Azathioprine (AZA) and its metabolite 6-mercaptopurine (6-MP) were developed >50 years ago and were initially used clinically in the management of childhood leukemia and organ transplantation. AZA is a prodrug that is quickly converted to 6-MP in erythrocytes and other tissues. 6-MP is metabolized in the liver and gut by 1 of 3 enzymes (1): thiopurine-S-methyltransferase (TPMT)—methylates 6-MP to the inactive metabolite 6-methyl-mercaptopurine; xanthine oxidase—to inactive thiourate; and hypoxanthine-guanine-phosphoribosyltransferase—to the active metabolite 6-thioguanine. The immunosuppressive action of thiopurine derivitives is thought to be via specific blockade of Rac1 activation leading to a mitochondrial pathway of apoptosis of T cells (2).
The first report of 6-MP used in the context of inflammatory bowel disease (IBD) was in 1962 (3), and since then the use of thiopurines has been well established in the management of moderate-to-severe IBD (4–6). Thiopurines offer an inexpensive and effective treatment option for maintenance of remission of IBD in comparison with biological agents, which may be up to 30 times more expensive (7).
Although 50% to 60% of patients with IBD respond well to thiopurines (5,6), a significant proportion of patients will not tolerate these drugs because of various adverse effects (8). The adverse effects of thiopurines may be dose related, patient related, or idiosyncratic.
One idiosyncratic adverse effect of thiopurines is acute pancreatitis (AP). Drug-induced AP is the assumed diagnosis when no other cause of AP can be found, the patient is taking a drug known to be associated with AP, and symptoms resolve after drug discontinuation. If pancreatitis reoccurs on reexposure, the drug is definitely considered the cause (9).
AP occurring soon after starting a thiopurine drug is usually considered as thiopurine-induced pancreatitis (TIP) and thiopurines are considered to be contraindicated in these patients. Often these patients have challenging disease, which requires various combinations of medications both in sequence and concurrently. Removing a critical medication from the treatment arsenal can make the challenge of managing these patients significantly more difficult.
There exists extremely little data in the literature on the safety of reintroducing a thiopurine after a suspected case of TIP. Anecdotal evidence from a few centers suggests that it has been tolerated in some patients with no recurrence of TIP. This report presents 4 cases in which a second thiopurine was successfully used in children with IBD, following a previous episode of suspected TIP.
This case series is a retrospective chart review of 4 patients with IBD in whom thiopurines were successfully reintroduced following suspected TIP. None of the patients in this series had a family history of recurrent pancreatitis, and none had previous or subsequent history of AP. None of these patients had any evidence of liver disease. The patients were all cared for in the pediatric IBD services at Sydney Children's Hospital, Sydney, Australia, and Christchurch Hospital, Christchurch, New Zealand.
This girl was diagnosed as having Crohn disease (CD) at 11 years of age with a 2- to 3-year history of lethargy, abdominal pain, diarrhea, and mouth ulcers. There was a history of a second-degree relative with CD, and subsequently her twin sister was also diagnosed as having CD. Investigation revealed patchy colitis with gastric and duodenal involvement. She had a normal barium follow-through. Following diagnostic investigations, she was commenced on exclusive enteral nutrition (EEN) to induce remission. AZA was commenced at 25 mg daily for 2 weeks after diagnosis. After 2 weeks of taking AZA, the dose was increased to 75 mg daily. Within 36 hours of the dose change, she developed nausea and abdominal pain and was found to have elevated amylase (332 U/L; normal range 36–128) and lipase (430 IU/L; normal range 8–57). Abdominal ultrasound (US) revealed an enlarged and bulky pancreas.
Her symptoms and biochemical changes resolved promptly with cessation of AZA. EEN was ceased and was instead commenced on oral prednisone. Four weeks later (8 weeks postdiagnosis), she started taking 6-MP 25 mg daily, with dose increased to 50 mg daily after 2 weeks. She remains clinically well on 6-MP with normal monitoring blood test results for >3 years.
