The incidence of inflammatory bowel disease (IBD) in children is increasing, with more than 25% of patients being diagnosed under the age of 16 (1,2). Children and adolescents with IBD are more likely to have Crohn disease (CD) than ulcerative colitis (UC). Both diseases tend to be more extensive and severe at diagnosis and more likely to extend in the first 3 years than in adults (3–5). Despite these factors, response rates to thiopurines in children (6–8) are reportedly better than those in adults (9–11). In CD at least, it has been postulated that this is because these drugs are used earlier in children than in adults, and at a time when the disease tends still to be inflammatory rather than stricturing or penetrating (12). No previous studies have controlled for differences in phenotype or directly compared response rates in adults with response rates in children. Suspecting that the response to thiopurines in children with IBD would be similar to that in adults when phenotype is matched between the 2 groups, we retrospectively evaluated the effects of thiopurine use in IBD, comparing steroid-free remission, response, and relapse rates in children with that in adults matched for sex, disease phenotype, and extent.
PATIENTS AND METHODS
Study Design and Setting
We conducted a retrospective evaluation of the response to thiopurines in children compared to adults matched for sex, disease type, and extent at Barts and the London National Health Service Trust.
Index Paediatric Cases
Using our electronic database we identified 65 children diagnosed as having IBD by conventional endoscopic, radiological, and histological criteria, who were started on a thiopurine before age 17 years.
Adults, defined here as starting on a thiopurine at 17 years old or older were identified using our therapeutic drug-monitoring database. This database is used by our IBD pharmacists to monitor all patients with IBD on immunosuppressants. Adult data were collected by researchers who were unaware of the paediatric data. Children were matched to adults for sex, disease type, location, behaviour, and/or extent according to the Montreal classification (13) at the time of azathioprine introduction. Where more than 1 adult match was identified, the patient with the closest disease duration was included.
Overall, we identified 65 children and 367 adults; 6 children and 130 adults were excluded because we could not accurately identify a start date for the thiopurine or they did not have 18 months of follow-up. We were unable to match 9 children for disease extent, leaving 182 adults unused (Fig. 1).
Definitions of Remission and Relapse
Demographic details were recorded at baseline (ie, the time at which they were started on azathioprine or mercaptopurine [MP]), and response rates at the closest follow-up visit to 6 months are reported here. Patients with normal thiopurine methyltransferase (TPMT) levels were given doses of 2 to 2.5 mg/kg, and 1 to 1.25 mg/kg for TPMT levels in the heterozygous range. Clinical remission was defined as the absence of symptoms, having withdrawn from steroids (prednisolone or budesonide) for more than 3 months, and having a C-reactive protein (CRP) <5 mg/L. Response was defined as 2 of the absence of symptoms, having withdrawn from steroids for more than 3 months, and having a normal C-reactive protein. Nonresponse was defined as 1 or none of the above (14).
Secondary outcomes recorded were primary intolerance requiring a switch to MP, as well as the proportions of patients who discontinued thiopurines in the first 6 months because of leucopenia, hepatitis, and pancreatitis. In those patients who had responded or were in full remission after 6 months of taking the thiopurine, time to relapse and subsequent treatments were recorded. Relapse was defined for UC as the presence of rectal bleeding and diarrhoea requiring an escalation in medication, and for CD as worsening abdominal pain, diarrhoea, and/or increase in CRP with a change in medication.
Statistical analyses were conducted using SPSS (version 16; SPSS Inc, Chicago, IL) and Prism (version 4; GraphPad, San Diego, CA) software. All of the analyses were two-tailed and P values <0.05 were considered significant. Baseline demographic data were analysed using the chi-square test for categorical data, the Mann-Whitney U test for discrete continuous variables, and the Student t test for normally distributed continuous variables. Differences in the proportions of adult and paediatric patients in clinical remission at 6 months were sought using the chi-square test. Survival analysis using Kaplan-Meier curves and the log-rank test were used to compare time to relapse during the following 1 year for those patients in remission and those who had at least a response at 6 months and for relapse rates in responders by age and disease type.
Because we were evaluating the outcome of thiopurine usage in our paediatric and adult services, we did not require formal ethical approval, according to the guidelines of the UK National Research Ethics Service (NRES) (15).
