Mucosal p-STAT1/3 correlates with histologic disease activity in Crohn's disease and is responsive to filgotinib.

Abstract

The validity and relevance of histologic disease activity in Crohn's disease (CD) is unclear, owing to disconnects with endoscopic pathology. Here, we explore relationships between endoscopic, histologic, and molecular activity. This post hoc analysis of the Phase 2 FITZROY trial (NCT02048618) assessed baseline and week 10 (W10) inflammation across matched ileal and colonic segments in CD patients receiving filgotinib 200 mg (n = 42) vs placebo (n = 18). Macroscopic and microscopic disease were assessed by Simple Endoscopic Score for CD ulceration subscore (uSES-CD) and Global Histologic Activity Score activity subscore (aGHAS), respectively. Molecular activity was quantified by phosphorylated signal transducer and activator of transcription (pSTAT)1 and pSTAT3 in epithelium and nonepithelium. Segments were classified as "low" or "high" activity; correlations and concordance were calculated. Logistic regression identified W10 outcome predictors. Overall, 300 segments in 60 patients were assessed. Baseline uSES-CD and aGHAS correlations were 0.72 and 0.53 in colon and ileum, respectively. pSTAT levels had poor-to-moderate concordance with uSES-CD (κ range, 0.11-0.49) but moderate-to-good concordance with aGHAS (0.43-0.77). With filgotinib vs placebo, uSES-CD and aGHAS decreased in significantly more segments with high baseline uSES-CD and aGHAS, and significantly more segments with high baseline pSTAT improved at W10. pSTAT1 was more sensitive to change than uSES-CD and aGHAS. Low baseline pSTAT3 in colon nonepithelium predicted W10 low uSES-CD (P = .044). There was better concordance between histologic and molecular disease activity associated with higher sensitivity to change vs endoscopic severity in ileocolonic CD. Our results suggest histologic activity be included in the assessment of CD inflammatory burden.