Abstract 3426: Genetic architecture of esophageal squamous cell carcinoma in African American veterans

Background: Esophageal carcinoma (EC) is one of the most common malignancies of the gastrointestinal tract worldwide and is unresponsive to therapy. Esophageal squamous cell carcinoma (ESCC) comprises the majority of EC in African Americans, Asians and other ethnic groups, while esophageal adenocarcinoma (EAC) predominates among Caucasians. Poor prognosis of EC in Caucasians and African Americans is reflected by five-year survival rates of 21% and 14 %, respectively. The aggressiveness and lower survival rate of African American ESCC patients than that from patients of other ethnic groups are evident even after adjusting for treatment modalities and socioeconomic factors. These characteristics suggest the existence of an ethnic- or race-dependent component of EC etiology. However, the genetic architecture of ESCC in AA is not well studied, and thus, mostly undefined. Our study aims to identify exonic mutations that may represent critical genetic changes in African American ESC carcinogenesis. Methods and Materials: Whole exome sequencing (WES) of matched tumor and normal tissue DNA from endoscopic biopsies from late-stage ESCC of nine African American Veteran male patients was conducted. This study was approved by the DC VAMC Institutional Review Board, and a written informed consent was obtained from each patient prior to biopsy. Biopsies were de-identified, freshly frozen and stored at -80 °C. To proceed to exon capture, we enriched our samples by using Agilent SureSelect XT Human All Exon V6+UTR. Paired-end sequencing at a read depth of 100X was performed on the exon libraries using the Illumina HiSeq 4000 sequencer. We applied Genome Analysis Toolkit's best practices WES pipeline to identify both germline and somatic variations in the dataset. The Seven Bridges Cancer Genomics Cloud platform was used for processing the data and annotating the variants. Variants were mapped to genomic regions and further aggregated at the gene level, pathways, and biological processes relevant to disease by using Reactome, Pathway Studio, and Ingenuity Variant Analysis. Results: In all samples, 27,586 variants consisted of 86% single nucleotide variation, 8.5% insertions and 5.5% deletions. Approximately half of the variants caused missense changes while 0.12% created nonsense mutations. High impact mutations occurred in genes that have a role in DNA damage repair, stress response, detoxification pathways, keratinization and immune surveillance. Conclusion: We found uni