Abstract Background Trials evaluating ulcerative colitis (UC) therapies necessitate instruments that are responsive to changes in disease status. We assessed responsiveness of the Mayo Endoscopic Subscore (MES), Ulcerative Colitis Endoscopic Index of Severity (UCEIS), Robarts Histopathology Index (RHI), Geboes Score (GS), Nancy Histological Index (NHI), and UC-100 score. Methods Using data from a phase 2 placebo-controlled trial of mirikizumab, responsiveness was quantified by the probability that a patient in the treatment group would have a better score than (win over) a control patient. Inference on win probability (WinP) was conducted without normality assumption of scores. Spearman rank correlations between each index and stool frequency (SF), rectal bleeding (RB), faecal calprotectin (FCAL), and C-reactive protein (CRP) were assessed. Results WinP estimates showed that magnitudes of responsiveness were “medium” for the UCEIS (0.67; 95% CI 0.59, 0.74) and UC-100 score (0.68; 95% CI 0.60, 0.75) and “small” for the RHI (0.62; 95% CI 0.54, 0.70), MES (0.61; 95% CI 0.53, 0.68), and GS (0.60; 95% CI 0.52, 0.68). The NHI showed no statistical evidence of responsiveness (0.57; 95% CI 0.48, 0.65). The UCEIS and MES had weak-to-moderate correlations with SF (0.21-0.26) and RB (0.29-0.33) but no correlation with biomarkers. The UC-100 had moderate correlations with RB (0.46) and FCAL (0.42) and weak correlation with CRP (0.19). Conclusions The UCEIS and UC-100 score were responsive to changes in endoscopic and global UC disease activity, respectively. Results may help inform future early phase UC drug trials but should be confirmed in larger studies.
Vuyyuru et al. (Mon,) studied this question.