Cost-effectiveness of Cyp2d6 genotyping in older depressed patients, starting with nortriptyline therapy

Objectives: Genotyping for the cytochrome P450-2D6 has the potency to predict differences in metabolism of nortriptyline. This information could optimize treatment. We explored if possible benefits could outweigh genotyping costs for Dutch depressed patients in clinical psychiatry. Methods: First, a decision-tree was created to model the first weeks of nortriptyline therapy. In the model, costs of hospitalization, therapeutic drug monitoring, and drug costs were captured. Based on the patients genetics, patients were distributed among three health states: correctly, sub-, or supra-therapeutically dosed. Utilities for each of these health states and at different points in time were obtained from an expert opinion (nine clinicians). Second, an improvement in sub or supra-therapeutically dosed patients to correctly dosed patients, was simulated, assuming genotyping would prevent under or overdosing. In the base case the improvement was 36%. In addition, we assumed genotyping could reduce hospitalization days with a maximum of 3.7 days (average: 28.6 days). Results from the model without genotyping were compared with the genotyping model. In a scenario analyses we varied the effects of genotyping to reach cost-effectiveness at € 20 000/quality adjusted life year (QALY) or € 50 000/ QALY. In a univariate sensitivity analysis, effects of lowering genotyping costs were examined. A probabilistic sensitivity analysis (PSA) was performed to investigate influence of parameter uncertainty. Results: In the base case, the incremental cost-effectiveness ratio (ICER) was € 32 697/QALY. For an ICER of € 20 000/QALY, a genotyping facilitated improvement of 45% was needed and for € 50 000/QALY this was 27%. Lowering the genotype price to € 162 made genotyping cost-saving. Results of the PSA indicated a probability of 0.95 for a willingness-to-pay threshold of € 46000/ QALY. Conclusions: Genotyping could be cost-effective and even be cost-saving when genotyping costs drops. However, there is a need for more clinical evidence to support assumptions made in this model.