Counterpoint: pharmacogenetic-based initial dosing of warfarin: not ready for prime time.

Availability of the human genome sequence offers the promise of personalized medicine through pharmacogenomics. Warfarin, a member of the coumarin family of oral anticoagulants used to prevent and treat thromboembolic disorders and one of the top 20 prescribed medications in the US, is an ideal drug for applying the principles of pharmacogenetics. Warfarin inhibits reduction of vitamin K epoxide by the vitamin K epoxide reductase complex, subunit 1 (VKORC1) enzyme, causing hypogammacarboxylation of vitamin K– dependent coagulation factors and an acquired coagulopathy. Warfarin therapy is monitored with the international normalized ratio (INR) derived from the prothrombin time. The INR therapeutic range is narrow, and the maintenance warfarin dose required to produce a therapeutic INR for an individual is both unpredictable and widely variable, leading to bleeding complications, especially during the initiation period when dose adjustments are made by trial and error (1 ). During the past 12 years, discoveries regarding the molecular basis of warfarin pharmacokinetics and pharmacodyanmics have been combined with clinical and demographic information from stably anticoagulated patients to generate many dosing algorithms. Up to 54% of the interpatient variation in therapeutic warfarin dose can be accounted for by the combination of patient age, body size, target INR, and use of amiodaron with the genotypes for 2 single-nucleotide polymorphisms (SNPs) in cytochrome 2C9 that reduce warfarin metabolism and 1 from a group of SNPs in the vitamin K epoxide reductase complex, subunit 1 (VKORC1) gene in high linkage disequilibrium and associated with increased sensitivity to warfarin (2 ). Few pharmacogenetic algorithms have been validated, however, and all are less accurate when used to predict therapeutic warfarin doses in African Americans (3 ), most likely because of currently unknown genetic mechanisms that affect warfarin sensitivity. Ongoing molecular and translation research has identified additional genetic variants affecting warfarin dosing (4 ), but to date they are rare or have modest impact on therapeutic dose prediction. In August 2007, the US Food and Drug Administration (FDA) added information to warfarin and Coumadin® package insert regarding lowering of therapeutic warfarin doses in cases involving the cytochrome P450, family 2, subfamily C, polypeptide 9 (CYP2C9) gene CYP2CP*2/*3 SNPs and the VKORC1 SNP, but the FDA did not recommend or require genotyping be performed before initiation of warfarin therapy. In response, to enable detection of warfarin pharmacogenetically relevant SNPs, the molecular diagnostic industry has developed reagents for real-time PCR instruments, fluorescent plate readers, and reagent-instrument platforms. To date, the FDA has licensed 5 tests, and others are currently under evaluation. In 2007, the College of American Pathologists added CYP2CP*2/*3 and VKORC1 SNPs to its pharmacogenetics proficiency testing panel. It would appear that there is considerable momentum from some stakeholders to adopt pharmacogenetic-based initial dosing of warfarin into routine clinical practice. It would be reasonable to expect that more accurate, pharmacogenetic-based initial dosing of warfarin would reduce the risk of serious bleeding and thrombotic complications. A few retrospective cohort studies support an association between bleeding complications and CYP2C9 *2/*3 (5–7 ) but not VKORC1 SNPs associated with increased warfarin sensitivity (7 ). Only 3 small prospective randomized control trials have compared pharmacogenetic-based initiation of warfarin to empiric dosing (8 –10 ) (Table 1), and the findings are not convincing. The studies used different algorithms and nomograms for genetic and control dosing arms, respectively, and measured different primary outcomes, including feasibility of rapid genotyping, time to first therapeutic INR, or percentage of outof-range INRs. Only 1 trial identified a significant 1 Department of Pathology and Immunology, Division of Laboratory and Genomic Medicine, Washington University School of Medicine, St. Louis, MO. * Address correspondence to this author at: Department of Pathology and Immunology, Division of Laboratory and Genomic Medicine, Washington University School of Medicine, 660 S. Euclid Avenue, St. Louis, MO 63110. Fax 314-362-1461; e-mail [email protected]. Received January 9, 2009; accepted January 16, 2009. Previously published online at DOI: 10.1373/clinchem.2008.115972 2 Nonstandard abbreviations: VKORC1, vitamin K epoxide by vitamin K epoxide reductase complex, subunit 1; INR, international normalized ratio; SNP: single nucleotide polymorphism; FDA, US Food and Drug Administration. 3 Human genes: VKORC1, vitamin K epoxide reductase complex, subunit 1; CYP2C9, cytochrome P450, family 2, subfamily C, polypeptide 9. Clinical Chemistry 55:4 712–714 (2009) Point/Counterpoint