关键词：基因；前列腺癌；风险；案例研究
摘 要：Prostate cancer (PCa) is the most common cancer and the second leading cause of cancer death among men in the United States. Considering that PCa development requires the coordination of many genes, it is expected that a simultaneous evaluation of multiple genetic variants can improve the statistical power to detect additional PCa risk variants. Recent improvements in analytical methods and computation make it feasible to search for gene-gene interaction of SNPs in the genome. We hypothesized that multiple sequence variants in the genome may interact to increase PCa risk. These variants may or may not have known main effect on PCa risk and can be better detected by systematically evaluating gene-gene interactions for SNPs in the genome. We utilized data from an existing GWAS of a large NCI Cancer Genetic Markers of Susceptibility (CGEMS) study to systematically discover genes that interacted with known PCa risk variants in the genome. We also evaluated the genes that interacted with known PCa risk variants in another two independent populations, including a population based PCa case-control study from Sweden (CAPS) and a PCa patient population from Johns Hopkins Hospital (JHH). In addition, we performed an exhaustive search for pair-wide SNP-SNP interactions without main effect in the JHH and CGEMS populations using a novel statistical approach of Boolean Operation-based Screening and Testing (BOOST). We identified thirty- five pairs of SNPs that significantly interacted with the thirty-two known risk variants on PCa risk at a P-value of 1E-05 in the combined analysis of three populations. The most significant interaction detected was between rs12418451 in MYEOV and rs784411 in CEP152, with a Pinteraction of 1.15E-07 in the meta- analysis.