Sustained Therapeutic Benefit of Vedolizumab Throughout 1 Year in Ulcerative Colitis in GEMINI I, a Randomized, Placebo-Controlled, Double-Blind, Multicenter Trial

Abstract

We assessed the effect of vedolizumab (VDZ), an investigational, gut-selective monoclonal antibody targeting α4β7 integrin, over 52 wks of therapy in patients with moderately to severely active ulcerative colitis in whom at least 1 prior therapy had failed. Eligible adult patients had a Mayo score ≥6 and endoscopic subscore ≥2 despite corticosteroids (CSs), purine antimetabolites, and/or TNFα antagonists. After 2 induction doses of VDZ (wk 0, wk 2), patients with clinical response (reduction in Mayo score of ≥3 points and ≥30% from baseline, plus decrease in rectal bleeding subscore of ≥1 point or absolute rectal bleeding subscore ≤1 point) to VDZ at wk 6 were randomized 1:1:1 to VDZ 300 mg IV q4wks, VDZ 300 mg IV q8wks, or placebo (PBO) for 46 wks. Clinical remission was defined as complete Mayo score (CMS) or partial Mayo score (PMS) ≤2 and no individual subscore >1; clinical response was based on PMS, as a reduction ≥2 points and ≥25% from baseline; durable mucosal healing, as Mayo endoscopic subscore ≤1 at both wks 6 and 52. Because the defined study endpoints used both CMS and PMS, analyses were conducted to evaluate agreement; moderate to substantial agreement was found between clinical response and remission endpoints as defined by CMS and PMS at wks 6 and 52. Among patients with a clinical response to VDZ at wk 6 (intention-totreat [ITT] population, N = 373), decreases in PMS occurred as early as wk 2 (Fig 1). After successful induction, PMS remained substantially lower than baseline in both VDZ groups through wk 52, whereas an increase in PMS was noted starting at wk 22 in the PBO group. Mean vedolizumab trough concentrations in the placebo group declined over time and were below quantitation limits at wk 22. In the ITT population, remission rates by PMS in the VDZ groups remained stable, whereas PBO remission rates decreased, over the course of the maintenance phase (Fig 2). Durable mucosal healing was achieved by 42.6%, 43.2%, and 17.5% of patients in the VDZ q8wks, VDZ q4wks, and PBO groups, respectively (P<0.0001 for both comparisons of VDZ vs PBO). Median CS dose declined over the maintenance phase in both VDZ groups, compared with rising CS use after wk 26 in the PBO group (Fig 3). In patients blinded to treatment who did not achieve a clinical response by wk 6 (n = 174), 25.0% of VDZ vs 14.6% of PBO patients achieved clinical response, and 8.7% vs 4.9% clinical remission, as determined by PMS by wk 10. In patients who responded to VDZ by wk 6, treatment effect was observed as early as wk 2; sustained reductions in PMS were observed in VDZ-treated patients over a 1-year period. Clinical remission rates in VDZ patients remained stable during the maintenance phase. Durable mucosal healing was achieved in more than twice as many VDZ patients as PBO patients. Continued dosing with VDZ induced response and remission among patients who did not achieve response at wk 6.