Presentation (CTI)

OBJECTIVES: Ustekinumab (UST) and vedolizumab (VED) are established advanced therapies for moderate-to-severe inflammatory bowel disease (IBD). Although both have demonstrated efficacy in randomized controlled trials, direct head-to-head comparisons in routine clinical practice are limited, and real-world data from Middle Eastern populations are particularly scarce. This study compared the real-world effectiveness of UST and VED in achieving corticosteroid-free clinical remission at 12 months in a Saudi cohort of IBD patients.
METHODS: Single-centre, retrospective cohort of 205 IBD patients initiating UST (n=105) or VED (n=100) at a tertiary referral center, identified from electronic medical records [Dec2016-May2026]. The primary outcome was corticosteroid-free clinical remission at 12 months; secondary at 6 months. Baseline characteristics were summarized descriptively, and group balance was evaluated using standardized mean differences. We estimated crude and multivariable-adjusted odds ratios (OR), with sensitivity analyses using multiple imputation for missing CRP and propensity-score average-treatment-effect-on-the-treated (ATT) weighting.
RESULTS: Ustekinumab-treated patients more often had Crohn's disease (78% vs. 30%), severe disease (86% vs. 67%), prior biologic exposure (91% vs. 61%), and prior bowel surgery (36% vs. 11%), suggesting substantial treatment channelling. Corticosteroid-free remission was assessable in 156 patients at 6 months and 135 at 12 months. In unadjusted analyses, remission favoured vedolizumab at 6 months (41.8% vs. 58.4%; OR 0.51, 95% CI 0.27-0.96) and 12 months (39.0% vs. 65.8%; OR 0.33, 0.16-0.67). After adjustment, the effects were reduced, with no significant difference at 6 months (adjusted OR 0.92, 0.41-2.10) or 12 months (adjusted OR 0.52, 0.21-1.29). Preliminary propensity-score ATT-weighted analyses with multiple imputed CRP showed consistent results; these analyses are ongoing.
CONCLUSIONS: The differences in corticosteroid-free clinical remission between ustekinumab and vedolizumab were reduced after adjustment for baseline characteristics, suggesting that confounding by indication contributed to the crude treatment differences. Ongoing sensitivity analyses using multiple imputation and propensity score weighting will further evaluate the robustness of these findings.