Y-ECCO Literature Review: Rawen Kader

Rawen Kader

Discontinuation of infliximab in patients with ulcerative colitis in remission (HAYABUSA): a multicentre, open-label, randomised controlled trial

Kobayashi T, Motoya S, Nakamura N, et al.

Lancet Gastroenterol Hepatol 2021;6:429–7. doi: 10.1016/S2468-1253(21)00062-5.

Rawen Kader
© Rawen Kader


The anti-tumour necrosis factor monoclonal antibody infliximab is one of the most widely used therapies for corticosteroid-refractory Ulcerative Colitis (UC). Long-term use of infliximab is associated with an increased risk of adverse events such as malignancies and infections, which is particularly concerning for those on concurrent immunosuppressive medications such as corticosteroids, thiopurines or calcineurin inhibitors [1–3]. With the number of patients with UC on long-term infliximab therapy continuing to rise, an important clinical question to address is whether these patients remain in remission upon discontinuing infliximab. Prospective studies have evaluated discontinuation of infliximab in patients with Crohn’s Disease, with deep (i.e. clinical, biological and endoscopic) remission thought to have a lower risk of relapse, but the evidence for patients with UC is limited to retrospective studies [4–6]. The HAYABUSA study aimed to address this issue with a randomised controlled trial (RCT) to evaluate discontinuing infliximab in patients with UC in remission.


The HAYABUSA study is a multicentre, open-label RCT carried out at 24 specialist centres in Japan. Patients with a definitive diagnosis of UC who were in clinical remission with the treatment of infliximab (5 mg/kg) every eight weeks and treated with infliximab for at least 14 weeks were enrolled into the study. They were then followed up for at least six months, and patients in sustained remission proceeded to be randomised in a 1:1 ratio to the infliximab-continued or infliximab-discontinued group, with randomisation stratified by the use of an immunomodulator (yes or no) and Mayo Endoscopic Subscore (MES) (0 or 1) at baseline. Sustained remission was defined as a clinical activity index (CAI) ≤4, corticosteroid-free, MES <2 and no additional UC-related medication since enrolment. Patients could continue to take 5-aminosalicylates or immunosuppressive agents at stable doses.

Patients in the infliximab-continued group received eight weekly infusions (5 mg/kg), with no change in dose or frequency of administration allowed, whilst the other group discontinued their infliximab treatment. If a patient relapsed in the infliximab-discontinued group, they were re-treated with infliximab (5 mg/kg) at 0, 2 and, if necessary, 6 weeks, alongside hydrocortisone (100 mg) or methylprednisolone (20–40 mg) upon re-treatment.

Patients had a total colonoscopy upon randomisation and again at 48 weeks. The MES was scored locally, and the histology was centrally read with the Nancy Histological Index graded by a histologist blinded to the clinical information. The CAI was scored locally at each patient visit by the attending clinician, and a blood trough concentration of infliximab was centrally measured at randomisation without any further longitudinal measurement.

The primary endpoint was the remission rate at week 48. The secondary endpoints included adverse events, factors associated with remission at week 48 and remission rate at week 8 in those who relapsed in the infliximab-discontinued group and were re-treated with infliximab. Remission was defined as CAI ≤4, whilst relapse was defined as the addition of new UC treatment (including topical 5-aminosalicylates or increasing the dose of current medication) for symptomatic worsening or an MES of ≥2.

Due to a slower than anticipated recruitment, the study sample size was recalculated during the study. The investigators initially estimated a 15% difference in remission rate between the infliximab-continued and infliximab-discontinued groups (90% vs 75%), which was re-estimated as a 24% difference (90% vs 66%) upon sample size recalculation.

Key Findings

A total of 122 patients were enrolled in the study between 16 June, 2014, and 28 July, 2017, with 95 of these eligible for randomisation. Three patients were excluded from the final analysis due to a lack of follow-up data, leaving 46 patients in each group. Infliximab was the first biologic treatment in all randomly assigned participants, and no positive antibodies to infliximab were detected during the study.

