A Narrative Review on Disease Models Used in the Study of Inflammatory Bowel Disease

Inflammatory bowel disease (IBD), which includes Crohn’s disease and ulcerative colitis, is a chronic, relapsing immune-mediated disorder of the gastrointestinal tract. Its pathogenesis involves complex interactions among genetic susceptibility, the intestinal microbiota, environmental factors and dysregulated mucosal immunity. The development and refinement of disease models have been central to mechanistic understanding and therapeutic innovation in IBD. This review provides a comprehensive overview of the major disease models currently employed in IBD research, including chemically induced rodent models (dextran sodium sulfate, trinitrobenzene sulfonic acid and oxazolone), genetic and immune-transfer models (IL-10 knockout, T-cell transfer colitis, MDR1a knockout, SAMP1/YitFc ileitis and Nod2/Nox2 knockout model mice), microbiota-associated models (germ-free, gnotobiotic and human microbiota-associated mice), humanized mouse models, and human-derived in vitro and ex vivo platforms (intestinal organoids, colonoids, explant cultures and organ-on-a-chip). We discuss the mechanistic basis, histopathological features, advantages and limitations of each model in the context of Crohn's disease or ulcerative colitis biology. We further highlight recent advances including patient-derived organoid biobanks, organoid-immune cell co-culture systems, CRISPR-engineered IBD knock-in models, the integration of single-cell and spatial transcriptomics and AI-assisted model prediction. Critical gaps remain, particularly in modeling intestinal fibrosis and accounting for sex as a biological variable. No single model fully recapitulates the multifaceted pathogenesis of human IBD; strategic selection and complementary use of multiple models across biological scales remain essential for translational progress. This review aims to guide researchers in selecting appropriate models and interpreting findings within their biological and translational context.