#NASPGHAN17 Is it time to stop using thiopurine therapy?

This blog entry has abbreviated/summarized this presentation. Though not intentional, some important material is likely to have been omitted; in addition, transcription errors are possible as well.

Safety in Pediatric IBD Therapy: Is it time to stop using thiopurines?

Jeffrey Hyams  Connecticut Children’s Medical Center

Key points from this lecture:

  • Dr. Hyams:  “There are better options than thiopurines in 2017 due to infrequent but serious risks”
  • The DEVEVOP study showed that anti-TNF agents did NOT increase the risk of lymphoma or hemophagocytic lymphohistiocytosis (HLH).  In contrast, these risks do occur with thiopurines –this is infrequent but remains significant.
  • Therapeutic drug monitoring may obviate the need for combination/dual therapy which has been shown to improve response rates to anti-TNF agents; methotrexate may work for combination therapy and may be safer than thiopurines
  • If a thiopurine is used as part of combination therapy, short duration (~6 months) is likely to have low risks
  • In addition to Dr. Hyams, Dr. Baldassano, in his discussion of treat to target (discussed in subsequent post), echoed the sentiment that he no longer recommends thiopurine therapy

Dr. Hyams slides list some of the relative risks of thiopurine therapy.  To understand these risks, the absolute risk is probably more helpful.

My take: This lecture did not focus on the main benefit of thiopurines which is its use in combination therapy. Many experts consider combination therapy to be the standard of care for adults with Crohn’s disease.  The advantages of combination therapy are mainly due to improved durability of anti-TNF therapy and lower antidrug antibodies.  How this benefit stacks up against the risks discussed in this lecture and whether this benefit can be supplanted by the use of therapeutic drug monitoring is uncertain.


Related blog posts:

Disclaimer: These blog posts are for educational purposes only. Specific dosing of medications (along with potential adverse effects) should be confirmed by prescribing physician.  This content is not a substitute for medical advice, diagnosis or treatment provided by a qualified healthcare provider. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a condition.


6-Thioguanine Levels in Autoimmune Hepatitis

A recent retrospective study (MA Sheiko et al JPGN 2017; 65: 80-5) examines the issue of azathioprine (AZA) metabolites and outcomes in pediatric autoimmune hepatitis (AIH).

Study characteristics:

  • 66 children
  • Mean age of diagnosis 9.6 years
  • Mean follow-up 2.9 years
  • Study period 2002-2013

Key findings:

  • 79% achieved biochemical remission (defined as ALT ≤50 U/L); mean time was 6.2 months
  • 6% required liver transplantation
  • 18% were weaned off immunosuppression and remained in remission
  • 6-thioguanine (6-TGN) levels ranging from 50 to 250 (pmol/8 x 10 to 8th red blood cell count) were associated with biochemical remission

Our study suggests that AZA dosing of approximately 1.2 to 1.6 mg/kg/day will achieve 6-TGN levels of 50 to 250 pmol, which is sufficient to maintain biochemical remission in the majority of patients.

This is significantly lower than dosing recommended for inflammatory bowel disease (recommended levels 250-450). The associated editorial (pg 2-3, N Kerkar) cautions that while “lower levels are sufficient for maintaining biochemical remission…higher levels, similar to that used in IBD, are required for inducing remission.”

My take: Lower doses of azathioprine are likely to maintain biochemical remission and cause fewer side effects.  Metabolite levels can be helpful to assure reasonable levels of 6-TGN and to assure medication adherence.

Related blog entries:

Disclaimer: These blog posts are for educational purposes only. Specific dosing of medications (along with potential adverse effects) should be confirmed by prescribing physician.  This content is not a substitute for medical advice, diagnosis or treatment provided by a qualified healthcare provider. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a condition.

