Etrolizumab Treatment Improves Histologic Activity as Assessed by Both the Robarts Histopathology and Nancy Histological Indices

Abstract

BACKGROUND: Etrolizumab, an anti-β7 monoclonal antibody, showed efficacy and safety during 10 weeks of induction in patients with moderate-to-severe ulcerative colitis in the Phase 2 EUCALYPTUS trial. Since a reduction in histologic inflammation has been linked with improved long-term clinical outcome, and the FDA recommends using both histologic and endoscopic assessments for efficacy evaluation, the effect of etrolizumab on histologic inflammation was evaluated in mucosal biopsies from EUCALYPTUS patients using the Robarts histopathology index (RHI) and Nancy histological index (NHI). METHODS: 124 patients were randomly assigned (1:1:1) to receive subcutaneous etrolizumab (100 mg at weeks 0, 4, and 8, with placebo at week 2, or 420 mg loading dose at week 0, followed by 300 mg at weeks 2, 4, and 8) or matching placebo. Biopsies were taken using flexible sigmoidoscopy/full colonoscopy from the most inflamed colonic area within 10-40 cm from the anal verge at baseline and at week 10. 62 patients provided consent for long-term sample storage for research; batched H&E-stained slides were scored by a single pathologist using the Geboes scale (later converted to RHI) and NHI. At week 10, mean changes in RHI and NHI scores for pooled etrolizumab or placebo were calculated. Subanalyses explored histologic response (reductions of ≥ 6 or 10 points or ≥ 50% improvement from baseline RHI and ≥ 1 or 2 points reduction from baseline NHI), remission (no neutrophils, RHI ≤ 4 and NHI=0, ≤ 1 or 2) and correlation with endoscopic improvement. RESULTS: Analysis included 56 patients with baseline data and baseline NHI>1. At week 10, RHI and NHI scores decreased by a greater extent with etrolizumab compared with placebo, regardless of anti–tumor necrosis factor α (aTNF) experience (RHI −8.4 vs −1.6; P=0.032 and NHI −1.2 vs −0.2; P=0.011 for all comers). A greater proportion of etrolizumab-treated patients achieved categorical histologic improvement and remission; specifically, histologic improvement was achieved in 49% (≥ 6 decrease from baseline RHI) and 58% (≥ 1 decrease from baseline NHI) of patients receiving etrolizumab compared with 30% and 14% receiving placebo, respectively. Of patients with an endoscopic subscore (ES) ≤ 1 at week 10 (n=6), 100% experienced histologic response as assessed by RHI (5/5 with RHI nonmissing at week 10), and 83% (5/6) by NHI. Mean (SD) RHI changes were −19.2 (10.0) in patients with an ES ≤ 1 at week 10 versus −4.4 (10.1) in patients with an ES>1. Mean (SD) NHI changes were −2.5 (1.5) in patients with an ES ≤ 1 at week 10 versus −0.6 (1.3) in patients with an ES>1. Spearman correlation coefficients between RHI and NHI were 0.82 at baseline and 0.91 at week 10, while both histologic scores were similarly correlated with ES (0.25-0.28 at baseline and 0.38-0.40 at week 10). CONCLUSION(S): Histologic activity assessment using RHI or NHI demonstrates improvement after week 10 with etrolizumab treatment and was greater in aTNF-naive patients. Importantly, RHI or NHI reductions were associated with improved ES at week 10.