Skip Over Navigation Links

Web Exclusives: Genetics

Math Gives Blood
By Emily Carlson
Posted February 25, 2009

Going to my grandmother's house was like trick or treating. Every time I walked in her door, I went straight to the fridge to find the candy. But she dished out as many sweets as she did stories.

Emily and Emily.
Emily and Emily.

When she was nine, the Great Depression hit and forced her family to leave the city and live and work on her uncle's farm in rural Iowa. My grandmother (for whom I'm named) had to help tend the vegetables in the garden.

"I loved to eat them, but I hated to grow them," she says, recalling one time when her father scolded her for planting a crooked row of seeds.

Now 88 and living through another economic crisis, she tells me she can't eat many of the foods she used to grow. Four years ago, her doctor said she had an abnormal heart rhythm that put her at greater risk of having a stroke. To help prevent that, he gave her a prescription for the blood thinner warfarin. This meant purging her diet of cabbage, spinach, broccoli and other foods high in vitamin K, which interferes with the drug.

"The last two times I've been in the 'safe zone,'" she tells me, explaining that a doctor checks her blood once a month for its clotting ability.

Every year, two million Americans start taking warfarin, making it one of the most widely prescribed drugs in the world. It's given to prevent dangerous blood clots that can lead to heart attacks, strokes or even death. People also take it after a major surgery to avoid other clotting problems.

Some people need stronger doses, and others need weaker ones. For this reason, doctors have a really hard time determining the right dose for each person. They typically use a trial and error approach that gets fine tuned through blood tests—lots of them.

My grandmother says it took about four months and more than eight trips to the doctor's office to learn that she needed only 2 milligrams a day to keep her blood at the right consistency. She lives alone and doesn't drive, so she had to ask my mother or others to take her.

In 2007, the FDA modified warfarin's label to indicate that genetic makeup may affect patient response to the drug. Credit: Alisa Machalek
In 2007, the FDA modified warfarin's label to indicate that genetic makeup may affect patient response to the drug. Credit: Alisa Machalek

New research and an upcoming clinical trial may speed the process and eliminate many extra needle pricks and hassles. By looking at people's genes, an international team of researchers has developed a mathematical tool that could help doctors accurately predict the ideal dose on the first visit.

Researchers know that two genes, CYP2C9 and VKORC1, and slight variations in them can influence warfarin's effectiveness. A 2005 study, for instance, found that certain population groups tend to have a higher prevalence of a particular VKORC1 variation and thus require a particular dose. While Asian Americans generally have the low-dose variation, African Americans have the high-dose version. European Americans, fall in the middle.

Based on other results like these, the U.S. Food and Drug Administration in 2007 worked with the makers of warfarin drug products to modify the product label to say that a patient's genetic makeup may affect how he or she responds to the drug.

To figure out whether information about these genes could improve initial dosing, researchers from more than 20 teams in nine countries on four continents formed the International Warfarin Pharmacogenetics Consortium.

By pooling their data, the consortium members had information from about 5,700 people worldwide who were taking stable doses of warfarin. The researchers had demographic information like age, gender and race as well as genetic information about CYP2C9 and VKORC1 variants. All of this, of course, was anonymous.

"The biggest challenge for researchers in this study was the agreement to share their data and then the aggregation of the data into one set," says Stanford University biomedical informatics scientist Teri Klein, one of the study's leaders. Klein adds that previous studies were based on smaller, less diverse data sets and that obtaining one of this size and breadth was crucial for drawing conclusions that could apply to a wide range of people.

The consortium members used this clinical and genetic information to develop a mathematical model for predicting each person's optimal dose. The researchers compared the predicted answers to medical records indicating the actual initial and stable dose.

Their predictions matched the stable doses, which had taken doctors months to determine. The results were even more accurate for patients at the low or high ends of the dosing range. Because these people are at greater risk for excessive bleeding from overdosing or clotting from underdosing, quickly optimizing their doses could minimize dangerous complications and improve the drug's effectiveness.

In an important step toward bringing these findings into the doctor's office, NIH is launching a major clinical trial next month. The goal is to test whether a dosing strategy similar to the one developed by the consortium improves patient outcomes. Considered the gold standard for testing medical therapies, this large, randomized clinical trial will determine if the more precise, gene-based prescribing strategy is the best option.

Researchers will have results from the clinical trial by 2013. That'll be too late to help my grandmother, who's already on a stable dose, but maybe not for yours.

Watch Drs. Teri Klein and Russ Altman discuss genetics and warfarin dosing.

Learn about related research

This page last reviewed on April 22, 2011