DNA tests to determine warfarin dose

At least 2 million Americans with an abnormal, clot-triggering heart rhythm take the pill warfarin, also sold as Coumadin.

dna 88 (photo credit: )
dna 88
(photo credit: )
Personalized medicine, the tailored treatments that a few patients now get based on their own DNA, is finally headed for the masses: the many heart patients at risk of deadly blood clots. At least 2 million Americans with an abnormal, clot-triggering heart rhythm take the pill warfarin, also sold as Coumadin. Getting too little can lead to a stroke, and too much can cause life-threatening bleeding. To find the right dose for each patient, doctors use trial and error - and the errors lead to tens of thousands of hospitalizations and deaths every year. Starting this month, about 1,000 patients who have a condition known as atrial fibrillation will take part in a project that will match their Coumadin dose to their specific genetic needs. This genetic fingerprinting should single out the many people whose bodies break down warfarin faster or slower than normal, and their doctors can immediately adjust their dosage to prevent dangerous complications. "Twenty to 30 percent of people are either very fast or very slow" to metabolize many drugs but don't know it, said Dr. Robert Epstein, chief medical officer at prescription benefit manager Medco Health Solutions of Franklin Lakes, N.J., which is collaborating in the effort with the Mayo Clinic, based in Rochester, Minn. Meanwhile, the federal government and researchers at Harvard University and elsewhere have begun or are planning similar studies. Epstein and other experts say the warfarin projects comprise the first broad use of personalized medicine, or targeted therapy, in which a person's genetic makeup is used to pick the best medicine or dose. This approach essentially adjusts for differences in body chemistry that explain why one pain reliever or allergy pill works great for you but not for your mom. "It's a big deal," said Edward Abrahams of the Personalized Medicine Coalition, which includes industry, government and patient advocacy groups as well as insurers and research centers. "Warfarin is a very widely used drug, it's been around for 50 years and it has all these adverse events associated with it." If the warfarin studies are successful, patients will start demanding personalized medicine, he predicted. Insurers will, too, if the Medco study proves it saves money and protects patients. That's fairly likely, given Medco data showing that if patients have their warfarin dose changed more than once - meaning doctors didn't get it right the first two tries - their risk of being hospitalized for complications jumps from 20 percent to 31 percent. Epstein said he's hit "a home run" with his pitch to get employers and insurance companies sponsoring the prescription plans to join the study. "Everyone we've talked to unanimously was in," said Epstein, who expects the reduction in medical costs will be triple the test price of a few hundred dollars per patient. He noted a couple dozen companies already are developing commercial tests for variations in the two genes crucial in warfarin dosing, the ones in the new studies. However, sales have been slow for the only government-approved testing device, Roche's Amplichip, which covers numerous gene variations. A November 2006 report by the American Enterprise Institute-Brookings Joint Center predicts using genetic information to prescribe warfarin would save an estimated $1.1 billion in US health care spending each year, while preventing about 17,000 strokes and 85,000 serious bleeding incidents. Bleeding complications alone kill about 8,000 people a year and are the No. 2 reason for medication-related ER visits, said Dr. Elizabeth Nabel, director of the National Heart, Lung, and Blood Institute. Still, warfarin is far better at preventing clots than aspirin or drugs like Plavix, which only affect part of the clotting system, she said. But patients on warfarin must have blood drawn and tested repeatedly to see if it clots too fast or too slow, initially every week or more often. "Most hospitals have a warfarin clinic, where people come in for adjustments," she noted. So along with the Food and Drug Administration, her institute now is planning one or two large studies on personalizing warfarin dosing. The FDA also is poised to alter the package insert for warfarin to note that gene variations in some patients mean they need a lower dose than normal. Meanwhile, five Harvard Medical School teaching hospitals just began a study including 500 warfarin patients to try to boost the percentage getting the optimal dose. Only about 60 percent do now, said Dr. Samuel Goldhaber, a Harvard cardiologist and professor. At the Marshfield Clinic Research Foundation in Wisconsin, a new study will compare results for 250 patients getting warfarin through standard trial and error with 250 getting doses that are genetically based, said Dr. Michael Caldwell, who started the personalized medicine program there in 2000. And in the one-year Medco project, DNA from more than 1,000 patients' blood samples will be shipped to the Mayo Clinic, which will do the genetic testing and send the results directly to the patient's doctor. Medco, which manages prescription benefits for one in five Americans, plans to use the results to speed up adoption of the strategy, according to Epstein. The studies will produce different, but complementary information that can be pooled for a complete picture, Goldhaber said. Dr. Franklyn Prendergast, director of Mayo's Center for Individualized Medicine, cautioned against assuming the results will be positive, because blood-clotting is a very complicated process. However, he noted that doctors are now able to use gene tests to find the right treatments for patients with breast and lung cancer, leukemia and a rare brain cancer. Uses in other areas of medicine, such as psychiatry, are generally limited to top academic medical centers, but likely will expand as research proves its value and testing becomes cheaper. Dr. Eric Braverman, director of integrated medicine at Cabrini Medical Center in New York, said more research is needed to prove personalized medicine improves care, but studies so far show promise in areas from Alzheimer's to addiction and obesity. He predicted personalized medicine will one day be used to forecast a child's potential future health problems, such as heart disease, so that preventive diet and lifestyle changes can be made early. As Prendergast put it: "The potential is unbelievable for doing good."