OP06Novel mouse models reveal therapeutic value of anti-IFN-g for perianal fistulizing CD

Abstract Background Perianal fistulizing Crohn’s disease (PFCD) is associated with significant morbidity and therapeutic resistance. Our recent multi-omics study identified hyperactivated IFN-g pathways in PFCD (Cao 2025 JCI). Further studies require clinically relevant and reliable in vivo models that allow easy drug screening and genetic modification. Methods We intended to establish novel PFCD models in mice with CD-like inflammation: 2,4,6-trinitrobenzenesulfonic acid (TNBS)-induced CD colitis in wild-type (WT) mice, Il10-/--mediated proctocolitis, and Tnfa overexpression (TNFΔ69AU unpublished)-mediated ileocolitis. A surgical fistula was created in the perianal area of WT-TNBS, Il10-/-, and TNFΔ69AU mice; a wire was kept in place to allow fistula tract to mature. After wire removal, fistulas were evaluated by examination under sedation and MRI weekly and by histology at the end of experiment. We modified the MAGNIFI-CD index for PFCD to assess disease activity and therapeutic response of perianal fistulas in mice. Fistula tracts, rectal mucosa, and perianal skin tissue were harvested at different time points for RNA-seq, flow cytometry, ELISA, and immunofluorescence (IF). Fistula and intestinal microbiome is analyzed using 16S rRNA-seq. The models were also treated with upadacitinib and anti-IFN-g. Results All three models form chronic and inflammatory perianal fistulas that closely resemble PFCD in patients based on clinical examination, MRI, histological assessment as well as molecular analyses. Importantly, mice sustain patent fistula tracts for at least 5 weeks after wire removal, making them reliable tools for mechanistic/functional studies and drug screening. Consistent with human PFCD data, IF and RNA-seq showed upregulated IFN-g signaling pathways, chronic inflammation, elevated chemotaxis, altered cellular stress response, heightened epithelial-to-mesenchymal transition, etc. Flow cytometry of perianal fistula tissues further revealed increased IFN-g+ immune cells and Th17 cells like in human PFCD. Furthermore, mice with perianal fistula exhibit altered intestinal microbiome compared to those without a fistula (Figure 1). Similar to some PFCD patients, our models respond therapeutically to upadacitinib. Importantly, the murine fistulas improved significantly after 1-2 doses of anti-IFN-g, which suppressed p-STAT1 downstream of IFN-g (Figure 2). Conclusion We have established three reliable, biologically and clinically relevant, therapeutically responsive mouse models for PFCD. We expect they will be new in vivo tools to study PFCD pathogenesis and develop novel therapies. The prompt improvement of murine perianal fistulas on anti-IFN-g supports IFN-g pathways as novel therapeutic targets for PFCD and warrants further investigation. Conflict of interest: Yao, Xin: No conflict of interest Zhu, Echo: No conflict of interest Ma, Kaiming: No conflict of interest Ballard, David: No conflict of interest Randolph, Gwendalyn: No conflict of interest Deepak, Parakkal: Grant: Johnson and Johnson, AbbVie, Takeda Pharmaceuticals, Prometheus Biosciences, Teva Pharmaceuticals, Sanofi, Merck, ExeGI Pharmaceuticals, Agomab, Tr1X, Direct Biologics, Astra Zeneca and Eli Lilly. Personal Fees: Johnson and Johnson, CorEvitas LLC, Eli Lilly, Sobi Cao, Siyan: No conflict of interest