Modeling the Outcome of Systematic TPMT Genotyping or Phenotyping Before Azathioprine Prescription: A Cost-Effectiveness Analysis

Abstract

Background: Thiopurine S-methyltransferase (TPMT) testing, either by genotyping or phenotyping, can reduce the incidence of adverse severe myelotoxicity episodes induced by azathioprine. The comparative cost-effectiveness of TPMT genotyping and phenotyping are not known.

Objective: Our aim was to assess the cost-effectiveness of phenotyping-based dosing of TPMT activity, genotyping-based screening and no screening (reference) for patients treated with azathioprine.

Methods: A decision tree was built to compare the conventional weight-based dosing strategy with phenotyping and with genotyping using a micro-simulation model of patients with inflammatory bowel disease from the perspective of the French health care system. The time horizon was set up as 1 year. Only direct medical costs were used. Data used were obtained from previous reports, except for screening test and admission costs, which were from real cases. The main outcome was the cost-effectiveness ratios, with an effectiveness criterion of one averted severe myelotoxicity episode.

Results: The total expected cost of the no screening strategy was €409/patient, the total expected cost of the phenotyping strategy was €427/patient, and the total expected cost of the genotyping strategy was €476/patient. The incremental cost-effectiveness ratio was €2602/severe myelotoxicity averted in using the phenotyping strategy, and €11,244/severe myelotoxicity averted in the genotyping strategy compared to the no screening strategy. At prevalence rates of severe myelotoxicity > 1%, phenotyping dominated genotyping and conventional strategies.

Conclusion: The phenotype-based strategy to screen for TPMT deficiency dominates (cheaper and more effective) the genotype-based screening strategy in France. Phenotype-based screening dominates no screening in populations with a prevalence of severe myelosuppression due to azathioprine of > 1%.