The median age at diagnosis (children age 10 years [5–15] versus adults age 27 years [10–67], P < 0.0001) was significantly lower in the children. As a consequence of our study design and matching, there were no significant differences between children and adults in terms of sex, ethnicity, disease type, location and behaviour, and/or extent (Table 2). Although not included as matching criteria between groups, there were no differences in disease duration (children 5.5 years [5–15] vs adults 6.8 [10–67], P = 0.09) or interval between diagnosis and the introduction of a thiopurine (children 1.3 years [0.3–6.7] vs adults 1.9 [0.2–20.9], P = 0.14).
More children (42% [13/31]) had documented upper gastrointestinal CD than adults (6% [2/31], P < 0.01). There was no difference in the proportions of patients who received steroids or in baseline steroid dose. Significantly more children with CD had received an exclusive enteral diet and had been treated with 5-aminosalicylic acids (5-ASAs). More adults had previous surgery (Table 1); in this group azathioprine was not used for secondary prophylaxis, and the mean (SEM) time to azathioprine for postoperative inflammatory recurrence was 6.2 (2.3) years. No differences in disease activity or medications were seen between children or adults with UC (Table 2).
Baseline Laboratory Tests
Paediatric patients were more anaemic at baseline (mean hemoglobin [SEM] 10.7 g/dL [0.2] vs adults 11.3 g/dL [0.2], P < 0.05), with significantly lower mean cell volume (MCV) (73.3 fL [0.2] vs adults 83.5 fL [1.2], P < 0.0001). Platelet counts and erythrocyte sedimentation rate (ESR) were significantly higher at baseline in children than in adults, but no differences were seen in CRP, neutrophil counts, or lymphocyte counts (Table 3).
Tolerance and Adherence
Overall, 1 child discontinued thiopurines before 6 months, significantly fewer than the 16% (8/50) adults who did so (P < 0.05). One child had pancreatitis compared to 3 adults; the remainder of the adults discontinued because of nausea and vomiting, nonadherence, and abnormal liver function tests (Table 4).
Response Rates After 6 Months of Thiopurine Therapy
At 6 months, steroid-free remission was achieved in 30% (15/50) of children, compared with 38% (19/50) of adults; a further 40% (20/50) of children and 26% (13/50) of adults had a response to the thiopurine (P = 0.33). Likewise, during the first 6 months of thiopurine therapy, platelet and neutrophil counts fell and MCV levels increased to similar degrees in both adults and children (Table 3). No differences were seen in response rates according to adherence (Table 4). Overall, 68% (21/31) of both adults and children with CD (P = 1) and 56% (10/18) of adults and 72% (13/18) of children with UC had at least a response after 6 months of treatment.
Phenotypic Factors Influencing Response After 6 Months of Thiopurine Therapy
There were no differences in response or remission rates at 6 months according to CD location or behaviour or to UC extent. In addition, no differences were seen in remission and/or response rates in those patients given thiopurines within 6 months of diagnosis (38% (9/24)) compared with those starting a thiopurine more than 6 months after diagnosis (33% [25/76], P = 0.81).
Relapse Rates and Subsequent Treatments in Responders Beyond 6 Months
Overall, patients who had a response to the thiopurine were significantly more likely to relapse and did so sooner than those who were in full clinical remission at 6 months, with no differences between adults and children (Fig. 1). No significant differences were seen in relapse rates between adults and children for UC and CD (Fig. 2). Similarly, there was no difference in the subsequent use of infliximab or adalimumab or in those undergoing surgery (Table 5), although children who had only a response at 6 months were more frequently admitted to hospital for further management.
Thiopurines are reportedly more efficacious in children than in adults with IBD. In relation to CD, it has been proposed (7) that this is because they are used earlier in the disease course in children when inflammatory disease predominates over stricturing and penetrating disease (12). No previous studies have directly compared the responses of children and adults to thiopurines. Nor has attention been given to the influence on response to thiopurines of differing phenotypes, where, especially in CD, noninflammatory complications such as bacterial overgrowth or bile salt malabsorption may cause symptoms that would not be expected to improve with immunosuppression.
Although our numbers are relatively small and because of our retrospective design the results are potentially subject to interpretation bias, we have used clinically relevant and observer-independent outcome criteria that are likely to be applicable in day-to-day practice. Thus, having controlled for disease phenotype, we report that the rates of corticosteroid-free remission and subsequent relapse rates in children are no different from those in adults.
