DOP020 Thiopurine monotherapy still has a place in the treatment of patients with mild-to-moderate Crohn’s disease in the biological era

Abstract

Background For more than half a century, thiopurines have been the first line maintenance therapy in patients with Crohn's disease (CD). With the increasing availability of biological drugs, thiopurines are often considered less potent, though in mild-to-moderate disease they may be a valid, safe and less expensive alternative. Genetic determinants including HLA, TPMT and NUDT15 have been associated with response and/or side effects to thiopurines. We here report outcome of thiopurine monotherapy in CD patients, and evaluated genetic associations with response. Methods The medical records of all genotyped CD patients (Illumina Immunochip) who ever received thiopurine monotherapy at our tertial referral centre were retrospectively assessed. All patients had TPMT-screening prior to thiopurine initiation. Response was defined as continuation of thiopurines for more than 1 year in monotherapy, with minimal corticosteroid use (max 1 course/year) and no need for other rescue therapy. Allelic association was assessed using PLINK v1.07. Results Over the past 18 years, 852 CD patients (median disease duration 4.0 years) received thiopurine monotherapy (99.8% azathioprine, 9.8% mercaptopurine), of whom a median (IQR) follow-up of 13.3 (8.3-18.3) years is available. One third of patients (35.5%) responded, whereas 35.5% experienced no response. Thiopurine withdrawal due to side-effects occurred in 29.0% of patients, early after initiation (29.2, 14.6-73 days), including pancreatitis (n=62), abnormal liver tests (n=18) and GI intolerance (n=52). Three lymphomas were diagnosed during follow-up. One quarter of responding patients (26.1%) never discontinued therapy during median follow-up of 7.9 (3.1-11.9) years. The other 73.9% initial responders stopped thiopurine therapy after 5.5 (2.8-9.2) years. Ileal disease location (p=0.001), older age (p=0.04), longer disease duration (p<0.001), absence of perianal disease behaviour (p=0.004) and absence of active perianal disease at time of thiopurine initiation (p<0.0001) were significantly associated with response. Genetic analysis revealed 3 loci significantly associated with response to thiopurines, including rs10196508 located at a regulatory region within chromosome 2 (p=2x10 -5, OR 2.6). Conclusion In this large retrospective series, thiopurine monotherapy could safely maintain clinical response in up to 35.5% of patients after one year, similar to previous prospective data observed during the SONIC-trial. We identified a genetic marker associated with response to thiopurines, which needs validation in an independent cohort before its clinical use can be evaluated. Copyright © 2018 AGA Institute. All rights reserved.