There can be little doubt that the thiopurines, azathioprine (AZA), 6-mercaptopurine (6-MP) and 6-thioguanine (6-TG), are effective therapies for inflammatory bowel disease (IBD). Despite this, clinicians still face a number of difficult clinical decisions when considering thiopurine therapy. When should AZA or 6-MP be started? Is there any difference between AZA and 6-MP? When should it be stopped? How can thiopurine therapy be tailored to the individual? What monitoring of therapy should be in place? Are thiopurines safe in pregnancy? What are the long-term risks of thiopurine therapy? The aim of this review is to address these challenging questions.
Evidence of efficacy
Thiopurines are effective both for active disease and for maintaining remission in both Crohn's disease (CD) and ulcerative colitis (UC). A Cochrane review of the efficacy of AZA and 6-MP for inducing remission in active CD demonstrates a benefit for thiopurine therapy compared to placebo with an odds ratio of 2.36 (95% CI 1.57–3.53) . This equates to a number needed to treat (NNT) of 5 (< 17 weeks’ treatment NNT 18, > 17 weeks’ treatment NNT 4) and a number needed to harm (NNH) of 14. As a result of the prolonged time to induction of remission, the thiopurines are more commonly used as maintenance therapy. Their efficacy at maintaining remission in CD is confirmed in another Cochrane review (OR 2.16 (CI 1.35–3.47), NNT 7)  and in a withdrawal study in UC from Nottingham, in which patients with UC in remission on AZA were randomized to placebo or continuing AZA . The 1 year relapse rate in the placebo and AZA groups respectively was 61% and 31% (P = 0.01, NNT 3).
When should thiopurines be considered?
Commencing thiopurine therapy for IBD could, logically, be considered at one of three points in the natural history of disease:
in newly diagnosed patients,
post-operatively (CD) or after a severe attack (UC and CD),
or in those who are steroid dependent or steroid refractory (UC and CD).
Thiopurines in newly diagnosed IBD patients
There are no data on the effectiveness of this approach in adults or in UC. A multi-centre study in paediatric CD suggests that initial treatment with thiopurine may be of benefit . Fifty-five children (< 18 years) with recently diagnosed CD and moderate-to-severe disease activity scores were randomized to either 6-MP (1.5 mg/kg/day) or placebo in addition to their standard tapering corticosteroid regimen. Forty-seven per cent of the placebo group relapsed, compared with only 9% in the 6-MP group (P = 0.007, NNT approximately 3) during the 18 month follow-up period. In addition, the cumulative steroid dose was lower at 6, 12 and 18 months (P = 0.01). 6-MP appeared safe and well tolerated, while growth was comparable in both groups. However, the data from this study suggest that the majority of patients (53%) are exposed to thiopurines unnecessarily (though this number may decrease with longer follow-up). Given that some long-term concerns on the safety of thiopurines remain, this approach may not be warranted, especially in children. Further studies on the course of disease to identify or predict those most likely to relapse and therefore benefit from a thiopurine would be of great value. Given the generally more benign course of disease in UC compared to CD, thiopurine therapy at first presentation is not warranted.
Thiopurines for the prevention of post-operative recurrence in Crohn's disease
There has only been one prospective study reported on the effectiveness of thiopurines in post-operative CD and, to date, this has only appeared in abstract . Following ileal or ileocolic resection for CD, 131 patients were randomized to receive 6-MP, mesalazine 3 g daily or placebo. The clinical relapse rates at 2 years were 50%, 59% and 69%, respectively. 6-MP was significantly better than mesalazine or placebo (P = 0.05). A criticism of this study may be the low dose of 6-MP (50 mg) and that the dosage was not calculated according to body weight (usually 1–1.5 mg/kg/day). It would also, of course, be reassuring to see this study published in a peer reviewed journal. Nevertheless, this study confirms that 6-MP is effective at preventing recurrence in a prospective randomized trial with an NNT of 5. The majority of patients, however, do very well after resection (72% and 64% in remission at 5 and 10 years, respectively)  and predictive risk factors for disease recurrence are needed. One suggestion is that 6 months after surgery all patients should have a small bowel barium or ileoscopy in order to identify macroscopic recurrence and hence those most likely to develop clinical recurrence . Those with macroscopic disease recurrence might be candidates for thiopurine therapy. This approach has not been validated and is difficult to justify in those who are asymptomatic after surgery. Other risk factors for post-operative disease recurrence of variable reliability include smoking, fistulating disease, extensive resection, concurrent perianal or jejunal disease and ileocolonic anastomosis [6,8–12]. It is reasonable to conclude that these patients, who appear to be at a high risk for recurrence, are most likely to benefit from post-operative thiopurine therapy.
