SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Background
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Authorship
  8. Acknowledgement
  9. References

Background

Ten percent of patients with autoimmune hepatitis (AIH) are nonresponsive or intolerant to thiopurine therapy. A skewed metabolism, leading to the preferential generation of (hepato)toxic thiopurine metabolites (6-MMPs) instead of the metabolic active 6-tioguanine (thioguanine) nucleotides (6-TGNs), may explain this unfavourable outcome. Co-administration of allopurinol to low-dose thiopurine therapy may effectively revert this deviant metabolism, as has been shown in inflammatory bowel disease.

Aim

To describe the effect of adding allopurinol to low-dose thiopurine therapy in patients with AIH with intolerance or nonresponse to normal thiopurine dosages due to a skewed metabolism.

Methods

We describe the clinical efficacy and tolerability of allopurinol–thiopurine combination therapy with allopurinol 100 mg and low-dose thiopurine (25–33% of the original dosage) in eight AIH patients with a skewed thiopurine metabolism. Patients were switched because of dose-limiting intolerance (n = 3), nonresponse (n = 3) or loss of response (n = 2) to conventional thiopurine treatment.

Results

All eight patients showed biochemical improvement with a reduction in median alanine aminotransferase (ALT) levels of 62 U/L at start to 35 U/L at 1 month (P = 0.03). This clinical benefit was sustained in seven patients. Allopurinol–thiopurine combination therapy effectively bypassed thiopurine side effects in four of five patients. Median 6-tioguanine nucleotides levels increased from 100 to 200 pmol/8 × 108 red blood cells (RBC) at 3 months (P = 0.04). Median 6-MMP levels decreased in all patients from 6090 to 175 pmol/8 × 108 RBC (P = 0.01).

Conclusion

Allopurinol safely and effectively optimises thiopurine therapy in patients with autoimmune hepatitis with intolerance and/or nonresponse due to an unfavourable thiopurine metabolism.


Background

  1. Top of page
  2. Summary
  3. Background
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Authorship
  8. Acknowledgement
  9. References

Autoimmune hepatitis (AIH) is a chronic autoimmune liver disease of unknown aetiology, characterised by hypergammaglobulinemia (IgG), serum autoantibodies and histologically interface hepatitis and plasmacellular infiltrates.[1] Current treatment strategies for AIH consist of an induction course with prednisone and frequently include subsequent addition of azathioprine (AZA) 1–2 mg/kg/day as corticosteroid-sparing maintenance therapy.[1] Unfortunately, in 10% of AIH patients, this therapeutic strategy proves ineffective, due to lack of clinical response or intolerable side effects of AZA.[1, 2] Before exerting its immunosuppressive potential, AZA needs to be metabolised into the pharmacologically active 6-tioguanine (thioguanine) nucleotides (6-TGN). During the complex metabolism, several other (toxic) thiopurine metabolites are produced (Figure 1).[3] The enzyme thiopurine methyltransferase (TPMT) plays a pivotal role in this process as its activity determines the level of generated methylated breakdown products, including 6-methylmercaptopurine (6-MMP).[4] The thiopurine metabolism varies across the population, possibly as the result of different TPMT phenotypes. Consequently, the levels of 6-MMP and 6-TGN vary between individuals.[5] High 6-MMP levels have been associated with the development of hepatotoxicity (especially elevation of transaminases) but also therapeutic failure, mainly due to concomitant lower levels of the biologically active 6-TGN.[6] The co-administration of allopurinol alongside low-dose thiopurine redirects the thiopurine metabolism towards 6-TGN formation instead of 6-MMPs.[7, 8] This strategy has been successfully applied to a subgroup of inflammatory bowel disease (IBD) patients, where nonresponsiveness to or side effects from thiopurine therapy were attributed to high 6-MMP levels.[9] Here, we report the first clinical experience on efficacy and safety of allopurinol salvage therapy in AIH patients.

