GLP-1 Receptor Agonists Confer No Increased Rates of IBD Exacerbation Among Patients With IBD
Levine I, Sekhri S, Schreiber-Stainthorp W, Locke B, Delau O, Elhawary M, Pandit K, Meng X, +1 more
Inflammatory Bowel Diseases (2025) · n=224
Among 224 IBD patients prescribed GLP-1 receptor agonists at a large US academic health network between 2009 and 2023, twelve-month rates of IBD exacerbation, IBD-related hospitalization, corticosteroid use, medication escalation, and surgery did not differ from pre-prescription baseline — a null safety signal that independently corroborates the Israeli Epi-IIRN findings.
This is the NYU Langone retrospective cohort analysis that constitutes the principal US-based independent replication of the favorable GLP-1-IBD signal observed in the Israeli Epi-IIRN cohort. The motivating clinical question was whether GLP-1-receptor-agonist prescription — increasingly common in IBD patients carrying comorbid obesity or type 2 diabetes — increases the rate of IBD exacerbation in the year following initiation, given the GLP-1-class biology on gastric emptying, gastrointestinal motility, and the unresolved theoretical questions about GLP-1-receptor effects on intestinal immune-mucosal crosstalk.
Investigators identified all IBD patients prescribed a GLP-1 receptor agonist at a large US academic healthcare network (NYU Langone) between 2009 and 2023 through pharmacy-claims and electronic-health-record linkage, yielding a cohort of 224 patients (a mix of ulcerative-colitis, Crohn's-disease, and indeterminate-colitis diagnoses). The pre-and-post analysis compared IBD-relevant clinical events in the 12-month window before GLP-1-agonist initiation versus the 12-month window after, using each patient as their own control. The primary composite outcome captured IBD-related hospitalization, corticosteroid prescription, IBD-medication escalation or change, and IBD-related surgery; secondary endpoints captured component-level events, BMI change, and disease-activity proxies.
Across the 224 patients, the 12-month post-initiation rate of the composite IBD-exacerbation outcome did not significantly differ from the 12-month pre-initiation rate. Component-level events — IBD-related hospitalization, corticosteroid prescription, medication escalation, IBD-related surgery — were each numerically unchanged or trended toward reduction. BMI decreased significantly across the cohort, consistent with the metabolic mechanism the prescriptions were issued to address. The authors framed the result as the first US-based observational evidence that GLP-1-agonist initiation in IBD patients is not associated with elevated rates of IBD exacerbation in the year following initiation, with the implication that the GLP-1-class metabolic-disease benefit can be pursued in IBD patients without an incremental flare-risk concern that would otherwise constrain prescribing decisions.
Cross-link to the semaglutide peptide page, to the IBD-and-peptides dossier where this paper sits alongside the Gorelik et al. 2025 Epi-IIRN cohort as the two independent observational signals that anchor the 2024–2025 IBD-and-GLP-1 literature, and to the Bayoumy et al. 2025 systematic review and meta-analysis that pooled this and ten further observational studies covering 16,242 IBD patients.
The 224-patient cohort is the smallest of the major 2024–2025 GLP-1-IBD observational analyses; statistical precision is lower than in the larger Israeli and pooled-meta-analysis datasets. The pre-and-post design with each patient as their own control reduces between-patient confounding but introduces a temporal-trend confound: any secular improvements in IBD-disease-activity standard of care across the 2009–2023 study window would manifest as a post-period improvement attributable to those background trends rather than to the GLP-1 prescription. The cohort is selected — every patient was prescribed a GLP-1-agonist for metabolic indications (obesity, type 2 diabetes), and the prescribing decision was conditioned on patient and clinician judgment in ways an electronic-health-record extraction cannot fully characterize. Disease severity and IBD-medication regimens at GLP-1 initiation are not as systematically reported as in a prospective design. Single-center US academic-health-network sampling constrains generalizability to non-academic and non-urban populations. The composite outcome combines events of varying clinical significance; component-level signals are directionally consistent but underpowered for definitive resolution at this sample size. No endoscopic, mucosal-healing, calprotectin, or CRP biomarker data are systematically reported. The null finding on exacerbation is informative — and is the primary clinical question motivating the analysis — but does not establish a positive disease-modifying effect; the question whether GLP-1 agonism actively reduces IBD-disease activity (as opposed to not worsening it) requires prospective interventional design that has not yet read out. Industry independence is high; the analysis was investigator-initiated and US-academic-funded with no GLP-1-manufacturer involvement disclosed.
07·Member discussion
No member discussion yet.
Member-only conversation lives here — cycle notes, practitioner commentary, pattern-matching. Be the first paying member to start the thread.