Upadacitinib ranked highest for inducing clinical remission and endoscopic improvement in patients with moderate-to-severe ulcerative colitis (UC), according to a systematic review and network meta-analysis published online in Clinical Gastroenterology and Hepatology.
Interleukin-23 p19 inhibitors ranked most favorably for serious adverse events and adverse events leading to treatment discontinuation, although the authors rated confidence in the safety estimates as lower. Treatment rankings also shifted depending on patients’ previous exposure to advanced therapies.
The analysis included 33 phase 3 randomized controlled trials involving 14,325 patients. Upadacitinib had the highest relative efficacy for clinical remission, with patients receiving the treatment about 6.9 times as likely to achieve remission as those receiving placebo. The researchers rated confidence in this finding as high.
Obefazimod, an investigational agent not yet approved by the U.S. Food and Drug Administration, ranked second for clinical remission at its 50-mg dose, followed by subcutaneous guselkumab.
The analysis was led by Pablo A. Olivera, MD, MSc, of the Division of Gastroenterology at St. Michael’s Hospital, University of Toronto. The corresponding author was Laurent Peyrin-Biroulet, MD, PhD, of Northwestern Medicine’s Digestive Health Institute in Chicago.
“Because randomized trials directly comparing advanced therapies are very expensive and scarce, network meta-analyses offer a valuable alternative,” Berkeley N. Limketkai, MD, PhD, clinical professor of medicine and director of clinical research at the UCLA Center for Inflammatory Bowel Diseases, who was not involved in the study, told GI & Hepatology News.
“An integrative analysis of both efficacy and safety is crucial to balance the trade-offs between these outcomes,” the authors wrote.
Study design and methods
The researchers conducted a systematic review and network meta-analysis to compare the efficacy and safety of biologics and targeted small-molecule therapies used for induction treatment of moderate-to-severe UC. They searched MEDLINE, Embase, and Cochrane CENTRAL through Oct. 23, 2025.
Eligible studies were phase 3 randomized controlled trials enrolling adults aged 18 years or older with moderate-to-severe UC. The trials defined moderate-to-severe disease as a Mayo score of 6 to 12 or a modified Mayo score of 4 to 9, plus an endoscopic subscore of at least 2. Trials could be placebo-controlled or head-to-head comparisons. Patients enrolled in these trials could be either advanced therapy-naive or advanced therapy-experienced.
The analysis included therapies approved by the FDA or the European Medicines Agency as of October 2025, along with agents in late-stage development that had available phase 3 results and ongoing registration efforts. Agents included infliximab, adalimumab, golimumab, vedolizumab, ustekinumab, tofacitinib, upadacitinib, filgotinib, ozanimod, etrasimod, risankizumab, mirikizumab, intravenous guselkumab, subcutaneous guselkumab, and obefazimod. For approved therapies, only on-label doses were analyzed; for agents in late-stage development, all doses tested in phase 3 were included. Induction periods in the included trials ranged from six to 14 weeks.
The primary outcome was induction of clinical remission, as defined in each original study. Secondary outcomes included endoscopic improvement, defined as a Mayo endoscopic subscore of 0 or 1. They also included three safety outcomes: serious adverse events, adverse events leading to treatment discontinuation, and serious infections.
The researchers used frequentist random-effects models for the network meta-analyses, which allowed both direct and indirect treatment comparisons. Treatments were ranked using P-scores, with higher scores indicating more favorable relative rankings. Confidence in the findings was evaluated using the Confidence in Network Meta-Analysis framework.
Clinical remission outcomes
The clinical remission analysis included 33 trials and 14,325 patients. Across the included trials, advanced therapies were associated with about a 2.5-fold greater likelihood of clinical remission compared with their respective comparators.
In the network analysis, all interventions except filgotinib and mirikizumab demonstrated superiority over placebo for clinical remission.
Upadacitinib ranked first, with a P-score of 0.95. Obefazimod 50 mg ranked second, with a P-score of 0.77 and low confidence. Subcutaneous guselkumab followed, with a P-score of 0.74 and high confidence.
Outcomes by previous advanced therapy exposure
Among patients without previous exposure to advanced therapies, upadacitinib ranked highest for clinical remission, with a P-score of 0.84. Obefazimod 50 mg and 25 mg followed, with P-scores of 0.82 and 0.79, respectively.
Indirect comparisons favored upadacitinib over filgotinib, adalimumab, tofacitinib, and mirikizumab. Obefazimod 50 mg also demonstrated superiority over filgotinib and adalimumab.
Among patients previously exposed to advanced therapies, upadacitinib again ranked highest for clinical remission, with a P-score of 0.94. Ustekinumab followed at 0.88 and tofacitinib at 0.85. Infliximab ranked lowest and adalimumab second lowest in this group, for both clinical remission and endoscopic improvement.
“To our knowledge, our evidence synthesis is the first to provide insights on efficacy according to the number of previously failed advanced therapies,” the authors wrote.
