Correspondence
Genetics of Warfarin Response
N Engl J Med 2008; 358:2741-2744June 19, 2008
- Article
To the Editor:
In the study reported by Schwarz et al. (March 6 issue),1 specific genetic variants related to warfarin metabolism reduced the time to the first therapeutic and supratherapeutic international normalized ratio (INR) tests. A pharmacogenetic-guided dosing algorithm for warfarin,2 however, did not increase the time in the therapeutic range or reduce the proportion of supratherapeutic INR values. An exploratory analysis suggested that the algorithm may be useful for genetic subtypes. Although scientific discourse about personalized medicine has focused on pharmacogenomics, this view limits the extent to which discovery can improve the public's health. My research group has shown that discordance with regard to the dose of warfarin (i.e., miscommunication regarding the regimen) is associated with variation in the INR3 and that visual aids improve concordance.4 A communication-based algorithm reduced the time to the first therapeutic INR5 and increased the time in the therapeutic range for the subgroup with discordance, suggesting that decision making and communication intensity should be tailored according to the degree of discordance. These disparate but complementary approaches to a common clinical problem highlight the need to expand the paradigm of personalized medicine (and associated funding streams) to include disciplines at the distal end of the translational research process.
5 ReferencesDean Schillinger, M.D.
University of California at San Francisco, San Francisco, CA 94143
dschillinger@medsfgh.ucsf. edu 1
Schwarz UI ,Ritchie MD ,Bradford Y , et al. Genetic determinants of response to warfarin during initial anticoagulation. N Engl J Med 2008;358:999-1008
Full Text | Web of Science | Medline2
Anderson JL ,Horne BD ,Stevens SM , et al. Randomized trial of genotype-guided versus standard warfarin dosing in patients initiating oral anticoagulation. Circulation 2007;116:2563-2570
CrossRef | Web of Science | Medline3
Schillinger D ,Wang F ,Rodriguez M , et al. The importance of establishing regimen concordance in preventing medication errors in anticoagulant care. J Health Commun 2006;11:555-567
CrossRef | Web of Science | Medline4
Schillinger D ,Machtinger EL ,Wang F ,Palacios J ,Rodriguez M ,Bindman A . Language, literacy, and communication regarding medication in an anticoagulation clinic: a comparison of verbal vs. visual assessment. J Health Commun 2006;11:651-664[Erratum, J Health Commun 2006;11:811.]
CrossRef | Web of Science | Medline5
Machtinger EL ,Wang F ,Chen LL ,Rodriguez M ,Wu S ,Schillinger D . A visual medication schedule to improve anticoagulation control: a randomized, controlled trial. Jt Comm J Qual Patient Saf 2007;33:625-635
Medline
To the Editor:
Schwarz et al. report that the vitamin K epoxide reductase (VKORC1) haplotype was more strongly associated with genetic variability than the cytochrome P-450 2C9 enzyme (CYP2C9) genotype in determining the early INR response to warfarin therapy in adults. This information and available computer-based, clinical pharmacogenetic programs for warfarin dosing ignore the implications for pediatric patients. Because of the intrinsic developmental differences in the coagulation systems of adults and children,1,2 guidelines for antithrombotic therapy in adults cannot be extrapolated to therapy in children and adolescents.
Although several studies of warfarin have been performed in children, there is a paucity of data about how developmental expression affects the pharmacokinetics and pharmacodynamics of warfarin. This is especially true with regard to VKORC1, for which the patterns of developmental expression are not known. In order to provide children and adolescents who need warfarin therapy with the same pharmacogenomic advances that are being integrated into adult care, it is imperative that pediatric studies be completed to develop and validate a dosing nomogram that considers the effect of both development and concomitant treatment on the disposition and action of warfarin in children and adolescents.
2 ReferencesKathleen A. Neville, M.D.
Brian M. Wicklund, M.D.C.M., M.P.H.
Gregory L. Kearns, Pharm.D., Ph.D.
Children's Mercy Hospital, Kansas City, MO 640791
Andrew M ,Vegh P ,Johnston M ,Bowker J ,Ofosu F ,Mitchell L . Maturation of the hemostatic system during childhood. Blood 1992;80:1998-2005
Web of Science | Medline2
Sosothikul D ,Seksarn P ,Lusher JM . Pediatric reference values for molecular markers in hemostasis. J Pediatr Hematol Oncol 2007;29:19-22
CrossRef | Web of Science | Medline
To the Editor:
The study reported by Schwarz et al. shows that genetic polymorphisms influence the initial response to warfarin therapy, suggesting that pharmacogenomic testing could guide warfarin dosing. However, a recent randomized, controlled trial1 concluded that although fewer dose changes were required to achieve the target INR with genotype-guided dosing than with standard warfarin dosing, the primary end point (the percent out-of-range INR) did not differ. A possible explanation for this finding is that the effect of warfarin is also influenced by the patient's age, body weight, smoking status, use or nonuse of concomitant drugs that modify CYP2C9 enzyme activity, and dietary vitamin K intake. Therefore, although genetic variations explain interindividual variations, intraindividual variations in response to warfarin may depend largely on acquired factors.2
Schwarz et al. found that the INR at the end of the first week correlated with VKORC1 genotypes. Conversely, one could conclude that measuring the INR at 1 week obviates the need to analyze VKORC1 polymorphisms. Phenotypic variables such as the INR and plasma concentrations of warfarin and 7-hydroxy warfarin may account for inherited as well as acquired factors affecting the warfarin dose–response relationship better than pharmacogenomic approaches do.3,4
4 ReferencesUday P. Kulkarni, M.B., B.S.
Dilip R. Karnad, M.D.
Nithya J. Gogtay, M.D., D.N.B.
