Effectiveness of anti-TNF therapy after prior ustekinumab exposure in Crohn’s disease: A propensity-weighted cohort study

Background As ustekinumab is increasingly used earlier in Crohn’s disease (CD), evidence regarding the effectiveness of subsequent anti-tumor necrosis factor (TNF) therapy remains limited. We compared anti-TNF outcomes after prior ustekinumab exposure with first-line anti-TNF therapy in biologic-naïve patients. Methods We conducted a registry-based, propensity score-weighted cohort study within the prospective Multi-omics Predicting Response to Biologics in Inflammatory Bowel Disease registry. Adults with active CD who initiated infliximab or adalimumab between September 2022 and May 2025 were classified as biologic-naïve patients receiving anti-TNF therapy as first advanced therapy or ustekinumab-exposed patients receiving anti-TNF therapy as second advanced therapy. Stabilized inverse probability of treatment weighting was used to reduce measured baseline imbalance. The primary endpoints were steroid-free clinical remission at weeks 14 and 52. Endoscopic and intestinal-ultrasound outcomes were considered exploratory assessed-subset endpoints. Results Among 173 patients, 117 were ustekinumab-naïve and 56 were ustekinumab-exposed. At week 14, weighted steroid-free clinical remission was 60.8% and 53.4%, respectively (adjusted odds ratio [aOR] for naïve vs exposed, 1.34; 95% confidence interval [CI], 0.54–3.31). Week-52 clinical assessments were available for 105/117 (89.7%) and 40/56 (71.4%) patients; corresponding steroid-free remission rates were 65.2% and 51.4% (aOR, 1.91; 95% CI, 0.69–5.28). No statistically significant between-group differences were detected in other clinical, biochemical, endoscopic, or transmural outcomes, although objective reassessment was selective and estimates were imprecise. IPTW-weighted 52-week treatment-persistence probabilities were 77.4% (95% CI, 67.4–88.8%) and 81.9% (95% CI, 69.2–96.9%), respectively (HR, 1.32; 95% CI, 0.46–3.74). At least one recorded adverse event occurred in 18.8% and 10.7% of patients, respectively; safety findings were descriptive. Conclusions Anti-TNF therapy achieved clinically meaningful responses in a substantial proportion of patients after prior ustekinumab exposure. However, prior exposure was inherently linked to treatment line, the exposed cohort was small, week-52 follow-up was incomplete, and effect estimates were imprecise. The study therefore does not establish equivalence with first-line anti-TNF therapy, and clinically meaningful differences cannot be excluded. Clinical trial number Not applicable