image

Figure 1. Azathioprine (AZA) is non-enzymatically degraded to mercaptopurine (MP). By several enzymatic steps [including the enzyme hypoxanthine phosphoribosyl transferase (HPRT)], 6-MP is ultimately metabolised via 6-thioinosine-monophosphate (6-TIMP) into the pharmacologically active 6-tioguanine nucleotides [6-TGN: 6-tioguanine-monophosphate (6-TGMP), 6-tioguanine-diphosphate (6-TGDP) and 6-tioguanine-triphosphate (6-TGTP)]. Alternatively, 6-MP and 6-TIMP can be methylated by the enzyme TPMT leading to the formation of 6-methylmercaptopurine (6-MMP) and 6-methylthioinosine-monophosphate (6-MTIMP) respectively. Xanthine oxidase (XO) can inactivate 6-MP by the generation of 6-thiouric-acid (6-TUA).

Download figure to PowerPoint

Materials And Methods

  1. Top of page
  2. Summary
  3. Background
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Authorship
  8. Acknowledgement
  9. References

Patients

Autoimmune hepatitis patients treated with allopurinol combined with thiopurines at the department of Gastroenterology and Hepatology of the VU University Medical Center (tertiary referral centre) were included in this retrospective case-series analysis. Patients were switched to allopurinol–thiopurine combination therapy, if they were either nonresponsive and/or were experiencing dose-limiting side effects from conventional thiopurine [either AZA or mercaptopurine (MP)] therapy. Moreover, thiopurine metabolite measurements had to display preferential 6-MMP formation, arbitrarily defined as 6-MMP levels >5700 pmol/8 × 108 red blood cells (RBC) and/or a 6-MMP:6-TGN metabolite ratio of ≥15.

Nonresponse and intolerance of thiopurine therapy

Nonresponse and loss of response were defined as persistently raised aminotransferases [>upper limit of normal (ULN); Females: 35 IU/L; Males: 45 IU/L] despite previous response to induction therapy and thiopurine therapy, respectively. Dose-limiting intolerance was defined as unbearable side effects related to the administration of thiopurines necessitating dose reduction or discontinuation of therapy.

Thiopurine metabolite measurements

Thiopurine metabolites were measured according to the method described by Dervieux-Boulieu.[10] The 6-TGN concentrations were divided by 2,6 for comparison with the more widely used method described by Lennard.[11, 12].

Follow-up

Patients were treated according to local protocol, receiving allopurinol 100 mg/day with a reduced dose (approximately 25–33% of original thiopurine dosage) of AZA or MP. Patients were seen at regular intervals at the out-patient clinic for clinical and laboratory parameter evaluation. Patient characteristics, reason for cessation of thiopurine monotherapy, allopurinol and thiopurine dose and duration of therapy, potential side effects, thiopurine metabolites, biochemical and haematological parameters were recorded. In case of myelosuppression, thiopurine dosage was reduced.

Statistical analysis

Statistical analysis was performed using PASW Statistics 18 (SPSS Inc., Chicago, IL, USA). Demographic and therapy-specific data are given descriptively. Grouped values were stated as median with range. Statistical testing between groups was performed with the Mann–Whitney U-test or Wilcoxon signed rank test.

Results

  1. Top of page
  2. Summary
  3. Background
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Authorship
  8. Acknowledgement
  9. References

Baseline characteristics

Eight AIH patients (five women) with a median age of 59 years (range: 25–66 years) started with allopurinol–thiopurine combination therapy between February 2011 and October 2012 (Table 1). One patient had an AIH overlap syndrome with primary sclerosing cholangitis (PSC) and one with primary biliary cirrhosis (PBC). The median diagnostic score according to the 1999 International AIH Group criteria was 19 points (range: 7–22).[13] The median time from the time of diagnosis prior to the switch to allopurinol–thiopurine combination therapy was 47 months (range: 17–206 months). At this time, all patients had received prednisone therapy according to standard induction tapering regime, starting with 30 mg/day.[1] The induction therapy was followed by either AZA (n = 4) or MP (n = 4) therapy for a median of 50 months (range: 3–140 months) and 23 months (range: 4–33 months) respectively. Median AZA and MP doses prior to the initiation of allopurinol–thiopurine combination therapy were 125 mg AZA (range: 75–150 mg) for 2 months (range: 2–13 months) and 50 mg MP (range: 37.5–150 mg) for 19 months (range: 3–21 months) respectively (Table 1). All patients remained corticosteroid-dependent with either prednisone (n = 5) or budesonide (n = 1) prior to the initiation of allopurinol–thiopurine combination therapy. Two additional patients (n = 2) used both prednisone and budesonide for a limited period during the transition from prednisone to budesonide prior to initiation of allopurinol–thiopurine combination therapy (Table 1).

