Role of Thiopurine Methyltransferase Activity in the Safety and Efficacy of Azathioprine in the Treatment of Pemphigus Vulgaris

Objective  To evaluate the relationship between thiopurine methyltransferase (TPMT) activity and the safety and efficacy of azathioprine sodium in the treatment of pemphigus vulgaris.

Design  Cross-sectional study.

Setting  Referral university hospital for autoimmune blistering diseases.

Participants  One hundred thirty-nine patients with pemphigus vulgaris treated with azathioprine.

Intervention  The TPMT activity in red blood cells was measured using high-performance liquid chromatography.

Main Outcome Measure  Severe adverse effects were defined as those judged serious enough that azathioprine therapy be discontinued in 139 patients treated with azathioprine. To evaluate the relationship of clinical response and TPMT concentration in 52 patients who had been treated with a combination of prednisolone and azathioprine only for at least 1 year were included in the study, and the clinical response was considered favorable if there was no recurrence of pemphigus vulgaris in the first year of treatment.

Results  The median activity of TPMT was 44.7 ng/mL/h (interquartile range, 28.7 ng/mL/h). Eleven patients (7.9%) had low TPMT activity (TPMT-HL), 127 patients (91.4%) had normal TPMT activity (TPMT-HH), and 1 patient (0.7%) had supranormal enzyme activity; TPMT activity was noted in all patients. Serious adverse effects occurred in 14 patients (10.1%). There was no relationship between development of adverse effects and TPMT activity (P = .29). Eleven patients with low TMPT activity had been treated with azathioprine for a mean (SD) of 10.2 (4.1) months. Only 1 patient exhibited serious adverse effects. The TPMT enzyme activity was not different in 28 patients with unfavorable clinical response compared with 24 patients with favorable clinical response (P = .09).

Conclusion  Larger prospective studies are needed to determine the clinical relevance of TPMT activity and to determine accurate azathioprine dosing guidelines based on TPMT activity.

Pemphigus vulgaris (PV) is a rare but serious autoimmune disease that causes blistering of the skin and oral cavity.¹ It is the most common form of autoimmune blistering disease of the skin in Iran, with incidence estimated at approximately 1/100 000 population per year and prevalence of 30/100 000.² Systemic corticosteroids are the mainstay of treatment for pemphigus, but the high and prolonged doses needed produce adverse effects that can be serious. Adjuvant immunosuppression therapies are, therefore, widely used.¹ Azathioprine sodium is a purine antagonist commonly used as an adjuvant immunosuppressive agent in treating PV.³ Mercaptopurine is the active compound of azathioprine, and thiopurine methyltransferase (TPMT) is the principal inactivation enzyme for this cytotoxic metabolite in hematopoietic tissues.⁴ The TPMT catalyzes S-methylation of 6-mercaptopurine to biologically inactive metabolites. The enzyme activity of TPMT varies according to a common allelic polymorphism. About 88.6% of the healthy population is homozygous for an allele for high TPMT activity (TPMT-HH). A heterozygous TPMT genotype is associated with 50% enzyme activity (TPMT-HL with intermediate activity), and approximately 11.1% of the population is affected. Homozygotes, with 2 TPMT mutant alleles, have deficient (undetectable) TPMT activity (TPMT-LL), which affects only 0.3% of the population.^5,6 The TPMT activity is usually determined in peripheral red blood cell (RBC) lysates, and this activity correlates well with enzyme activity in the liver.⁷ Some studies have shown that the efficacy and adverse effects (primarily myelosuppression) of azathioprine are dependent on the activity of TPMT. That is, the toxic effect of azathioprine is more common in individuals with TPMT-LL because of accumulation of toxic metabolites in the tissues. In contrast, the drug is rapidly metabolized in patients with TPMT-HH; thus, they cannot have a favorable response to azathioprine therapy.^5,8 By assessing TPMT activity before starting azathioprine therapy, a toxic effect can be anticipated and suboptimal doses of azathioprine can be averted in individuals with very high TPMT activity.⁹ Despite the common use of azathioprine in the treatment of PV, to our knowledge, only 1 study has addressed the role of the TPMT assay in the management of pemphigus with this drug.¹⁰ Nevertheless, many have recommended routine measurements of TPMT activity before the administration of azathioprine to patients with PV.^9,11,12 The purpose of the present study was to determine whether inherited variations of TPMT activity might be a factor influencing response to therapy and adverse effects in patients with PV receiving azathioprine therapy.

