A multicenter assessment of liver toxicity by MRI and biopsy in IBD patients on 6-thioguanine

Background/Aims

Although 6-thioguanine (6-TG) has been suggested as an effective treatment option for patients with inflammatory bowel disease (IBD), the recent description of its hepatotoxicity has led to the recommendation not to consider this drug. We initiated a multicenter safety study in IBD-patients treated with 6-TG to investigate hepatic changes by liver biopsy and magnetic resonance imaging (MRI).

Methods

Forty-five patients from three European centers treated with 6-TG (40–80 mg/d) at least for 8 weeks were enrolled. In all patients liver biopsy and MRI were performed. Slides and MR images were independently read by two pathologists and radiologists, respectively, and interpreted according to predefined criteria by consent.

Results

In 8 patients nodular regenerative hyperplasia (NRH) was diagnosed by liver biopsy, in 8 additional patients NRH could not be excluded due to equivocal pathological findings. MRI demonstrated a sensitivity of 77% and a specificity of 72% in the detection of pathohistological findings consistent with and/or possibly related to NRH.

Conclusions

Our study suggests that 6-TG therapy in IBD patients is associated with NRH of the liver. Based on a special MRI protocol, non-invasive diagnosis of NRH with promising sensitivity and specificity was demonstrated.

Introduction

In patients with inflammatory bowel disease (IBD), long-term immunosuppressive therapy with azathioprine (AZA) or 6-mercaptopurine (6-MP) has shown to be an effective and steroid-sparing option for treatment, which has led to their recommendation as first-line immunosuppressants [1], [2], [3]. However, late onset of action [4], as well as intolerance and resistance to therapy in up to 28 and 30% of treated patients, respectively, are limiting issues [1], [5]. Since clinical efficacy of AZA/6-MP is supposed to be based on the formation of 6-thioguanine nucleotides (6-TGNs), directly used 6-thioguanine (6-TG) may be more rapidly metabolised to the active compounds than the well-established pro-drugs [6]. Different modes of actions of thiopurines were described. 6-TGNs are proposed to be incorporated into DNA and RNA inhibiting replication, repair mechanisms and protein synthesis [7]. In activated T-lymphocytes, 6-thio-guanosyltriphosphate induces apoptosis by inhibition of the Rac1 pathway [8]. Classically, 6-TG has been applied in children with acute lymphoblastic leukaemia. First uncontrolled clinical trials have revealed a promising clinical efficacy of 6-TG in IBD patients resistant or intolerant to AZA/6-MP, reporting clinical response rates up to 75%, combined with a solely low frequency of side effects [6], [9], [10].

Despite these first promising clinical data, there has been a raising concern regarding potential hepatotoxic side effects of 6-TG. A recent report of a potential hepatotoxicity has darkened some bright hopes about 6-TG and even has led the authors to the recommendation not to consider this drug in IBD [11], [12]. The observed entity named nodular regenerative hyperplasia (NRH) represents a non-malignant hepatic disorder reported rarely as spontaneous finding in association with autoimmune and haematological disease, or after bone marrow transplantation [13], [14], [15], [16]. Histologically, NRH is characterized by a local hyperplasia of hepatocytes and atrophy of the intervening cells resulting in nodule formation, overall thought to be due to vascular flow abnormalities [17]. However, the exact pathogenesis of NRH has not been disclosed so far. Only very limited data on the natural progress of NRH is available [18].

The aim of our safety assessment was to clarify potential liver-directed side effects of 6-TG treatment in a study across three European IBD centers and to identify clinical or laboratory parameters with predictive value for the risk of such hepatic changes. Regarding the demand for non-invasive diagnosis of NRH and the lack of radiographic imaging studies, we further evaluated the sensitivity and specificity of magnetic resonance imaging (MRI) in detecting hepatic side effects of 6-TG therapy.