Abstract 2227: Genomic study on prostate cancer disparities reveals the differential transcriptomes, microRNA profiles, alternative splicing patterns and copy number variations between Caucasian and African American populations

Abstract

Abstract Prostate cancer (PCa) is a disease conferred by gene mutations, numerous alternations in gene expression and aberrant changes in genome composition/architecture. An area of research that continues to garner attention is PCa health disparities, wherein the African American (AA) population exhibits higher incidence and mortality rates compared to Caucasian Americans (CA). To identify the genetic predispositions and oncogenetic networks associated with the observed PCa disparities, we applied integrated genomic technologies to investigate RNA and microRNA expressions, alternative splicing events, and DNA copy number variations (CNVs) in AA and CA populations. RNA and DNA purified from PCa (Gleason score > 6) and paired adjacent normal prostate needle biopsies from AAs and CAs were processed and hybridized onto Affymetrix human Exon 1.0 ST and SNP 6.0 arrays. A 4-way statistical design (t-test with 10% false discovery rate) was employed to identify differentially expressed mRNAs in the following comparisons: AA normal vs. CA normal, AA cancer vs. CA cancer, AA cancer vs. AA normal, and CA cancer vs. CA normal. Pathway analyses of the differentially expressed genes (275 to 987 genes in the 4-way comparison) revealed network-level rewiring of gene interactions (e.g. gene networks or circuits) accounting for the PCa disparities between AA and CA. For example, the mis-regulated testosterone metabolism network in normal AA compared with normal CA tissues may represent a predisposition factor in the AA population; whereas the activated inflammatory response (NF-κB network), up-regulated oncogenic signaling pathways (ERK, JNK and p38) and more aggressive cancer invasion and metastasis (highly expressed RHOA and STAT1) in AA carcinomas may associated with the higher recurrence and death rates in AA patients. In addition, our expression and CNV atlas have identified hundreds of genes exhibiting differential splicing patterns or copy number aberrations (deletions or amplifications) between the AAs and CAs, representing candidate genes mediating PCa disparities. Notably, at least 13 genes residing within the 5 oncogenic signaling pathways have been identified as exhibiting either differential splicing (in FGFR3, PDGFRA, MET, EPHA3, NF1, RASGRP2, GSK3, TSC2, ATM, RAF, and RB1) or CNVs (at EPHA, PTEN and APC) between AA and CA PCa specimens. Taken together, our data suggest that gene-network rewiring, mRNA splicing and CNV may play important roles in the PCa health disparities between AA and CA populations. Further identification of these critical genetic elements related to PCa disparities may facilitate the development of biomarkers for screening, early detection, and prediction of clinical outcome. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2227.