasma protein carbonyl concentrations. However, intake of supplemental antioxidants appeared to modify the effect of treatment status. Low-dose users of supplemental vitamin C (LRT p=0.01) and vitamin E (LRT p=0.07) who had received chemotherapy had lower concentrations of plasma protein carbonyl than those who had not received chemotherapy. High-dose users of zinc (LRT p=0.09) who had received radiation therapy had higher plasma protein carbonyl concentrations than those who had not received radiation therapy. Our findings suggest that high doses of antioxidant supplementation are associated with increased plasma protein carbonyl concentrations, that this measure of oxidative stress may be modified by supplemental antioxidant intake during cancer treatment, and that these associations appear to vary with the dose and types of supplemental antioxidants used.