Clearly there is no indication for thiopurines after colectomy in UC. However, thiopurine therapy should be considered for those who have a severe, colon threatening attack requiring intravenous corticosteroids and/or ciclosporin rescue therapy. While there is little objective evidence to support this approach, other than a lower rate of colectomy in those given AZA after receiving ciclosporin, it is a pragmatic interpretation of other data.
Thiopurine therapy for steroid refractory or steroid dependent disease
The thiopurines are steroid sparing agents (NNT 3) in CD  and should be considered for those patients whose symptoms recur as the dose of prednisolone is tapered or recur soon after prednisolone is stopped. Given the long-term sequelae of recurrent/prolonged corticosteroid therapy thiopurines should be considered at an early stage. For arbitrary but practical purposes, this means patients who require two or more corticosteroid courses within a calendar year and those whose disease relapses as the dose of steroid is reduced below 15 mg, or within 6 weeks of stopping steroids.
Azathioprine or 6-mercaptopurine?
For historical reasons AZA has been the choice of clinicians in Europe whilst in parts of America, 6-MP is used exclusively. There are no direct comparisons of the efficacy of AZA and 6-MP in IBD. Nevertheless, patients who are intolerant of AZA can frequently tolerate 6-MP . AZA intolerance is independent of metabolizing enzyme (thiopurine methyl transferase (TPMT)) activity and may be related to the imidazole derivative that is non-enzymatically cleaved from AZA during conversion to 6-MP . However, the imidazole derivative may possess independent immunosuppressive properties  and AZA may be a more potent immunosuppressive than 6-MP . There may also be significant differences in the pharmacokinetics between branded (Imuran) AZA, generic AZA and 6-MP . Further work on the relative efficacies and pharmacokinetics of the thiopurines is warranted.
For how long should thiopurine therapy be continued?
The published evidence suggests that thiopurines may be effective as maintenance therapy for CD for up to 4 years. This retrospective study examined 157 patients and suggested that AZA remained effective at maintaining remission up to, but not beyond, 4 years. However, at 4 years the numbers included were small and only 28 patients remained in the analysis. More recently, the same group presented (Digestive Diseases Week 2002) the results of a prospective randomized trial . Eighty-three patients with CD who had been in remission for 312 years on AZA were randomized to either continue AZA or placebo and followed up for a further 18 months. The relapse rates were 21.3% and 7.9% in the placebo and AZA groups respectively (P = 0.0195). Risk factors for relapse included young age (≤30 years), raised C-reactive protein and anaemia, the latter presumably indicating persistent disease activity. There are two ways that these data can be interpreted: AZA continues to be effective for up to 5 years (NNT 8), or that approximately 80% of patients remain in remission without additional therapy 18 months after stopping AZA. Whether your glass is half full or half empty, these data show AZA to be effective for up to 5 years. Practical advice for patients who are started on AZA is to plan to continue treatment for 3–4 years and then to stop treatment except in those with evidence of continuing disease activity. For the 20% who relapse, AZA should be restarted and continued. In a large audit from Oxford of AZA therapy for IBD, of 324 patients who achieved remission on AZA, 63% and 62% were still in remission at 4 and 5 years, respectively .
Individual tailoring of thiopurine therapy
Tailoring or optimization of thiopurine therapy can occur prior to or during treatment. Clinicians should aim for a maintenance dose of AZA of 2–2.5 mg/kg/day and 6-MP of 1–1.5 mg/kg/day in IBD. Higher doses are more effective (Table 1) and, if patients are tolerating but not responding to the thiopurine, the dose should be increased up to the ‘maximum’ dose. However, in one study of 51 IBD patients who had not responded to 6-MP only 14 (27%) responded to simple dose escalation . Interestingly, most (13/14) of those who responded to dose escalation were male. The ‘maximum’ dose will differ between individuals and effectively means that level at which leucopenia develops. It may be as high as 3–4 mg/kg, but careful blood count monitoring is essential when such large doses are used. Since leucopenia is a myelotoxic side effect of thiopurines, it may reassure the clinician to define the metabolic phenotype of the individual by measuring TPMT activity or the TPMT genotype. Measuring TPMT activity or genotype in these circumstances is different to using these assays to predict general tolerability or efficacy of thiopurines.
