Predictors of Placebo Induction Response and Remission in Ulcerative Colitis

Abstract

High placebo response rates in clinical trials of ulcerative colitis (UC) have been reported previously. However, data from patient-level analyses are lacking. We assessed factors associated with clinical and endoscopic placebo response among placebo-treated patients in clinical trials of UC. We performed a post hoc analysis of pooled clinical trial data from GEMINI-1, ACT-1, ACT-2, PURSUIT, ULTRA-2, OCTAVE-1, and OCTAVE-2. Predictors were assessed in placebo-treated patients for their association with end of induction (week 6 of 8) clinical response (reduction in total Mayo score of ≥3 and ≥30% from baseline with ≥1 point decrease in rectal bleeding subscore [RBS] or absolute RBS ≤1); clinical remission (total Mayo score ≤2 and no subscore >1); endoscopic healing (Mayo endoscopic subscore ≤1); partial Mayo score of 0; patient-reported outcome 2-item remission (RBS of 0 and stool frequency ≤1), resolution of rectal bleeding, and stool frequency normalization. Predictors on univariate analyses with P < .05 were included in multivariate logistic regression models. Placebo-treated patients with normal serum C-reactive protein and albumin levels were more likely to attain clinical response (71 of 437 [16.3%] vs 49 of 660 [7.4%]; adjusted odds ratio, 2.76; 95% confidence interval, 1.19-5.41; P= .018). Compared with patients with a Mayo endoscopic score of 2, patients with a Mayo endoscopic score of 3 were less likely to attain clinical response (105 of 556 [18.8%] vs 179 of 675 [25.9%]; adjusted odds ratio, 0.33; 95% confidence interval, 0.16-0.68; P= .003). Similar findings were observed for clinical remission and resolution of rectal bleeding. Biomarkers such as normal serum C-reactive protein and albumin and baseline endoscopic severity were found to affect placebo response rates in clinical trials of UC. These findings have implications for clinical trial design in UC.