Abstract 5073: Integrated genomics of high-resolution esophageal adenocarcinomas identify candidate driver genes in regions of large copy number aberrations

Abstract

Abstract Introduction: Esophageal cancer occurs with two distinct histologies_ squamous cell carcinoma (SCC) and esophageal adenocarcinoma (EA). The incidence of EA has risen over the past few decades and is now the most common esophageal malignancy in the United States and Western Europe. The genome of esophageal adenocarcinoma is highly unstable due to the large number of amplification and deletion events. Presumably, such genomic events drive tumor formation and disease progression by altering the expression of genes within these regions. Several studies have reported large regions of genomic instability especially spanning chromosomal arms or in gene dense regions. Using a large patient cohort and high-resolution data for DNA copy number and expression, this study aims to identify potential driver genes and prognostic markers especially in whole chromosomal aberrations and gene dense regions. Materials & Methods: DNA copy number aberrations were explored in 116 esophageal adenocarcinomas (EA) using Affymetrix SNP 6.0 microarrays. Additionally, gene expression data from a subset of 110 of these tumors using Affymetrix U133 plus 2.0 microarrays were analyzed. DNA copy number aberrations were analyzed using Nexus 5.0 and GISTIC software while the gene expression data was processed in Partek Genomic Suite. Copy number and expression correlations for each gene in the regions of copy number aberrations were calculated using t-test to compare samples with the changes versus samples without the genomic aberrations. Results: We identified at least 50 regions of gains and losses of which 12 regions are either whole chromosomal arm aberrations or aberrations in gene dense regions. Some regions with large number genes include gains at 1q (460 genes), 7q21-22 (80 genes), 10q21-22 (100 genes), and losses at 4q (98 genes), 5q (283 genes), 21q (82 genes). Copy number and expression correlations for each gene in these regions considerably reduced the number of genes that could be considered as candidate driver genes. Furthermore, scouring the literature helped identify potential candidate driver genes such as ANP32E (1q), VCL (10q), CDKN2AIP (4q), and ING2 (4q). Additionally, we report identifying copy number and expression correlations of potential prognostic markers such as MCM4 (8q), CA9 (9p), KLF5 (13q) that lie in focal/smaller regions of genomic instability but at high frequencies. Conclusion: Analysis of the DNA copy number changes identified several regions of frequent gains and losses. At least 12 regions span the chromosomal arm or are in the gene dense regions. Using an integrated genomics approach with high-resolution DNA copy number and expression data, we are able to identify potential driver genes and prognostic markers that lie in these large amplification/deletion regions in EA. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5073. doi:1538-7445.AM2012-5073