Precision Medicine: Making Warfarin Safer
Posted on by Dr. Francis Collins
Every year, thousands of older Americans require emergency treatment to stop bleeding caused by taking warfarin, a frequently prescribed blood-thinning pill. My own mother received this drug in her later years, and her doctors encountered significant challenges getting the dose right. The problem is too much warfarin causes potentially serious bleeding, while too little leaves those who need the drug vulnerable to developing life-threatening clots in their legs or heart. The difference between too little and too much is distressingly small. But what if before writing a prescription, doctors could test for known genetic markers to help them gauge the amount of warfarin that a person should take?
Such tests have been available to doctors and patients for a few years, but they have not been widely used. The recent results of a national clinical trial offer some of the most convincing evidence that it’s time for that to change. In this study of 1,650 older adults undergoing elective hip or knee surgery, patients whose genetic makeup was used to help determine their dose of warfarin were less likely to suffer adverse events, including major bleeding. This trial marks an encouraging success story for the emerging field of pharmacogenomics, the study of how the variations in our genes affect our responses to medicines.
Most drugs are now prescribed in a “one-size-fits-all” fashion—as though they should work the same way in all people. But they don’t. Differences in our genes, sex, age, physiology, diet, and environment affect how our bodies respond to drugs.
That’s certainly the case with warfarin. With growing evidence that specific gene variations could affect benefits and risks, the Food and Drug Administration revised its product label in 2007 to encourage doctors to order pharmacogenomic tests when prescribing it. The suggestion, however, has gone largely unheeded. Prior studies hadn’t conclusively demonstrated that adjusting dosage on the basis of pharmacogenomic testing prevented adverse events among patients starting warfarin, and obtaining gene test results might take days or weeks—so doctors justifiably wondered whether this fine tuning of warfarin dosage was really necessary .
That’s what makes the results of the latest clinical trial, published recently in JAMA, so interesting . They come from the NIH-supported Genetic InFormatics Trial (GIFT) of Warfarin Therapy to Prevent Deep Venous Thrombosis, led by Brian F. Gage at Washington University, St. Louis.
Unlike the previous clinical studies, GIFT enrolled more participants to detect possible differences. The trial also tested for variants in three genes instead of two, and used the genetic results to guide warfarin dosing for a longer period of 11 days. These gene variants include: CYP4F2, involved in the metabolism of vitamin K to help clot blood; CYP2C9 and its influence on warfarin metabolism in the liver; and VKORC1, involved in warfarin sensitivity.
In the study, Gage and colleagues enrolled healthy older adults (average age 72 and 64 percent women) planning to undergo elective hip or knee surgery. Participants were randomly assigned to one of two groups. Those in one group were dosed according to clinical factors, such as age, height, and weight. The other participants arrived at a dose based on those same clinical factors, plus testing of the gene variants.
Of the 789 patients prescribed warfarin based on clinical factors alone, 116 (14.7 percent) had at least one adverse event: major bleeding, a warfarin overdose, or a blood clot. But of the 808 patients who also received pharmacogenomic testing, only 87 (10.8 percent) had an adverse event. Indeed, pharmacogenomic dosing led to a reduction in the number of bleeding events, from 74 to 57. It nearly halved the number of symptomatic adverse events overall.
While the findings are encouraging, their immediate impact will depend, in part, on whether insurance companies decide to cover the cost of testing for these genes, which is nearly $200 today, but may decrease in the future. Meanwhile, efforts like the NIH All Of Us Research Program—which will soon enroll 1 million Americans to improve health and health care—offer the chance to further explore the use of pharmacogenomic testing and potentially to expand its use to many drugs.
As the evidence builds and genetic testing becomes less expensive, some may choose to undergo pharmacogenomic testing before they get sick or need surgery. Gage says he hopes to integrate genetic information into patients’ electronic medical records so that it’s available anytime a new drug is prescribed. The goal is fewer adverse reactions to medicines, for older Americans and younger ones, too.
 Warfarin pharmacogenomics: current best evidence. Kimmel SE. J Thromb Haemost. 2015 Jun;13 Suppl 1:S266-71.
 Effect of genotype-guided warfarin dosing on clinical events and anticoagulation control among patients undergoing hip or knee arthroplasty: The GIFT Randomized Clinical Trial. Gage BF et al. JAMA. 2017;318(12):1115-1124.
Pharmacogenomics Fact Sheet (National Institute of General Medical Sciences/NIH)
Warfarin Dosing (Washington University, St. Louis)
Brian F. Gage (Washington University, St. Louis)
NIH Support: National Heart, Lung, and Blood Institute; National Center for Advancing Translational Sciences