Skip to main content
Premium Trial:

Request an Annual Quote

Study Provides 'Strong Evidence' Backing Warfarin Initiation PGx Testing; Adoption Impact Unclear


NEW YORK (GenomeWeb) – A large pharmacogenetic analysis published in The Lancet today has shown that atrial fibrillation patients with certain CYP2C9 and VKORC1 genotypes are at heightened risk for bleeding during the first 90 days of starting treatment with the anticoagulant warfarin and could avoid these adverse events if they instead receive a newer drug called Savaysa.

In the study, researchers led by Jessica Mega of Brigham and Women's Hospital genotyped more than 14,000 patients who took part in the 21,000-patient randomized, double-blind ENGAGE AF-TIMI 48 study, which found that a high and low dose of Savaysa (edoxaban) was non-inferior to warfarin in preventing stroke and systemic embolism and was linked to significantly lower rates of bleeding and deaths due to heart complications. The US Food and Drug Administration approved Daiichi-Sankyo's Savaysa in January based on the results of ENGAGE AF-TIMI 48.

Mega and colleagues conducted the prespecified pharmacogenetic analysis, in which patients were prospectively followed for a median of three years and nearly 5,000 genotyped patients received warfarin. The study authors compared clinical outcomes with warfarin and Savaysa according to genotype in order to determine whether a subset of patients derived differential benefit with Savaysa. Researchers grouped patients according to their genotypes into normal responders to warfarin (62 percent), sensitive responders (35 percent), and highly sensitive responders (3 percent).

They found that in the first 90 days of starting on warfarin, the risk of overt bleeding was 1.3 times greater among sensitive responders and 2.7 times more among highly sensitive responders compared to normal responders. During the same time period, sensitive and highly sensitive patients who received Savaysa had fewer bleeding complications compared to those who received warfarin; for normal responders either treatment seemed to have comparable safety.

Mega told GenomeWeb that the study could change clinical practice in terms of adopting PGx testing to guide the best anticoagulant option for patients starting such treatment.

"In cases where there is a plan to use warfarin, genotyping could identify close to 40 percent of patients in whom there is an early increased risk of over-anticoagulation and bleeding with use of standard dosing practices,”" she said. “"Our findings show that the risk could be substantially mitigated by using edoxaban, or potentially another direct factor Xa or thrombin inhibitor, instead of warfarin."

After PGx testing identifies sensitive and highly sensitive responders to warfarin, if doctors still want to go ahead with prescribing the older anticoagulant, then these patients should receive more INR monitoring and precision dosing, she added.

Warfarin has been on the market for 60 years and is available in generic form. When patients are given warfarin, it reduces the amount of vitamin K-dependent clotting factor made by the liver. The VKORC1 gene is involved in this process, and the blood of patients with mutations in this gene doesn't clot properly. Too much or too little warfarin places these patients at heightened risk of experiencing adverse reactions.

Warfarin is also metabolized by CYP450 enzymes, mainly CYP2C9. Patients with variations in this gene — particularly CYP2C9*2 or *3 alleles — can't metabolize warfarin well. According to the published literature, VKORC1 polymorphisms account for 30 percent of the warfarin dose variation in patients, while CYP2C9 variations contribute to 10 percent of the differences in therapeutic dose. When people who have these gene variations receive warfarin, they are at a heightened risk of life-threatening bleeding events or blood clots if they receive the wrong dose.

The US Food and Drug Administration updated the label for warfarin in 2007 and 2010, informing doctors that patients with certain variants of CYP2C9 and VKORC1 genes may require non-standard warfarin doses when they start therapy. In the drug label, the FDA recommends maintenance dosing ranges based on patients' CYP2C9 and VKORC1 genotypes. However, genotyping strategies haven't caught on in clinical practice and guidelines don't recommend genotyping patients ahead of prescribing them warfarin.

Meanwhile, given the dosing variability and bleeding risk with warfarin, drugmakers have been focused on launching new anticoagulants with reduced bleeding risk. The approval of Savaysa was the third Xa inhibitor approved by the FDA in recent years, after Bayer's Xarelto (rivaroxaban) and Bristol Myers-Squibb's Eliquis (apixaban); the agency has also approved a direct thrombin inhibitor, Boehringer Ingelheim's Pradaxa (dabigatran). However, as noted by Mega and colleagues in The Lancet, "US guidelines are largely neutral in terms of which type of oral anticoagulant to use."

The latest analysis, according to Mega et al. suggests that with the help of genetic testing, sensitive and highly sensitive responders to warfarin might do better with an Xa inhibitor in the early days of starting an anticoagulant. But some in the medical community might interpret the study findings as evidence that they should start all patients on the newer anticoagulants and not bother at all with PGx testing or warfarin's troublesome dosing variability.

Previously, the suggestion that people with CYP2C9 and VKORC1 genotypes might be at increased bleeding risk at standard warfarin doses came from retrospective and observational trials. However, in late 2013, the New England Journal of Medicine published two randomized-controlled studies which arrived at conflicting conclusions, making it even more difficult for doctors to decide whether to adopt PGx testing for warfarin.

One study, conducted by the EU-PACT group and involving 455 patients, compared a cohort given warfarin with the help of a point-of-care genetic test and a PGx-based algorithm against a cohort given standard dosing. Researchers led by University of Liverpool's Munir Pirmohamed found in this study that in the first 12 weeks of warfarin therapy, patients in the PGx-guided group stayed longer within therapeutic range — the dose range required for desired treatment effect — compared to those in the control arm (67.4 percent versus 60.3 percent). 

Another study, known as COAG, randomized approximately 1,000 patients to receive either a warfarin dose as determined by an algorithm that factored in clinical variables and patients' genotypes, or dosing only by a clinical algorithm. Led by University of Pennsylvania's Stephen Kimmel, COAG investigators found that in the first four weeks of treatment, genotype-guided warfarin administration was no better than the clinical algorithm in terms of the mean time that the two groups remained in therapeutic range. 

Neither of these studies followed patients for outcomes. Still, after these publications, some experts focused on the COAG study to say there's no reason to perform PGx testing for warfarin, while others looked at EU-PACT as more closely aligning with real-world practices and remained hopeful that a more definitive study focusing on outcomes would provide support for warfarin PGx testing.

The latest Lancet study may provide more backing for warfarin PGx testing supporters. "The large sample size and preplanned inclusion of genotype make this an important study. This will be enough to drive some programs to start or restart the use of CYP2C9/VKORC1 [testing] in their practices," Howard McLeod, medical director of the Personalized Medicine Institute at the Moffitt Cancer Center, told GenomeWeb.

But the study may not be a total home run in swaying the warfarin PGx testing doubters. The conflicting COAG and EU-PACT data in NEJM "caused some folks to blindly reject the notion of genotyping and others to adopt the practice from the paper that was most like their setting," McLeod said. Based on the latest study, "it is still not clear that prospective use of this [genotyping] data will decrease the bleeding risk," he added. "So, there will be folks that won't be moved."

Warfarin will continue to be used because of low cost and wide availability.

In the paper, Mega and colleagues say they achieved three things – they validated the normal, sensitive, and highly sensitive responder groups by genotype included in warfarin's label; showed that CYP2C9/VKORC1 genotypes are significantly associated with warfarin-related bleeding outcomes; and demonstrated that genetic data were in line with standard clinical factors for bleeding risk. Regarding this last point, researchers compared the HAS-BLED clinical bleeding score against bleeding outcomes according to genotype and found that normal warfarin responders had low HAS-BLED scores and the lowest bleeding rates, while highly sensitive responders had the highest HAS-BLED scores.

Although Mega et al. said their study provided "strong evidence" that genotyping can help tailor anticoagulant treatment for atrial fibrillation patients in the first 90 days of starting therapy, beyond 90 days the safety profile or Savaysa versus warfarin was similar across genotyped categories in terms of bleeding and cardiac related deaths. As such, some might argue, given the superiority of Savaysa over warfarin in the early initiation period in certain genotypes and Savaysa's comparable safety longer term, why not just give the newer, less variable agents for all patients?

McLeod acknowledged that there are those who believe that doctors should just abandon warfarin for the newer agents. Makers of newer anticoagulants have, as expected, conducted studies to demonstrate the overall lower bleeding events with their drugs compared to warfarin. Daiichi-Sankyo, which funded ENGAGE AF-TIMI 48, didn't respond to questions ahead of press time.

The problem with just going with a new anticoagulant, according to McLeod "is that they do not have a biomarker for monitoring, have significant bleeding risk, and are much more expensive than warfarin." Experts have noted that with warfarin, bleeding can be stopped by Vitamin K injections, but such strategies don't stop bleeding with newer anticoagulants. Moreover, warfarin costs pennies per day, while the newer agents can cost as much as $8 per day.

"The aggregate data from multiple clinical trials in patients with atrial fibrillation suggest that direct factor Xa and thrombin inhibitors are at least as efficacious as warfarin, reduce mortality, and reduce intracranial bleeding by around 50 percent," Mega said. "However, warfarin will continue to be used because of low cost and wide availability."

If warfarin is here to stay, then genotyping could be useful, according to the study authors, for managing bleeding risk for 40 percent of patients who are sensitive or highly sensitive to the drug. The study authors noted that because their analysis was done on atrial fibrillation patients who were receiving anticoagulants to prevent stroke, the utility of warfarin genotyping might be different in patients with other disorders.