DOP05 Single-cell RNAseq temporal analysis of ulcerative colitis patients undergoing tofacitinib treatment reveals a shift in myeloid cells towards pro-inflammatory phenotypes in refractory patients

Abstract

Abstract Background Tofacitinib is an oral JAK1 and JAK3 inhibitor approved for the treatment of moderate to severe ulcerative colitis (UC). The aim of this study is to identify the relevant cellular subsets and genes involved in response and/or resistance to tofacitinib using temporal single-cell RNA sequencing (scRNA-seq). Methods Colon biopsies from patients with active moderate to severe UC starting tofacitinib 10 mg bid for 8 weeks and 5 or 10 mg bid thereafter were collected at baseline (n=11) and during follow-up (week 8, 16, 24 and/or 48; n=12 samples) and processed by scRNA-seq. Patients were classified as responders (R) or non-responders (NR) based on the decrease in the endoscopic Mayo score of at least 1 point from baseline during follow-up. Additional biopsies were used for bulk RNA qPCR analysis (baseline n=24, follow-up n=39) and histology. Blood samples were also collected at baseline (n=15) and during follow-up (n=30) for transcriptional analysis. Results Fourteen out of twenty-four patients enrolled in this study responded to tofacitinib treatment. No baseline demographics or clinical differences were observed between R and NR. In blood, tofacitinib did not significantly modulate the expression of genes related to total T cells, Tregs, B cells, neutrophils or inflammatory monocytes independently of response. However, in R, JAK/STAT pathway targets SOCS1 and SOCS3 were significantly decreased both in blood and in tissue. ScRNA-seq analysis demonstrated partial recovery of the epithelium and a decrease of plasma cells, inflammatory fibroblasts and neutrophils in R. Importantly, the proportion of T cells remained unaltered. These findings were transcriptionally and histologically validated. In contrast, in NR, tofacitinib did not drive epithelial repair or significant remodeling of the stromal compartment, instead ongoing inflammation induced a small but significant increase in the proportions of neutrophils and inflammatory fibroblasts. Furthermore, NR M2 macrophages upregulated the expression of M1-related genes (CLEC5A, SPP1 and INHBA) during follow-up, while in R tofacitinib induced upregulation of IGF1 and IL10RA in this population of macrophages. Conclusion Our study suggest that tofacitinib has little effect on systemic cells. Beyond changes in the mucosa of R that are associated to mucosal healing (increased epithelial and decrease inflammatory cells), scRNAseq analysis reveals previously undescribed effects of tofacitinib on macrophage populations in NR. This suggests a bias towards pro-inflammatory phenotypes in these patients as a potential cause of sustained inflammation in tofacitinib refractory patients. This is an independent study promoted by IDIBAPS and financed by a Pfizer Independent Medical Grant (grant 54549477).