BPC 3 Aims: 1. To comprehensively survey candidate genes involved in steroid hormone metabolism, receptor proteins, and genes in the IGF pathway using a two-pronged approach: first, perform deep resequencing of putative functional regions (exons, 5' upstream regions and conserved non-coding regions) in DNA from 190 patients with advanced breast or prostate cancer to identify possible functional mutations: second, perform haplotype analysis across the entire genomic locus of each gene to search for unrecognized regulatory variants 2. To assess the associations of putative functional SNPs (identified by resequencing) and "haplotype tag SNPs" (identified by ancestral segment mapping) in these genes with breast and prostate cancer. 3. To assess, in a subset of these prospective studies, the association of these variants with plasma steroid hormones, IGF-1 levels, and cancer risk. 4. For polymorphisms in receptor proteins associated with risk of breast and prostate cancer, to examine whether these associations vary according to plasma levels of steroid hormones. 5. To assess whether these gene variants interact with lifestyle and anthropometric factors known to be associated with breast and prostate cancers. Specific aims 2 and 5 are addressed in each cohort, while specific aims 3 and 4 are examined in a subset of cohorts. Where appropriate, results will be pooled to increase statistical power. A final specific aim pertains to assessment of the feasibility and value of the consortial approach. 6. To establish a network of epidemiologists, human geneticists, and statisticians to develop collaborative analyses of prospective studies with biologic samples. Following the present approach as "proof of principle", the network will continue on a long-term basis to apply expanding knowledge and new technologies to the assessment of gene-environment interrelationships and cancer risk. Methodologic Aims: 1. To determine if undetected breast tumors influence levels of existing hormone in the months or years prior to diagnosis. 2. To assess whether use of multiple samples over time substantially improves the precision of estimates of hormone levels and relative risk. Other Aims: 1. To assess the relationship of the gene variants studied with a number of different hormonal metabolites in controls. 2. To assess breast cancer risk according to levels of total prolactin and its glycosylated and nonglycosylated forms.

Note: Please see full application and addendums for full list of aims and collaborators.

Stephen Chanock (NCI, DCEG)
Roni Falk (NCI, DCEG)
Thomas Fears (NCI, DCEG)
Michal Freedman (NCI, DCEG)
Richard Hayes (NCI, DCEG)
Wen-Yi Huang (NCI, DCEG)
Michael Leitzmann (NCI, DCEG)
Catherine Schairer (NCI, DCEG)
Rebecca Troisi (NCI, DCEG)
Sholom Wacholder (NCI, DCEG)
Regina Ziegler (NCI, DCEG)
Regina Ziegler (NCI, DCEG)
Robert Hoover (NCI, DCEG)