Shem Creek, SC

What I did not know …a few items

  1. 6-Mercaptopurine (6-MP) often can be used when patients are intolerant of azathioprineS Hubener et al. Clin Gastroenterol Hepatol 2016; 14: 445-53.  This retrospective study showed that 15 of 20 patients with autoimmune hepatitis and prior azathioprine intolerance responded to 6-MP.  This is somewhat unexpected as azathioprine is metabolized into 6-MP.  However, rather than 6-thiouracil, the “imidazol component of azathioprine, which is cleaved off…might trigger adverse reactions.”  Another artical on thiopurines (Aliment Pharmacol Ther 2016; 43: 863-883) (thanks to Ben God for this reference) provides a thorough review of the pharmacogenetics and pharmocokinetics of these medications.  While this review reinforces the recommendation to check TPMT before treatment, it notes that only a small proportion of thiopurine toxicity is related to deficient TPMT activity.
  2. There is no formal validated or consensus definitions of mild, moderate, or severe IBD. L Peyrin-Biroulet et al. Clin Gastroenterol Hepatol 2016; 14: 348-54.  While the Lemann index measures the cumulative structural bowel disease, the authors propose criteria which involves three areas of severity: impact of the disease on the patient (eg. clinical symptoms), inflammatory burden (eg. biomarkers, mucosal disease, disease extent), and disease course (eg. structural disease, intestinal resection, perianal disease, extraintestinal manifestations)
  3. Fundic gland polyps (often associated with proton pump inhibitor therapy) are not premalignant lesions. There is an “inverse correlation between the FGPs and gastric neoplasia.” S Varghese et al. Gastroenterol Hepatol; 2016; 12: 153-4.
  4. Parents of newborns do not know how to use car seats.  BD Hoffman et al (J Pediatr 2016; 171: 48-54) showed that 95% of car seats were misused (291 families).  Serious misuse was present in 91%.


Related blog posts: Lemann index: Short Takes on IBD Articles | gutsandgrowth

Gibbs Gardens has >20 million daffodils

Gibbs Gardens has >20 million daffodils



The results of the “UC SUCCESS” trial show that combination therapy with infliximab and azathioprine is more effective than either medication as monotherapy in ulcerative colitis (UC) (Gastroenterol 2014; 146: 392-400). This study findings are similar to the SONIC trial in Crohn’s disease (CD).

Study Design: randomized, double-blind trial with evaluation at 16 weeks with a total of 239 patients.  In patients assigned to infliximab (IFX) alone, they were given daily oral placebo pills. In patients with azathioprine monotherapy (AZA), dosed at 2.5 mg/kg/day, they also received placebo infusions.  Patients had moderate to severe UC as defined by Mayo scores at baseline and had not responded adequately to a course of corticosteroids.  All patients were naive to tumor necrosis factor α antagonists (anti-TNFα).  Mean age was approximately 40 years.


  • IFX/AZA had a 39.7% corticosteroid-free remission at week 16 compared with 22.1% with IFX monotherapy and 23.8% with AZA monotherapy.
  • Mucosal healing at week 16 was evident in 62.8% of combination group compared with 54.6% IFX monotherapy and 36.8% with AZA monotherapy.
  • Serious adverse events were noted more frequently in the AZA monotherapy group, though this did not reach statistical significance.
  • A subset of patients had antibodies to infliximab (ATIs) measured.  ATI-positivity was more common with IFX monotherapy (19%, 7 of 37) than for IFX/AZA combination (3%, 1 of 31)

While this study indicates that for moderate to severe UC combination therapy with IFX/AZA was superior in this age group, there were several limitations.  Given the slow onset of action of azathioprine, more patients may have responded to this therapy if longer treatment duration was studied.

Take-home message: Combination therapy for UC, like CD, is more effective.  In this small study population, the adverse events were not increased. In the pediatric population, particularly males, the concern for malignancy in patients (especially males) treated with combination therapy may limit the frequency of combination therapy.

Related blog posts:

Other recent IBD articles of interest:

Inflamm Bowel Dis 2014; 20: 291-300. “Malignancy and Mortality in Pediatric Patients with Inflammatory Bowel Disease”  This article presented the results of a survey of 20 European countries and Israel.  Key finding: 18 cases of cancer and 31 deaths in 44 children. 5 of the deaths were due to cancer; the most common cause of mortality was infectious (n=14).  In this cohort, all HSTCL or EBV-positive lymphomas were treated with thiopurine monotherapy.

Inflamm Bowel Dis 2014; 20: 196-212.  “Opportunistic Infections Due to Inflammatory Bowel Disease Therapy”  This review article covers a broad range of pathogens and includes recommendations for prophylaxis and treatment (Table 3).  In addition the authors  provide suggestions for checking for several infections prior to treatment and vaccinations.


AGA Guidelines for the Use of Thiopurines and Anti-TNF Agents for Crohn’s

The link (from KT Park’s twitter feed): gastrojournal.org/article/S0016-5085(13)01521-7/fulltext …

Some of the key points/recommendations for adults with Crohn’s disease:

  • In clinical practice, CD of moderate severity is defined as disease requiring systemic corticosteroids for symptom control.

