We propose to analyze blood samples in a nested case-control manner from the 32,826 participants in the Nurses' Health Study (NHS) who provided samples in 1989-90 and were 43 to 69 years of age at that time. The samples have been stored at greater than or equal to 130 C in liquid nitrogen freezers since collection. We will assay samples from women who were diagnosed after blood collection and matched controls who remained disease-free, thus efficiently utilizing these prospectively collected samples. We propose to build upon our recent positive findings for several plasma hormones and nutrients in relation to breast and colon cancer risk, and to address new hormonal hypotheses in relation to risk of breast, ovarian and colon cancers and myocardial infarction. Specifically, we will examine (1) plasma estrogens, androgens and prolactin levels in relation to breast cancer risk in postmenopausal women, (2) a polymorphism in the catechol-O- methyl transferase (COMT) gene and risk of breast cancer, (3) DHEA, DHEAS and 5-androstene-3beta,17beta-diol and risk of breast and ovarian cancers and myocardial infarction, (4) insulin-like growth factor I and its binding proteins 1 and 3, plasma vitamin D and polymorphisms in the vitamin D receptor, and plasma antioxidant levels - all in relation to risk of breast and colon cancers. The ongoing NHS (CA40356 and HL34594) will provide follow-up and documentation of disease outcomes in addition to providing information on important covariates (such as exogenous hormone use, diet, smoking status, among others) for the proposed study. Participation in the NHS has been high: among the 32,826 women who provided a blood sample, 98 percent continue to complete questionnaires, and vital status has been documented for 99 percent. Overall, the large size of the cohort, the prospective design, the high follow-up rate, the detailed exposure data, and the availability of archived blood specimens provide a unique opportunity to test several hypotheses of public health importance. We also propose to collect a second blood sample from about 18,000 of the women who provided a first blood sample in 1989-90. This will increase our statistical power in future analyses, and allow us to assess in detail the temporal relationships between biomarkers and disease risk and to reduce the attenuating effects of measurement error.
Showing the most recent 10 out of 526 publications
