Letter: successful mercaptopurine therapy after azathioprine-related pancreatitis in patients with IBD – authors' reply



The encouraging data on pancreatitis presented by Beswick and colleagues[1] are roughly similar to the ones we previously reported.[2] However, the patients in our study with azathioprine (AZA)-induced pancreatitis presented more heterogeneous symptoms during mercaptopurine (MP) challenge, and two patients without pancreatitis on AZA developed this when shifted to MP.

In our experience, it is not possible to predict whether a shift to MP will be tolerated or a new pancreatitis will develop. The patients' TPMT status is probably not helpful, as pancreatitis was observed in both the normal and intermediate TPMT range.[3] Fortunately, thiopurine-induced pancreatitis is almost always mild and resolves quickly if the drug is discontinued.[4]

We thus agree with the authors that treatment with MP may be considered in selected patients, in whom the clinical indication for a continued thiopurine treatment is strong. However, factors like patient discomfort, delay in effective treatment and the risk of a more severe pancreatitis during MP challenge have to be taken into consideration when deciding to treat with MP. Furthermore, at least two other treatment options are available in this situation. Treatment with ‘tioguanine (thioguanine)’, which in our experience was tolerated by four out of four patients with an AZA-induced pancreatitis,[5] is considered safe and without risk for nodular regenerative hyperplasia if given in doses of 20 mg daily and if TGN levels remain ≤600 pmol/8 × 108 RBC.[6, 7] Combination therapy with low-dose AZA and allopurinol is another possibility, which might work, as it produces a similar metabolic pattern as tioguanine. So far, we have successfully treated one patient with a skewed metabolism (high methylated metabolites together with low TGN metabolites) and an AZA-induced pancreatitis with this regimen (own unpublished data).

In conclusion, data on the best way to treat these patients are scarce and at the end of the day the decision of how to treat has to rely on clinical judgement. Nevertheless, our understanding of the pharmacokinetic and pharmacogenetic basis of thiopurines have increased during the last years, which may lead to a more advanced and ‘modern’ application of these important drugs.


The authors' declarations of personal and financial interests are unchanged from those in the original article.2