Abstract
BACKGROUND: Ulcerative colitis (UC) is a chronic inflammatory disease of the colonic mucosa, extending proximally from the rectum. However, the segmental pattern of healing in UC remains unclear. Endoscopic improvement (EI), a key therapeutic endpoint, is typically assessed using the Mayo endoscopic score (MES), which scores the worst affected area and may miss partial/segmental healing. This study evaluates healing patterns in UC and compares conventional MES with a 3-segment MES approach for detecting treatment effects in clinical trials. METHODS: A post hoc analysis of HIBISCUS I/II and GARDENIA trials was conducted in UC patients with moderate-to-severe disease (MES >2 up to the descending colon). The primary outcome was the proportion of anti-tumor necrosis factor-treated participants achieving MES ≤1 in the descending colon, sigmoid colon, and rectum at week 10. Secondary outcomes included conventionally measured EI, segmental MES improvements, clinical response, and Patient-Reported Outcome 2 (PRO2) normalization. Outcomes were compared between adalimumab, infliximab, and placebo groups. RESULTS: Among 300 participants, 217 received infliximab or adalimumab, while 83 received placebo. Healing followed a proximal-to-distal pattern, with the highest EI in the descending colon and the lowest in the rectum. Infliximab-treated patients continued this trend at week 54. Anti-tumor necrosis factor therapy significantly improved EI vs placebo (42.9% vs 19.3%; P < .001). No segmental MES approach outperformed conventional MES for detecting treatment effects. Combined endpoints (MES ≤1 + PRO2 normalization) better captured therapeutic benefits than PRO2 alone (28.6% vs 13.3%; P = .006). CONCLUSIONS: UC healing follows a proximal-to-distal pattern. Conventional MES remains superior for detecting treatment effects over segmental MES. Further studies should explore alternative endoscopic scoring methodologies.