233. ENDOSCOPIC REFERENCE SCORE AND HISTOLOGICAL SCORING SYSTEM ARE COMPLEMENTARY ASSESSMENTS FOR CHARACTERIZING FIBROSTENOTIC OR INFLAMMATORY PHENOTYPES IN EOSINOPHILIC ESOPHAGITIS PATIENTS

Abstract

Abstract The Histologic Scoring System (HSS) assesses severity/extent of histologic features (eosinophil density [ED]/basal zone hyperplasia [BZH]/eosinophil abscesses [EA]/eosinophil surface layering [SL]/surface epithelial alteration [SEA]/dyskeratotic epithelial cells [DEC]/dilated intercellular spaces [DIS]/lamina propria fibrosis [LPF]) in eosinophilic esophagitis (EoE). The Endoscopic Reference Score (EREFS) assesses endoscopic feature severity (edema/exudates/furrows/rings/strictures). We correlated HSS components with inflammatory/remodeling EREFS components in placebo-treated EoE patients from 2 trials—phase 2 proof-of-concept (POC; NCT02379052), phase 3 TREET part A (Phase3-PartA; NCT03633617). POC was a 12-week trial in adults, and Phase3-PartA was a 24-week trial in adults and adolescents. Proximal/mid/distal esophagus biopsies were collected at baseline/end of treatment (EOT) and scored by HSS for grade (severity)/stage (extent)/components (higher scores = greater severity). Endoscopies were scored by EREFS (higher scores = greater severity) and inflammatory (edema+exudates+furrows) and remodeling (rings+stricture) subscores calculated. Pearson correlation and single co-variate regression model analyses were performed between baseline HSS grade/stage, total/individual components scores and EREFS total/components scores and subscores, and for the change in scores from baseline at EOT (ΔEOT). Moderate-to-strong correlations (r > 0.3) were observed between total EREFS and HSS grade/stage for baseline/ΔEOT in POC and baseline in Phase3-PartA (Table). Moderate correlations were observed between BZH and EREFS remodeling subscore and rings for ΔEOT in Phase3-PartA; DIS grade and EREFS remodeling subscore and strictures for baseline and ΔEOT in POC but not Phase3-PartA; and EREFS inflammation and remodeling subscores and components for SEA and SL at baseline and ΔEOT in POC. Moderate-to-strong correlations were observed between EREFS inflammation subscore and DIS stage and between edema and DIS grade/stage at baseline in both studies. Similar results were observed using regression analyses. Specific components of HSS and EREFS, reflecting inflammatory and remodeling processes, are closely correlated, when examined by Pearson correlations or regression analyses, and can aid identification of fibrostenotic or inflammatory phenotypes in EoE. Correlations of DIS with endoscopic inflammatory and remodeling features reflect the role of barrier dysfunction in maintaining chronic inflammation that ultimately leads to remodeling.