Variants in inflammation-related genes and risk of prostate cancer: The PLCO study Free

4119

Chronic inflammation has been hypothesized to increase prostate cancer risk. A role for inflammation is supported by the decreased risk of prostate cancer observed with use of aspirin and the increased risk associated with prostatitis and obesity. To examine whether common genetic variants in inflammation genes are associated with prostate cancer risk, we conducted a nested case-control study within the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. Cases (n=1,265) and controls (n=1,797) were frequency-matched on age, race/ethnicity, time since initial screening, and date of blood draw. We genotyped 15 single nucleotide polymorphisms (SNPs) in 4 candidate genes, PTGS2 (7 SNPs), PPARD (1 SNP), PPARG (2 SNPs), and TNF-α (5 SNPs). Multivariate odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using conditional logistic regression for individual SNPs and haplotype analyses. Of the 15 variants examined, two, both in the PTGS2 gene, were associated with an increased risk of prostate cancer. Specifically, the CC genotype of Ex10 + 837T>C (rs5275) was associated with a 30% increased risk of prostate cancer, compared with the TT genotype (95% CI 1.03-1.64, p-trend=0.03). In addition, the GG genotype of the IVS5 -275G>T marker (rs20432) was associated with a 45% increased risk (95% CI 1.02-2.06) of prostate cancer compared with the TT genotype. Two other PTGS2 SNPs (rs5277 and rs4648276) showed borderline significant trends (p=0.06 for both) with prostate cancer risk. Haplotype analysis provided further support of a role for PTGS2 in prostate cancer etiology. Based on 5 SNPs from the PTGS2 gene (rs689470, rs5275, rs4648276, rs20432, rs5277), four common haplotypes were inferred that explained over 95% of the variation. Haplotype frequency was significantly different between cases and controls (p<0.01). Relative to the most frequent haplotype (G-T-T-T-G), the other 3 common haplotypes were each significantly associated with a 24-25% increase in prostate cancer (G-T-T-T-C haplotype, OR=1.24, 95% CI 1.06-1.45; G-C-T-T-G haplotype, OR=1.25, 95% CI 1.07-1.45; G-C-C-G-G haplotype, OR=1.24, 95% CI 1.04-1.47). Results were similar when analyses were stratified by regular NSAID use vs. not and obesity vs. not; results also persisted after adjustment for multiple comparisons. In contrast, SNPs in the PPARD, PPARG, and TNF-α genes were not associated with prostate cancer risk. In summary, our results suggest that polymorphisms in the PTGS2 gene could be associated with prostate cancer risk, thus supporting a role for inflammation in prostate cancer etiology. Further analysis in the Cancer Genetic Markers of Susceptibility (CGEMS) project should provide sufficient power to determine if these findings robustly replicate.