Identification of Genes Associated With Cancer in Patients and Siblings Who Have Cancer

OBJECTIVES: * Gather allele-sharing statistics at approximately 100 candidate loci throughout the human genome most likely to influence genetic risk of cancer. * Use these allele-sharing statistics to test the interaction of each locus individually with cancer-associated, rare alleles of HRAS1. * Generalize this approach for one of four cancers (breast, colon, lung, or prostate), using allele-sharing statistics to test the interaction of each locus with every other locus. * Replicate positive results in a distinct set of sibling pairs with cancer. * Examine, when loci contributing to risk are detected, the influence of this genetic background on clinical outcomes, such as survival. OUTLINE: Each patient-sibling pair completes a family history questionnaire about the incidence of cancer in the family. Blood samples are obtained from both the patient and the sibling (and both living parents, if available). The blood samples are genotyped using approximately 300 micro-satellite markers flanking 100 candidate genes previously implicated in genetic risk for cancer. Certain loci are a priority due to their association with HRAS1: BRCA1 and all known mismatch repair loci; other repair genes, such as ATM; the Bloom's syndrome locus; and the XRCC group. Other genes are also mapped. Patients do not receive the results of the genetic testing and the results do not influence the type and duration of treatment. Patients only are followed annually. PROJECTED ACCRUAL: This study will accrue 1,000 patient-sibling pairs for breast cancer. After 18 months, another 1,000 patient-sibling pairs will be accrued for breast cancer. Therefore, up to 2,000 patient-sibling pairs will be accrued over 5 years. (Feasibility of accruing pairs for lung, colon, and prostate cancer is being assessed.)