For Induction of Remission:

  • We Suggest Against Using Thiopurine Monotherapy to Induce Remission in Patients With Moderately Severe CD (Weak Recommendation, Moderate-Quality Evidence)
  • We Suggest Against Using Methotrexate to Induce Remission in Patients With Moderately Severe CD (Weak Recommendation, Low-Quality Evidence)
  • We Recommend Using Anti–TNF-α Drugs to Induce Remission in Patients With Moderately Severe CD (Strong Recommendation, Moderate-Quality Evidence)
  • We Suggest Using Anti–TNF-α Drugs in Combination With Thiopurines Over Anti–TNF-α Drug Monotherapy to Induce Remission in Patients Who Have Moderately Severe CD (Weak Recommendation, Moderate-Quality Evidence)

Maintenance of Remission:

  • We Recommend Using Thiopurines Over No Immunomodulator Therapy to Maintain a Corticosteroid-Induced Remission in Patients With CD (Strong Recommendation, Moderate-Quality Evidence)
  • We Suggest Using Methotrexate Over No Immunomodulator Therapy to Maintain Corticosteroid-Induced Remission in Patients With CD (Weak Recommendation, Low-Quality Evidence)
  • We Recommend Using Anti–TNF-α Drugs Over No Anti–TNF-α Drugs to Maintain Corticosteroid- or Anti–TNF-α—Induced Remission in Patients With CD (Strong Recommendation, High-Quality Evidence)
  • We Make No Recommendation for or Against the Combination of an Anti–TNF-α Drug and a Thiopurine Versus an Anti–TNF-α Drug Alone to Maintain Remission Induced by a Combination of These Drugs in Patients With CD (No Recommendation, Low-Quality Evidence)

Related blog posts:


Adult versus Pediatric Data: Autoimmune Hepatitis

Recent data in adults indicate that budesonide may be more effective than prednisone for treating autoimmune hepatitis (AIH) (Gastroenterol 2010; 139: 1198-206).  Is this true in pediatrics? A small study and an associated editorial indicate that budesonide may be inferior (J Pediatr 2013; 163: 1347-53 & editorial 1246).

The study consisted of 46 pediatric patients with AIH (35 females) who were enrolled in a prospective double-blind, randomized, active-controlled trial with budesonide at a dose of 3 mg TID or prednisone (starting at 40 mg and tapered to 10 mg); all patients received concomitant treatment with azathioprine (1-2 mg/kg/day).  After the initial 6 months, a further 6-month open-label treatment with budesonide (n=42) followed.  Approximately 70% of the patients had type 1 AIH.


  • Normalization of aminotransferases occurred in only 16% of budesonide group and 15% of prednisone group after 6 months.
  • At 12 months, 46% of those on budesonide had biochemical remission

The editorial explains why these results are unlikely to affect current management and provides succinct summary of AIH management.

Key points:

  • “The juvenile form of AIH is particularly aggressive…between 40% and 80% of children are reported to have cirrhosis at diagnosis”
  • Standard treatment for juvenile AIH: Prednisone 2 mg/kg/day (max 60 mg) which is tapered over 4-8 weeks with the decline of aminotransferase levels to a maintenance dose of 2.5-5 mg/day.  80% of the improvement (in ALT values) occurs within the first two months.  Typically, azathioprine is added after some improvement in aminotransferases.  In the absence of toxicity, the dose is increased to 2-2.5 mg/kg/day.
  • Remission is defined as complete normalization of aminotransferases along with normalization of serum immunoglobulin G levles, negative or very low autoantibody titer, and histologic resolution of inflammation.
  • Histologic remission lags biochemical response, “though 95% of patients have a marked histologic improvement after a mean duration of 4 years of effective therapy.”
  • Budesonide is not used in the presence of cirrhosis
  • Problems with current study are detailed in the editorial.  The design likely contributed to remission rates which were significantly lower than reported with standard prednisone/azathioprine.

Related posts:

Disclaimer: These blog posts are for educational purposes only. Specific dosing of medications (along with potential adverse effects) and specific medical management interventions should be confirmed by prescribing physician.  Application of the information in a particular situation remains the professional responsibility of the practitioner.