One hundred thirty-nine patients with PV attending the Pemphigus Clinic at Razi Skin Hospital, Tehran, Iran, who had previously received azathioprine (68 patients) or were currently receiving azathioprine (71 patients) were considered consecutively for inclusion in the study. The diagnosis of PV was confirmed at histopathologic and direct immunofluorescence studies. Patients were excluded if reliable data for clinical response or adverse effects could not be obtained or if they had other variants of pemphigus. The usual treatment regimen for PV in this clinic consists of prednisolone, 2 mg/kg/d, combined with azathioprine sodium, 2 to 3 mg/kg/d, tapered gradually after control of the disease. After obtaining written consent, 10 mL of blood was obtained from each patient at the clinic visit for TPMT measurement. The TPMT activity in RBC lysates was measured using high-performance liquid chromatography according to the method of Khalil et al¹³ and was categorized into 4 groups: (1) undetectable, compatible with TPMT-LL; (2) intermediate (<20 ng/mL/h of RBCs), compatible with TPMT-HL; (3) normal (20-130 ng/mL/h of RBCs), compatible with TPMT-HH; and (4) very high (>130 ng/mL/h of RBCs), compatible with supranormal enzyme activity. For this study, adverse effects were defined as those judged severe enough by the attending physician to discontinue azathioprine therapy. To assess the relationship of clinical response and TPMT level, 52 patients receiving the combined treatment of prednisolone and azathioprine and no other adjuvant therapy for at least 1 year were included. The standard treatment regimen for PV in our clinic is to start with prednisolone, 2 mg/kg/d, plus azathioprine, 2 to 3 mg/kg/d, and taper the dosage according to patient clinical response. Clinical response was considered favorable if the patient had no recurrence of disease in the first year of treatment and unfavorable if the patient had recurrence of disease with development of new lesions leading to an increase in the dosage of prednisolone. In addition, duration needed to taper prednisolone from 2 mg/kg/d at the start of therapy to 30 mg/d was considered another variable for evaluation of the efficacy of azathioprine therapy. To determine clinical response, patient medical records were reviewed by a dermatologist (N.G.) who was unaware of the results of TPMT activity assay.

Statistical analysis was performed with commercially available software (SPSS version 13.0; SPSS, Inc, Chicago, Illinois). Because of nonnormal distribution of TMPT activity (demonstrated using the Kolmogorov-Smirnov test with Lillieferos significance correction), nonparametric tests (Mann-Whitney test) and linear regression analysis were used. For analyzing quantitative data, variables were described as mean (SD), median, and interquartile range. P < .05 was considered statistically significant.

The mean (SD) age of the patients with PV was 42.3 (14.1) years, and duration of disease was 3.4 (3.2) years. Eighty-eight patients (63.3%) were women. The Figure shows the TMPT activity in patients with PV. The median activity of TPMT was 44.7 ng/mL/h (interquartile range, 28.7 ng/mL/h; minimum, 3.8 ng/mL/h; maximum, 257.8 ng/mL/h). Eleven patients (7.9%) had low TPMT activity (TPMT-HL), 127 patients (91.4%) had normal TPMT activity (TPMT-HH), and 1 patient (0.7%) had supranormal enzyme activity. None of the patients had TPMT-LL. No significant difference was noted between TPMT activity in male and female patients (P = .35), and no relationship was found between TPMT activity and age (r = .02; P = .84). In patients with low TPMT activity (n = 11, including 8 women), the mean (SD) duration of azathioprine treatment was 10.2 (4.1) months (median, 12 months). Only 1 of these patients exhibited adverse effects severe enough that azathioprine therapy was discontinued. This patient was a 27-year-old man treated with azathioprine sodium, 2 mg/kg/d, who demonstrated a more than 3-fold increase in the upper limit of normal liver enzyme levels after 10 days. The TPMT activity was 18.5 ng/mL/h in this patient. One patient, a 35-year-old woman with TMPT activity of 15.5 ng/mL/h, developed anemia, leukopenia, and elevated liver enzyme levels after 2½ months of treatment with azathioprine sodium, 2 mg/kg/d, which resolved after the drug dose was decreased to 1 mg/kg/d. Adverse effects leading to discontinuation of azathioprine therapy occurred in 14 of 139 patients (10.1%) treated with azathioprine. These adverse effects included hepatotoxic reaction in 11 patients and leukopenia, anemia, and pancreatitis in 1 patient each. Demographic data, reason for withdrawal, and TPMT activity in these patients are given in Table 1. Only 1 patient had low TPMT activity. No relationship was noted between the development of these adverse effects and TPMT activity (P = .29). The relationship between TPMT activity and efficacy of azathioprine therapy was evaluated in 52 patients treated for at least 1 year with a combination of azathioprine plus prednisolone therapy only. The demographic data, clinical response, time to tapering of prednisolone from 2 mg/kg/d to 30 mg/d, and TPMT activity in these 52 patients are given in Table 2. There was no correlation between clinical response and TPMT activity (P = .09). In addition, no relationship was observed between enzyme activity and the time needed to taper prednisolone from 2 mg/kg/d at the start of therapy to 30 mg/d (r = .06; P = .72).

