In this issue of Intestinal Research, Yamada et al. tried to assess the long-term efficacy and safety of maintenance treatment with thiopurines in UC.3 They enrolled a total of 59 bio-naive patients who maintained remission by treatment with thiopurine and 5-aminosalicylate after induction. The cumulative remission-maintenance rates at 24, 48, and 84 months were reported as approximately 69%, 55%, and 44%, respectively. Recent meta-analyses of trial data revealed remission maintenance rates of 56-76% during thiopurine treatment.12 The remission maintenance rate of Yamada et al. was similar to that of meta-analyses, considering that the mean follow-up period of these studies was shorter. Yamada et al. also demonstrated that high-dose corticosteroid treatment (≥40 mg/day) was a negative predictive factor for maintenance treatment with thiopurine. This could be inferred from the fact that other maintenance treatments such as biologics should be considered in refractory UC patients treated with high-dose corticosteroids.

Mucosal healing (MH) is becoming more widely accepted as a relevant outcome marker for UC. In a Norwegian population-based cohort study examining effects between the presence or absence of MH and long-term outcomes, MH at 1 year was confirmed in 50% of UC patients, and the colectomy rate at 5 years was 1.7%. MH at 1 year significantly decreased the need for colectomy at 5 years,4 and this association was significant after adjusting for factors that may influence colectomy, such as age, smoking, education level, and disease extent. Therefore, the previous study provides evidence that MH is strongly associated with the clinical outcomes of UC. Yamada et al. reported MH rates of >60% at 26.8 months, and a colectomy rate of 3.6%. Compared to that of Western studies, this result was not significantly different. However, only two-thirds of study patients underwent colonoscopy to evaluate MH and the time interval between initiating thiopurine therapy and performing the colonoscopy to assess MH was different in each patient due to the retrospective design of the study. Therefore, the results of Yamada et al. should be interpreted while considering the limitations of the study, such as the small number of patients included and the lack of adjustment for potential confounding factors.

Efficacy must be balanced against safety and tolerability. In the Yamada et al. study, the median dose of thiopurine was 50 mg. Traditionally, the dose-escalating approach is used in East Asian patients.56 Several trials, including this study, have suggested that lower doses of azathioprine or 6-mercaptopurine may be effective and safe for the treatment of UC patients,78 contrary to the results of Western studies. According to previous studies, approximately 10-28% of patients reported adverse events (AEs), and 25.6% of patients had to discontinue thiopurine therapy due to AEs.910 A similar rate of AEs (28.8%) was reported in the study by Yamada et al. In this study, the discontinuation rate due to AEs was reported as low as 10.2%, probably due to low thiopurine dose they used. Although there were no severe AEs such as lymphoma and non-melanoma skin cancer in the Yamada et al. study, and the absolute risks of these malignancies remain low, the risks and benefits of continuing or ceasing thiopurine therapy should be weighed for each individual patient.

Despite the fact that clinical trials (including the Yamada et al. study) concerning the use of thiopurines for UC have been heterogeneous in design, the evidence is sufficient to conclude that thiopurines are effective for maintenance of remission.