Declaring War on Warfarin Misdosing

About 20 million Americans are on warfarin, and another 2 million start taking the drug each year. So, chances are, you know someone who is on Coumadin, Jantoven, Marevan, or Waran. If that’s the case, chances are you also know that getting the dose just right is a tricky business involving frequent office visits for blood tests and dosage adjustments.

This monitoring is important. Too much warfarin and you can bleed to death. Too little, and a blood clot may get you. The widespread use of warfarin and the hassle of getting the dose just right makes this drug an ideal candidate for pharmacogenetic research to find out if a person’s genes can help predict his or her best initial dose.

THE OPTIMAL DOSE

Warfarin is an anticoagulant, indicated after major surgery, such as hip replacement, and for people with artificial heart valves, or for those with conditions such as chronic atrial fibrillation or deep vein blood clots. A synthetic derivative of coumarin, a chemical found in many plants, warfarin is available in tablet form and is effective in preventing or eliminating clots. Warfarin’s drawbacks are its narrow therapeutic range and the fact that the optimal dose varies greatly from person to person. Some people metabolize warfarin slowly, while others do so quickly.

Throughout most of warfarin’s 50-year history, physicians have tried to calculate the optimal dose for individuals by using algorithms that factor into account gender, age, height, weight, race, evidence of liver disease, other drugs being taken, and the baseline International Normalized Ratio (INR), a measure of how quickly a person’s blood clots.

But that approach leaves a lot of room for improvement. More than a third of patients do not respond to their initial dose of warfarin as expected and end up in the emergency room. In 2004 and 2005, according to a study published in JAMA the following year, warfarin was implicated in an annual average of 43,000 ER visits for adverse drug events (ADEs). That’s second only to ADEs involving insulin, and it doesn’t include warfarin-related ADEs in hospitals, nursing homes, clinics, offices, or homes. Because of the high rate of ADEs associated with it, warfarin is underprescribed.

A few years ago, when researchers began figuring out which genes do what, they discovered that cytochrome P450 liver enzymes are responsible for metabolizing many drugs, including warfarin. Within this family, one gene, CYP2C9, explains about 10 percent of the variation in warfarin dosing. In 2005, the vitamin K epoxide reductase enzyme, encoded by the VKORC1 gene, was found to explain another 30 percent of dose variations. Together with clinical variables such as gender, age, height, and weight, the CYP2C9 and VKORC1 genes account for about 54 percent of the variation in how individuals respond to warfarin.

Numerous companies have developed commercially available tests to detect variations in the CYP2C9 and VKORC1 genes. Last year, the U.S. Food and Drug Administration updated labeling information for warfarin and suggested that physicians order these new genetic tests to help them establish more precise warfarin doses in patients.

A WIN FOR EVERYONE

One of the experts who testified before the FDA in support of genotyping patients for warfarin dosing was Michael D. Caldwell, MD, PhD, a vascular surgeon and director of the wound healing program at the Marshfield Clinic, in Marshfield, Wis. In February, Caldwell and colleagues published an article in Blood1 describing their discovery of a new cytochrome P450 genetic variant, CYP-4F2, that accounted for a difference in stable warfarin dose of about 1 milligram per day in three independent cohorts of patients.

Jim Varney/Photo Researchers Inc.

One of the simplest applications of personalized medicine could prevent more than 17,000 strokes a year: genetic testing for warfarin metabolism.

One study estimated that integrating genetic testing into warfarin therapy could save more than $1 billion a year in adverse event-related spending.

“The addition of the CYP4F2 explains only about 2 additional percent of the interindividual variability in therapeutic dose in the population as a whole, but it can make a huge difference in people who have the variant,” Caldwell explains. “A person with full enzyme function gets an average dose of 5 milligrams per day, so 1 milligram amounts to a 20 percent difference.”

Last November, Marshfield Clinic licensed the new 4F2 warfarin biomarker to Pasadena, Calif.-based Osmetech Molecular Diagnostics. Osmetech already has an FDA-cleared cystic fibrosis test on the market, and CEO James White expects FDA clearance of its new eSensor warfarin sensitivity test by the middle of this year. The test will run on Osmetech’s patented XT-8 platform, and it will include the CYP2C9 *2, *3, and VKORC1 markers. A second version of the warfarin sensitivity test is being developed that includes CYP4F2 and other important markers to address interindividual variability.

“Our belief is that it’s in everyone’s interest to get to a standard of care that is an improvement over where we are today,” says White. “We’re certainly trying to get across the improved levels of patient care that this test will provide, the cost-effectiveness of having fewer patients turning up at ERs with adverse drug reactions, and shortening the time it takes to get them on the right dosage. These kinds of things should have significant benefits to payer and employer groups.” The resulting cost savings should also be good news for healthcare providers whose payment is capitated or case based.

Michael D. Caldwell, MD, PhD, founded the nationally acclaimed Personalized Medicine Research Project at Marshfield Clinic, in Wisconsin.

A 2006 paper by the AEI-Brookings Joint Center for Regulatory Studies, “Health Care Savings from Personalizing Medicine Using Genetic Testing: The Case of Warfarin,” estimates that “Formally integrating genetic testing into routine warfarin therapy could allow American warfarin users to avoid 85,000 serious bleeding events and 17,000 strokes annually. We estimate the reduced healthcare spending from integrating genetic testing into warfarin therapy to be $1.1 billion annually.”

NOT “FAR-OUT SCIENCE”

Potential savings notwithstanding, integrating genetic testing into routine warfarin therapy may take a while. For one thing, studies of warfarin genetic markers have all been retrospective. Several prospective studies — the gold standard for demonstrating that an intervention results in improved health outcomes — are planned or under way.

FDA’s Critical Path Initiative and the National Heart, Lung, and Blood Institute are planning a prospective clinical trial that will study warfarin dosing based on genetic test information. Critical Path also is helping to finance two other projects — one, a clinical study by Harvard Partners that will develop personalized dosing algorithms for patients newly started on warfarin, and the other, a research project by the University of Utah and the Critical Path Institute of Tucson, Ariz., to develop genetically based dosing instructions. Caldwell says that Marshfield Clinic is halfway through a prospective trial funded by the Agency for Healthcare Research and Quality.

Medco Health Solutions, the large pharmacy benefit manager, is collaborating with Mayo Clinic’s Mayo Collaborative Services to evaluate patient safety and financial savings derived from the use of genetic testing for warfarin dosing. The study has enrolled more than 1,000 patients who are new to warfarin therapy and who will receive care in community settings.

Teresa M. DeLuca, MD, MBA, vice president in Medco’s Department of Personalized Medicine, is the principal investigator of the Mayo/Medco study. Although she believes the jury is still out on CYP4F2, she says early results from Medco’s collaboration with Mayo suggest significant gains in safety and efficacy.

“This is not far-out science,” DeLuca adds. “Genotyping gives you the ability to use information that you can gather painlessly, easily, and not that expensively; that actually will [result in] true cost savings; and, for the patient, adds safety and efficacy. We take this technology very seriously, and we stand behind this as the way medicine needs to be practiced.”