Table 1. Patient characteristics and outcome of allopurinol–thiopurine combination therapy
PatientSexAge (years)Diagnosis (IAIHG Score)Prior to combination therapyClinical reason for CTOutcome of combination therapy
Treatment (mg/day)Time on thiopurinea (months)Time on CT (Months)Clinical ResponsePrednisone reduction
  1. AIH, autoimmune hepatitis; ALT, alanine aminotransferase; AZA, azathioprine; B, budesonide; CT, allopurinol–thiopurine combination therapy; IAIHG, international AIH group diagnostic score; F, female; M, male; MP, mercaptopurine; N, no; P, prednisone; PBC, primary biliary cirrhosis; PSC, primary sclerosing cholangitis; Y, yes.

  2. a

     Time on maximum (displayed) dose of thiopurines prior to initiation of allopurinol–thiopurine combination therapy.

  3. b

     Patient 6 discontinued allopurinol–thiopurine combination therapy at 4 months because of side effects.

1F66AIH-PSC (19)P(2,5) + AZA(50)13Thiopurine intolerance8YY
2F66AIH (18)P(10) + MP(50)3Thiopurine intolerance16YY
3M60AIH (14)P(5)  + B(9)  + MP(38)21Thiopurine intolerance15YY
4M58AIH (20)P(12,5)  + AZA(150)2Nonresponse18YY
5F61AIH-PBC (7)P(12.5)  + MP(50)18Loss of response16YY
6bF53AIH (19)B(3)  + AZA(100)2Nonresponse4Y-
7F40AIH (22)P(2,5)  + B(9)  + AZA(150)2Nonresponse7YN
8M25AIH (22)P(15)  + MP(150)21Loss of response11YN

Reasons for initiation of allopurinol–thiopurine combination therapy: nonresponse and intolerance

Five patients were switched to allopurinol–thiopurine combination therapy after a median of 33 months (range: 3–140 months) because of nonresponse (n = 3) or loss of response (n = 2) on final conventional thiopurine dose with raised ALT levels (median: 79 U/L, range 60–111 U/L). Three patients were switched to allopurinol–thiopurine combination therapy after a median of 27 months (range: 4–56 months) due to development of one or more dose-limiting side effects of AZA or MP. These included gastrointestinal complaints (nausea and vomiting; n = 2), arthralgia (n = 1) and headache (n = 1). Two of the previously mentioned nonresponders also reported persistent, but tolerable thiopurine-related side effects of myalgia and arthralgia. Prior to allopurinol–thiopurine combination therapy, patients with nonresponse and loss of response had both higher thiopurine doses (Table 1) and metabolite levels when compared with the intolerance group [6-TGN: 139 vs. 92 pmol/8 × 108 RBC, P = 0.2; 6-MMP: 7430 vs. 1860 pmol/8 × 108 RBC, P = 0.07]. At the start of allopurinol treatment (100 mg per day), the median daily dose of AZA (n = 4) was reduced from 125 mg (range: 75–150 mg) to 25 mg (range: 25–50 mg). The median daily MP dose (n = 4) was reduced from 50 mg (range: 37.5–150 mg) to 25 mg in all four patients.

