November 7, 2007 (Orlando, FL) – Using genetic tests as well as clinical factors improved the accuracy and efficiency of warfarin dose initiation, but the primary end point of a reduction in out-of- range INRs was not significantly changed in the overall population in the Couma-Gen study [1]. Subgroup analysis suggested, however, that pharmacogenetic guidance might reduce out-of-range INRs in certain patient groups--those without any variant genotypes or those with multiple variant genotypes.The study was presented today at the American Heart Association (AHA)2007 Scientific Sessions and published simultaneously online in Circulation. Lead investigator Dr Jeffrey Anderson (University of Utah, Salt Lake City) said: "In our study, patients using genotype- guided warfarin dosing needed fewer dose changes and were able to get to their optimum INR in a more efficient way. But the primary end point of out-of-range INRs was not statistically different in the genotype-guided treatment." He noted, however, that the study was conducted on an inpatient basis with careful management by a dedicated anticoagulation service, leading to rapid correction of inappropriate warfarin doses, which likely contributed to better-than- expected outcomes in the standard dose arm.

Anderson explained that several gene variants have been discovered that affect interindividual dose variability of warfarin and that the FDA has already endorsed the idea of using these gene variants to guide warfarin dosing. "There is very strong evidence that genetics influence the correct dose. The FDA has just taken this to the next logical step, saying that lower doses should be considered if the patient has multiple variants. This may be somewhat premature in many people's eyes, as it hasn't been shown to really affect outcomes yet, and our study hasn't shown this definitively, either. I think this approach has a lot of promise for the future, but it's maybe not ready for right now," he commented.

Anderson stressed that larger trials are needed and are now being planned, with a National Institutes of Health trial in 2000 patients to start next year.

Former AHA president Dr Raymond Gibbons (Mayo Clinic, Rochester, MN) shared Anderson's views. "This is an area of enormous potential importance, but we need more time and more evidence before it becomes part of clinical practice. I definitely do not think doctors should rush out there and start giving genetic tests to all the patients they want to put on warfarin at the moment. Maybe one day this will happen, and yes, it does make sense, but we need evidence that it will have a real benefit, and that's not there yet."

Gibbons added that for warfarin treatment, the potential of genetic testing would be useful only in the initial dose selection, which is important, but that it would not help in the chronic management of these patients. "People's genetics don't change over time, but what they eat, what they weigh, and how they metabolize drugs do, and it is this that affects the long-term variability in warfarin levels,"

he said.

"A very promising first step"

Discussant of the trial at the late-breaking clinical-trials session, Dr Julie Johnson (University of Florida, Gainesville), described the study as "a very promising first step. This trial did suggest a reduction in adverse events such as INRs over 4 and minor bleeds, which, while not significant, probably would have been in a larger study. It also showed a significant reduction in the number of dose changes needed to get to a stable INR, which would probably reduce the number of clinic visits necessary and therefore reduce costs,"she noted.

The Couma-Gen study randomized 206 patients being initiated on warfarin to pharmacogenetic-guided or standard dosing. The genetic tests were performed on buccal swab samples with a rapid assay.

"There are companies out there that offer this test, but you usually have to send off the sample and will not get the results for a few days. We managed to achieve this turnaround in just one hour. That is what is needed for this test to be useful in a clinical setting,"

Anderson said. The researchers then entered the genetic information along with data on weight, age, and sex into a computer program, which provided an individualized dose for each patient. Results showed that age, sex, and weight explained 12% of the variability of warfarin dosing, while genetics explained 35% of the variability.

Patients were followed for three months, and INR was measured on days 0, 3, 5, 8, 21, 60, and 90. While out-of-range INRs did not differ significantly between arms in the whole population, exploratory subset analysis suggested that those with the wild-type genotype (no genetic variants) and carriers of multiple variant alleles showed promising reductions in out-of-range INRs with pharmacogenetic guidance.

Per-patient percentage of out-of range INRs

Patient group Pharmacogenetic-guided dosing (%) Standard dosing (%) p

Whole population 30.7 33.1 NS

Patients with 0 or multiple variants 29 39 0.03

Explaining this subset analysis finding, Anderson noted that about 40% to 45% of patients have one genetic variant and 70% of patients have one or more variants, so the wild-type patients are in the minority. "The average patient therefore has one variant, and this is who we are targeting our standard dose of warfarin at. That is probably why patients in this study with one variant did not show any benefit of this genetic approach--they are already treated well with the standard dose. But wild-type patients and those with more than one genetic variant did show benefit. When we just use standard-dose warfarin we are underdosing wild-type patients and overdosing patients with more than one genetic variant," he added.

1. Anderson JL, Horne BD, Stevens SM, et al. Randomized trial of

genotype-guided versus standard warfarin dosing in patients initiating oral anticoagulation. Circulation 2007; DOI: 10.1161/ CIRCULATIONAHA.107.737312. Available at: http://circ.ahajournals.org.