Feb. 18 (Bloomberg) -- A formula combining a patient’s genetic profile with data such as height, weight and medical condition may make it easier for doctors to give the right dose of the blood thinner warfarin, an international study found.

The formula predicted the proper amount of medicine 49 percent of the time, compared with 33 percent accuracy using just measurements and medical conditions, for patients needing lower-than-average doses of the drug, according to researchers.

Warfarin is commonly prescribed to prevent blood clots that can lead to heart attacks and stroke, and getting the right dose can be tricky because it can vary widely from patient to patient. The research, published in tomorrow’s New England Journal of Medicine, points the way for doctors to use genetic information to increase the effectiveness of drugs while reducing side effects and saving money.

“Your genetics can influence in an important way the drugs that would be best for you and the doses that would be best for you,” said one of the study authors, Michael Wagner, a research professor at the University of North Carolina in Chapel Hill, in a Feb. 17 telephone interview. “This will probably become a part of practice over the next couple of years, and this will improve the health care that we can offer to people and make it more cost effective, we hope.”

Dose Crucial

Too much warfarin can cause people to bleed easily and too little can fail to stop blood clots. In the U.S., about 800,000 people, or about 46 percent of those taking warfarin, require doses either higher or lower than the average, said one of the study’s lead authors, Russ Altman, a professor of bioengineering, genetics and medicine at Stanford University, near Palo Alto, California.

Doctors often start patients on the average dose of warfarin and then check their blood weekly until the ideal amount of medicine is achieved. That process can take months, researchers said.

The study was undertaken by researchers from nine countries using clinical and genetic data from 4,043 patients to create and compare three formulas to see which best determined the appropriate warfarin dose. One formula was based solely on clinical data such as height, weight and medical conditions. A second formula combined genetic information with the clinical data, and a third used a fixed daily dose.

Predictions

They used the formulas on 1,009 people who were already taking warfarin to determine how near they were to predicting each patient’s optimum dose. Because of their individual genetic and physical makeup, some patients needed lower-than-average amounts and some higher.

Adding genetics to clinical data in the formula produced a more accurate answer in those who needed the smallest and the largest doses, the research showed. For those in the middle, little difference was seen among the results from the formulas.

In those patients requiring lower-than-average doses, the genetics formula predicted the right amount 49 percent of the time, compared with 33 percent of the time with the clinical formula. In those who needed higher-than-average doses, the genetics formula predicted the right amount 25 percent of the time compared with the clinical formula’s 7 percent accuracy.

Doctors could start on the average dose as soon as they saw the patient and use the genetic information in the second visit, Altman said in a Feb. 17 telephone interview. That approach could save money as the price of genetic testing drops and serious side effects requiring hospital stays are avoided.

Individual Responses

“What we’d like is to have a higher chance of people responding well and a lower chance of being harmed by any given drug,” said Janet Woodcock, director for the Center for Drug Evaluation and Research at the U.S. Food and Drug Administration, who also wrote an accompanying editorial in the journal.

Less than half of the drugs approved in the U.S. would be able to be tailored to individuals based on genetics, she said in a Feb. 17 telephone interview. Patients now undergo gene tests before they receive GlaxoSmithKline Plc’s AIDS drug Ziagen to prevent serious side effects. Asian patients are screened before they receive the seizure medicine carbamazepine, she said.

More Than 30 Million Prescriptions

Warfarin, sold under the brand name Coumadin by Bristol- Myers Squibb Co., was approved in the 1950s. More than 30 million prescriptions for the drug are written each year.

The U.S. National Institutes of Health will study 1,200 people in a trial set to begin next month that will test whether adding genetic screening to determine the warfarin dose improves results from the onset. About half of those in the study will have their initial dose of warfarin determined by clinical data, and the other half by genetic and clinical data.

For Related News and Information:

To contact the reporter on this story: Nicole Ostrow in New York at nostrow1@bloomberg.net.

Last Updated: February 18, 2009 17:00 EST