Gene-environment Interaction in Pediatric Eosinophilic Esophagitis

Abstract

Introduction: The rapid increase in the incidence of eosinophilic esophagitis (EoE) may be due to interactions between a genetic predisposition and changing environmental factors. While EoE has been associated with several early life factors, gene-environment interactions have not been explored. Here, we test for interactions between SNPs or SNPs in high linkage disequilibrium with SNPs previously associated with EoE (TSLP, LOC283710/ KLF13, CAPN14, and CCL26) and early life factors (antibiotic use in infancy, cesarean delivery, breastfeeding, NICU admission, and absence of pets in the home). Methods: We conducted a case-control study using hospital-based cases identified from a registry of EoE patients (n=127) and population-based controls representative of the hospital catchment area (n=121). Early life environmental factors were assessed through a self-administered questionnaire provided to mothers. Cases and controls had been previously genotyped. We computed case-only interaction tests and where there was evidence of possible interaction, secondary analyses were used to describe the combined and independent effects of genotype and environmental factor on risk of EoE. Models were adjusted for maternal education and population stratification. Results: The case-only analyses identified interactions between CAPN14 and breastfeeding (p=0.02), TSLP and NICU admission (p=0.06), TSLP and not having a furry pet in infancy (p=0.05), and LOC283710/ KLF13 and NICU admission (p=0.02), but not with CCL26 and any of the early life factors examined. Case-control analyses of these associations showed the interactive effects for all associations were antagonistic; specifically the risk of having EoE for those with both the susceptibility genotype and early life factor was attenuated as compared to the risk of disease among those with only the genotype or early life factor (Figure). Examination of the effect of early life factors, among those with the CAPN14 susceptibility SNP, identified a strong protective association for breastfeeding (OR: 0.09, 95% CI: 0.01-0.63). Conclusion: In this first study of gene-environment interactions in EoE, there is preliminary evidence that certain early life factors may interact with susceptibility genotypes to modify risk of disease. In particular, interactions between CAPN14 and breastfeeding, LOC283710/ KLF13 and NICU admission, TSLP and NICU admission, and TSLP and pets merit further examination.Figure 1