The use of 6-mercaptopurine in patients with inflammatory bowel disease after failure of azathioprine therapy


J. R.Boulton-JonesDr Kingsmill Hospital, Mansfield Road, Sutton-in-Ashfield, UK. E-mail:



Azathioprine is a useful therapy in patients with inflammatory bowel disease that is difficult to control. However, 10% of patients are unable to tolerate azathioprine, and the best form of treatment for this group of patients is unknown. The azathioprine metabolite 6-mercaptopurine may be a useful therapy for these patients.


To review our clinical experience of the use of the 6-mercaptopurine in inflammatory bowel disease patients who are intolerant of azathioprine.


All patients who were prescribed 6-mercaptopurine in a 2-year period were identified from pharmacy records. The case notes were reviewed and those who had previously been intolerant of azathioprine were included.


A total of 19 with either ulcerative colitis and Crohn’s disease were included. The reasons for discontinuing azathioprine were side-effects (13 patients), failure of efficacy (four patients) and leucopenia (two patients). Eleven of the 19 patients (68%) tolerated 6-mercaptopurine, including seven out of 13 patients (54%) who discontinued azathioprine due to side-effects. The length of follow-up of patients on 6-mercaptopurine was between 126 and 780 days (median 390 days).


6-mercaptopurine should be considered in patients with inflammatory bowel disease who require continuing immunosuppressive therapy, but are intolerant of azathioprine.


Azathioprine and its metabolite 6-mercaptopurine are used in patients with inflammatory bowel disease who have frequent relapses requiring steroids or in whom remission cannot be obtained without using unsustainable doses of steroids.1[2]–3 These drugs have significant side-effects that necessitate withdrawal of the drug in 8–10% of patients.2, 4, 5 The further management of patients in whom these drugs have been withdrawn is unclear. De-sensitization techniques have been used to overcome the side-effects, with mixed results.4, 6 An alternative approach may be to use 6-mercaptopurine when a side-effect has developed to azathioprine (or vice versa). In this article, we report our experience of using 6-mercaptopurine in patients who have been intolerant of or have had no clinical response to azathioprine.


All patients in our unit who were prescribed 6-mercaptopurine during a 2-year period from March 1997 to March 1999 were identified through pharmacy records. All the patient case notes were reviewed and those who had failed azathioprine therapy before commencing 6-mercaptopurine were included in this study. Data were collected as to the reason for stopping azathioprine, the length of time on azathioprine, the length of time on 6-mercaptopurine therapy and the outcome of 6-mercaptopurine therapy. The data obtained from case note review were supplemented by a face to face interview in the out-patient department.


A total of 19 patients were identified who had been prescribed 6-mercaptopurine after failing azathioprine treatment. The patient characteristics and results are summarized in Table 1. The patients selection was typical of those with inflammatory bowel disease: 12 of the 19 (61.2%) were female; the age range was 24–67 years (mean 42.3 years); 10 patients had Crohn’s disease and nine had ulcerative colitis. The disease distributions are shown in Table 1.

Table 1.  The disease distribution, reason for and timing of AZA withdrawal and outcome of 6-MP treatment are shown for each patientThumbnail image of

The reasons for discontinuing azathioprine fell into three groups. First, side-effects necessitated stopping the drug in 13 patients. The commonest side-effects experienced were abdominal pain and vomiting in eight patients. Interestingly the only patient that developed a documented rise in amylase was asymptomatic. Other common side-effects recorded were skin rash (two patients) and headaches (three patients). The second reason for stopping azathioprine was leucopenia (two patients) and the third was lack of efficacy (four patients). The dose of azathioprine that our patients were taking prior to discontinuation ranged from 50 to 150 mg (median dose 100 mg). The length of time for which our patients tolerated azathioprine before developing side-effects ranged from 4 to 1642 days, with a median of 26 days. There was a marked difference in the length of time for which these subgroups were on azathioprine. Those that experienced side-effects remained on azathioprine for between 4 and 178 days (median 18 days), whereas those that stopped due to failure of efficacy or leucopenia, were on azathioprine for between 78 and 1542 days (median 291 days).