This girl, with a maternal history of CD, was investigated at 3 years of age following presentation with chronic diarrhea from 18 months of age. Her assessment revealed colitis with significant protein losing enteropathy. A white cell scan revealed increased uptake in the ileocecal area. She was initially treated with corticosteroids and 5-aminosalicylates (5-ASA). Upon reassessment at 13 years of age, endoscopic findings revealed colonic and upper gastrointestinal CD, with normal small intestine seen on magnetic resonance enterography. She was then commenced on AZA 25 mg daily and dose was increased to 75 mg daily 2 weeks later. Three days later, she developed abdominal pain and vomiting with a lipase measured at 934 IU/L. Symptoms settled rapidly with cessation of AZA and lipase also normalized within 2 days. 6-MP was commenced 6 weeks later without any difficulty and is well tolerated 12 months later with normal monitoring blood test results.
This boy presented at 13 years of age with a history of diarrhea, anorexia, weight loss, lethargy, and pallor. Endoscopic assessment demonstrated patchy ileocolitis with more marked changes at the rectosigmoid junction. Magnetic resonance elastography demonstrated small bowel involvement in the terminal ileum only. TPMT activity was low and genotyping showed him to be a TPMT heterozygote.
He entered clinical and biochemical remission after 8 weeks of EEN. AZA 50 mg daily was introduced 6 weeks after diagnosis, and 3 weeks following, he presented with severe epigastric pain and vomiting. Blood tests showed amylase of 158 U/L and lipase of 829 IU/L and his abdominal US demonstrated changes consistent with AP. His symptoms settled with 24-hour bowel rest and cessation of AZA. Despite commencing maintenance 5-ASA therapy, he had recurrence of symptoms with elevated inflammatory markers. Subcutaneous methotrexate was commenced 6 months after diagnosis with clinical and biochemical remission for several months; however, his symptoms progressively worsened during subsequent months.
6-MP was commenced 18 months after diagnosis at an initial dose of 12.5 mg daily and then increased to 25 mg 2 weeks later. He remains well with no further clinical features of pancreatitis.
This boy presented at 14 years of age with 8 months of abdominal pain, lethargy, weight loss, and diarrhea. There was no family history of IBD. Endoscopic assessment revealed granulomatous ileocolonic disease and focal enhancing gastritis. He was initially treated with EEN and AZA 25 mg daily was introduced after 3 weeks. Two weeks later, he represented with abdominal pain and nausea. He was found to have an amylase level of 226 IU/L and lipase of 876 U/L. AZA was immediately ceased with prompt resolution of pancreatitis. He was unable to tolerate nutritional therapy and was then commenced on prednisone and 5-ASA. Four months later, because of poor control, he started taking methotrexate. Despite a 6-month period of methotrexate, he had ongoing symptomatic disease and was then commenced on 6-MP at 25 mg daily. After 2 weeks, the 6-MP was increased to a full therapeutic dose with no pain or nausea and normal biochemistry. Two months later, he continues to tolerate 6-MP with no recurrence of pancreatitis.
This case series documents the successful introduction of a second thiopurine subsequent to documented TIP in 4 children with CD (Table 1). These children were able to tolerate the re-introduction of a thiopurine without recurrence of AP.
Drugs are considered a rare cause of AP in the general population, and most cases that do occur are mild and self-limited (10). There is, however, an 8-fold higher risk of AP in patients with IBD treated with thiopurines (11). The overall incidence of TIP in IBD is approximately 3% (12).
Thiopurine-induced AP usually occurs within 4 weeks of starting treatment and occurs more frequently in girls (13). Pediatric patients are mentioned in the literature as at-risk cohorts, but this may be conferred because other causes of pancreatitis (eg, hyperlipidemia, gallstones, alcohol) are less common in children. Hence, in children, a diagnosis of TIP, which is in part a diagnosis of exclusion, is easier to arrive at. In addition, viral etiology is fairly common in pediatric pancreatitis, and this is difficult to fully exclude in most cases (14,15).
There are various other causes of pancreatitis in IBD besides thiopurines. Other IBD therapies have also been implicated in AP including 5-ASA compounds, corticosteroids, and metronidazole (16). Common bile duct obstruction as a complication of duodenal CD has also been implicated in the development of AP (17). Autoimmune pancreatitis, immunoglobulin G4 positive or negative, may be associated with IBD and provide another pathway for the development of AP in CD (18). Furthermore, primary sclerosing cholangitis, which is strongly associated with IBD, can cause AP either via a direct obstructive effect or as the result of the same autoimmune disturbance seen in AIH (19).
