Abstract Background Tofacitinib (TOF) is an effective treatment for ulcerative colitis (UC). A relationship between dose and both response and side-effect profile is recognised for JAK inhibitors. Dose reduction from 10mg bd to 5mg bd is therefore associated with improved safety but also with a risk of relapse. The BRITE study aimed to identify clinical predictors and biomarkers of relapse in a cohort of patients with stable UC undergoing TOF dose reduction. Methods Steroid-free patients with stable UC responding to TOF 10mg bd for at least 8 weeks were followed from dose reduction to time of relapse or for one year. Relapse was defined by worsening modified Mayo score, including an increase in the rectal bleeding and endoscopic subscores of ≥ 1 point. Clinical and endoscopic assessment with biopsy was performed prior to dose reduction (baseline) and at end of trial. PBMCs were isolated at baseline and were cryopreserved. Flow cytometry analysis of phospho-STAT levels of STAT1, STAT3, STAT4 and STAT5 was performed in T cells from unstimulated and cytokine-stimulated baseline PBMC. In addition, CD4 T cells were isolated from PBMC and were stimulated with plate-bound anti-CD3 and anti-CD28 (TCR activation) for 5 days; cell supernatants were analysed for 17 CD4 T-helper (Th) cell effector cytokines using a 17-cytokine Luminex assay. Results 50 patients were enrolled across 3 centres; baseline characteristics are displayed in table 1. 23 (46%) patients relapsed after a median duration of 23 weeks (IQR 9-45). Demographics, disease duration, prior advanced therapy exposure (including duration on TOF), and baseline clinical, biochemical (calprotectin/CRP), and histo-endoscopic disease severity were not associated with risk of relapse. In contrast, molecular differences in CD4 T-cell function were identified in relapsers and non-relapsers. pSTAT5 levels were significantly induced by IFNa in relapse patients (p<0.05), but not in remission patients (Fig 1a), while canonical pSTAT5 activation by IL2 showed no significant differences between the two cohorts of patients. Furthermore, TCR activation of baseline CD4 T cells showed significantly higher expression of IL10 (p=0.01) in remission patients compared to relapse (Fig 1b). Higher levels of resting pSTAT1High expression that was refractory to IFNa stimulation in relapse patients but inducible in remission patients was also identified (p=<0.01). Conclusion In this large prospective study of TOF dose reduction, in contrast to previous studies, no baseline clinical or endoscopic predictors of relapse were found. However, potentially clinically useful association with markers of T cell activation were identified and should be investigated further.
Read full abstract