Thiopurines = Low Efficacy for Crohn’s

The enthusiasm for thiopurine therapy for Crohn’s disease (CD) had already dropped a lot before two pivotal articles were recently published that confirmed the modest efficacy:

  • Cosnes J, et alGastroenterol 2013; 145: 758-65
  • Panes J, et alGastroenterol 2013; 145: 766-74
  • Gastroenterol 2013; 145: 714-16 (editorial)

The Cosnes study (from the GETAID group) reports an open-label randomized trial in 147 adult patients with newly diagnosed CD (from 2005–>2010) and risk factors for disabling disease who were recruited from 24 French centers.  Risk factors for disabling disease:

  • Age <40 years
  • Active perianal lesions
  • Corticosteroid use within 3 months of diagnosis

The characteristics and Paris classification are detailed in the paper’s Table 1. Patients were divided into early azathioprine or “conventional” treatment. Patient’s were followed for 3 years.  Azathioprine was dosed at 2.5 mg/kg/day.  The primary endpoint was the proportion of trimesters spent in corticosteroid-free and anti-tumor necrosis factor (TNF)-free remission.


  • 67% of azathioprine group achieved the primary endpoint compared with 56% in the conventional group.  The difference in achieving the primary endpoint was not statistically significant between the two groups.  Also, 41 (61%) of the conventional group were placed on azathioprine (mean time 11 months after enrollment).
  • The azathioprine group patients were more likely to not have perianal surgery (96%) compared with the conventional group patients (82%).
  • Adverse events included pancreatitis in 7 (10%) of azathioprine group compared with 1 (1%) of conventional group.  Elevated liver function tests were noted in 3 (4%) compared with 1 (1%) respectively.  No cases of neutropenia were noted in early azathioprine group.

The latter finding is interesting especially as the authors did not check thiopurine methyltransferase (TPMT) assays in a systematic manner.

Panes et al (for the AZTEC study group) performed a prospective double-blind trial of adult patients with a recent CD diagnosis (<8 weeks).  This study enrolled 131 patients from 31 centers in Spain.  68 received azathioprine (2.5 mg/kg/day) and 63 received placebo.


  • After 76 weeks of treatment, 30 (44.1%) azathioprine patients and 23 (36.5%) placebo-treated patients were in sustained corticosteroid-free remission (P= .48).
  • Relapse rates were lower in azathioprine group compared with placebo: 11.8% vs 30.2%.
  • Serious adverse effects were more frequent in the azathioprine group compared with placebo: 20.6% vs. 11.1% (P= .16).  In the azathioprine group, 7 (10%) developed pancreatitis, 16 (24%) developed leukopenia, 9 (13%) developed anemia, and 9 (13%) developed abnormal liver function tests.  Infections were more common in the placebo-treated group (24%) compared with 12% in the azathioprine group

The accompanying editorial should be mandatory reading for all health care providers who help manage inflammatory bowel disease (IBD) patients.  The editorial traces how thiopurines (azathioprine and 6-mercaptopurine) became an accepted cornerstone of IBD treatment.  In adults, after initial disappointing results from Summers et al (Gastroenterol 1979; 77: 847-69), efficacy was demonstrated in a seminal study by Present et al (NEJM 1980; 302: 981-87).  However, the authors note that a recent Cochrane review reported that “thiopurines are not effective for induction of remission, but are effective for maintenance of remission.”

The editorial notes that a high degree of efficacy was demonstrated from a small but influential pediatric study of 55 patients.  After a high remission rate (89%) for all patients, this study showed that among those in remission, “1 patient (4%) in the 6MP group had a relapse within 180 days of achieving remission, compared with 7 patients (28%) in the placebo group.”

The editorial draws the following conclusions from the current studies:

  • “The remaining indications for primary therapy with thiopurines are maintenance of steroid-induced remission/steroid sparing in patients with CD that is not newly diagnosed, and prevention of postoperative recurrence.”
  • “The strongest indication for thiopurines may be as part of combination therapy.”
  • “If TNF antagonists had a similar low cost as generic thiopurines, there would likely be little debate regarding an evolution toward treatment of CD with TNF antagonists, either as monotherapy or ideally as combination therapy.  Currently, the annual costs of TNF antagonists is 10-20 times that of thiopurines.”
  • Because of the difference in cost.., “the use of thiopurines in CD is likely to persist, despite the shrinking number of indications, the modest effect size, and the suboptimal safety profile.”

Related blog posts: