Genotype-Guided Use of Oral Antithrombotic Therapy: A Pharmacoeconomic Perspective

Taylor Sweezy; Shaker A Mousa


Personalized Medicine. 2014;11(2):223-235. 

In This Article

Abstract and Introduction


Pharmacogenomics focuses on tailoring therapy to the individual as opposed to the historical model of fitting the individual to the therapy, and it offers the potential to maximize medication efficacy while reducing adverse events. By its very nature, personalized medicine is conducive to a patient-centered care model. Oral antithrombotics as a class could benefit immensely from this type of approach because an imbalance of safety and efficacy in either direction can yield deadly consequences. Since the current healthcare climate in the USA requires thoughtful allocation of resources, pharmacoeconomic analysis has become critical for all stakeholders, and the adoption of new technologies hinges upon economic impact. This article summarizes the current state of genetics in oral antithrombotic therapy, including clinical relevance as well as cost–effectiveness from a US healthcare system perspective, and provides insight into the future of pharmacogenomics in treating and preventing thromboembolic disorders.


This article presents oral antithrombotics used in the prevention and treatment of thromboembolic disorders, such as venous thromboembolism (deep vein thrombosis and pulmonary embolism), stroke and heart attack. Two major subclasses of oral antithrombotics discussed include the oral anticoagulants, which prevent and treat venous clot (mainly fibrin) formation and the oral antiplatelet agents, which prevent and treat arterial blood clot (mainly platelets) formation. Antithrombotics are vital in the treatment and prevention of an array of cardiovascular diseases, yet the individual response varies widely. This introduces new risks due to subtherapeutic or supratherapeutic drug levels, which would impact safety or efficacy. Several factors such as age, race, gender, smoking and diet are known to affect the response to antithrombotic therapy. However, the most elusive factor is genetics, which is responsible for a considerable percent of interindividual variation.

Pharmacogenomics is a relatively new field with the purpose of maximizing safety and efficacy of medication through personalization based on genetics.[1] The uncovering of genetic determinants in oral antithrombotic response variability has led to US FDA warnings and has created a market for novel antithrombotic drugs that do not succumb to genetic variants. The topic has led to controversy as we continue to elucidate the clinical utility of genetic testing in antithrombotic therapy.

The economic impact of genetic dosing strategies is just as important as the clinical consideration. The field of pharmacoeconomics has become a driving force for formulary managers, third-party payers and pharmaceutical companies. A recent review of third-party payers in the USA found that of 65 insurers with publicly available coverage policies, 58 and 57% addressed policies regarding genetic testing for CYP2C9 and VKORC1, respectively, in relation to the oral anticoagulant warfarin. Of the same 65 insurers, 46% addressed policies on CYP2C19 testing in relation to the oral antiplatelet clopidogrel.[2] If genotype-guided antithrombotic therapy has a positive impact on patient outcomes, it will have to be proven cost effective for widespread adoption to occur.

Oral Anticoagulants

Warfarin. Warfarin (Coumadin®, Bristol-Myers Squibb) was the only oral anticoagulant therapy for over 50 years before the recent introduction of the new oral anticoagulants. However, its serious adverse effect profile and necessary close monitoring of the international normalized ratio (INR) led to the development of several novel agents that do not require such stringent monitoring. Mechanistically, warfarin acts by inhibiting the vitamin K conversion cycle, subsequently inhibiting an array of vitamin K-dependent clotting factors in the coagulation cascade.[3] In spite of warfarin's proven efficacy as an anticoagulant, it falls prey to substantial dosing issues due to drug interactions, diet, age and genetic makeup.[3,4] The serious threat of bleeding risk and the need for therapeutic monitoring has made room for new drugs within the class, which do not require routine laboratory monitoring. New oral anticoagulants include dabigatran, rivaroxaban and apixaban. These agents have the benefit of convenience; however, they still carry many of the same risks as warfarin, but perhaps to a lesser extent.

Warfarin is metabolized in the liver and is subject to the catalyzing effects of CYP2C9, which is heavily responsible for converting warfarin into inactive metabolites.[5] As such, genetic polymorphisms in CYP2C9 lead to large variations in dose response (Table 1). Still, CYP2C9 polymorphisms did not account for differences in response to individuals with the same CYP2C9 genotype. It was subsequently discovered that VKORC1 also plays a part in the genetic variability in warfarin therapeutic response (Table 1).[6,7]VKORC1 is responsible for the conversion of vitamin K from its oxidized, inactive form to the reduced, active form.[8] Warfarin inhibits VKORC1, which subsequently inhibits several vitamin K-dependent clotting factors, thereby disrupting normal coagulation.[9] Historically, initiating proper warfarin dosing has been mostly a guessing game, but genetic testing and resulting interpretation could potentially give clinicians a necessary tool for dose selection. In 2007, the FDA modified the warfarin product label to reflect CYP2C9 and VKORC1 variability and recommend using this information when available for dosing purposes.

