OBJECTIVES : The field of pharmacogenomics (PGx) studies how inherited genetic variations in individuals affect drug absorption, distribution and metabolism in a patient body. PGx panel testing can potentially help improve efficiency and accuracy in individualizing therapy. However, the cost-effectiveness of PGx panel testing strategies regarding the timing of the tests has not been assessed. This study aims to compare the cost-effectiveness between preemptive PGx testing, reactive PGx testing and usual care (no PGx testing) in cardiovascular disease management.

METHODS : We developed a decision analytic model that composed of a short-term decision tree and life-time Markov model to compare the three testing strategies. The study sample included a hypothetical cohort of 10,000 patients that represents a target US population of age 45 years or older. The testing panel assumed to include tests for the following genes and corresponding drug choices: CYP2C19 (clopidogrel vs. ticagrelor), CYP2C9/VKORC1 (warfarin vs. novel oral anticoagulants), and SLCO1B1 (statins vs. PCSK9). Scenario and probabilistic sensitivity analyses were performed to assess the impact of age, sex, race, risk level, and time frame duration. The study reported incremental cost-effectiveness ratios (ICERs) as the main outcome while costs, quality-adjusted life years (QALYs), incremental costs, and incremental QALYs were also reported for the base-case and sensitivity analyses.

RESULTS : The results suggested that preemptive testing was cost-effective compared to usual care (ICER $71,540/QALY) at the willingness-to-pay (WTP) of $100,000/QALY while reactive testing was not (ICER $155,870/QALY). Preemptive testing gained a patient an additional 0.74 QALYs compared to usual care, while reactive testing lead to a 0.12 QALYs increase. The cost-effectiveness was not sensitive to age, sex, race and risk-level, but was sensitive to longer time frame.

CONCLUSIONS : This study concludes that preemptive PGx panel testing for CVD diseases management was cost-effective and should be implemented in a system wide manner for the target population.