For decades, vitamin K antagonists (VKAs) such as warfarin have been the standard oral anticoagulant used to prevent or treat thromboembolic disease. Although the VKAs are effective, their narrow therapeutic window, combined with the significant inter-individual variability in dose response, makes this class of medications challenging to use. Once the decision has been made to initiate treatment with a VKA, the therapeutic focus is on maximizing the percentage of time in the therapeutic range (TTR), as a higher percentage of TTR correlates with better clinical outcomes.1, 2
In the last 10 years, scientists have identified common genetic polymorphisms that impact the pharmacodynamic effect of VKAs. Specifically, testing for single-nucleotide polymorphisms in the genes VKORC1 (Vitamin K epoxide reductase complex subunit 1) and/or CYP2C9 (cytochrome P450 2C9) improves the accuracy of warfarin dose prediction, raising the hope that genotype-guided dosing might increase the safety and efficacy of coumarin therapy. Until now, however, sufficiently powered clinical studies designed to test whether genotype-guided dosing improves TTR or clinical outcomes have been lacking.
Table. Randomized Controlled Trials of Pharmacogenetic Testing
| . | Number of Patients . | Follow-Up(weeks) . | Clinical Factors Used to Estimate Dose for Control Group . | TTR (%) . | ||
|---|---|---|---|---|---|---|
| PG | Control | P value | ||||
| Kimmel, et al. | 1015 | 4 | Yes | 45.2 | 45.4 | 0.91 |
| Verhoef, et al. | 548 | 12 | Yes | 61.6 | 60.2 | 0.52 |
| Pirmohamed, et al. | 455 | 12 | No* | 67.4 | 60.3 | <0.001 |
| . | Number of Patients . | Follow-Up(weeks) . | Clinical Factors Used to Estimate Dose for Control Group . | TTR (%) . | ||
|---|---|---|---|---|---|---|
| PG | Control | P value | ||||
| Kimmel, et al. | 1015 | 4 | Yes | 45.2 | 45.4 | 0.91 |
| Verhoef, et al. | 548 | 12 | Yes | 61.6 | 60.2 | 0.52 |
| Pirmohamed, et al. | 455 | 12 | No* | 67.4 | 60.3 | <0.001 |
PG - group of patients whose dose was estimated with knowledge or their VKORC1 and CYP2C9 genotype.
TTR - time in therapeutic range, the primary endpoint for all three studies.
*All patients were given 10 mg on day 1 and 5 mg on days 2 and 3 except for patients > 75 years old who received 5 mg daily on days 1-3.
Recently, Stephen E. Kimmel et al. of the COAG group and Talitha I. Verhoef et al. of the EU-PACT group compared dosing algorithms based on both clinical and genotypic variables with algorithms based solely on clinical variables. Patients in the COAG trial had INRs checked twice a week for two weeks, then weekly for two weeks. Patients in the EU-PACT trial had a similar evaluation timeline with the addition of INR checks at eight and 12 weeks. Neither study found a significant difference between genotype-guided and control dosing groups for the primary endpoint of TTR during warfarin initiation (Table).
Munir Pirmohamed et al. used a slightly different trial design, comparing a dosing algorithm based on genotype and clinical variables with a three-day standardized loading dose regimen in which all control-group patients received an identical dose for the first three days (except those older than 75 years who received a smaller dose on day 1). In this trial, the genotype-guided group had a modestly greater TTR (67.4% vs. 60.3%, P<0.001). The experimental group also experienced fewer supratherapeutic INRs and, on average, reached a first therapeutic INR more quickly (21 vs. 29 days, P<0.001).
In Brief
Two of these important clinical trials suggest that genotype-guided dosing of VKAs will not improve upon traditional VKA management. The statistically significant TTR difference seen in the Pirmohammed study is intriguing, but difficult to interpret, because the initial dose for patients in the experimental group was based on both their genotype and clinical characteristics, whereas dosing for the control group was pre-defined and independent not only of pharmacogenetic but also of most clinical information.
Together, these three randomized, controlled studies support the use of clinically based dose calculators (e.g., www.warfarindosing.org) to optimize TTR during VKA initiation; however, they also indicate that, in practice, adding pharmacogenetic testing to clinical dose estimation plus frequent INR measurement will not effect a better clinical outcome.
References
Competing Interests
Dr. Samuelson and Dr. Garcia indicated no relevant conflicts of interest.