Dabigatran. Dabigatran (Pradaxa®, Boehringer Ingelheim Pharma GmbH & Co. KG) is an oral direct thrombin inhibitor used in the prevention of stroke and systemic embolism in patients with nonvalvular atrial fibrillation. One of the most touted benefits of dabigatran, and of all of the novel oral anticoagulants, is the lack of need for laboratory testing. Still, major bleeding rates for dabigatran and warfarin were similar in the RE-LY study (hazard ratio [HR]: 0.72; 95% CI: 0.45–1.48),[17] which begs the question as to whether or not genetic testing could affect outcomes for dabigatran. It is a prodrug that is supplied as dabigatran etexilate and converted to active dabigatran by hepatic esterases including CES1.[12,13] Unlike warfarin, the highly polymorphic CYP enzymes are not involved in dabigatran metabolism. Despite this, recent evidence suggests that genetics may play a role in dabigatran response. The ABCB1 gene encodes the P-glycoprotein drug efflux transporter.[12,14] Dabigatran etexilate is an ABCB1 substrate however, and the active dabigatran is not.[13] The effect of ABCB1 influence on dabigatran etexilate absorption and availability is still unclear. Recent data suggest the CES1 SNP rs2244613 minor allele is responsible for lower exposure to dabigatran resulting in a reduced risk of bleeding events, yielding an odds ratio of 0.67 per minor allele (95% CI: 0.55–0.82).[13] This genetic determinant was discovered using a genome-wide association study in a subset of patients from the RE-LY trial.[17] The genetic effect of CES1 SNP rs2244613 was found to be greater than the effect of drug dosage (either 110 or 150 mg) in the parent study, and 32.8% of patients from the RE-LY study had the presence of this polymorphism and did not affect the warfarin arm. This is the first study to suggest a potentially clinically significant genetic determinant in dabigatran therapeutics. These results indicate that genetic testing may eventually be of some utility for dabigatran therapy and illustrate that unknown genetic factors may influence even the novel oral antithrombotics. Subsequent studies are certainly warranted to further evaluate this finding before any clinical relevance can be ascertained.

Rivaroxaban. Rivaroxaban (Xarelto®, Bayer Pharma AG) is a direct factor Xa inhibitor that was approved in 2011 for deep vein thrombosis prophylaxis. The ROCKET AF trial demonstrated that rivaroxaban was noninferior compared with warfarin for prevention of stroke and systemic embolism and was associated with fewer intracranial or fatal bleeds.[18] The pharmacogenetic impact on rivaroxaban is not well understood; however, its metabolism may provide some insight into potential interests for future investigators. Rivaroxaban is metabolized by CYP3A4, CYP2J2 and is a substrate for P-glycoprotein systems.[19] Given that rivaroxaban is a P-glycoprotein substrate, it may be inferred that as with dabigatran, the ABCB1 gene may have a role in rivaroxaban availability and disposition. An ongoing clinical trial, New Oral Anticoagulant Drugs Dabigatran Etexilate and Rivaroxaban: Influence of Genetic Factors in Healthy Volunteers, aims to build upon current knowledge.[20]

Apixaban. Apixaban (Eliquis®, Britsol-Myers Squibb and Pfizer), like rivaroxaban, is a direct factor Xa inhibitor. Apixaban was approved in the USA in 2012 for stroke risk reduction and systemic embolism risk reduction in patients with atrial fibrillation not caused by a heart valve problem. Apixaban was found to be superior to warfarin in prevention of stroke and systemic embolism in atrial fibrillation patients, causing less bleeding events along with a reduction in mortality.[21] Apixaban is metabolized by a combination of CYP3A4, CYP-independent hepatic pathways and renal elimination.[22] It is not believed to inhibit or induce CYP enzymes, making it an unlikely culprit of drug interactions. Furthermore, apixaban is not believed to be affected by genetic differences, but as the newest approved oral anticoagulant, this is not certain. A potential genetic influence on apixaban response is the sulfotransferases (SULTs), which are responsible for conjugating apixaban into O-demethyl-apixaban, a major metabolite. The two SULTs most involved in apixaban conjugation are SULT1A1 and SULT1A2.[23] No current evidence suggests that the SULTs are implicated in the pharmacogenetic response of apixaban; however, further investigation would be prudent.

