DISEASE MODELING IN INFLAMMATORY BOWEL DISEASE (IBD) USING PATIENT-DERIVED EXPLANTS (PDE) REVEALS CHEMOKINES AS A RELEVANT TARGET OF JAK INHIBITION

Abstract Despite a recent explosion of approved drugs for IBD that primarily target cytokines and T-cell trafficking, therapeutic response rates remain low (30-50 %), underscoring the need for more efficacious treatments with fewer side effects. From a drug development perspective, it takes over 10 years for a new drug to reach the market with >90% attrition rates. The realization that drug testing in animal models does not always translate to humans, particularly for complex diseases of high heterogeneity like IBD, leaving a significant gap in pre-clinical drug evaluation. Technologies such as organoids or “gut on a chip” attempted to fulfill this gap, but among their limitations are lack of tissue complexity and an externally induced, and often biased, “disease state” that fails to capture patient heterogeneity and underlying disease mechanisms. We developed a Therapeutic Response Predictor Assay for IBD (TheRPA _ibd) which is based on establishing short term explant cultures from the patient’s diseased tissue (PDE). PDEs can be treated with various candidate drugs or their combinations and inform whether an individual patient is a responder or non-responder to the particular treatment. TheRPA_ibd has been clinically validated for anti-TNF with a predictive accuracy of 0.87 (p<0.0001). TheRPA_ibd was used to comparatively evaluate 6 JAK inhibitors (JAKi) with different JAK isoform selectivity (tofacitinib, baricitinib, upadacitinib, ritlecitinib, filgotinib, and peficitinib). PDEs were derived from mucosal inflamed biopsies obtained during colonoscopy or surgical resection (26 CD/11 UC). JAKi responses were evaluated using targeted and bulk gene expression, antibody arrays and multiplex ELISA. Our analysis, in addition to previously described cytokines, identified several chemokines as downstream targets of JAK inhibition: CXCL1; CXCL2; CXCL3; CXCL5; CXCL8; CXCL9; CXCL10; CXCL11; CXCL16; CCL10; CCL20. Of clinical importance, interrogation of publicly available gene expression IBD datasets confirmed a robust dysregulation of those targets in both UC and CD. Such findings open new avenues for exploring chemokine-driven pathways as potential therapeutic targets in IBD and potentially additional inflammatory conditions. From a mechanistic perspective, JAK1 or JAK3 selective inhibitors appeared to lack broader cytokine and chemokine inhibition compared to pan-JAKi. Notably, as a target, TYK2 was found to be critical for maximizing the anti-inflammatory effects of JAKi. The PDE-based assay (TheRPA) not only enables selection of patient-tailored treatments but also as a “clinical trial in a dish” platform can efficiently contribute to candidate drug de-risking and prioritization. As a proof of concept, using this technology we have been able to demonstrate efficacy in CD of certain JAK inhibitors prior to their FDA approval for this indication.