S794 Pathway Enrichment Analysis Reveals Ulcerative Colitis Patients With Non-Response to TNFi Therapy May Have More Biological Dysregulation

Abstract

Introduction: Tumor necrosis factor-α inhibitors (TNFi) are essential to ulcerative colitis (UC) management, however only 50% of patients will achieve a clinical response during induction therapy. There is an unmet need to identify and predict UC therapy response. This study aims to analyze potential etiologies for lack of TNFi response using pathway enrichment analysis. Methods: Differential expression analysis between TNFi responders, non-responders, and healthy controls was conducted from Gene Expression Omnibus UC patient cohorts (Table) using the limma package. All datasets contained post-treatment mucosal gene expression data and corresponding response, defined as endoscopic healing at 4 to 8 weeks after treatment initiation. Enrichment analysis of signaling pathways from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database was performed to compare the gene expression profile dysregulation severity between responders and non-responders. Genes exclusive to each group for select significantly enriched pathways were identified. The difference between the number of responder and non-responder exclusive genes was computed and assessed using the random label permutation test. Significantly different pathways between non-responder-exclusive genes and responder-exclusive genes were reported. Results: Several differentially expressed genes between responders and healthy controls (R-set) and between non-responders and healthy controls (NR-set) were identified (Figure a). Multiple overlapping dysregulated genes were observed between the two groups. Non-responders had significantly more differentially expressed genes (Figure b). Pathway enrichment analysis on the R- and NR-sets demonstrated that 40 of 282 KEGG pathways were significantly enriched with non-responder genes (p< 0.05). Of the 40 pathways, 28 had significantly more NR-exclusive genes than R-exclusive genes (p< 0.05) (Figure c). The NR-set included unique differentially expressed genes involved in cytokine signaling, receptor mediation, and signal transduction. Conclusion: UC-relevant KEGG pathways are significantly more enriched and disrupted in non-responders compared to responders, suggesting that non-responders have more dysregulated biological pathways. Other enriched pathways highlight the role of inflammation, barrier integrity, and the intestinal microbiome in UC. This study illustrates the potential value of precision medicine to predict clinical response to TNFi in UC patients.Figure 1.: KEGG pathway enrichment analysis for genes differentially expressed in responders and non-responders at baseline with respect to healthy controls. (a) schematic illustration of the differential expression gene sets obtained by comparing different pairs of responders, non-responders, and healthy controls; (b) Venn diagram showing responders’ and non-responders’ differentially expressed genes at baseline with respect to healthy controls after merging infliximab- and golimumab-based cohorts. Merging is done by taking the intersection of differential expression gene sets from infliximab- and golimumab-based cohorts; (c) KEGG pathways significantly enriched with NR-gene set that also have significantly more NR-exclusive genes than R-exclusive genes. UC-related pathways are shown here. Table 1. - Number of patients from Gene Expression Omnibus UC patient cohort datasets included in analyses Cohort Dataset Control Responders Non-Responders Infliximab GSE16879 6 8 16 GSE23597 - 24 7 Golimumab GSE92415 21 32 27