Eosinophilic esophagitis in adults is associated with IgG4 and not mediated by IgE.

Eosinophilic esophagitis is usually triggered by foods, by unclear mechanisms. We evaluated the roles of IgE and IgG4 in the development of eosinophilic esophagitis. We performed a prospective, randomized, double-blind, placebo-controlled trial of adults with eosinophilic esophagitis given an antibody against IgE (omalizumab, n = 16) or placebo (n = 14) every 2-4 weeks for 16 weeks, based on weight and serum level of IgE. Endoscopy was performed, esophageal biopsy specimens were collected, and symptoms were assessed at baseline and at 16 weeks. Maximum numbers of eosinophils/high-power field were determined. Homogenates of esophageal biopsy specimens from 11 subjects with eosinophilic esophagitis and 8 without (controls) were assessed for IgM, IgA, and IgG subclasses. In a retrospective analysis, we performed immunofluorescence analysis of IgG4 in fixed esophageal tissues from 2 patients with eosinophilic esophagitis who underwent esophagectomy and 47 consecutive autopsies (controls). We also performed immunofluorescence analysis of IgG4 in esophageal mucosal biopsy specimens from 24 subjects with eosinophilic esophagitis and 9 without (controls). Finally, sera were collected from 15 subjects with eosinophilic esophagitis and from 41 without (controls), and assayed for total and food-reactive IgG4. Omalizumab did not alter symptoms of eosinophilic esophagitis or eosinophil counts in biopsy samples compared with placebo. Homogenates of esophageal tissues from patients with eosinophilic esophagitis had a 45-fold increase in IgG4 compared with controls (P < 3 × 10(-5)), but no significant increases in other IgG subclasses, IgM, or IgA. Sparse stromal deposits resembling immune complexes were found in 2 of 5 eosinophilic esophagitis biopsy specimens based on ultrastructural analysis. Esophagectomy samples from 2 patients with eosinophilic esophagitis contained 180 and 300 IgG4 plasma cells/maximal high-power field, mainly in the deep lamina propria; these levels were greater than in tissues from controls. Fibrosis essentially was exclusive to the lamina propria. Granular extracellular IgG4 was detected in biopsy specimens from 21 of 24 patients with eosinophilic esophagitis, but in none of the specimens from 9 controls (P = 6 × 10(-6)). The total serum level of IgG4 increased only slightly in patients with eosinophilic esophagitis, compared with controls. Subjects with eosinophilic esophagitis had increased serum levels of IgG4 that reacted with milk, wheat, egg, and nuts-the 4 foods that most commonly trigger this condition (P ≤ 3 × 10(-4) for each food). In a prospective trial, omalizumab did not reduce symptoms of eosinophilic esophagitis or tissue eosinophil counts compared with placebo. This finding, along with observed granular deposits of IgG4, abundant IgG4-containing plasma cells, and serum levels of IgG4 reactive to specific foods, indicate that, in adults, eosinophilic esophagitis is IgG4-associated, and not an IgE-induced allergy. ClinicalTrials.gov number: NCT 00123630.