Abstract
BACKGROUND: Eosinophilic esophagitis (EoE) is a T-cell-mediated disease that is caused by specific foods and results in esophageal dysfunction. Existing allergy testing modalities are not helpful when attempting to identify EoE-causal foods necessitating empiric food elimination and recurrent endoscopy. The goal of this study was to identify and compare allergen-specific immune features that can be assayed in a minimally invasive manner to predict clinical food allergy in EoE. METHODS: We obtained blood samples from control subjects (n = 17), subjects with clinical EoE milk allergy (n = 17), and subjects with immunoglobulin E-mediated milk allergy (n = 9). We measured total and milk-specific plasma immunoglobulin G (IgG)4 levels and peripheral memory CD4(+) T helper (T(H) ) cell proliferation and cytokine production after stimulation with endotoxin-depleted milk proteins. Sensitivity and specificity for predicting clinical EoE milk allergy were calculated and compared between approaches. RESULTS: Total and milk-specific IgG4 levels were not significantly different between control subjects and subjects with clinical EoE milk allergy. Stimulation with milk proteins caused T(H) lymphocytes from subjects with clinical EoE milk allergy to proliferate more (%P1 of 38.3 ± 4.6 vs. 12.7 ± 2.8, p < 0.0001), and produce more type 2 cytokines (%IL-4(+) of 33.7 ± 2.8 vs. 6.9 ± 1.6, p < 0.0001) than cells from control subjects. Milk-dependent memory T(H) -cell proliferation (sensitivity and specificity of 88% and 82%, respectively) and interleukin 4 (IL-4) production (sensitivity and specificity of 100%) most strongly predicted clinical EoE milk allergy. CONCLUSIONS: Peripheral markers of allergen-specific immune activation may be useful in identifying EoE-causal foods. Assaying milk-dependent IL-4 production by circulating memory T(H) lymphocytes most accurately predicts clinical EoE milk allergy.