Background & aims: Thioguanine (6-TG) has been studied as an alternative thiopurine in inflammatory bowel disease (IBD). Short-term safety and efficacy data were favorable. Experience with 6-TG in patients with acute lymphoblastic leukemia raised long-term safety concerns when implicated in nodular regenerative hyperplasia (NRH) of the liver and portal hypertension. The aim of this study was to describe the association between 6-TG and NRH in IBD.
Methods: Liver chemistries and complete blood counts were monitored, and patients were encouraged to undergo liver biopsy. Clinical data were collected by chart review, and associations were tested by univariate and multivariable analyses. Patients were classified based on the presence (group 1) or absence (group 2) of laboratory abnormalities.
Results: Laboratory abnormalities occurred in 29 of 111 patients (26%). Elevations of liver enzymes and a decrease in platelet counts (<200,000) were most commonly observed. Male gender (odds ratio, 2.9; 95% CI, 1.1-7.3; P < 0.03) and preferential 6-methylmercaptopurine production on 6-mercaptopurine/azathioprine (odds ratio, 3.0; 95% CI, 1.2-7.4; P < 0.04) were independently associated with laboratory abnormalities. No association was seen with duration of 6-TG treatment, cumulative dose, or 6-TG nucleotide levels. The median increase in alanine aminotransferase, aspartate aminotransferase, and alkaline phosphatase levels was 39, 30, and 75 U/L, respectively, in group 1, and the median decrease in platelet count was 115,000 in group 1 versus 7000 in group 2 (P < 0.001). NRH occurred in 76% of patients undergoing biopsy in group 1 and 33% in group 2.
Conclusions: NRH is a common finding in 6-TG-treated patients with IBD. The progression or reversibility of NRH remains unknown. Our findings suggest that 6-TG should not be considered as therapy for patients with IBD.