Early Assessment of Thiopurine Metabolites Identifies Patients at Risk of Thiopurine-Induced Leukopenia in Inflammatory Bowel Disease

Abstract

BACKGROUND AND AIMS: Only a quarter of thiopurine-induced myelotoxicity in IBD patients is related to TPMT deficiency. We determined the predictive value of 6-thioguanine nucleotide (6TGN) and 6-methylmercaptopurine ribonucleotide (6MMPR) concentrations one week after initiation (T1) for development of leukopenia during the first eight weeks of thiopurine treatment. METHODS: The study was performed in IBD patients starting thiopurine therapy as part of the Dutch randomised controlled TOPIC trial (ClinicalTrials.gov NCT00521950). Blood samples for metabolite measurement were collected at T1. Leukopenia was defined by leukocyte counts <3.0x10*(9)/L. For comparison, patients without leukopenia who completed the eight weeks on stable dose were selected from the first 272 patients of the TOPIC trial. RESULTS: Thirty-two patients with, and 162 patients, without leukopenia were analysed. T1 threshold 6TGN concentrations of 213 pmol/8x10*(8) erythrocytes and 3,525 pmol/8x10*(8) erythrocytes for 6MMPR were defined: patients exceeding these values were at increased leukopenia risk (OR 6.2 (95%CI: 2.8-13.8) and 5.9 (95%CI: 2.7-13.3), respectively). Leukopenia rates were higher in patients treated with mercaptopurine, compared to azathioprine (OR 7.3 (95%CI: 3.1-17.0)), and concurrent anti-TNF therapy (OR 5.1 (95%CI: 1.6-16.4)). Logistic regression analysis of thiopurine type, threshold concentrations, and concurrent anti-TNF therapy revealed that elevations of both T1 6TGN and 6MMPR resulted in the highest risk for leukopenia, followed by exceeding only the T1 6MMPR or 6TGN threshold concentration (AUC 0.84 (95%CI: 0.76-0.92)). CONCLUSIONS: In approximately 80% of patients, leukopenia could be explained by T1 6TGN and/or 6MMPR elevations. Validation of the predictive model is needed before implementing in clinical practice.