Remission and Relapse Rates in CD
The rates of thiopurine-induced steroid-free remission reported here are similar to the 30% reported in the recent SONIC trial (16), and 27% to 38% at 4 to 7 months in previous studies in adults that have used CD activity index (CDAI) to define remission (17–19). Our remission rates are, however, lower than in other studies (6,20,21) that did not use complete steroid withdrawal as an outcome criterion. The marked disparity between the response rates in the only preceding randomised controlled trial of 6-mercaptopurine (6-MP) in CD in children is probably explained primarily by differences in definitions of remission: Markowitz et al (6) defined remission by CDAI and did not require discontinuation of steroids.
It is conceivable that the shorter the interval between diagnosis and the introduction of a thiopurine, the better its effect (6,7), because a higher proportion of patients treated earlier would have inflammatory rather stricturing or penetrating disease (12). In our albeit small subsample the rates of clinical remission were no different in those patients treated with a thiopurine within 6 months of diagnosis than in those treated at a later stage in the more usual “bottom-up” manner. In addition, and acknowledging the risk of making a type II error, we found no differences in response according to disease behaviour or location.
Remission and Relapse Rates in UC
Our results for UC are consistent with those published in randomised trials in adult studies. Timmer et al, in the most recent Cochrane meta-analysis (11) of 4 trials (22–25), reports a pooled odds ratio of 0.41 (95% CI 0.24–0.70), equivalent to a 30% (range 17%–41%) chance of failing to maintain remission at 12 months or longer. No randomised controlled studies of thiopurines have been conducted exclusively in children with UC. Retrospective studies, again reporting steroid dose reduction but not withdrawal, suggest remission rates of 50% to 75% (26–28). More recent data presented in abstract form, from a large cohort of children from North America, report steroid-free survival rates of about 70% at 1 year (29), findings that resemble ours.
As reported in a previous series of different patients from our centre, the rates of intolerance and subsequent discontinuation because of adverse effects of thiopurines are higher in adults than in children (5). Drug adherence could account for the difference in tolerance, because adherence is reportedly worse in children and adolescents than in adults (30,31) and individuals not taking a drug cannot have adverse effects from it and will not therefore be recorded as being intolerant to it. Against significant nonadherence in the paediatric group here is the observation that surrogate markers of thiopurine use, including the degree to which the neutrophil count decreased and the MCV increased, were similar between groups.
In summary, it is likely that the perceived superior response rates to thiopurines in children relate not to differences in interval to treatment, and therefore disease behaviour at the time of their introduction, but to differences in definitions of remission and/or relapse employed in different studies. Corticosteroid-free remission and subsequent relapse rates are similar in children and adults when their disease phenotype is matched, although children do seem to tolerate thiopurines better than adults.
We are grateful to Dr James Lindsay, Dr Louise Langmead, Dr Sandhia Naik, and Dr Nigel Meadows for their constructive criticism whilst reviewing the work and for allowing us to include their patients. We also thank the National Association of Crohn's and Colitis and CORE/British Society of Paediatric Gastroenterology, Hepatology, and Nutrition for funding J.R.G.'s research.
1. Sawczenko A, Sandhu BK, Logan RF, et al
. Prospective survey of childhood inflammatory bowel disease in the British Isles. Lancet 2001; 357:1093–1094.
2. Armitage E, Drummond HE, Wilson DC, et al
. Increasing incidence of both juvenile-onset Crohn's disease and ulcerative colitis in Scotland. Eur J Gastroenterol Hepatol 2001; 13:1439–1447.
3. Vernier-Massouille G, Balde M, Salleron J, et al
. Natural history of pediatric Crohn's disease: a population-based cohort study. Gastroenterology 2008; 135:1106–1113.
4. Van Limbergen J, Russell RK, Drummond HE, et al
. Definition of phenotypic characteristics of childhood-onset inflammatory bowel disease. Gastroenterology 2008; 135:1114–1122.
5. Goodhand J, Dawson R, Hefferon M, et al
. Inflammatory bowel disease in young people: the case for transitional clinics. Inflamm Bowel Dis 2010; 16:947–952.
6. Markowitz J, Grancher K, Kohn N, et al
. A multicenter trial of 6-mercaptopurine and prednisone in children with newly diagnosed Crohn's disease. Gastroenterology 2000; 119:895–902.
7. Markowitz J. Early inflammatory bowel disease: different treatment response to specific or all medications? Dig Dis 2009; 27:358–365.
8. Punati J, Markowitz J, Lerer T, et al
. Effect of early immunomodulator use in moderate to severe pediatric Crohn disease. Inflamm Bowel Dis 2008; 14:949–954.