King Edward Memorial Hospital, Mumbai 400012, India
upk30@rediffmail.com 1
Anderson JL ,Horne BD ,Stevens SM , et al. Randomized trial of genotype-guided versus standard warfarin dosing in patients initiating oral anticoagulation. Circulation 2007;116:2563-2570
CrossRef | Web of Science | Medline2
Wadelius M ,Pirmohamed M . Pharmacogenetics of warfarin: current status and future challenges. Pharmacogenomics J 2007;7:99-111
CrossRef | Web of Science | Medline3
Millican EA ,Lenzini PA ,Milligan PE , et al. Genetic-based dosing in orthopedic patients beginning warfarin therapy. Blood 2007;110:1511-1515
CrossRef | Web of Science | Medline4
Kulkarni UP ,Swar BD ,Karnad DR , et al. A pilot study of the association of pharmacokinetic and pharmacodynamic parameters of warfarin with the dose in patients on long-term anticoagulation. Br J Clin Pharmacol 2008;65:787-790
CrossRef | Web of Science | Medline
To the Editor:
The finding of Schwarz et al. that the anticoagulant response during the initiation of warfarin therapy is more strongly associated with VKORC1 than with CYP2C9 genetic variants is of uncertain clinical impact. In patients who are sensitive to warfarin, determination of the response to therapy (e.g., a high INR measured 2 to 3 days after the initial dose) enables the practitioner to adjust subsequent doses. What is clinically important but still unknown is whether genetic variants associated with warfarin hypersensitivity confer a higher risk of bleeding complications and whether this risk is diminished by pretreatment genotyping. Although genetic tests for warfarin sensitivity are licensed by the Food and Drug Administration, we think that their use is still clinically unjustified, costly, and associated with the emotional implications of genetic testing. The National Heart, Lung, and Blood Institute is planning a randomized trial1,2 to establish whether initial warfarin dosing based on genotyping is superior to standard prescription strategies. Unfortunately, the primary outcomes of this trial are not bleeding and thrombosis, but surrogates such as the percentage of time that the INR is within the therapeutic range.
2 ReferencesPier M. Mannucci, M.D.
Marta Spreafico, Ph.D.
Flora Peyvandi, M.D., Ph.D.
Maggiore University Hospital, 20122 Milan, Italy
pmmannucci@libero.it 1
Department of Health and Human Services. Federal business opportunities. (Accessed May 30, 2008, at http://www.fbo.gov/?s=opportunity&mode=form&id=0aaf5fd39b29a0e8f175fa0122d519f3&tab=core&_cview=1.)
2
Shurin SB ,Nabel EG . Pharmacogenomics -- ready for prime time? N Engl J Med 2008;358:1061-1063
Full Text | Web of Science | Medline
Author/Editor Response
Schillinger suggests that careful communication with patients improves the outcomes of warfarin therapy, and Kulkarni and colleagues suggest that a phenotypic outcome such as the INR will capture variability in the patient's response to therapy due to acquired factors. Indeed, the concept of personalized medicine should not be applied exclusively to the genetic components of variation in response to a drug. Factors such as adherence to the medication regimen, diet, and monitoring, as well as expert adjustment of the warfarin dose according to the INR, are important determinants of variation in the response to therapy. In addition, the patient's age affects his or her response to warfarin, and as Neville and colleagues point out, pediatric studies will be of interest.
We showed that the VKORC1 haplotype contributes to the initial variability in the INR response; Kulkarni and colleagues ask the obverse question: does consideration of the initial variability in the INR response eliminate the contribution of the VKORC1 haplotype? As we discussed in our article, in 92 patients with orthopedic conditions who received their initial dose of warfarin tailored according to clinical and genetic factors, much of the contribution of VKORC1 to the prediction of warfarin sensitivity was captured by the INR after three doses of warfarin.1 However, in that statistical model, the initial dose of warfarin (determined in part by the VKORC1 genotype in some patients) did contribute significantly. In another study of 132 hospitalized patients, the INR on day 4 accounted for 31.0% of the variability in the warfarin dose on day 14, and the CYP2C9 and VKORC1 genotypes accounted for 6.5% of the variability in the warfarin dose.2 Thus, additional studies will be required to define the independent contributions of the genotype and phenotype for guiding warfarin dosing.
Mannucci and colleagues suggest that genotyping patients who are treated with warfarin is not clinically justified until randomized trials have shown that it decreases the outcome of bleeding or thrombosis. Evidence this strong would be ideal; however, many recommendations regarding the modification of the dose of a drug for an individual patient are based on pharmacokinetic or pharmacodynamic observations. For example, few (if any) randomized trials with clinical end points have provided data on the adjustment of drug doses in renal or hepatic dysfunction, despite the widespread and appropriate use of this information in patient care. Thus, there are advocates both for the introduction of genotyping into warfarin-dosing algorithms and for randomized clinical trials to compare genotyping with the standard of care, and both approaches are being studied.
2 ReferencesC. Michael Stein, M.B., Ch.B.
Dan M. Roden, M.D.
Vanderbilt University School of Medicine, Nashville, TN 37232-6602
michael.stein@vanderbilt. edu Ute I. Schwarz, M.D.
University of Western Ontario, London, ON N6A 5A5, Canada1
Millican EA ,Lenzini PA ,Milligan PE , et al. Genetic-based dosing in orthopedic patients beginning warfarin therapy. Blood 2007;110:1511-1515
CrossRef | Web of Science | Medline2
Michaud V ,Vanier MC , Brouillette D et al. Combination of phenotype assessments and CYP2C9-VKORC1 polymorphisms in the determination of warfarin dose requirements in heavily medicated patients. Clin Pharmacol Ther 2008;83:740-748
CrossRef | Web of Science | Medline