TPMT activity and genotype
There has been considerable recent interest in thiopurine pharmacogenetics. Individuals who are TPMT deficient are at increased risk of thiopurine bone marrow suppression . However, in one study the majority (77%) of 41 patients with AZA induced bone marrow suppression did not carry a TPMT mutation . Patients carrying two TPMT mutations all developed their bone marrow toxicity within 2 months, while all but one of the heterozygotes developed toxicity within 7 months. In contrast, wild type homozygotes (the majority of the population) developed toxicity at any time up to 7 years after starting AZA therapy. This suggests that the TPMT status may predict early bone marrow toxicity. However, in another study, only one out of 13 IBD patients with thiopurine associated leucopenia carried a TPMT mutation and 97% of patients with thiopurine toxicity were wild type for TPMT . There is one small study suggesting that TPMT activity assessment can predict response to AZA  but further work is needed on the role of TPMT status in predicting response to and toxicity from thiopurines. From a practical point of view, establishing reliable biochemical assays for TPMT enzyme activity is more difficult than identifying the genotype, which any transplant immunology laboratory should be able to offer.
Concentration of 6-thioguanine nucleotides
The major active thiopurine metabolites are the 6-thioguanine (6-TG) nucleotides and some groups have assessed the role of the concentration of 6-TG nucleotides in red cells as a method of optimizing thiopurine therapy. In a study of 92 paediatric patients treated with 6-MP, Dubinsky et al. demonstrated that there was a benefit in patients who achieved a 6-TG nucleotide concentration above 235 pmol per 8 × 108 erythrocytes (OR 5 (2.6–9.7)) . Cuffari et al. found a similar result in 82 IBD patients who achieved a concentration above 250 pmol per 8 × 108 erythrocytes . In addition, they identified 22 thiopurine non-responders, 18 of whom responded to dose escalation, dictated by the objective of increasing the concentrations of 6-TG nucleotides above 250 pmol per 8 × 108 erythrocytes. In contrast, Lowry et al. found no correlation between the 6-TG nucleotide concentration and response in a larger (170 patients) study . It is fair to say that the field is still evolving and that no explicit guidelines on measuring TPMT activity, TPMT genotype or the concentration of 6-TG nucleotides during therapy are yet possible.
To summarize, there can be little doubt that higher thiopurine doses are more effective and dose escalation can induce remission in some ‘non-responders'. Whilst TPMT genotyping may predict early bone marrow toxicity, the role of measuring TPMT activity or the concentration of 6-TG nucleotide metabolites in thiopurine therapy has yet to be fully assessed.
Blood count monitoring during thiopurine therapy
The manufacturers of AZA recommend weekly full blood counts for the first 8 weeks of therapy followed by bloods at least every 3 months. There is no evidence that this is effective. The authors’ practice is to perform a full blood count every 2–4 weeks for 2 months and then every 4–8 weeks. The rationale for this approach is that, of patients who develop thiopurine associated myelotoxicity, approximately half will develop it within 2 months and nearly two thirds within 4 months . Just as important is the advice to patients to report promptly should a sore throat or any other evidence of infection occur. Profound leucopenia can develop suddenly and unpredictably, in between blood tests, although patients (and their general practitioners!) should be reassured that it is rare (around 3% overall). Bone marrow toxicity has been reported to occur up to 11 years after starting AZA  and blood monitoring should continue throughout thiopurine therapy. Thiopurines may also cause hepatotoxicity (prevalence approximately 2.5% ) and this may be related to elevated TPMT activity . Thiopurine doses should be reduced in renal impairment. For these reasons monitoring electrolytes and liver function tests throughout treatment is appropriate.
The safety of thiopurines during pregnancy
For women with IBD who are maintained on a thiopurine and are considering conception, our advice is usually to continue therapy through pregnancy. This advice is based on a study (as yet published in abstract form only) of 155 men and women with IBD ‘involved’ in 347 pregnancies whilst taking 6-MP. The authors found no difference in miscarriage, congenital abnormality or infection rate in the thiopurine group compared to a control group . More recent data have been published suggesting that 6-MP taken by the father within 3 months of conception may be associated with an increased risk of complications in pregnancy . The thiopurines are excreted in high concentrations in breast milk, and breast feeding should be avoided in those taking 6-MP or AZA. These issues should be discussed with the patient and partner: the balance of risk between a relapse of previously refractory CD should thiopurines be stopped, usually favours continuing treatment.
