The effect of epistatic interactions between genetic variants located in microRNA and silencing complex genes on prostate cancer progression risk

Previous studies conducted in Asian and European populations have provided evidence of the association between microRNA-related genetic variants and prostate cancer (PCa) risk and/or progression. Nevertheless, the results obtained in these studies are inconsistent, which could be explained by the limitations of single-locus main effect evaluations to detect joint effects of multiple genetic variants, reflected in statistical epistases. Therefore, we conducted the analysis of potential epistatic interactions between variants located in microRNA genes and in genes encoding the components of RNA-induced silencing complex (RISC) in relation with PCa risk/aggressiveness. Raw data on genotyping results from our previous studies involving four microRNA polymorphisms and five variants in RISC genes were subjected to the exclusion of samples based on missing data criterion, followed by the re-evaluation of Hardy-Weinberg equilibrium. Afterwards, these genotyping results were included in the Multifactor dimensionality reduction (MDR) analysis. Permutation testing was conducted in order to assess statistical significance of the best models from MDR tests. MDR tests on the risk of developing PCa yielded statistically insignificant results. Nevertheless, the MDR results for comparison of PCa patients with high and low cancer progression risk were statistically significant for the analysis that included rs11614913, with the 3-locus best model comprising this genetic variant, rs7813 and rs784567. We conclude that statistical epistasis between rs11614913 in hsa-miR-196a2, rs7813 in GEMIN4 and rs784567 in TARBP2 shows association with the invasiveness of PCa.