Genotype-guided dosing of warfarin.

Vitamin K antagonists (VKAs) have been used to treat thromboembolic disease for over 60 years, and they continue to be the most commonly prescribed anticoagulants worldwide. However, VKAs such as warfarin have a narrow therapeutic index, and the mean daily dose of warfarin varies widely from patient to patient. Underdosing can leave patients undertreated for their thromboembolic diseases; overdosing can result in bleeding. As such, patients are forced to undergo frequent blood testing, and dose adjustments are made on the basis of the resulting international normalized ratio (INR) values. To achieve a therapeutic INR, which is defined as a value between 2.0 and 3.0 for conditions such as atrial fibrillation and venous thromboembolism, the daily dosing of warfarin can range over 20fold, from 1 to over 20 mg (1 ). The variability in warfarin dosing depends on several factors, including demographics, the environment, and genetics. Specifically, variants in the genes that encode an enzyme responsible for the metabolism of warfarin [cytochrome P450, family 2, subfamily C, polypeptide 9 (CYP2C9)] and the molecular target of warfarin [vitamin K epoxide reductase complex, subunit 1 (VKORC1)] influence warfarin dosing. The Food and Drug Administration has noted these findings, and the warfarin label recommends that if a patient’s genotype is known, then this information should be considered when initiating therapy. Nonetheless, debate continues about the utility of routine genetic testing in this setting, a topic that has been addressed by several studies, including the EUropean Pharmacogenetics of AntiCoagulant Therapy (EUPACT) and Clarification of Optimal Anticoagulation through Genetics (COAG) trials (2, 3 ). The EU-PACT trial enrolled 455 patients initiating warfarin treatment, of whom approximately 72% had atrial fibrillation and 28% venous thromboembolism. In this open-label study, patients were randomized to either warfarin dosing based on an algorithm that included genetic information or a warfarin loading dose consistent with the local standard of care. Genotype information was available at the time that the therapies were started. The study concluded that the primary endpoint of the time in therapeutic INR range between 2.0 and 3.0 during the first 12 weeks of warfarin therapy was significantly higher among the genotype-guided group than the standard-of-care group (67.4% vs 60.3%, P 0.001), and the genotype-guided group also had fewer supratherapeutic values and a shorter time to achieve a therapeutic INR. A second contemporaneous trial conducted in the US, COAG, included 1015 patients requiring warfarin therapy; 22% of the patients had atrial fibrillation, 58% venous thromboembolism or pulmonary embolus, and 20% a combination or another indication. In total, 67% of the population had inpatient initiation of warfarin treatment. Patients were randomly assigned in a double-blind fashion to receive warfarin dosing using either a clinical algorithm or one that incorporated both clinical and genetic factors. Genotype data were available for 45% of the participants before the first dose of warfarin and 94% before the second dose. The primary endpoint of the trial was time in therapeutic INR range between 2.0 and 3.0 during the first 4 weeks of warfarin therapy. The study concluded that the addition of genetics to a warfarin-dosing algorithm based on clinical factors did not improve coagulation control in the first 4 weeks (45.2% vs 45.4%, P 0.91). Why did EU-PACT and COAG generate different conclusions? First, the warfarin dosing strategies varied in the 2 trials. COAG assessed whether an algorithm that included genetics in addition to clinical variables could outperform an algorithm with clinical variables alone. Ultimately, the impact of genotype on dosing in this situation was small. For example, the mean (SD) difference between the dose calculated for patients without genotype data on day 1 vs the dose they would have received if the genotype data were available was 0.1 (0.4) mg per day. It follows that there would be minimal impact of genotyping on coagulation control. In contrast, EU-PACT compared a genetic-guided dosing algorithm to a standard, fixed loading-dose regimen. In this case, patients in the genotype-guided group achieved a stable warfarin dose more quickly 1 Thrombolysis in Myocardial Infarction Study Group, Cardiovascular Division, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA. * Address correspondence to this author at: Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115. E-mail [email protected]. Received March 10, 2014; accepted March 20, 2014. Previously published online at DOI: 10.1373/clinchem.2013.220004 © 2014 American Association for Clinical Chemistry 2 Nonstandard abbreviations: VKA, Vitamin K antagonists; INR, international normalized ratio; EU-PACT, EUropean Pharmacogenetics of AntiCoagulant Therapy; COAG, Clarification of Optimal Anticoagulation through Genetics. 3 Human genes: CYP2C9, cytochrome P450, family 2, subfamily C, polypeptide 9; VKORC1, vitamin K epoxide reductase complex, subunit 1. Clinical Chemistry 60:7 920–922 (2014) Perspective