Eosinophilic esophagitis (EoE) genetic susceptibility is mediated by synergistic interactions between EoE-specific and general atopic disease loci

Abstract

Background: Eosinophilic esophagitis (EoE) is an esophageal inflammatory disease associated with atopic diseases. Thymic stromal lymphopoietin (TSLP) and calpain 14 (CAPN14) genetic variations contribute to EoE, but how this relates to atopy is unclear.

Objective: The purpose of this study was to explore the relationship between EoE, atopy, and genetic risk.

Methods: EoE-atopy enrichment was tested by using 700 patients with EoE and 801 community control subjects. Probing 372 single nucleotide polymorphisms (SNPs) in 63 atopy genes, we evaluated EoE associations using 412 nonatopic and 868 atopic disease control subjects. Interaction and stratified analyses of EoE-specific and atopy-related SNPs were performed.

Results: Atopic disease was enriched in patients with EoE (P < .0001). Comparing patients with EoE and nonatopic control subjects, EoE associated strongly with IL-4/kinesin family member 3A (IL4/KIF3A) (P = 2.8 × 10^-6; odds ratio [OR], 1.87), moderately with TSLP (P = 1.5 × 10^-4; OR, 1.43), and nominally with CAPN14 (P = .029; OR, 1.35). Comparing patients with EoE with atopic disease control subjects, EoE associated strongly with ST2 (P = 3.5 × 10^-6; OR, 1.77) and nominally with IL4/KIF3A (P = .019; OR, 1.25); TSLP's association persisted (P = 4.7 × 10^-5; OR, 1.37), and CAPN14's association strengthened (P = .0001; OR, 1.71). Notably, there was gene-gene interaction between TSLP and IL4 SNPs (P = .0074). Children with risk alleles for both genes were at higher risk for EoE (P = 2.0 × 10^-10; OR, 3.67).

Conclusions: EoE genetic susceptibility is mediated by EoE-specific and general atopic disease loci, which can have synergistic effects. These results might aid in identifying potential therapeutics and predicting EoE susceptibility.

Keywords: Genetic association; atopy; eosinophilic esophagitis.

FIG 1

Genetic associations with EoE differ by control group. IL4 is a predominant association when using nonatopic control subjects. TSLP is a predominant association when using atopic disease control subjects. The y-axis is the −log₁₀ P value. Numbers on the top x-axis denote chromosomes. Within a chromosome, variants are ordered by genomic position.

FIG 2

Interaction between IL4 and TSLP. IL4 and TSLP variants exhibit significant interaction (top panel). In subjects with at least 1 copy of the IL4 risk variant (rs2243250), TSLP is strongly associated with EoE; however, in subjects lacking this variant, the association is only nominal (middle panel). Subjects with IL4 risk variants have substantially larger effect sizes (OR) than subjects lacking the IL4 risk variant (bottom panel).

FIG 3

Synergistic effect of IL4 and TSLP risk alleles for patients with EoE versus population-representative control subjects (Cincinnati Genomic Control Cohort). ^†Reference category, subjects with neither risk allele. IL4 risk allele, rs2243250; TSLP risk allele, rs3806933. *P < .05 and **P < .001. ORs were adjusted for sex, age, and PCs.