Eleven of the 19 (68%) patients were able to tolerate 6-mercaptopurine, with follow-up on 6-mercaptopurine ranging from 126 to 780 days (median 390 days). The 11 patients who tolerated 6-mercaptopurine were taking a dose ranging from 50 to 100 mg (median dose 100 mg). Those who failed to tolerate 6-mercaptopurine were on a dose of 25–100 mg (median dose 50 mg). Of those who stopped azathioprine due to side-effects, seven out of 13 (54%) were able to tolerate 6-mercaptopurine. Eight patients failed 6-mercaptopurine treatment; one patient required emergency surgery for obstruction whilst on 6-mercaptopurine and was therefore considered a treatment failure; seven developed side-effects. Two patients had been able to tolerate azathioprine but stopped due to lack of efficacy, illustrating that there are patients who can tolerate azathioprine but not 6-mercaptopurine. Of the remainder, four had similar side-effects on both medications, whilst one patient suffered different side-effects to each drug.

All patients were subsequently commenced on 6-mercaptopurine between 1 and 547 days after stopping azathioprine. There was no obvious relationship between the length of time between stopping azathioprine and starting 6-mercaptopurine and side-effects related to 6-mercaptopurine, which does not suggest that a wash-out period between the use of these two drugs is essential.

Four patients discontinued azathioprine because it was judged clinically ineffective. Two of these patients subsequently failed to tolerate 6-mercaptopurine. The clinical condition of the two other patients was judged to have improved on 6-mercaptopurine. One patient with perianal Crohn’s disease had a persistently leaking perianal fistula, sigmoidoscopic appearances of active disease and raised inflammatory markers including a C-reactive protein (CRP) of 79 (normal range < 10) and a platelet count of 591 (normal range 150–400), despite 4 months treatment with azathioprine. After commencement of 6-mercaptopurine, both patient and physician felt that he had symptomatically improved, the rate of drainage from his fistula decreased (although it did not heal completely) and his CRP fell to < 10 and platelet count to 326. Sigmoidoscopic appearances were not reassessed. The second patient had ulcerative proctitis with persistent rectal bleeding and active inflammation on sigmoidoscopy despite 9 months treatment with azathioprine; CRP was 10. Following conversion to 6-mercaptopurine, the rectal bleeding resolved, both patient and physician felt that he was symptomatically better and CRP became undetectable. Sigmoidoscopic appearances were not reassessed. These cases suggest that a proportion of patients may have a better therapeutic response to 6-mercaptopurine than azathioprine.


Azathioprine and 6-mercaptopurine are related purine analogues that have been successfully used to treat inflammatory bowel disease.1, 3 However, around 10% of patients are unable to tolerate these drugs due to side-effects.4 Common side-effects include nausea, diarrhoea, abdominal pain, pancreatitis, fever, muscular pains, arthralgia and disturbed liver function. These side-effects are due to hypersensitivity reactions and immediate withdrawal of the drug is recommended;7 dose-related leucopenia can also occur.8 6-Mercaptopurine is an azathioprine metabolite that is produced by a non-enzymatic reaction of azathioprine with amino- or sulfhydral-containing compounds.9 It has a similar range of side-effects to azathioprine although it is unclear whether patients who fail to tolerate azathioprine can be successfully treated with 6-mercaptopurine.

A total of 140 patients in our practice were treated with azathioprine during the study period and in 20 patients the azathioprine was discontinued. Four patients were discontinued because azathioprine was judged clinically ineffective, one patient elected to discontinue azathioprine and the remaining 15 (10.7%) had to discontinue azathioprine due to side-effects. The rate of side-effects is similar to previous reports.4, 5 All 20 patients were offered treatment with 6-mercaptopurine, and all accepted except the patient who had chosen to discontinue azathioprine.