For the primary outcome, 37 [80.4% (95% CI 66.1%–90.6%)] of 46 patients in the infliximab-continued group and 25 [54.3% (95% CI 39.0%–69.1%)] of 46 patients in the infliximab-discontinued group remained in remission at week 48. After adjusting for the stratified factors (use of immunomodulators and MES 0 or 1), the difference in remission rate between the groups was 27.3% (95% CI 8.0%–44.1%; p=0.0059). Adverse events were 3.9% (95% CI −10.3% to 18.1%; p=0.59) higher in the infliximab-continued group (17%) compared to the infliximab-discontinued group (13%), but this did not reach statistical significance. Twelve patients in the infliximab-discontinued group relapsed and were retreated with infliximab. Eight [66.7% (95% CI 34.9%–90.1%)] of these patients were in remission within eight weeks of re-treatment with no infusion reactions reported.

In the multiple logistic regression analyses adjusted for the treatment group of all 92 randomised participants, both a lower CRP [OR 0.0016 (95% CI 0.00–0.73); p=0.039] and a Nancy index score of <2 [0.46 (0.24–0.88); p=0.019] at randomisation were shown to be significantly associated with remission. Other factors predictive of deep remission, such as the CAI, MES (0 or 1), infliximab trough concentrations and the use of immunomodulators before randomisation, were not found to have a significant effect on remaining in remission.


The investigators should be congratulated on delivering the first RCT to evaluate the remission rates when discontinuing infliximab in patients with UC. The study findings are in keeping with the evidence from retrospective studies showing higher relapse rates upon discontinuation of infliximab. Interestingly, the study suggests that CRP and the Nancy index score could be predictive factors for remission in patients who discontinue infliximab, while no significant effect was observed for MES, CAI or use of immunomodulators. These findings were obtained in the context of a study sample size smaller than that initially planned and some patients did not provide serum samples or biopsies (or both). Nonetheless, these findings suggest that further adequately powered RCTs are warranted to identify the risk factors for relapse when discontinuing infliximab in UC patients. Such studies would help the development of an exit strategy for patients with UC on long-term infliximab.


  1. Sebastian S, Neilaj S. Practical guidance for the management of inflammatory bowel disease in patients with cancer. Which treatment? Ther Adv Gastroenterol. 2019;12:1756284818817293.
  2. Long MD, Martin CF, Pipkin CA, Herfarth HH, Sandler RS, Kappelman MD. Risk of melanoma and nonmelanoma skin cancer among patients with inflammatory bowel disease. Gastroenterology. 2012;143:390–9.
  3. Ford AC, Peyrin-Biroulet L. Opportunistic infections with anti-tumor necrosis factor-α therapy in inflammatory bowel disease: meta-analysis of randomized controlled trials. Am J Gastroenterol. 2013;108:1268–76.
  4. Louis E, Mary J, Vernier–Massouille G, et al. Maintenance of remission among patients with Crohn’s disease on antimetabolite therapy after infliximab therapy is stopped. Gastroenterology. 2012;142:63–70.
  5. Kennedy NA, Warner B, Johnston EL, et al. Relapse after withdrawal from anti-TNF therapy for inflammatory bowel disease: an observational study, plus systematic review and meta-analysis. Aliment Pharmacol Ther. 2016;43:910–23.
  6. 6. Steinhart H. Exit strategies for biologic therapy in IBD: is it too soon to stop? Inflamm Bowel Dis. 2018;24:669–70.

Rawen Kader - Short Biography

Rawen Kader is a gastroenterology trainee and clinical research fellow at University College London, UK. His research interests are in post-colonoscopy colorectal cancer and the use of artificial intelligence (AI), specifically deep learning, to improve the quality of colonoscopy. His experience includes both the pre-clinical development of AI algorithms and their evaluation through randomised controlled trials. He is the secretary of the British Gastroenterology AI Task Force and the chair of GLINT (Gastro London Investigative Network for Trainees).

Posted in ECCO News, Y-ECCO Literature Reviews, Committee News, Y-ECCO, Volume 16, Issue 4