Background: Gastroesophageal reflux disease (GERD), the regurgitation of gastric acids often accompanied by heartburn, affects up to 20% of the general population. Genetic predisposition is suspected from twin and family studies but gene-hunting efforts have so far been scarce and no conclusive genome-wide study has been reported. We exploited data available from general population samples, and studied self-reported reflux symptoms in relation to genome-wide single nucleotide polymorphism (SNP) genotypes.
Methods: We performed a GWAS meta-analysis of three independent population-based cohorts from Sweden, Finland, and UK. GERD cases (n=2247) and asymptomatic controls (n=4503) were identified using questionnaire-derived symptom data. Upon stringent quality controls, genotype data for more than 2.5M markers were used for association testing. Bioinformatic characterization of genomic regions associated with GERD included gene-set enrichment analysis (GSEA), in silico prediction of genetic risk effects on gene expression, and computational analysis of drug-induced gene expression signatures using Connectivity Map (cMap).
Key results: We identified 30 GERD suggestive risk loci (P≤5×10-5 ), with concordant risk effects in all cohorts, and predicted functional effects on gene expression in relevant tissues. GSEA revealed involvement of GERD risk genes in biological processes associated with the regulation of ion channel and cell adhesion. From cMap analysis, omeprazole had significant effects on GERD risk gene expression, while antituberculosis and anti-inflammatory drugs scored highest among the repurposed compounds.
Conclusions: We report a large-scale genetic study of GERD, and highlight genes and pathways that contribute to further our understanding of its pathogenesis and therapeutic opportunities.
Keywords: GERD; SNP; genetics; genome-wide association studies; meta-analysis; risk loci.
© 2016 John Wiley & Sons Ltd.