Various pathophysiologic mechanisms of drug-induced AP are described: these include a direct toxic effect (as with metronidazole), an accumulation of toxic metabolites (as with tetracycline), or an immunologic mechanism (9); however, there has been no clear understanding of the mechanism of TIP. One study suggested that a deficiency in inosine triphosphate pyrophosphohydrolase, an enzyme involved in metabolism of thiopurines, may increase the risk of TIP, but this has not been supported in other studies (20). Another hypothesis of the etiology of idiosynchratic reactions of AZA, such as dyspepsia and part of the spectrum of liver injury, is from the release of imidazolic compounds resulting from the conversion of AZA to 6-MP (21); however, whether this has any role in TIP is merely speculation.
Although AP is a well-recognized adverse effect in patients with IBD treated with thiopurines, it is much rarer or not associated with AZA therapy in many other conditions, including rheumatoid arthritis, systemic lupus erythematosis, or in posttransplant patients. This was best demonstrated in a large retrospective review of 1564 patients treated with AZA at a single university hospital between 1995 and 2002; AP was detected as a complication of AZA in 4.9% of patients with CD, 1.1% with ulcerative colitis, and 1.5% with autoimmune hepatitis, compared with 0.5%, 0.4%, 0%, and 0% in patients postkidney transplant, liver transplant patients, and patients with systemic lupus ertythematosus and Wegener granulomatosis, respectively (22).
TIP is considered a non–dose-dependent idiosyncratic complication. It has been the practice in our clinic to start AZA at low doses and monitor for signs of myelosuppression. If this test dose is tolerated, the dose is increased to 2.5 mg/kg, usually rounded off to the nearest 25 mg for ease of administration. This practice may have been partially supplanted by TPMT testing.
TIP is generally considered an indication to cease therapy because of the assumed risk of recurrence of AP on reintroduction. Although the arsenal of medication available for the treatment of IBD is growing, it is still somewhat limited, and the effect of excluding an effective, generally well tolerated, and relatively inexpensive medication from the list can at times be significant. 6-Thioguanine, another thiopurine derivative, could also be considered an alternative to 6-MP; however, given the added expense and more limited experience, TIP would not be its most appropriate indication.
At the time of writing, there existed only rare case reports and anecdotal evidence that reintroducing thiopurines following an assumed TIP could be well tolerated. Alexander et al (23) describes 3 patients, 2 with convincing pancreatitis, who tolerated a change from AZA to 6-MP. Interestingly, in 1 case, a second bout of pancreatitis occurred after an accidental dose of AZA, despite 6-MP being well tolerated. Lees et al (24) presented mixed data with 1 patient with possible pancreatitis because of AZA tolerating 6-MP, and another patient who developed a second attack following 6-MP.
This case series is a small uncontrolled study, but it does include patients with well-defined CD who developed well-delineated AP appropriately related temporally to the initiation of thiopurine therapy. Although an extensive investigation of other causes of AP was not undertaken in these patients, with the exception of an abdominal US in most of them, all settled promptly with cessation of AZA. The clinical presentation of each case is typical of the common reactions seen in children with TIP. No patient had a personal or family history of AP, and none had any subsequent episodes.
Ideally, a study aimed at assessing the safety and success of introducing a second thiopurine after TIP would be a controlled prospective trial, and this would be a useful future study to be undertaken. This case series is merely a presentation of 4 successful cases of thiopurine introduction within a short period of time in our IBD clinics, which should serve as a demonstration of the possibility of this course of action.
There exists extremely inadequate data to prognosticate the risk of developing a further bout of pancreatitis from repeat exposure to a thiopurine. Because most cases of TIP are mild and self-limited (14), the risk of an unsuccessful reintroduction could be considered relatively low, particularly in the otherwise healthy younger patient.
This case series should call into question the assumption that suspected TIP is an absolute contraindication to the future use of a second thiopurine. Cautious reintroduction of a thiopurine, in a controlled setting, should be considered in certain circumstances. The clinical relevance of this option is most marked in patients with complicated disease who require long-term immunosuppressive therapy, in whom other therapies are poorly tolerated or contraindicated. The benefit of this course of action would be limiting the need to escalate therapy, with potential added toxicity, until required for the management of more aggressive disease, rather than a lack of treatment options.
Additionally, thiopurines are extensively used in combination with antitumor necrosis factor-α therapy to enhance and prolong the effect of the biological agent. This combination effect is more pronounced with thiopurines than methotrexate (25). Hence, the option of continued use of this effective medication may be of great benefit in the long-term outcome of CD and ulcerative colitis in the 3% of patients who will have TIP.