Long-term risks of thiopurine therapy
There have been conflicting data about the long-term risk of thiopurine therapy in IBD. Large audits from Oxford  (622 patients) and St Mark's  (755 patients) have shown no increased risk of lymphoma or other cancers in IBD patients treated with AZA. However, it is worth noting that the median follow-up of these two studies was approximately 2 years and 1 year, respectively. A primary care prescribing database study of nearly 1500 IBD patients who received at least one prescription of AZA/6-MP also showed no overall risk (relative risk 1.6 (95% CI 0.1–8.8)) of lymphoma but little is known about the duration or dose of therapy of this cohort . In contrast, a study from Dublin reported an increased risk of non-Hodgkin's lymphoma (relative risk 59) in IBD patients treated with immunomodulators . An accompanying editorial to this article suggests that, whilst there may be an increased risk of lymphoma in patients receiving immunosuppression (including thiopurines), as seen in other inflammatory conditions, this study probably exaggerates the risk . There have also been a number of reports of Epstein–Barr virus (EBV) positive lymphoma in IBD thiopurine recipients and in one study seven IBD patients with EBV positive and 11 IBD patients with EBV negative lymphomas were identified. Five out of the seven EBV positive lymphoma patients had been prescribed a thiopurine compared to only one of 11 in the EBV negative lymphoma cohort . A decision analysis suggested that, in IBD, the benefits of AZA outweighed the small risk of lymphoma .
The thiopurines are very effective therapies for IBD and are probably under-utilized by clinicians. Higher doses are more efficacious and AZA is effective for at least 5 years in maintaining remission in CD. The role of assessment of TPMT status prior to commencing thiopurine therapy remains controversial, but it may predict early bone marrow toxicity, and assays to determine the concentration of 6-TG nucleotides in red cells may help identify non-compliance and inadequate dosage. Blood monitoring should continue throughout thiopurine therapy and patients need to understand the importance of this. The thiopurines are safe in pregnancy but may be associated with a small risk if taken by fathers within 3 months of conception. Finally, there may be a small increased risk of lymphoma following thiopurine therapy, though this remains controversial, and the benefits appear to outweigh the risks.
1.Sandborn W, Sutherland L, Pearson D, May G, Modigliani R, Prantera C. Azathioprine
for inducing remission of Crohn's disease. Cochrane Database Syst Rev
2.Pearson DC, May GR, Fick G, Sutherland LR. Azathioprine
for maintaining remission of Crohn's disease. Cochrane Database Syst Rev
3.Hawthorne AB, Logan RFA, Hawkey CJ, Foster PN, Swarbrick ET, Scott BB, et al
. Randomised controlled trial of azathioprine
-withdrawal in ulcerative colitis. BMJ
4.Markowitz J, Grancher K, Kohn N, Lesser M, Daum F. A multicenter trial of 6-mercaptopurine
and prednisone in children with newly diagnosed Crohn's disease. Gastroenterology
5.Korelitz B, Hanauer S, Rutgeerts P, Present DH, Peppercorn M. Postoperative prophylaxis with 6-MP, 5-ASA or placebo in Crohn's disease: a 2 year multicenter trial. Gastroenterology
6.Bernell O, Lapidus A, Hellers G. Risk factors for surgery and recurrence in 907 patients with primary ileocaecal Crohn's disease. Br J Surg
7.Rutgeerts P. Protagonist: Crohn's disease recurrence can be prevented after ileal resection. Gut
8.Rutgeerts P, Geboes K, Vantrappen G, Beyls J, Kerremans R, Hiele M. Predictability of the postoperative course of Crohn's disease. Gastroenterology
9.Moskovitz D, McLeod RS, Greenberg GR, Cohen Z. Operative and environmental risk factors for recurrence of Crohn's disease. Int J Colorectal Dis
10.Cosnes J, Carbonnel F, Beaugerie L, Le Quintrec Y, Gendre JP. Effects of cigarette smoking on the long-term course of Crohn's disease. Gastroenterology
11.Aeberhard P, Berchtold W, Riedtmann HJ, Stadelmann G. Surgical recurrence of perforating and nonperforating Crohn's disease. A study of 101 surgically treated patients. Dis Colon Rectum
