Abstract 2780: Chemokine and chemokine receptor genetic variants as predictors of aggressive prostate cancer among European-American men

Abstract

Abstract Background: Chemokines are essential mediators in the pathogenesis of inflammatory, autoimmune, vascular and neoplastic diseases. Consequently, investigators have evaluated the relationship between sequence variants in chemokines in relation to autoimmune or inflammatory disorders. However, the impact of inheriting multiple chemokine-chemokine receptor genetic susceptibilities on prostate cancer (PCa) clinical outcomes remains uncharacterized. Objectives & Hypothesis: Consequently, we systematically evaluated the joint modifying effects of 78 single nucleotide polymorphisms (SNPs) detected in 43 chemokine and chemokine receptor-associated genes. We hypothesized that individuals inheriting multiple chemokine-associated alleles linked with increased tumor metastasis, cell survival, or inflammation would have an increased risk of developing PCa or aggressive disease relative to those with the referent genotypes. Methods: To test the hypothesis, we used SNP profile data collected from 2277 European-American male participants of the Cancer Genetic Markers of Susceptibility (CGEMS) project, involving 1176 PCa cases and 1101 controls. Using a case-control and case-only designs, we evaluated the independent and joint modifying effects of 78 SNPs detected in chemokines, chemokine receptors and downstream signaling targets in relation to PCa risk and aggressive disease (tumor stage >2 and tumor grade >2). Main effects and complex interactions were assessed using logistic regression analysis, multi-factor dimensionality reduction (MDR) with covariate adjustment (i.e., age-group and family history of prostate cancer), and hierarchical entropy graphs. To control for multiple comparisons, we used 1000-fold permutation testing. Results: A three-SNP interaction [CCL21 promoter SNP at position -191 (rs11574914), CCR6 promoter SNP at position -397 (rs2023305), CCL17 intron 1+2217 (rs223895)] was the best prediction model in relation to aggressive disease, following restriction of the data analysis to the top 25th percentile of markers based on ReliefF scores (TBA = 56%, CVC = 90%, p = 0.024). Verification of this interaction using other bioinformatic techniques, including symbolic modeling, classification and regression trees (CART) and computational evolutionary systems modeling is currently underway. Conclusions: Our preliminary findings support the role of chemokine-related markers as significant predictors of aggressive prostate cancer among European-American men. Future Directions: Additional studies are needed to determine whether CCL21-CCR6-CCL17 interactions are predictive of disease prognosis within ethnically diverse sub-populations. In addition, the biological implications of the interaction require further investigation. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 2780. doi:10.1158/1538-7445.AM2011-2780