Clinical and biochemical effectiveness

The median follow-up after initiation of allopurinol–thiopurine combination therapy was 13 months (range: 7–18 months). The treatment regimen was clinically effective in all but one patient with improvement of median baseline ALT levels from 62 U/L (range: 26–111 U/L) to 35 U/L (range: 26–48 U/L) at 1 month (P = 0.03), 24 U/L (range: 26–48 U/L) at 3 months (P = 0.08) and 29 U/L (range: 19–112 U/L) at 6 months (P = 0.03; Figure 2). Four of five nonresponders did sustain the biochemical improvement during follow-up. One patient developed raised aminotransferases at 3 months (ALT: 122 U/L) after initial improvement at 1 month (ALT: 48 U/L). Despite this, allopurinol–thiopurine combination therapy was continued in this patient, showing spontaneous biochemical improvement (ALT: 41 U/L) at 6 months. The group of AIH patients (n = 3) that started allopurinol–thiopurine combination therapy because of intolerance reported disappearance of side effects and had disease follow-up without incidents. After initiation of allopurinol–thiopurine combination therapy, the prednisone dosages could be lowered in five patients, whereas in the other patients, the steroid dosages remained unchanged. No flares of AIH occurred during follow-up necessitating dose escalation of steroids.

image

Figure 2. Overall reduction in median ALT levels (U/L) during 6 months of allopurinol–thiopurine combination therapy.

Download figure to PowerPoint

Tolerability and adverse events during allopurinol–thiopurine combination therapy

Overall, the treatment regimen was well tolerated in all but two patients. One prior nonresponder, with concomitant multiple sclerosis (MS), reported progressive myalgia and worsening complaints of pre-existent hyperesthesia and numbness of hands and feet 4 months after commencement of combination therapy. She associated these complaints with the novel treatment regimen and stopped allopurinol–thiopurine combination therapy and started taking MP in the original dosage (100 mg/day). Subsequently, her complaints of neuropathy improved. Overall white blood cell counts (WBC) (normal range: 4–10 × 109/L) showed a mild decrease after initiation of allopurinol–thiopurine combination therapy from 6.3 × 109/L (range: 3.6–9.0 × 109/L) to 4.9 × 109/L (range: 4.4–7.6 × 109/L) at 6 months (P = 0.2). One patient developed a mild pancytopenia [haemoglobin: 7.1 mmol/L (normal range: 8.5–11.0 mmol/L), white blood cell count: 2.9 × 109/L and thrombocytes: 145 × 109/L (normal range: 150–400 × 109/L)] at 13 months of therapy with 6-TGN levels of 196 pmol/8 × 108 RBC. It should be noted that this patient was also using hydroxychloroquine as treatment for Sjögren's disease. As blood cells counts remained stable during 4 months of subsequent follow-up, he continued allopurinol–thiopurine combination therapy without dose adjustment of AZA (75 mg/day) and allopurinol (100 mg/day).

Thiopurine metabolites levels

Median levels of 6-TGN increased from 100 pmol/8 × 108 RBC (range: 50–185 pmol/8 × 108 RBC) at baseline to 200 pmol/8 × 108 RBC (range: 54–265 pmol/8 × 108 RBC) at 3 months (P = 0.04) (Figure 3). Simultaneously, levels of 6-MMP decreased from a median of 6090 pmol/8 × 108 RBC (range: 1700–9000 pmol/8 × 108 RBC) at baseline to 175 pmol/8 × 108 RBC (range: 0–490 pmol/8 × 108 RBC) at 1 month (P = 0.01) (Figure 4). This steep decrease was observed in all patients (Figure 4). The observed median 6-MMP/6-TGN ratio decreased in all patients from 55.7 (range: 18–122) at baseline to 1.3 (range: 0–2) during allopurinol–thiopurine combination therapy (P = 0.01).

image

Figure 3. 6-tioguanine nucleotide levels (pmol/8 × 108 RBC) before and at 12 weeks of allopurinol–thiopurine combination therapy.