The overall prevalence of lower than normal TPMT activity (TPMT-HL) in this cross-sectional study was approximately 7.9%, and no patients were completely enzyme deficient. Only 1 patient with low TPMT activity exhibited a more than 3-fold increase in liver enzyme levels leading to discontinuation of azathioprine therapy, and myelosuppression did not develop in any patients. The TPMT activity did not correlate with age or sex. In addition, there was no correlation between TPMT activity and either adverse effects of azathioprine therapy or patient clinical response. Patients with low TPMT activity might be at increased risk of adverse effects (primarily myelosuppression) after treatment with azathioprine. In contrast, individuals with supranormal TPMT activity should, in theory, respond less well to standard dosages of azathioprine, because the metabolism of azathioprine is preferentially shunted to the production of inactive metabolites, whereas the level of active metabolites is low or they are not produced at all. Therefore, patients with higher than normal TPMT activity might be at risk of poor clinical response to standard dosages of azathioprine, and higher dosages of azathioprine may be necessary to prevent therapeutic failure.

Several studies in patients with inflammatory bowel disease have shown the relationship between TPMT activity and efficacy and adverse effects of azathioprine^14-16; however, others have failed to confirm this association.^17-19 Although azathioprine is widely used in the treatment of various skin diseases, few studies have examined the effect of TPMT polymorphism in patients with dermatologic diseases. Snow and Gibson¹⁰ evaluated 28 patients with dermatologic diseases, including 5 patients with PV, who had received azathioprine therapy. They observed that patients with higher TPMT activity had poorer response to azathioprine therapy. In a study of atopic dermatitis, Murphy and Atherton²⁰ could not show any significant difference between TPMT activity in good responders to azathioprine therapy compared with poor responders.

To our knowledge, the present study is the largest to evaluate the relationship between TPMT activity and response to azathioprine therapy in patients with dermatologic diseases, in particular, those with PV. We could not show any relationship between TPMT activity and either toxic effects or efficacy of azathioprine therapy. Azathioprine therapy was discontinued in 14 of 139 patients (10.1%), primarily because of the hepatotoxic effects of the drug. The level of enzyme activity in this group of patients was not different from that in those who did not experience severe adverse events with azathioprine therapy. Only 1 of these patients had lower than normal TPMT (TPMT-HL), and the remaining patients exhibited normal enzyme activity. On the other hand, among 11 patients with low TPMT activity (TPMT-HL), only 1 developed a hepatotoxic reaction severe enough that treatment had to be discontinued. Our results show that inherited variations in TPMT activity may not be the only determinant of individual differences in the response to azathioprine therapy in patients with PV. Although this test is expensive and not readily available, it is still cost-effective.²¹ We failed to confirm this test as a good predictor of effectiveness of azathioprine therapy in patients with PV and to determine the appropriate azathioprine dosage. An alternative to obtaining TPMT activity is to start therapy with lower dosages of azathioprine and check the complete blood cell count and the results of liver function tests at lower dosages. Future larger prospective studies are needed to determine the clinical relevance of the TPMT test.

Correspondence: Alireza Firooz, MD, Center for Research & Training in Skin Diseases & Leprosy, Tehran University of Medical Sciences, 79 Taleqani Ave, Tehran 14166, Iran (firozali@sina.tums.ac.ir).

Accepted for Publication: October 5, 2007.

Author Contributions: Dr Firooz had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Firooz, Ghandi, Hallaji, Chams-Davatchi, and Karbakhsh Davari. Acquisition of data: Ghandi, Hallaji, and Valikhani. Analysis and interpretation of data: Firooz, Ghandi, and Karbakhsh Davari. Drafting of the manuscript: Firooz, Ghandi, and Karbakhsh Davari. Critical revision of the manuscript for important intellectual content: Firooz, Ghandi, Hallaji, Chams-Davatchi, Valikhani, and Karbakhsh Davari. Statistical analysis: Karbakhsh Davari. Obtained funding: Firooz and Ghandi. Administrative, technical, and material support: Ghandi, Hallaji, Chams-Davatchi, and Valikhani. Study supervision: Firooz, Ghandi, Hallaji, Chams-Davatchi, and Valikhani.

Financial Disclosure: None reported.

Funding/Support: This study was supported by research grant 423/791 from the Center for Research & Training in Skin Diseases & Leprosy (Dr Firooz).

Role of the Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.

Previous Presentations: This study was presented in part at the 15th Congress of the European Academy of Dermatology and Venereology; October 4, 2006; Rhodes, Greece; and a part of the article was published in Farsi in Iranian J Dermatol. 2006;9:204-210.

Additional Contributions: Mehdi Hedayati, PhD, Shaheed Beheshti Medical Sciences University, Tehran, assisted in the measurement of TPMT concentrations.