Among patients previously exposed to at least two advanced therapies, only upadacitinib, guselkumab, and obefazimod 50 mg demonstrated superiority over placebo. The researchers noted that estimates in these smaller subgroups were less precise.
Endoscopic improvement
The endoscopic improvement analysis included 32 trials and 13,556 patients.
In the overall network, upadacitinib ranked first for endoscopic improvement, with a P-score of 0.97 and moderate confidence in the estimate. Patients receiving upadacitinib were about 5.3 times as likely to achieve endoscopic improvement as those receiving placebo.
Obefazimod 50 mg ranked second, with a P-score of 0.91 and about a 4.2-fold relative likelihood of improvement compared with placebo. Obefazimod 25 mg ranked third, with a P-score of 0.83 and about a 3.6-fold relative likelihood.
Among patients without previous advanced therapy exposure, obefazimod 25 mg ranked highest for endoscopic improvement, followed by obefazimod 50 mg and upadacitinib. Their respective P-scores were 0.94, 0.92, and 0.87.
Among previously exposed patients, upadacitinib ranked first, with a P-score of 0.99. Tofacitinib followed at 0.78 and obefazimod 50 mg at 0.77.
Safety outcomes
Safety was assessed in the overall population only.
Interleukin-23 p19 inhibitors generally had the most favorable rankings for serious adverse events and treatment discontinuation.
For serious adverse events, risankizumab ranked first for safety. It had a P-score of 0.92, a risk ratio of 0.23 compared with placebo, and high confidence in the estimate. Subcutaneous guselkumab ranked second, with a P-score of 0.80 and low confidence. Intravenous guselkumab ranked third, with a P-score of 0.73 and moderate confidence, although its estimate did not differ significantly from placebo.
For adverse events leading to treatment discontinuation, risankizumab ranked first, followed by subcutaneous guselkumab and mirikizumab. The corresponding risk ratios compared with placebo were 0.17, 0.19, and 0.22, respectively. Confidence in these estimates was low for risankizumab and subcutaneous guselkumab and high for mirikizumab.
For serious infections, infliximab ranked first for safety, followed by ustekinumab and golimumab. However, the researchers rated confidence in the infliximab estimate as very low. None of the three top-ranked therapies differed significantly from placebo.
The authors emphasized that confidence in the safety outcomes was generally limited. They also noted that few treatment comparisons demonstrated statistically significant differences.
Study limitations
The authors noted several limitations. Trial designs changed over two decades. Older trials were typically powered for clinical response and used looser remission definitions. Newer trials required a rectal bleeding subscore of 0 and centrally read endoscopy. These differences may have favored higher remission rates in older trials. The authors noted that upadacitinib was evaluated under the stricter definitions.
Steroid dosing and tapering also varied across trials. Upadacitinib trials allowed tapering during induction, whereas most others required stable doses. The authors wrote that upadacitinib’s top ranking across most outcomes and subgroups suggested this difference was unlikely to have changed the overall findings.
The safety rankings rested on less certain evidence. The analysis also covered induction only, because maintenance trial designs differ substantially.
The authors urged caution with comparisons involving tumor necrosis factor (TNF) inhibitors in previously treated patients. EFFICACI, a head-to-head trial of infliximab vs. vedolizumab, enrolled patients who had previously failed a subcutaneous TNF inhibitor, which may have lowered infliximab’s ranking. Excluding that trial improved infliximab’s estimates without changing the overall treatment hierarchy.
The authors wrote that, as with any indirect comparison, the results “should be interpreted as hypothesis-generating.”
Conclusion
The authors concluded that upadacitinib has high efficacy for inducing clinical remission and endoscopic improvement. In patients without previous advanced therapy exposure, they described obefazimod and ozanimod as good alternatives. Ustekinumab and tofacitinib remained highly ranked in previously exposed patients. The authors called for head-to-head trials to confirm the findings.
The results “aid individualized treatment selection by integrating efficacy, safety, prior therapy exposure, and confidence, rather than relying solely on rankings,” the authors wrote.
“Rankings should be considered alongside the magnitude of effect, certainty of evidence, and clinical context,” they added.
Expert insight
“The growing number of advanced therapies for ulcerative colitis has provided patients with a menu of safe and effective treatment options,” Limketkai said. “However, these therapies differ in their safety profiles and performance, so understanding these nuances can help inform treatment selection.”
“Olivera et al. were able to stratify advanced therapies according to their relative safety and efficacy,” he said. “These findings are helpful for guiding treatment decisions, although the choice of therapy should ultimately be made collaboratively by the clinician and patient, while taking into consideration the individual’s disease characteristics, comorbidities, preferences, and values.”
The study received no funding. All eight authors reported consulting fees, speaker honoraria, research support, or other financial relationships with drugmakers, including AbbVie, Abivax, Janssen, Pfizer, Takeda, and Eli Lilly.