Nitin Rai, MD1, Vithyaa Premjeyanth, MD2, Nasruddin Sabrie, MD2, Parul Tandon, DO, PhD3, Valentin Ritschl, PhD4, Tanja Stamm, PhD4, Walter Reinisch, MD, PhD5 1McMaster University, Hamilton, ON, Canada; 2University of Toronto, Toronto, ON, Canada; 3UHN, Toronto, ON, Canada; 4Medical University of Vienna, Spitalgasse, Wien, Austria; 5Medical University of Vienna, Department of Internal Medicine III, Division of Gastroenterology and Hepatology, Spitalgasse, Wien, Austria Introduction: Placebo-controlled trials in Crohn’s disease (CD) remain standard but raise ethical concerns, particularly for prolonged maintenance phases which are designed to determine clinical remission. Identifying early predictors of long-term outcomes could guide trial design to minimize placebo exposure and potential harm from lack of treatment. We conducted a systematic review and multiple meta-regression to evaluate whether induction phase clinical endpoints predict respective maintenance phase clinical endpoints in phase 2 and 3 randomized controlled trials (RCTs) of advanced CD therapies like biologics and small molecules. Methods: We searched Medline, Embase, major conference abstracts, and ClinicalTrials.gov up to 2022. Eligible studies were phase 2/3 RCTs of advanced therapies in moderate-to-severe CD reporting both induction and maintenance clinical outcomes. We extracted trial characteristics, induction and maintenance outcomes (clinical response and remission per study-defined criteria), and baseline patient data. We used the random effects model to pool effect sizes for achieving clinical response and clinical remission followed by calculation of odds ratios and risk ratios. A multiple meta-regression using a mixed-effect model assessed variables in induction trials predictive of maintenance outcomes, including induction response, induction remission, baseline Crohn’s Disease Activity Index (CDAI), biologic dose, and demographics. Results: Nine RCTs (n=3327 induction, n=1146 maintenance) were included, covering seven biologics: ustekinumab, risankizumab, vedolizumab, adalimumab, upadacitinib, CDP571, and vercirnon. Pooled risk ratios for achieving clinical response and remission were 1.49 (95% CI 1.37-1.62, I²=38.0%) and 1.61 (95% CI 1.41-1.84, I²=37.0%) in induction, and 0.96 (95% CI 0.87-1.05, I²=42.0%) and 1.00 (95% CI 0.92-1.09, I²=0%) in maintenance, respectively. Achieving clinical remission in induction, baseline CDAI, and biologic dose were significant predictors of maintenance response and remission, while mean age and disease duration were neither predictive. Discussion: Clinical remission achieved during induction predicts both clinical response and remission in maintenance phases of CD trials. These findings support limiting placebo exposure to induction phases and using early outcomes to guide trial continuation. Further research is needed to assess the predictive value of endoscopic and histologic endpoints, given inconsistent reporting across trials.