12.Lautenbach E, Berlin JA, Lichtenstein GR. Risk factors for early postoperative recurrence of Crohn's disease. Gastroenterology
13.McGovern DP, Travis SPL, Duley J, el Shobowale-Bakre M, Dalton HR. Azathioprine
intolerance in patients with IBD may be imidazole-related and is independent of TPMT activity. Gastroenterology
14.Crawford DJ, Maddocks JL, Jones DN, Szawlowski P. Rational design of novel immunosuppressive drugs: analogues of azathioprine
lacking the 6-mercaptopurine
substituent retain or have enhanced immunosuppressive effects. J Med Chem
15.Calne RY, Alexandre GPJ, Murray JE. A study of the effects of drugs in prolonging survival of homologous renal transplant in dogs. Ann NY Acad Sci
16.Cuffari C, Hunt S, Bayless TM. Enhanced bioavailability of azathioprine
compared to 6-mercaptopurine
therapy in inflammatory bowel disease
: correlation with treatment efficacy
. Aliment Pharmacol Ther
17.Lemman M, Bouhnik Y, Colombel J, Duclos B, Soule J, Lerebours E, et al
. Randomized, double blind, placebo-controlled, multicentre, azathioprine
withdrawal trial in Crohn's disease. Gastroenterology
18.Fraser AG, Orchard TR, Jewell DP. The efficacy
for the treatment of inflammatory bowel disease
: a 30 year review. Gut
19.Dubinsky MC, Yang H, Hassard PV, Seidman EG, Kam LY, Abreu MT, et al
. 6-MP metabolite profiles provide a biochemical explanation for 6-MP resistance in patients with inflammatory bowel disease
20.Lennard L, Gibson BE, Nicole T, Lilleyman JS. Congenital thiopurine methyltransferase deficiency and 6-mercaptopurine
toxicity during treatment for acute lymphoblastic leukaemia. Arch Dis Child
21.Colombel JF, Ferrari N, Debuysere H, Marteau P, Gendre JP, Bonaz B, et al
. Genotypic analysis of thiopurine S
-methyltransferase in patients with Crohn's disease and severe myelosuppresion during azathioprine
22.Dubinsky MC, Lamothe S, Yang HY, Targan SR, Sinnett D, Theoret Y, et al
. Pharmacogenomics and metabolite measurement for 6-mercaptopurine
therapy in inflammatory bowel disease
23.Campbell S, Kingstone K, Ghosh S. Relevance of thiopurine methyltransferase activity in inflammatory bowel disease
patients maintained on low-dose azathioprine
. Aliment Pharmacol Ther
24.Cuffari C, Hunt S, Bayless T. Utilisation of erythrocyte 6-thioguanine metabolite levels to optimise azathioprine
therapy in patients with inflammatory bowel disease
25.Lowry PW, Franklin CL, Weaver AL, Pike MG, Mays DC, Tremaine WJ, et al
. Measurement of thiopurine methyltransferase activity and azathioprine
metabolites in patients with inflammatory bowel disease
26.Connell WR, Kamm MA, Ritchie JK, Lennard-Jones JE. Bone marrow toxicity caused by azathioprine
in inflammatory bowel disease
: 27 years of experience. Gut
27.Francella A, Rubin P. 6-Mercaptopurine
is a safe therapy for child bearing patients with inflammatory bowel disease
28.Rajapakse RO, Korelitz BI, Zlatanic J, Baiocco PJ, Gleim GW. Outcome of pregnancies when fathers are treated with 6-mercaptopurine
for inflammatory bowel disease
. Am J Gastroenterol
29.Connell WR, Kamm MA, Dickson M, Balkwill AM, Ritchie JK, Lennard-Jones JE. Long-term neoplasia risk after azathioprine
treatment in inflammatory bowel disease
30.Lewis JD, Bilker WB, Brensinger C, Deren JJ, Vaughn DJ, Strom BL. Inflammatory bowel disease
is not associated with an increased risk of lymphoma. Gastroenterology
31.Farrell RJ, Ang Y, Kileen P, O'Briain DS, Kelleher D, Keeling PW, et al
. Increased incidence of non-Hodgkin's lymphoma in inflammatory bowel disease
patients on immunosuppressive therapy but overall risk is low. Gut
32.Kinlen LJ. Non-Hodgkin's lymphoma after immunosuppressive therapy. Gut
33.Dayharsh GA, Loftus EV Jr, Sandborn WJ, Tremaine WJ, Zinsmeister AR, Witzig TE, et al
. Epstein–Barr virus-positive lymphoma in patients with inflammatory bowel disease
treated with azathioprine
34.Lewis JD, Schwartz JS, Lichtenstein GR. Azathioprine
for maintenance of remission in Crohn's disease: benefits outweigh the risk of lymphoma. Gastroenterology