OP12Predictors of response to anti-TNF treatment in Inflammatory Bowel Disease: A registry-based approach

Abstract Background In patients with IBD treated with an anti-TNF (tumour necrosis factor) drug like infliximab or adalimumab, between 13-40% of patients experience primary non-response (PNR) and around 50% of patients experience an eventual secondary loss of response (LOR).1 None of the biomarkers previously presented have been sufficiently predictive of treatment outcome for clinical use,2 and only few genetic associations remain significant following meta-analysis.3 As many previous studies have been limited by small sample sizes, we aimed to establish a phenotype for PNR and secondary LOR to anti-TNF therapy in patients with IBD utilizing large-scale, Danish nationwide registry-based data, to analyse numerous potential associations with treatment response. Methods We included all patients with IBD in Denmark who received infliximab or adalimumab as their first biologic therapy and completed therapy induction. Events considered indicative of PNR or LOR were prescription of systemic corticosteroids, major IBD-related surgery or hospitalization, lack of reduction in CRP, or need for intensification of treatment or a different biologic. Variables tested for association with treatment response included demographic information, biochemical samples, concurrent medication during induction, severity and behaviour of IBD, genetic data, and polygenic scores. Results In total, 14,072 patients were included, with 19.0% experiencing PNR, 43.2% LOR, and 37.8% remaining in remission. We found an increased hazard of LOR with longer time to biologic therapy, a protective effect of taking an immunomodulator during induction, a protective effect of male sex, multiple associations with baseline biochemical samples, and an apparent genetic gradient in treatment outcomes based on polygenic score ratios. Finally, we present a model with moderate abilities in predicting PNR to infliximab (UC AUC: 0.72, CD AUC: 0.74), possibly allowing clinicians to initiate a change in treatment strategy earlier than today. Conclusion In this study, we present a large-scale, registry-based approach to determining associations with anti-TNF treatment outcomes in patients with IBD, including over 14,000 patients in a population-based cohort. We were able to analyse far more patients than what is possible in clinical studies, confirming numerous previously reported associations, presenting novel biochemical and genetic associations, and showing findings in support of the recent shift from a bottom-up to a top-down approach in treating IBD. We believe our study to be the largest population-based analysis of associations with treatment response in IBD so far. References: 1.Ballesta-López O, Gil-Candel M, Centelles-Oria M, et al. Pharmacogenetics in Response to Biological Agents in Inflammatory Bowel Disease: A Systematic Review. Int J Mol Sci. 2025;26(4). doi:10.3390/IJMS26041760 2.Stevens TW, Matheeuwsen M, Lönnkvist MH, et al. Systematic review: predictive biomarkers of therapeutic response in inflammatory bowel disease—personalised medicine in its infancy. Aliment Pharmacol Ther. 2018;48(11-12):1213-1231. doi:10.1111/APT.15033, 3.Al-Sofi RF, Bergmann MS, Nielsen CH, Andersen V, Skov L, Loft N. The Association between Genetics and Response to Treatment with Biologics in Patients with Psoriasis, Psoriatic Arthritis, Rheumatoid Arthritis, and Inflammatory Bowel Diseases: A Systematic Review and Meta-Analysis. Int J Mol Sci. 2024;25(11). doi:10.3390/IJMS25115793 Conflict of interest: Lie, Mads: No conflict of interest Vestergaard, Marie Vibeke: No conflict of interest Asaad, Adam Kamakh: No conflict of interest Larsen, Lone: Speaker fee from Takeda, Eli Lilly, Abbvie. Share holder Novo Nordisk A/S. Advisory Board for Tillotts, Abbvie, Eli Lilly, Celltrion. Consultancy for Eli Lilly. Sazonovs, Aleksejs: No conflict of interest Jess, Tine: Consultancy for Ferring, Pfizer, and Johnson&Johnson.