Cost-Utility Analysis of Pharmacogenetic Testing Based on CYP2C19 or CYP2D6 in Major Depressive Disorder: Assessing the Drivers of Different Cost-Effectiveness Levels from an Italian Societal Perspective

Abstract

Background and objectives: Major depressive disorder (MDD) is a common and severe psychiatric disorder that has enormous economical and societal costs. As pharmacogenetics is one of the key tools of precision psychiatry, we analyze the cost-utility of test screening of CYP2C19 and CYP2D6 for patients suffering from major depressive disorder (MDD) and try to understand the main drivers that influence the cost-utility.

Methods: We developed two pharmacoeconomic nonhomogeneous Markov models to test the cost-utility, from an Italian societal perspective, of pharmacogenetic testing genetic to characterize the metabolizing profiles of cytochrome P450 (CYP) 2C19 and CYP2D6 in a hypothetical case study of patients suffering from major depressive disorder (MDD). The model considers different scenarios of adjustment of antidepressant treatment according to the patient's metabolizing profile or treatment over a period of 18 weeks. The uncertainty of model parameters is tested through both a probabilistic sensitivity analysis and a one-way deterministic sensitivity analysis, and these results are used in a post-hoc analysis to understand the main drivers of three alternative cost-effectiveness levels ("poor," "standard," and "high"). These drivers are first evaluated from an exploratory multidimensional perspective and next from a predictive perspective as the probability that a patient belongs to a specific cost-effectiveness level is estimated on the basis of a restricted set of parameters used in the original pharmacoeconomic model.

Results: The models for CYP2C19 and CYP2D6 indicate that screening has an incremental cost-effectiveness ratio of 60,000€ and 47,000€ per quality-adjusted life year (QALY), respectively. The probabilistic sensitivity analysis shows that the treatments are cost-effective for a 75,000€ willingness to pay (WTP) threshold in 58% and 63% of the Monte Carlo replications, respectively. The post-hoc analysis highlights the factors that allow us to clearly discriminates poor cost-effectiveness from high cost-effectiveness scenarios and demonstrates that it is possible to predict with reasonable accuracy the cost-effectiveness of a genetic test and the associated therapeutic pattern.

Conclusions: Our findings suggest that screenings for both CYP2C19 and CYP2D6 enzymes for patients with MDD are cost-effective for a WTP threshold of 75,000€ per QALY, and provide relevant suggestions about the most important aspects to be further explored in clinical studies aimed at addressing the cost-effectiveness of genetic testing for patients diagnosed with MDD.

Fig. 1

The proposed nonhomogeneous Markov model. PM poor metabolizer, IM intermediate metabolizer, EM extensive metabolizer, UM ultra-metabolizer

Fig. 2

Deterministic sensitivity analysis results. DSA deterministic sensitivity analysis, Prob. probability, PM poor metabolizer, IM intermediate metabolizer, EM extensive metabolizer, UM ultra-metabolizer, PGx pharmacogenetically screened cohort, NoPgx standard guidelines cohort, QALY quality-adjusted life year, ICER incremental cost-effectiveness ratio, WTP willingness to pay, CYP cytochrome P450

Fig. 3

Probabilistic sensitivity analysis: cost-effectiveness plane. PGx pharmacogenetically screened cohort, NoPgx standard guidelines cohort, QALY quality-adjusted life year, WTP willingness to pay, CYP cytochrome P450

Fig. 4

Probabilistic sensitivity analysis: acceptability curve. PGx pharmacogenetically screened cohort, NoPgx standard guidelines cohort, QALY quality-adjusted life year

Fig. 5

Canonical discriminant factors for cytochrome P450 (CYP) 2D6: the “poor” cost-effectiveness group is labeled “a,” the “standard” group “b,” and the “high” group “c”. PGx pharmacogenetically screened cohort, NoPgx standard guidelines cohort, Pr_SE probabilities of side effects, Pr_Impr probability of having an improvement (effect), Pr_SWT probability of switch/titration, PM poor metabolizer, IM intermediate metabolizer, EM extensive metabolizer, NoIm Ut utility for no effect

Fig. 6

Decision boundary obtained by linear discriminant factors for CYP2D6: the “poor” cost-effectiveness group is in the red area (circular points), the “standard” group in the green area (triangular points), and the “high” group in the blue area (squared points). LD1 and LD2 are the two linear discriminant functions. CYP cytochrome P450