1. Nielsen OH, Vainer B, Rask-Madsen J. Review article: the treatment of inflammatory bowel disease
with 6-mercaptopurine or azathioprine. Aliment Pharmacol Ther
2. Tiede I, Fritz G, Strand S, et al. CD28-dependent Rac1 activation is the molecular target of azathioprine in primary human CD4+ T lymphocytes. J Clin Invest
3. Bean RH. The treatment of chronic ulcerative colitis with 6-mercaptopurine. Med J Aust
4. Fraser AG, Orchard TR, Jewell DP. The efficacy of azathioprine for the treatment of inflammatory bowel disease
: a 30 year review. Gut
5. Pearson DC, May GR, Fick GH, et al. Azathioprine and 6-mercaptopurine in Crohn disease. A meta-analysis. Ann Intern Med
6. Pearson DC, May GR, Fick GH, et al. Azathioprine for maintenance of remission in Crohn's disease. Cochrane Database Syst Rev
7. Yu P, Johnson S, Wang ST, et al. Cost utility of adalimumab versus infliximab maintenance therapies in the United States for moderately to severely active Crohn's disease. Pharmacoeconomics
8. Gearry RB, Barclay ML, Burt MJ, et al. Thiopurine
drug adverse effects ina population of New Zealand patients with inflammatory bowel disease
. Pharmacoepidemiol Drug Saf
9. Balani AR, Grendell JH. Drug-induced pancreatitis
: incidence, management and prevention. Drug Saf
10. Nitsche CJ, Jamieson N, Lerch MM, et al. Drug induced pancreatitis
. Best Pract Res Clin Gastroenterol
11. Floyd A, Pedersen L, Nielsen GL, et al. Risk of acute pancreatitis
in users of azathioprine: a population-based case-control study. Am J Gastroenterol
12. Ooi CY, Bohane TD, Lee D, et al. Thiopurine
metabolite monitoring in pediatric inflammatory bowel disease
. Aliment Pharmacol Ther
13. Bermejo F, Lopez-Sanroman A, Taxonera C, et al. Acute pancreatitis
in inflammatory bowel disease
, with special reference to azathioprine induced pancreatitis
. Aliment Pharmacol Ther
14. Parenti DM, Steinberg W, Kang P. Infectious causes of acute pancreatitis
15. Sibert JR. Pancreatitis
in children. A study in the North of England. Arch Dis Child
16. Pitchumoni CS, Rubin A, Das K. Pancreatitis
in inflammatory bowel disease
. J Clin Gastroenterol
17. Foutch PG, Ferguson DR. Duodenal Crohn's disease complicated by common bile duct obstruction: report of a case and review of the literature. Am J Gastroenterol
18. Ravi K, Chari ST, Vege SS, et al. Inflammatory bowel disease
in the setting of autoimmune pancreatitis
. Inflamm Bowel Dis
19. Saeki K, Hozawa S, Miyata N, et al. IgG4-negative autoimmune pancreatitis
with sclerosing cholangitis and colitis: possible association with primary sclerosing cholangitis? Intern Med
20. Van Dieren JM, Hansen BE, Kuipers EJ, et al. Meta-analysis: inosine triphosphate pyrophosphatase polymorphisms and thiopurine
toxicity in the treatment of inflammatory bowel disease
. Aliment Pharmacol Ther
21. McGovern DP, Travis SP, Duley J, et al. Azathioprine intolerance in patients with IBD may be imidazole-related and is independent of TPMT activity. Gastroenterology
22. Weersma RK, Peters FT, Oostenbrug LE, et al. Increased incidence of azathioprine-induced pancreatitis
in Crohn's disease compared with other diseases. Aliment Pharmacol Ther
23. Alexander S, Dowling D. Azathioprine pancreatitis
in inflammatory bowel disease
and successful subsequent treatment with mercaptopurine. Intern Med J
24. Lees CW, Maan AK, Hansoti B, et al. Tolerability and safety of mercaptopurine in azathioprine-intolerant patients with inflammatory bowel disease
. Aliment Pharmacol Ther
25. Sokol H, Seksik P, Carrat F, et al. Usefullness of co-treatment with immunomodulators in patients with inflammatory bowel disease
treated with scheduled infliximab maintenance therapy. Gut