Safety profile and potential favorable reporting signals of liraglutide in patients with ulcerative colitis: A FAERS pharmacovigilance and network pharmacology study

Abstract: Background: Liraglutide, a glucagon-like peptide-1 receptor agonist (GLP-1 RA), has preclinical anti-inflammatory and barrier-protective effects, but its safety profile and potential favorable reporting associations in ulcerative colitis (UC) remain uncertain. Objectives: To characterize liraglutide-associated reports in patients with UC and explore biological pathways that may contextualize the observed pharmacovigilance signals. Methods: Reports submitted to the U.S. Food and Drug Administration Adverse Event Reporting System (FAERS) from January 1, 2013, to December 31, 2025, were analyzed in which liraglutide was recorded as the primary suspected drug and UC as the indication. Disproportionality was assessed using the reporting odds ratio (ROR), proportional reporting ratio (PRR), Bayesian confidence propagation neural network (BCPNN) and multi-item gamma Poisson shrinker (MGPS). Network pharmacology identified shared targets, constructed a protein-protein interaction network and performed Gene Ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analyses. Results: Among 5,903 unique cases and 28,607 preferred-term event entries, the most frequent categories were general disorders and administration-site conditions, gastrointestinal disorders, injury or medication-related complications, infections and investigations. The term “condition aggravated” was reported in 1,512 cases. Medication management, metabolic, infectious, neuropsychiatric and gastrointestinal signals were also detected. Herpes zoster was reported in 154 cases and pneumonia-related reports included eight fatal outcomes, although concomitant immunosuppressive therapy substantially limited attribution. The reporter-coded term “pre-existing condition improved” occurred in 38 cases (ROR 25.63; 95% confidence interval 16.48-39.85) and was interpreted only as an exploratory favorable reporting association. Network pharmacology identified 84 shared candidate targets and prioritized 10 hub genes. Enriched pathways included phosphoinositide 3-kinase/protein kinase B, tumor necrosis factor and Janus kinase/signal transducer and activator of transcription signaling.  Conclusion: FAERS findings generated safety and favorable-reporting hypotheses but did not establish incidence, causality, or therapeutic efficacy. The predicted pathways provide biological context only and require confirmation in comparator-based longitudinal studies, prospective clinical investigations and targeted experimental models.