Introduction
Combination therapy targeting pathogenic pathways may overcome the current therapeutic efficacy ceiling and benefit patients with inflammatory bowel disease (IBD). Emerging preclinical evidence suggests that the costimulatory molecule OX40-ligand (OX40L) represents a promising target in both Crohn’s disease (CD) and ulcerative colitis (UC), particularly when inhibited in combination with anti-TNF agents. Despite these data, a comprehensive understanding of OX40L and its receptor, OX40, in IBD patients is lacking, especially related to potential synergy of OX40L and TNF blockade, particularly in previous anti-TNF non-responders.
Aims & Methods
We investigated OX40L biology in IBD using single-cell transcriptomic data from the terminal ileum of CD patients [n=73] and control subjects [n=106] (Open Targets Consortium) to query cell-specific gene expression profiles for OX40L and OX40 (TNFSF4, TNFRSF4). We also leveraged bulk transcriptomic studies profiling CD terminal ileum [n=99] and both UC [n=272] and control [n=18] colon, as well as available clinical metadata. We assessed enrichment of OX40L biology by gene set variation analysis (GSVA) using OX40L gene signatures. Patterns of gene and signature expression were then associated with endoscopic disease activity and longitudinal response to anti-TNF therapy.
Results
In cross-sectional analyses, increased mucosal OX40 gene expression was associated with higher endoscopic disease activity scores in both CD and UC patients (Simple Endoscopic Score, Mayo endoscopic score; p<0.05). Cellular estimates of lymphocyte subsets associated with resistance and/or residual disease post-anti-TNF therapy were found to correlate with mucosal enrichment of OX40L gene signature in both CD and UC (FDR<0.05; bicor range UC: 0.94-0.21, CD: 0.89-0.67). In longitudinal analyses, UC patients demonstrated elevated expression of OX40L gene signature in colon at baseline which, after anti-TNF therapy initiation, was progressively reduced in responders while persistently elevated in non-responders (p<0.05) throughout induction (week 4-8) and maintenance (week 30) phases. Exploratory single-cell analyses in CD terminal ileum showed cell type-specific increases in OX40L and OX40 gene expression relative to control, with enrichment in T-cell subsets (both), NK cells and dendritic cells (OX40L), myeloid and epithelial cells (OX40). In UC, stronger associations were found between OX40L gene signature enrichment and cellular estimates for multiple lymphocyte subsets (ILC1, PD1+CD4+ T-cells, plasma cells, mature and germinal center B-cells) as compared to TNF signature (FDR<0.05; bicor range OX40: 0.94-0.67, TNF: 0.69-0.37).
Conclusion
Expression of OX40L and its receptor OX40 is increased in pathogenic cell types from IBD intestinal tissue relative to control and operates in distinct cellular subnetworks compared to TNF. Additionally, increased OX40 gene expression is associated with higher disease activity in both CD and UC. OX40L signatures, which correlate with presence of lymphocyte subsets that may promote TNF resistance, are found to be persistently elevated in UC anti-TNF non-responders. These results provide a novel, data-driven rationale for combining anti-OX40L therapy with anti-TNF therapy in future experiments and clinical trials. This combination may optimize treatment efficacy in IBD by targeting mechanisms that are associated with anti-TNF resistance as well as residual disease following anti-TNF therapy.
Disclosure
CAR, RM, MRD, GF, MT, SK, EdR, CB, IW and HM are employees of Sanofi and may hold stock options or equity in Sanofi; ZH, HHB, RS, AC, ASM, YC are employees of CytoReason and may hold stock options or equity; SSO holds equity and is a consultant of CytoReason and holds an unpaid position with the Human Immunome Project
