A multi-faceted approach to analyzing gene-gene interactions: Variant angiogenesis-related genes and prostate cancer risk

561

Background: Prostate cancer (PCA) incidence and mortality are disproportionately high among African-American (AA) men. Its detection and perhaps its disparities could be improved through the identification of genetic susceptibility biomarkers within essential biological pathways. Interactions among highly variant genes, central to angiogenesis, may modulate susceptibility prostate cancer, as demonstrated in previous reports.
 
 Research Goal & Objective: With the ultimate goal of identifying an important panel of genetic markers capable of predicting prostate cancer risk, this study evaluates the interplay among three highly variant genes and their respective receptors (i.e., IL-10, IL-10R, TGFβ1, TGFβR-1, VEGF, and VEGFR) and their influence on prostate cancer risk within a large and unique PCA case-control study of 923 AA men.
 
 Hypothesis: We hypothesize that variations within regulatory or coding regions within the selected angiogenesis biomarkers will individually or jointly modulate PCA risk, presumably due to alterations in the capacity to form tumor vasculature pivotal to tumor growth.
 
 Methods: In order to evaluate single gene effects as well as joint modifying effects on PCA risk, we conducted a prostate cancer case-control study comprised of 923 African-American men. Genetic alterations detected in germ-line DNA collected from 224 incident prostate cancer cases and 699 healthy age-matched controls were assessed using high-throughput TaqMan polymerase chain reaction (PCR) allelic discrimination strategies. The intricate interaction among polymorphic IL-10, IL-10R, TGFβ1, TGFβR-1, VEGF, and VEGFR was analyzed using conventional logistic regression analysis (LR) models combined with multi-factor dimensionality reduction (MDR) and hierarchical interaction graphs.
 
 Results/Conclusions: With the exception of IL-10 -819T and IL-10 241G genotype we did not observe any main effects among the remaining eleven loci. In fact, we observed a 1.6 - 6.8 fold increase in the risk of developing prostate cancer among carriers of the IL-10 -819T (OR = 1.95; 95% CI = 1.10-3.44) and IL-10 241G alleles (OR = 8.30; 95% CI = 3.01-22.89). However, these two main effects were not confirmed by MDR analysis (P = 0.172). The overall best three factor model, containing IL10 241, IL-10 -819 and IL-10 -1082, was predictive of PCA with a cross validation consistency and average testing accuracy of 100% and 71.3%, respectively (P = .001). This model requires further validation using permutation testing.
 Clinical Relevance: Ultimately, our research will be able to classify a panel of SNPs that will optimize the detection and treatment of prostate cancer.