The outcome of Infliximab induction in patients with severe ulcerative colitis may be related to intestinal microbiota

Background and objective Although infliximab (IFX) is endorsed for induction therapy in severe ulcerative colitis (SUC), its therapeutic response remains heterogeneous. We conducted a comprehensive analysis of fecal and mucosal microbiota, coupled with targeted metabolomics, to delineate microbial and metabolic signatures predictive of IFX induction efficacy and to explore mechanistic pathways underlying differential treatment responses. Methods This study was a prospective cohort study (clinical study registration number: ChiCTR2300071816). It enrolled adult patients aged ≥ 18 years who were first diagnosed with SUC at the First Affiliated Hospital of Nanjing Medical University and Northern Jiangsu People’s Hospital between February 2022 and December 2023. None of these patients had received any medication for UC, including antibiotics. Clinical data, fecal samples, and rectal mucosal samples were collected for analysis. High-throughput sequencing of the 16S rRNA gene and non-targeted metabolomics analysis were performed on both fecal and rectal mucosal samples. All patients underwent IFX (5 mg/kg) induced remission therapy at weeks 0, 2, and 6. Based on their clinical response at week 14, they were categorized into two groups: response and unresponse. Significant differences in bacterial composition between the two groups were identified by screening fecal and mucosal samples. The gut microbiota of feces and intestinal mucosa were combined with clinical data to create four prediction models and conduct comparisons. Furthermore, different metabolites from fecal and mucosal samples of the two groups were screened and compared with KEGG and PubChem databases to identify metabolic pathways associated with the efficacy of IFX-induced therapy. Results Compared with non-responders, IFX responders harbor distinct gut microbiota. Clinical indices alone poorly forecast induction response (AUC 0.6429). Augmenting clinical variables with fecal or mucosal microbiota improves prediction to AUC 0.795 and 0.900, respectively; combining both compartments further elevates performance to AUC 0.964, indicating that integrated microbiome profiling is essential for optimal IFX response prediction. IFX responders and non-responders differ metabolically, with more discriminatory features in mucosa than feces. The most enriched differential pathways in feces included nicotinate and nicotinamide metabolism, butyrate metabolism, biosynthesis of valine, leucine, and isoleucine, pantothenate and coenzyme A biosynthesis, histidine metabolism, and alanine, aspartate, and glutamate metabolism. In mucosa, the most enriched differential pathways included alanine, aspartate, and glutamate metabolism, sphingolipid metabolism, ascorbate and aldarate metabolism, tryptophan metabolism, and D-glutamine and D-glutamate metabolism. The common pathway enriched in both feces and mucosa was alanine, aspartate, and glutamate metabolism. Conclusion The intestinal microbiota may be a predictive factor for IFX induction outcome in patients with SUC. The metabolic pathway of alanine, aspartate, and glutamate may be associated with the 14-week clinical response to IFX treatment in SUC.