UW News

March 5, 2009

Genetic profile may improve warfarin dosing

Including genetic information in a patient’s clinical profile might help determine the optimal starting dose of the common blood-thinner warfarin, according to findings from a large-scale study published Feb. 19 in the New England Journal of Medicine.


Researchers from more than 20 scientific teams in nine countries on four continents collaborated on the project. Stanford University helped coordinate the study. The findings are described in the article, “Warfarin Dosing Using Clinical and Pharmacogenetic Data.”


Warfarin is often prescribed to prevent blood clots in people with certain heart conditions or other serious diseases. Getting the right starting dose can be tricky because people vary in their responses to the drug. A patient taking too high a dose may begin bleeding. If the dose is too low, deadly blood clots might form. The safest, most effective dose can vary by a factor of 10 among different patients.


In 2002 and again in 2005, UW pharmacy and genome science researchers made major contributions in the field of genetic variations that influence how an individual responds to warfarin. David Veenstra of the UW School of Pharmacy and his colleagues found that patients with genetic variants in the human enzyme that metabolizes warfarin required lower doses and were more likely to suffer bleeding problems from the drug.


Later, Allan E. Rettie, professor and chair of medicinal chemistry in the UW School of Pharmacy, Mark Rieder, research professor of genome sciences in the UW School of Medicine, and their colleagues showed that warfarin response can also be affected by variations in the gene encoding vitamin K epoxide reductase. Retrospective studies suggested that variations in this gene stratified patients into low-dose, intermediate-dose, and high-dose warfarin groups.


Their work helped set the stage for future research on what is often called personalized medicine, in which a person’s genetic makeup might be used to individualize and optimize drug therapy. Veenstra, Rettie, Rieder and Lisa Meckley, also of the UW School of Pharmacy, contributed data and helped in the interpretation of the data analysis for the new study.


The U.S. Food and Drug Administration in 2007 modified the warfarin product label and suggested using genotype information in initial dosing for some, but not all, patients. However, most genetic-based dosing methods relied on geographically and ethnically distinct populations. The International Warfarin Pharmacogenetics Consortium was created to develop a dosing algorithm, or mathematical model, based on a wide variety of patients from many countries and ethnic backgrounds.


“It became clear that larger groups of patients must be studied,” Veenstra said, “to understand the effect of genomic variation on warfarin-related outcomes.” The consortium combined anonymous clinical and genetic data from 5,052 patients taking warfarin. They created two dosing algorithms: one with only clinical variables, such as height, weight, diabetes and other conditions, and additional medications taken, and another using both clinical variables and genetics, which they called the pharmacogenetic algorithm.


They compared these algorithms with each other and with a fixed-dose strategy, and then compared the algorithm-derived dose estimates with the dosages actually received by patients.


They found that the pharmacogenetic model was significantly better than both the clinical algorithm and the fixed dosage strategy in predicting starting dosages that were closer to the required stable therapy warfarin dose, especially for those patients in the low-dose or high-dose groups. The pharmacogenetic algorithm was less likely to overestimate the dose for the low-dose group, or underestimate the dose in high-dose group.


“These findings will help refine the use of genomics to guide warfarin therapy,” Veenstra said, “although additional studies are still needed to evaluate the benefit, if any, of doing so.”


In a step toward determining if these findings are useful in clinical practice, next month the National Institutes of Health’s Heart, Lung and Blood Institute will launch the largest prospective, multi-center, randomized clinical trial in the United States to test whether a gene-based strategy for prescribing the starting warfarin dose will improve patient outcomes. The clinical trial, called Clarification of Optimal Anticoagulation through Genetics or COAG, will use a dosing strategy similar to the pharmacogenetics model developed in the international study.


“I hope that personalized medicine will provide significant benefits to many patients as well as reasonable economic value,” Veenstra noted, “But I think it will be a long time, perhaps decades, before personalized medicine can accomplish this, and it will not necessarily be widespread. Consideration of clinical and economic outcomes, as is done in the UW School of Pharmacy’s Pharmaceutical Outcomes Research and Policy Program, will help ensure that advances in personalized medicine benefit both individuals and society as a whole.”


The new study was supported by the National Institute of General Medical Sciences, the National Heart, Lung, and Blood Institute, the National Institute of Neurological Disorders and Stroke, and the National Center for Research Resources, all part of the National Institutes of Health. The UW work on the study was also supported by the Centers for Disease Control and Prevention National Office of Public Health Genomics, and UW Drug Metabolism, Transport, and Pharmacogenetics Research.