Azathioprine and 6-mercaptopurine are equally effective in thiopurine naïve IBD patients

Introduction: 6-Mercaptopurine (6-MP) and its pro-drug azathioprine (AZA) are frequently prescribed for the treatment of inflammatory bowel diseases (IBD), however evidence for the preference for one of both agents is scarce. Our aim was to compare the incidence of side effects and efficacy of AZA and 6-MP in thiopurine naïve patients. Methods: We analyzed data from the TOPIC-study (”Thiopurine response Optimization by Pharmacogenetic testing in Inflammatory bowel disease Clinics” (ClinicalTrials.gov: NCT00521950)). In this multicentre trial, 768 thiopurine naïve IBD patients were randomized for thiopurine dosing based on thiopurine S-methyltransferase (TPMT) genotype versus standard thiopurine dosing. In this study physicians were free in their choice for AZA or 6-MP. Patients were followed for 20 weeks and treatment response was assessed with Harvey-Bradshaw index in Crohn's disease (CD) and the partial Mayo score in Ulcerative colitis (UC). Gastro-intestinal side effects were defined as the occurrence of nausea, vomiting or decrease of appetite. Hepatic side effects were defined as an increase above two times the upper limit of normal (ULN) of either ALT, AST, GGT or AP, with no other explanation. Thiopurine induced pancreatitis was defined as amylase or lipase above two times the ULN with clinical symptoms or radiological findings suggestive for pancreatitis. Results: AZA was prescribed in 495 (64.5%) patients (mean dose: 2.05 mg/kg) and 6-MP in 273 (35.5%) patients (mean dose: 1.14 mg/ kg). The majority of patients had CD (n=464) compared to UC (n=297) and undetermined colitis (n=4). AZA was prematurely withdrawn in 42.4% and 6-MP in 44.0% of the patients (P=0.4). The incidence of gastro-intestinal side effects was similar in both groups; 47.2% (AZA) vs. 50.5% (6-MP); P=0.38). In total, 26 (3.4%) patients developed pancreatitis that was attributed to thiopurine use, 18 (3.7%) AZA and 8 (2.9%) 6-MP (P=0.59). Hepatotoxicity was reported in 99 (20%) patients with AZA and 71 patients (26.0%) taking 6-MP (P= 0.06). 6-Thioguanine nucleotide levels measured at week 8 were significantly higher in 6- MP users (314 pmol/10e8 vs. 276 pmol/10e8 per red blood cell) (P=0.03). Treatment response was known for 273 patients with CD and 132 with UC and showed no significant differences (P=0.79 and P=0.46) between AZA and 6-MP. Conclusion: This clinical trial suggests that AZA and 6-MP are similar in terms of efficacy, and safety as initial thiopurine in the treatment of IBD.