Download figure to PowerPoint

image

Figure 4. 6-Methylmercaptopurine levels (pmol/8 × 108 RBC) before and at 12 weeks of allopurinol–thiopurine combination therapy.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Background
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Authorship
  8. Acknowledgement
  9. References

Clinical and biochemical effectiveness

In this study, we demonstrate that in AIH patients suffering from ineffectiveness or intolerance to AZA or MP due to an unfavourable thiopurine metabolism, the combination therapy of low-dose thiopurine combined with allopurinol proved an effective and well-tolerated alternative immunosuppressive maintenance strategy. Despite the heterogeneity and small size of the studied patient group, regarding type and duration of prior thiopurine therapy, the clinical benefit was marked by a sustained reduction and normalization of ALT levels in seven of eight patients. The majority of patients with previous intolerable adverse events during initial thiopurine therapy were able to tolerate allopurinol–thiopurine combination therapy without development of major drug-related side effects. These observations are consistent with reports on allopurinol–thiopurine combination therapy in IBD patients displaying (non)hepatotoxic adverse events.[9]

Adverse events

Two patients developed mild adverse reactions, being neuropathy and myelodepression, during allopurinol–thiopurine combination therapy. Reversible peripheral neuropathy is a rarely reported side effect of allopurinol and although in our patient these complaints may also be attributed to disease activity of MS, this led to early cessation of allopurinol–thiopurine combination therapy.[14] Development of uncomplicated, mild pancytopenia was noted in the second patient after 15 months of follow-up. Despite the overall mild decrease in WBC at 6 months, this case illustrates the need for continued monitoring of blood cell counts in all patients during thiopurine therapy.

Changes in thiopurine metabolites levels

The co-administration of allopurinol in these AIH patients with a deviant thiopurine metabolism led to a steep decrease in 6-MMP levels and a mild increase in 6-TGN levels. A 6-TGN level of 220 pmol/8 × 108 RBC has been reported as the optimal cut-off value as predictor for remission in AIH (sensitivity: 83% specificity: 62%).[15] Although most of our patients did not reach this threshold, the observed promising clinical outcome after switching to allopurinol–thiopurine combination therapy underlines reported associations between inefficacy and/or toxicity on the one hand and elevated 6-MMP levels combined with low 6-TGN levels on the other.[15] The pharmacological explanation of this thiopurine metabolism modulation after co-adminstration of allopurinol is still enigmatic, as the activity of the methylating enzyme TPMT appears to be unaffected by allopurinol.[7]

Implications for second-line therapy

Currently, there is no established second-line maintenance therapy for AIH. Several alternative drugs, such as mycophenolate mofetil (MMF), tacrolimus or cyclosporine, have been studied in small series of patients.[16] Although showing good results (70% remission) in treatment-naïve AIH patients, administration of MMF as salvage therapy led to an improvement of reported side effects in patients with dose-limiting intolerance to AZA, but it was considerably less effective in nonresponders.[17-19] Ciclosporine and tacrolimus both are calcineurin inhibitors, which have primarily been studied as remission induction agents in AIH patients with lack of response or intolerance to prednisone.[16] Recently, it has also been suggested that calcineurin inhibitors might be effective as long-term alternative to thiopurines in cases of refractory disease.[2] Yet, in thiopurine-refractory patients with preferential 6-MMP metabolism, thiopurine–allopurinol combination therapy might prove to be beneficial over the more costly and nephrotoxic calcineurin inhibitors.[20, 21]

In conclusion, allopurinol in combination with low-dose thiopurine might be an effective and relatively safe alternative immunosuppressive strategy for AIH patients failing standard thiopurine therapy due to preferential 6-MMP metabolism. The present report is limited by its small and heterogeneous patient group and therefore larger and controlled studies are needed to confirm the promising outcomes of this combination therapy.

Authorship

  1. Top of page
  2. Summary
  3. Background
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Authorship
  8. Acknowledgement
  9. References

Guarantor of the article: YS de Boer.

Author contributions: YS de Boer performed data collection, analysis and wrote the manuscript. NMF van Gerven performed analysis and critically revised the manuscript. NKH de Boer and C Mulder critically revised the manuscript. G Bouma had the original idea and critically revised the manuscript. CMJ van Nieuwkerk treated the patients at the out-patient clinic and critically revised the manuscript. All authors approved the final version of the manuscript.

References

  1. Top of page
  2. Summary
  3. Background
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Authorship
  8. Acknowledgement
  9. References