Practical pharmacogenetics: the cost effectiveness of screening for thiopurine s-methyltransferase polymorphisms in patients with rheumatological conditions treated with azathioprine.

OBJECTIVE: Thiopurine S-methyltransferase (TPMT), which catalyzes the inactivation of azathioprine (AZA), exhibits genetic polymorphism that results in dose related, serious toxicities (mainly hematological cytopenias) in 10-15% of individuals treated with AZA. Polymerase chain reaction (PCR) tests provide a sensitive, specific means of prospectively identifying these patients before AZA therapy and minimizing toxicity through dosage reduction. Our objective was to model the cost effectiveness of the 2 alternative AZA treatment strategies in rheumatologic conditions: (1) utilizing PCR to determine polymorphisms leading to TPMT deficiencies prior to AZA therapy with a reduction in dose; and (2) no testing. The analysis was conducted from a third party payer perspective over one year. METHODS: A decision analytic model was applied to map the costs and outcomes of patients under both strategies. Data applied to the model included the positive and negative predictive values of the PCR, the probabilities of adverse events due to AZA, and the costs associated with their management. Sources of data included published clinical trials, diagnostic test evaluations, surveillance trials, and economic evaluations. RESULTS: Dose related toxicities resulted in AZA discontinuation rates of 10-20%. The usual dosing strategy cost $677 Cdn per patient, whereas the genotype directed dosing strategy cost $663 Cdn per patient. In the genotype dosing strategy, the number needed to treat to avoid one adverse event over 6 months was 20. Thus, the genotype based dosing strategy dominated the usual dosing strategy. One-way sensitivity analyses revealed that the estimates were robust to ranges of +/- 30% for the costs, the properties of the PCR test, and the probability of adverse events. CONCLUSION: The introduction of PCR testing to identify TPMT polymorphisms prior to AZA treatment may represent good value in certain health care settings.

Azathioprine and allopurinol: the price of an avoidable drug interaction.

OBJECTIVE: To report the price of a drug interaction between azathioprine and allopurinol that resulted in pancytopenia in a patient who had undergone a heart transplant. CASE SUMMARY: A 63-year-old white man who received an orthotopic heart transplant in 1987 was hospitalized in June 1991 with a diagnosis of pancytopenia. His immunosuppressive medications on admission included cyclosporine 125 mg bid, azathioprine (AZA) 200 mg/d, and prednisone 2.5 mg/5 mg every other day. Six weeks prior to admission, the patient's local physician prescribed allopurinol for left wrist pain suspected to be gout. It was determined that the pancytopenia was caused by the drug interaction between AZA and allopurinol, both of which were withheld on admission. During hospitalization, the patient's white blood cell count dropped to 1.1 x 10(3)/mm3 with an absolute neutrophil count of less than 0.5 x 10(3)/mm3, a platelet count of less than 20 x 10(3)/mm3, and a hemoglobin of 3.7 g/dL. Four units of packed red blood cells were transfused and regramostim (GM-CSF) therapy was begun on hospital day 3 to speed the marrow recovery process. The patient was discharged on hospital day 8 and AZA, which had been withheld since admission, was restarted. The dosage was titrated to 200 mg/d over the following 2 weeks. The price of this patient's hospital stay was $13,042. DISCUSSION: Not included in this price was the effect this drug interaction had on the patient's quality of life. Even after discharge from the hospital, it was estimated that it would take up to 3 months for the patient to fully recover his previous level of strength and functional capability. This interaction between AZA and allopurinol could easily have been avoided. Both the physician and the pharmacist missed this well-documented and potentially life-threatening drug interaction. Also, the patient failed to notify the transplant team when allopurinol was prescribed by his local physician. The importance of patient responsibility for medication therapy must be stressed to help avoid unnecessary drug interactions. CONCLUSIONS: Undetected drug interactions can be life-threatening to patients as well as costly to the healthcare system. Drug interactions also can have a profound negative effect on the patients' quality of life, the price of which cannot be measured in dollars alone. It is vital that the physician, pharmacist, and patient work together to optimize therapeutic outcomes and avoid unnecessary drug interactions.