Dear Warfarin Study: Don’t Neglect Patient Safety
On December 12, 2013, the New England Journal of Medicine published the results of one of the largest clinical warfarin trials ever undertaken. Written by Dr. Stephen Kimmel et al, the paper questions the clinical utility of genotyping prior to warfarin administration.
The pharmacogenetic testing industry is one of the next-generation industries most at need for comprehensive studies due to its large amount of information and constant new discoveries. As such, Kimmel’s paper is an extremely valuable resource. However, there are a few points that deserve more investigation before any conclusive statements can be made with regards to warfarin genotyping.
Kimmel et al examined two groups of patients—one whose warfarin dose was adjusted based on a clinical algorithm and one based on a genotype-powered algorithm—and concluded there was no significant difference overall between the mean percentage of time in the therapeutic range.
However, while the time in therapeutic range is very useful for prescribers, Kimmel’s study was not powered to examine the incidence of adverse events, which may in fact be the more important endpoint for patient safety. Warfarin has been known to cause severe, sometimes fatal bleeding events in some patients and genetic variants have been previously implicated in the risk of these events. Prescribers should keep this in mind.
There may also be some hidden effects of how the study was set up. In 55% of the genotyped group, the individuals received their first dose of warfarin without using genetic information. It’s possible that the genotyped group would have reached therapeutic dose faster if all 100% were genotyped prior to administration. This may seem like a minor point, but methodological discrepancies should be scrutinized in any clinical trial.
Kimmel’s paper examines the benefit of genotype-guided warfarin dosing over clinical algorithm-based dosing and found little benefit. Both the genotyping and clinical algorithms included patient age, race, and other demographic or health information.
This comparison is relatively unique—many previous warfarin studies examined the effectiveness of genotyping versus a standardized dosing strategy. In these previous trials, there was a significant difference between genotyped vs. non-genotyped groups. Because there isn’t a single universally agreed-upon way to dose warfarin, more studies should investigate multiple dosing methods. In the meantime, clinics using standardized dosing strategies may benefit from using genotype-based dosing tools such as YouScript.
Overall, Kimmel’s paper is very valuable and interesting, but don’t throw out the clinical utility of genotyping just yet—more endpoints need to be investigated, trial methods need to be refined, and more people need to take a look at it before conclusions can be drawn from Kimmel.
The most important point, however, is patient safety. Given the role genetic variants may play in the incidence of bleeding events, more large-scale studies are needed before definite statements can be made about patient safety.
To read the full Kimmel article, go to http://bit.ly/1cLpF6x.