Tofacitinib as Salvage Therapy for 55 Patients Hospitalised With Refractory Severe Ulcerative Colitis


Mathieu Uzzan; Clément Bresteau; David Laharie; Carmen Stefanescu; Christophe Bellanger; Franck Carbonnel; Mélanie Serrero; Stéphanie Viennot; Maria Nachury; Aurélien Amiot; Romain Altwegg; Laurence Picon; Stéphane Nahon; Lucine Vuitton; Philippe Ah Soune; Julien Kirchgesner; Laurent Peyrin-Biroulet; Yoram Bouhnik


Aliment Pharmacol Ther. 2021;54(3):312-319. 

In This Article


Study Design and Patients

We performed an observational and multicentre study with both retrospective and prospective collections. The cohort was built from a centralised GETAID registry from French IBD centres where consecutive patients were enrolled using an online registry. Data were then retrospectively and prospectively collected from medical records. Patients recruited retrospectively were included from January 2017 to May 2020. Prospective inclusions and data collection started from 1 June 2020. The date of inclusion corresponded to the first day of tofacitinib. Follow-up was censored at colectomy, or the date of point (19 April 2021) or at the date of last news. Patients who did not undergo colectomy had all at least 3 months of follow-up until last news. Inclusion criteria were as follows: (a) age >16 years old; (b) diagnosis of ulcerative colitis based on ECCO criteria;[8] (c) admission in hospitalization for a flare-up of the IBD; (d) initiation of tofacitinib during the hospital stay; (e) inadequate response with, loss of response to, and/or intolerance to at least one anti-TNF agent, vedolizumab or ustekinumab. Patients with Crohn's disease, IBD unclassified, previous colectomy, a history of colonic cancer were excluded. The study was approved by local ethic committee review board.

As routine care, usual bacterial intestinal infections and Clostridium difficile infection were ruled out before tofacitinib induction. Flexible sigmoidoscopy was performed in all patients to assess for severity and exclude cytomegalovirus infection.

Aims and Endpoints

The primary objective was colectomy-free survival after the first administration of tofacitinib in patients hospitalised for a flare of UC.

The secondary objectives were to determine (a) factors associated with colectomy, (b) survival without tofacitinib discontinuation, (c) clinical response, clinical remission and steroid-free clinical remission at weeks 6 and 14 and (d) drug safety in this population especially regarding infections, major adverse cardiovascular events and venous thrombotic events.

The primary endpoint was defined as the occurrence of colectomy during follow-up. Clinical response was defined as a decrease of the partial Mayo score of at least 3 points, with a decrease of at least 30% and a reduction of at least 1 point on the bleeding subscore with a subscore of 0 or 1, as compared with inclusion. Clinical remission was defined as a partial Mayo score of less than 3, without any of the subscores having a value higher than 1.

Adverse effects were classified as severe when they lead to treatment interruption, rehospitalization for other reason than UC flare, disability or persistent damage, colectomy or death.

Statistical Analysis

Quantitative variables are expressed as median (interquartile range [IQR]). Qualitative variables are given as numbers (percentages). Colectomy-free survival was assessed using a Kaplan-Meier analysis. In order to identify predictors of colectomy, a Cox proportional hazards regression analysis was performed. Variables significant at P < 0.20 in the univariate analysis were entered into a multivariate Cox proportional hazards regression model. The results are reported with hazard ratios (HR) and their 95% confidence interval (95 CI). Alpha risk was set at 5% for statistical significance level. Calculations were performed using R software (The R Project for Statistical Computing—