Pooled analysis of association between a genetic variant in the 3'-untranslated region of Toll-like receptor 4 and cancer risk.

Many epidemiological studies have shown the association between certain genetic variations in the Toll-like receptor 4 (TLR4) gene (for example, rs4986790 and rs4986791) and cancer risk. However, the results from investigations into the association between rs11536889, a genetic variant in the 3'-untranslated region of TLR4, and cancer risk lack consensus. We performed a meta-analysis to investigate the effect of rs11536889 on cancer risk. A total of 12 relevant case-control studies were included in this analysis (6222 cases and 7948 controls). The pooled ORs with their corresponding 95%CIs were estimated. We did not detect any association between rs11536889 and overall cancer risk (P = 0.13). However, stratification analysis by cancer type revealed a borderline statistically significant increased risk in both genotype comparison (OR for the variant genotype in the dominant model = 1.17; 95%CI = 0.98-1.40; P = 0.08) and allele comparison (OR for variant allele = 1.14; 95%CI = 0.99-1.32; P = 0.07) for prostate cancer. In contrast, no statistically significant or borderline result was found for gastric cancer. These findings indicate that rs11536889 in TLR4 might be specifically associated with prostate cancer. However, owing to the modest and underpowered results, and the limitations of the original studies included for analysis, further prospective studies with larger sample sizes are required to confirm our findings.