Assessment of the one-year Efficacy and Safety of Tofacitinib in biologic-refractory patients with Ulcerative Colitis: a real-world Belgian cohort study

Introduction: Tofacitinib, an oral Janus kinase inhibitor, has been approved in 2018 for the treatment of moderate to severe ulcerative colitis (UC) in Europe. Efficacy and safety data from long-term real-world studies are scarce.

Aim: The aim of the study was to evaluate the real-world long-term efficacy and safety of tofacitinib in a Belgian cohort of patients with refractory UC who were exposed to both anti-TNF and vedolizumab.

Methods: We performed an observational, national, retrospective multicenter study including all patients whith active UC who started on tofacitinib between November 2018 and August 2019 in 26 Belgian centers (Ethics Committee approval number P2019/429, date: 10/12/2019; amendment for study extension received on 16/07/2020). Data were prospectively collected and retrospectively analyzed according to intention-to-treat. Clinical response (decrease from baseline in partial adapted Mayo score (stool frequency and rectal bleeding) by ≥ 1 and ≥ 30%), clinical remission (Partial Adapted Mayo score ≤ 1), steroid-free clinical remission, endoscopic response (decrease from baseline in Mayo endoscopic subscore of ≥ 1), endoscopic remission (endoscopic Mayo subscore of 0), drug survival, need for colectomy and adverse events (AEs) were assessed at week 8, 16 and 52.

Results: A total of 75 patients were included with a median follow-up of 45 weeks (IQR: 19-51). Patients were predominately men (59%), and median age at baseline was 44 years (IQR: 31-59). Median disease duration was 8 years (IQR: 4-17). Fifty-five percent had left-sided UC and 45% had pancolitis. Overall, 72 (96%), 51 (68%) and 6 (8%) patients were exposed to at least 1, 2 or 3 antiTNFs respectively, and 73 (97%) to vedolizumab. At baseline, 42 (56%) patients were under steroids. Median Partial Adapted Mayo score at baseline was 4 (IQR: 3-5), and median endoscopic Mayo subscore was 2 (IQR: 2-3). Thirty-nine (52%) patients required prolonged induction at 10mg twice daily for 8 additional weeks. Dose optimization was needed in 12 (16%) patients due to disease relapse after induction response, 9 (12%) of which could regain response. After 1 year, 52% and 43% of patients had experienced clinical response and clinical remission respectively, and 39% achieved steroid-free clinical remission. Endoscopic response and remission were observed in 37% and 9% of patients. Faecal calprotectin < 250 µg/g at week 16 (OR: 0.03(95%CI: 0.003-0.4)) was a positive predictor of clinical remission at week 52. Overall, 34 (45%) patients discontinued tofacitinib (22 due to primary non response, 11 due to secondary loss of response and 1 due to AE) during follow-up with a median exposure duration of 17 weeks (IQR: 11-42). Among these patients, 6 underwent colectomy for disease worsening after a median treatment duration of 15 weeks (IQR: 11-19) and a median follow-up of 20 weeks (IQR: 12-25) , while the others switched to another medical therapy. Forty AEs were reported in 25 (33%) patients, with only one (pneumonia) leading to treatment discontinuation. The most common AEs were arthralgia and lower respiratory tract infections followed by herpes zoster, urinary tract and upper respiratory tract infections. Two cases of prostate cancer have been reported. No opportunistic infection, venous thromboembolism, pulmonary embolism or cardiovascular event were reported. A statistically significant increase of 43% in the low-density lipoprotein (LDL) level was observed between baseline and week 52 among patients in clinical remission at week 52 , with no significant changes in total cholesterol and high-density lipoprotein (HDL).

Conclusions: Tofacitinib effectively induced long-term clinical and endoscopic response and remission in a refractory cohort of patients with UC in a real-world clinical setting. During this one-year follow-up, tofacitinib was relatively well tolerated with respect to adverse events.