The reasons for withdrawal of azathioprine in our study are typical of those reported in the literature. The majority of patients who developed hypersensitivity reactions did so soon after commencing azathioprine, whereas those that developed leucopenia did so after having been on azathioprine for some time. Hypersensitivity reactions do appear to occur early, although one meta-analysis suggests that there may be a trend towards increasing side-effects with increasing cumulative dose of azathioprine or 6-mercaptopurine.3, 4, 6 Previous reports suggest that patients who are maintained on azathioprine for more than 6 months rarely have to stop the drug due to side-effects.10 Leucopaenia can develop at any time after commencing therapy, although severe leucopenia does tend to occur early in the course of treatment.8

Our findings suggest that approximately half of patients who cannot tolerate azathioprine, can take 6-mercaptopurine. The length of patient follow-up, whilst on 6-mercaptopurine, of between 126 and 780 days is adequate for most side-effects to have developed and does suggest that these patients can genuinely tolerate 6-mercaptopurine. Furthermore, these results are similar to a previous report where four out of seven patients with Crohn’s disease were able to tolerate 6-mercaptopurine after experiencing side-effects to azathioprine.11 Other authors report that five out of six patients who suffered allergic complications with 6-mercaptopurine and were treated with azathioprine, subsequently developed the same side-effect.6 These conflicting reports may be because patients who experience side-effects on 6-mercaptopurine cannot tolerate azathioprine, since azathioprine is rapidly converted to 6-mercaptopurine in the body.9 If azathioprine is given as the initial therapeutic agent, it may be azathioprine itself that is responsible for the side-effect in a proportion of patients, and that giving 6-mercaptopurine may bypass the adverse reaction. Alternatively, the side-effects reported in the second paper were primarily of fever or joint pain and may be due to an immunological hypersensitivity reaction caused by either drug. Conversely, in our report the patients suffered primarily from gastrointestinal side-effects which may be caused by a different mechanism. This latter possibility may explain why the two patients in our study who developed fevers (which are almost certainly immunologically mediated) whilst on azathioprine, had an identical reaction to 6-mercaptopurine. Further work in this area is needed.

Our finding that some patients appear to have a better therapeutic response to 6-mercaptopurine is interesting. There have been no studies directly comparing the immunosuppressive activity of azathioprine and 6-mercaptopurine in humans. Animal studies have suggested that azathioprine is a more potent immunosuppressant than 6-mercaptopurine.12, 13 In clinical practice, however, some reports suggest that 6-mercaptopurine may be more effective than azathioprine. One report details four patients with Crohn’s disease who failed azathioprine therapy but improved on 6-mercaptopurine.14 Another paper reports a patient with autoimmune chronic active hepatitis who did not respond to azathioprine, but entered complete remission with 6-mercaptopurine.15 Therefore, there may be a group of patients with inflammatory or autoimmune conditions in whom 6-mercaptopurine is a better immunosuppressant than azathioprine. Why this should be so is unclear, but it has been suggested that a proportion of azathioprine may be metabolized via an alternative pathway that does not result in the production of 6-mercaptopurine and is therefore not therapeutically effective.9 The use of 6-mercaptopurine in patients in whom azathioprine has failed due to lack of efficacy is an area that should be further studied by a prospective, preferably randomized trial.

In summary, our experience confirms that 6-mercaptopurine can be tolerated by about half of patients who are intolerant of azathioprine. 6-Mercaptopurine should be considered in inflammatory bowel disease patients intolerant of azathioprine who require continued immunosuppressive therapy. Patients who have to discontinue azathioprine due to severe hypersensitivity reactions should probably be excluded. Furthermore, 6-mercaptopurine may be useful in a proportion of patients in whom azathioprine has failed to achieve an adequate clinical response. Further work on this question is required.