Comparing JAK inhibitors in ulcerative colitis: which one truly leads in the real world? Results from the ENEIDA registry of GETECCU

Abstract

Background: The addition of selective Janus kinase inhibitors (JAKi) –tofacitinib (TOFA), upadacitinib (UPA), and filgotinib (FILGO)– to the therapeutic arsenal for ulcerative colitis (UC) has provoked the need for comparative studies to establish their positioning in clinical practice. Aim: To compare the durability and the short- and long-term effectiveness of TOFA, UPA, and FILGO in UC, and to evaluate their real-world safety. Methods: Adult patients who started JAKi treatment for UC (from April 2021) at least 8 weeks prior to data collection and were part of the prospectively maintained ENEIDA registry of GETECCU were included. Patients receiving concomitant biologic agents or with a history of colectomy were excluded. Patients were followed from the first JAKi dose until treatment discontinuation or their last visit. Effectiveness analysis was restricted to patients with active disease [Partial Mayo Score (PMS)>2] at JAKi initiation. Clinical effectiveness was measured using PMS, and concomitant steroid use was assessed at each visit. Patients who discontinued JAKi before their last visit were considered not in remission at subsequent time points (negative imputation). The propensity score was calculated using the inverse probability weighting (IPTW) method to balance confounding factors. Results: A total of 611 patients were included (74 on FILGO, 369 on TOFA, and 168 on UPA). Baseline characteristics are summarised in Table 1. Discontinuation rates per patient-year were 57% for FILGO, 40% for TOFA, and 27% for UPA, with UPA showing significantly lower cumulative discontinuation rates compared to FILGO and TOFA (Fig. 1a). Reasons for discontinuation are detailed in Fig. 1b. Adjusted analysis indicated that baseline disease activity and prior JAKi exposure were associated with a higher discontinuation risk. Both TOFA and FILGO had significantly higher discontinuation risks than UPA. Steroid-free clinical remission (SFCR) rates during follow-up are shown in Fig. 2, with UPA demonstrating higher SFCR rates at most time points. At week 8, disease activity, prior JAKi exposure, and JAKi type were significantly associated with SFCR, with UPA having a higher probability of SFCR than both TOFA and FILGO (Table 2). The IPTW method confirmed all these results. Adverse events occurred in 12 (16%) patients with FILGO, 85 (23%) with TOFA, and 45 (27%) with UPA (p=0.2), with a minority leading to discontinuation (Fig. 1b). Conclusion: In real-world practice, UPA shows superior effectiveness and drug survival compared to TOFA and FILGO in UC patients, with a comparable safety profile. These findings support the positioning of JAKi treatments in clinical practice and may guide treatment decisions in UC management.