Barrett's esophagus (BE) occurs in 2-5% of the US population, and is the only known precursor to esophageal adenocarcinoma (EA). Screening BE cases by surveillance endoscopy is costly, invasive, and largely unsuccessful since most BE cases do not progress. Since EA is quickly fatal, with only 15% survival at 5 years, our goal is toinform both basic biology and translational strategies for improved prevention and detection of BE to prevent progression to EA. Epidemiologic research has determined that gastroesophageal reflux (GER), obesity, and smoking increase risk of BE and EA, and non-steroidal anti-inflammatory (NSAID) use decrease risk. These shared factors all impact local and systemic inflammation that promotes cell proliferation and genetic instability. Given that the inflammation-related environmental exposures are common but the outcomes are rare, we propose to identify the contribution of genetic variation in inflammation pathways to identify those at highest risk of disease. We will use the rich data resource available to us through the BEACON consortium that pools extensive genotype (Illumina Omni-1M array) and environmental exposure data on approximately 2400 BE cases, 1500 EA cases, and 2200 controls from 14 studies. In Aim 1 we will determine whether variation in genes comprising five key inflammation pathways (COX, oxidative stress, cytokines, HLA/KIR, and NF?B) is associated with the development of BE and EA. Independent datasets will be used to validate the overall genetic findings. In Aim 2 we will assess the extent to which the impact of genetic variation may vary by established inflammation-related exposures, including reflux, obesity, smoking, and NSAID use. The inflammation pathway genes we propose to explore represent a synthesis of information derived from functional studies and public repositories of pathway knowledge in curated databases. Standard agnostic approaches to analyzing GWAS data may miss some genetic signals that are below the high statistical threshold required for genome- wide significance in association studies. Pathway analysis is complementary to GWAS, and we plan to use a principal component analysis (PCA) approach that will reduce data complexity by combining numerous individual variants into a few integrated factors. Positive results from this secondary analysis of GWAS data will identify new genetic signals, and determine if examining those signals within strata of established risk factors can identify those most likely to progress.Future studies will build on this study by extending our findings through tissue-based epigenetic analysis of BE and EA cases.