Richard Hayes

NCI, DCEG, OEEB

PLCO (Learn more about this study)

2007-0036

Jun 12, 2007

Genetic Polymorphisms of Selenoproteins and Prostate Cancer

Selenium is a promising nutrient for the prevention of prostate cancer, and is being tested in the Selenium and Vitamin E Cancer Prevention Trial (SELECT), one of the largest ongoing prevention trial for cancer. The anti-carcinogenic properties of selenium have primarily been attributed to a small group of selenium containing proteins. Selenium in the form of selenocysteine (Sec) is incorporated into the active center of these selenoproteins. The incorporation of selenium as Sec into selenoproteins is highly complex and requires an RNA stem loop (called a SECIS element) present in the 3' untranslated region of each selenoprotein and several trans-acting factors that that assemble on the SECIS. The most important of these trans factors are a selenocysteine specific elongation factor (SELB), a specific SECIS binding protein (SECISBP2), tRNA-selenocysteine, and several recently identified proteins that also interact with the complex (nucleolin and ribosomal protein L30). Variations in genes encoding for these selenoproteins or trans-acting factor required for incorporation of Sec have the potential to modify the production or activity of some or all selenoproteins and potentially the risk of prostate cancer. The here proposed study builds on an ongoing study funded by an Intramural Research Award (PI Peters), which measured serum selenium concentration in the same men as included in CGEMS prostate cancer study (Peters et al. AJCN 2007). As part of the ongoing study we also resequenced six important selenoproteins, which will be used to identify gene regions that are insufficient covered by CGEMS and identify tagging single nucleotide polymorphisms (SNPs) for complete coverage (r2 minimum =0.8). Independent from CGEMS we plan to genotype these additional tagging SNPs in the PLCO study population.

Results of this study will have immediate application to the ongoing SELECT trial that investigates the chemopreventive effect of selenium and vitamin E on prostate cancer in 32,400 men, allowing us to monitor the effect of selenium supplementation based on the genetic make-up of the participants.

To adjust for multiple comparisons we will apply the False Discovery Rate including all genotyped SNPs of the selenoenzyme. To test for significance of interaction with serum selenium concentrations we will conduct a global omnibus test for interaction by simultaneously including all of the cross-product terms of the serum selenium (coded as continuous variable) with the different genotypes for a gene (coded as dummy variables) and comparing it to a null model that includes only main effects for selenium and the genotypes. These multi-locus global tests automatically adjust for multiple testing based on the degrees of freedom of the corresponding chi-square test. Moreover, the multi-locus tests can efficiently capture the multivariate linkage disequilibrium (LD) pattern within a gene and hence can be more efficient than tests based on single SNPs for detecting associations or interactions when the true causal variant in the region may not have been genotyped.

We are planning to publish these results even if no association or interaction remains significant after adjusting for multiple comparisons as this study is focusing on a potentially important pathway, which is currently tested within SELECT, one of the largest ongoing cancer intervention trials.