Search in the Abstract Database

Abstracts Search 2019

P469 Effectiveness of dose optimisation by pre-genotyping NUDT 15 R139C on reducing thiopurine-induced leucopoenia in Chinese patients with Crohn’s disease: a randomised controlled trial

K. Chao*1, L. Lin1, Y. Huang2, C. Zhang3, J. Huang1, Q. Cao2, X. Gao1, K. Chao1

1The Sixth Affiliated Hospital, Sun Yat-sen University, Department of Gastroenterology, Guangzhou, China, 2Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Department of Gastroenterology, Hangzhou, China, 3School of Pharmaceutical Sciences, Sun Yat-Sen University, Guangzhou, China


More than 20% Chinese patients with IBD develop thiopurine-induced leucopoenia. Recent retrospective studies have confirmed that NUDT15 R139C variant is a reliable marker of thiopurine-induced leucopoenia in Asian population. Thus we conduct this prospective study to explore whether an optimising strategy based on NUDT15 R139C genotypes affect outcomes of Chinese patients with Crohn’s disease (CD).


A prospective, randomised study was conducted in two tertiary hospitals in China (NCT02929706). CD patients (18–65 years old) with indication of the use of thiopurine were included. The exclusion criteria were: Contraindication of thiopurine, previous use of azathiopurine and co-treatment with 5-ASA or allopurinol. The Patients were randomly assigned to the intervention group (pre-genotyping NUDT 15 R139C) or control group (receive standard dosage of azathiopurine with a target dosage of 2–2.5 mg/kg/day). Patients in the intervention group found to be wild-type carriers were prescribed with standard dose of azathiopurine, while the heterozygotes received 50% of the standard dosage. Considering all the variant of homozygotes develop leucopoenia in the previous study, these patients in the intervention group did not receive thiopurine. NUDT15 R139C genotypes were determined with PCR-RFLP and sequencing. Patients were followed for 48 weeks. The primary endpoint was the differences of incidence of leucopoenia (white blood cell <3500 mm−3).The secondary outcomes were other adverse events and the efficacy (evaluated by CDAI, CRP, and mucosal healing) between the two optimisation strategies.


A total of 400 Chinese CD patients were randomised. The frequency of NUDT15 R139C variant, sex, age, baseline CDAI were similar in the two groups. The rate of thiopurine-induced leucopoenia is significantly lower in the intervention group (20.8% vs. 29.7%; p = 0.041; relative risk = 0.619; 95% confidence interval (0.389–0.982).The difference is more significant in patients with NUDT15 variant [29.6 vs. 65.7 (RR 0.220, 95% CI 0.07–0.65)]. The patients develop leucopoenia in the intervention group seems milder the control group (Grade 1; 68.6% vs. 45.8%, p = 0.04). No differences of other adverse events were found. We compare the efficacy in patients treated with corticosteroids and maintain remission with AZA only. One hundred and twenty-one patients were included and no differences were found in CDAI, CRP level, and the rate of mucosal healing at the 48th week.


Pre-genotyping NUDT15 R139C before starting thiopurine could be a promising strategy to reduce the rate of leucopoenia. This optimisation strategy does not seem to influence the clinical efficacy in 48 weeks of follow-up.