A Model Based Cost-Effectiveness Analysis of Routine Genotyping for CYP2D6 among Older, Depressed Inpatients Starting Nortriptyline Pharmacotherapy

Objective

Genotyping for CYP2D6 has the potential to predict differences in metabolism of nortriptyline. This information could optimize pharmacotherapy. We determined the costs and effects of routine genotyping for old aged Dutch depressed inpatients.

Methods

With a decision-tree, we modelled the first 12 weeks of nortriptyline therapy. Direct costs of genotyping, hospitalization, therapeutic drug monitoring and drugs were included. Based on genotype, patients could be correctly, sub-, or supratherapeutically dosed. Improvement from sub-or supratherapeutically dosed patients to correctly dosed patients was simulated, assuming that genotyping would prevent under-or overdosing of patients. In the base case, this improvement was assumed to be 35%. A probabilistic sensitivity analysis (PSA) was performed to determine uncertainty around the incremental cost-effectiveness ratio (ICER).

Results

In the base case analysis, costs for genotyping were assumed sic200 per test with a corresponding ICER at sic1 333 000 per QALY. To reach a sic50 000 per QALY cut-off, genotyping costs should be decreased towards sic40 per test. At genotyping test costs <sic35 per test, genotyping was dominant. At test costs of sic17 per test there was a 95% probability that genotyping was cost-effective at sic50 000 per QALY.

Conclusions

CYP2D6 genotyping was not cost-effective at current genotyping costs at a sic50 000 per QALY threshold, however at test costs below sic40, genotyping could be costs-effective.