9. Prefontaine E, Macdonald JK, Sutherland LR. Azathioprine or 6-mercaptopurine for induction of remission in Crohn's disease. Cochrane Database Syst Rev
10. Prefontaine E, Sutherland LR, Macdonald JK, et al. Azathioprine or 6-mercaptopurine for maintenance of remission in Crohn's disease. Cochrane Database Syst Rev
11. Timmer A, McDonald JW, Macdonald JK. Azathioprine and 6-mercaptopurine for maintenance of remission in ulcerative colitis. Cochrane Database Syst Rev
12. Cosnes J, Cattan S, Blain A, et al
. Long-term evolution of disease behavior of Crohn's disease. Inflamm Bowel Dis 2002; 8:244–250.
13. Silverberg MS, Satsangi J, Ahmad T, et al
. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 2005; 19(Suppl A):5–36.
14. Wahed M, Louis-Auguste JR, Baxter LM, et al
. Efficacy of methotrexate in Crohn's disease and ulcerative colitis patients unresponsive or intolerant to azathioprine mercaptopurine. AP&T 2009; 30:614–620.
16. Colombel JF, Sandborn WJ, Reinisch W, et al
. Infliximab, azathioprine, or combination therapy for Crohn's disease. N Engl J Med 2010; 362:1383–1395.
17. Summers RW, Switz DM, Sessions JT, et al
. National Cooperative Crohn's Disease Study: results of drug treatment. Gastroenterology 1979; 77:847–869.
18. Sandborn WJ, Tremaine WJ, Wolf DC, et al
. Lack of effect of intravenous administration on time to respond to azathioprine for steroid-treated Crohn's disease. North American Azathioprine Study Group. Gastroenterology 1999; 117:527–535.
19. Lemann M, Mary JY, Duclos B, et al
. Infliximab plus azathioprine for steroid-dependent Crohn's disease patients: a randomized placebo-controlled trial. Gastroenterology 2006; 130:1054–1061.
20. Candy S, Wright J, Gerber M, et al
. A controlled double blind study of azathioprine in the management of Crohn's disease. Gut 1995; 37:674–678.
21. Present DH, Korelitz BI, Wisch N, et al
. Treatment of Crohn's disease with 6-mercaptopurine. A long-term, randomized, double-blind study. N Engl J Med 1980; 302:981–987.
22. Sood A, Kaushal V, Midha V, et al
. The beneficial effect of azathioprine on maintenance of remission in severe ulcerative colitis. J Gastroenterol 2002; 37:270–274.
23. Sood A, Midha V, Sood N, et al
. Role of azathioprine in severe ulcerative colitis: one-year, placebo-controlled, randomized trial. Indian J Gastroenterol 2000; 19:14–16.
24. Hawthorne AB, Logan RF, Hawkey CJ, et al
. Randomised controlled trial of azathioprine withdrawal in ulcerative colitis. BMJ 1992; 305:20–22.
25. Jewell DP, Truelove SC. Azathioprine in ulcerative colitis: final report on controlled therapeutic trial. BMJ 1974; 4:627–630.
26. Verhave M, Winter HS, Grand RJ. Azathioprine in the treatment of children with inflammatory bowel disease. J Pediatr 1990; 117:809–814.
27. Kader HA, Mascarenhas MR, Piccoli DA, et al
. Experiences with 6-mercaptopurine and azathioprine therapy in pediatric patients with severe ulcerative colitis. J Pediatr Gastroenterol Nutr 1999; 28:54–58.
28. Tajiri H, Tomomasa T, Yoden A, et al
. Efficacy and safety of azathioprine and 6-mercaptopurine in Japanese pediatric patients with ulcerative colitis: a survey of the Japanese Society for Pediatric Inflammatory Bowel Disease. Digestion 2008; 77:150–154.
30. Ediger JP, Walker JR, Graff L, et al
. Predictors of medication adherence in inflammatory bowel disease. Am J Gastroenterol 2007; 102:1417–1426.
31. Greenley RN, Stephens M, Doughty A, et al
. Barriers to adherence among adolescents with inflammatory bowel disease. Inflamm Bowel Dis 2010; 16:36–41.
Keywords:Copyright 2011 by ESPGHAN and NASPGHAN
azathioprine; children compared to adults; Crohn disease; phenotype; ulcerative colitis