Cancer Control Research5R03CA128073-02
Eliassen, A. Heather
PLASMA CAROTENOIDS AND BREAST CANCER RISK-POOLED ANALYSIS OF PROSPECTIVE STUDIES
DESCRIPTION (provided by applicant): The epidemiology of breast cancer suggests that lifestyle factors play a role in the etiology. Diet has received much attention as a possible factor in population differences in breast cancer incidence, but results are not entirely consistent. Carotenoids, prominent micronutrients in fruits and vegetables, have been of specific interest because of their antioxidant properties. Weak inverse or null associations have been observed with dietary intake of carotenoids, with stronger associations observed in subgroups, such as premenopausal women and smokers. However, dietary intake assessment is potentially misclassified due to errors in dietary recall as well as individual differences in absorption and metabolism of carotenoids. To date, only six prospective studies with more than 60 cases have been published on plasma carotenoids and breast cancer risk. Most of these studies have reported significant inverse associations with one or more carotenoids, but the specific carotenoid is not consistent across studies. In this application, we propose a collaborative analysis of existing data from these six well- conducted prospective studies to allow a more comprehensive understanding of the association between carotenoids and breast cancer. In this study, we will utilize existing data from six prospective case-control studies of plasma carotenoids and the risk of breast cancer. The following studies are included in this application: Columbia, Missouri; Ume¿, Sweden; New York University Women's Health Study; Washington County, Maryland; Nurses' Health Study; and Women's Health Study. These studies represent >90% of the world's published literature on plasma carotenoids and breast cancer risk. The combined analysis will include 2,348 cases and 2,535 controls. To address the potential of laboratory differences between studies, we will re-assay 20 samples from each study at a central location and use these to recalibrate original carotenoid values to a common scale. Although each individual study was limited by small numbers for subgroup analyses, we will have the power to examine the association of carotenoids with breast cancer risk by menopausal status, smoking status and alcohol consumption, as well as by tumor characteristics (e.g., ER+/PR+ vs. ER-/PR-). This collaborative analysis will allow us to get a clearer answer to the question of whether carotenoids impact breast cancer risk. We will be able to explore the associations of specific carotenoids, as well as examine the associations among subgroups that have been suggested, but underpowered, in previous literature. Given there are few breast cancer risk factors that are modifiable, it is of public health importance to pursue the role of potentially modifiable factors, such as diet, in breast cancer incidence.