Oral Antiplatelets

Aspirin (acetylsalicylic acid) is perhaps the most widely used antiplatelet because of its low cost and over-the-counter availability. Aspirin exhibits its antiplatelet effect by irreversible acetylation of COX-1, reducing thromboxane A2 synthesis, which inhibits thrombus formation on arterial walls.[15] Data published in the past decade have indicated that a number of patients are 'resistant' to aspirin's antiplatelet effect, which garnered significant attention by media outlets. It is believed to be multifactorial, stemming from a combination of drug interactions, poor compliance, increased platelet turnover and, perhaps, genetic polymorphisms.[24,25] The term resistance is, in truth, a misnomer; it implies that nonresponse to aspirin is due to an inability of aspirin to block platelet aggregation caused by arachidonic acid conversion to thromboxane A2 via COX-1.[25] This nonresponse has been reported as widespread. However, further evaluation has indicated that nonresponse is likely rare, with more liberal estimates likely confounded by issues such as noncompliance.[26] Aspirin nonresponse is not seen as a pressing issue because it tends to be overcome by increasing the dose or adding a P2Y12-receptor inhibitor for a synergistic antiplatelet effect. As such, clinical guidelines suggest dual treatment with aspirin and a thienopyridine agent.[27,28] That is not to say that patients relying solely on aspirin for its antiplatelet properties would not benefit from baseline and postdosing thromboxane levels to determine response to aspirin.

Clopidogrel. Clopidogrel (Plavix®, Bristol-Myers Squibb and Sanofi) is a thienopyridine that acts by irreversible P2Y12-receptor inhibition, which replaced the prototypical ticlopidine due to its superior safety profile and more rapid platelet inhibition.[29] Clopidogrel has recently become available generically, making it an attractive choice for formulary committees and third-party payers. Clopidogrel is a prodrug that is activated in the liver by the CYP enzymes. Of particular importance, CYP2C19 plays a role in the two-step activation of clopidogrel.[25] A list of CYP2C19 variables and effect of clopidogrel response is listed in Table 2. However, the CYP2C19 loss-of-function alleles account for only up to 20% of response variability, with the remaining percentage potentially due to other genes or variations of the CYP2C19 enzyme.[30] The reduced platelet inhibition seen in this population appears to be able to be overcome by increasing dose; however, the clinical relevance of this finding is uncertain.[31] Another potential genetic variant causing a decreased response to clopidogrel is the ABCB1 gene (Table 2). The ABCB1 gene encodes the P-glycoprotein efflux pump and is involved in the absorption of clopidogrel, as well as many other antithrombotic agents. However, the clinical relevance of ABCB1 polymorphisms has since been disputed.[32]

Prasugrel. Prasugrel (Effient®, Eli Lilly and Company) is another P2Y12 receptor inhibitor approved after clopidogrel in 2009 for the reduction of cardiovascular events in patients with acute coronary syndrome who are managed with percutaneous coronary intervention. Prasugrel's landmark study, the TRITON-TIMI 38 trial, found that prasugrel use led to fewer cardiovascular-related deaths (HR: 0.81; 95% CI: 0.73–0.90) compared with clopidogrel, with an increase in major bleeding events.[45] Prasugrel, like clopidogrel, is a prodrug that requires activation by CYP enzymes. A key difference is that clopidogrel is majorly shunted to a dead-end inactive pathway, with the remainder requiring two CYP-dependent steps for activation of the prodrug, while prasugrel is activated without a dead-end pathway in one CYP-dependent step.[46,47] Furthermore, metabolism of prasugrel is mostly mediated by CYP3A4 and CYP2B6, further explaining the lack of association with CYP2C19 polymorphism complications.[40] Owing to these fundamental differences in metabolism, it appears that prasugrel is not subject to the genetic response variation seen with clopidogrel. Using results of the TRITON-TIMI 38 trial, investigators completed a genetic substudy to determine association between ABCB13435C→T and CYP2C19 loss-of-function alleles with cardiovascular death, myocardial infarction or stroke.[38] No association was found with the genetic variants and an increase in major cardiovascular events with prasugrel. While prasugrel does indeed utilize the P-glycoprotein efflux system, the authors attributed the lack of genetic effect to prasugrel's rapid metabolism; however, this hypothesis requires further investigation. Further evaluation has since confirmed no effect of CYP2C19 phenotype on prasugrel response.[48]

Ticagrelor. Ticagrelor (Brilinta®, AstraZeneca) is a novel reversible, direct acting P2Y12 receptor inhibitor approved in 2011 by the FDA in the USA for prevention of thrombotic events in patients with acute coronary syndrome or ST elevated myocardial infarction (STEMI). In the PLATO trial, ticagrelor was associated with fewer deaths from cardiovascular causes than clopidogrel (HR: 0.84; 95% CI: 0.77–0.92), with an increase in minor bleeding events.[49] Ticagrelor does not require metabolic activation and thus appears to be unaffected by the genetic variations associated with clopidogrel.[41,42] However, as with prasugrel, recent evidence suggests wide variability in absorption in STEMI patients, which warrants genetic investigation.[43]