From the Journals

Genotype-guided warfarin dosing reduced adverse events in arthroplasty patients


Key clinical point: Presurgical genotyping can guide warfarin dosing and reduce anticoagulation-related adverse events.

Major finding: Genotype-guided dosing reduced adverse events – primarily elevated INRs – by almost 4% compared to clinically based warfarin dosing.

Data source: A randomized trial comprising 1,650 elderly patients undergoing elective knee or hip arthroplasty.

Disclosures: Dr. Gage had no financial disclosures, but several of his coauthors noted relationships with pharmaceutical and imaging companies.

View on the News

Questions of practicality remain

Warfarin is the most commonly used anticoagulant in the world, and a significant cause of emergency department visits and hospitalizations, especially among older patients. Walking the fine line between dosing too little and too much is not an easy task – especially since warfarin response is influenced by diet, comorbidities, interactions with other medications and – as studies over the last 20 years have confirmed – many genetic variants.

Also, the practicality of genotyping every patient who needs anticoagulation therapy must be questioned. Based on the results of GIFT, 26 patients would need to be genotyped to prevent one event, typically an INR of 4 or greater. Although the cost of genotyping continues to decline, health insurers and publicly funded health systems have not yet been convinced that genotype-guided warfarin prescribing is a cost-effective strategy.

The benefits of genotyping would likely be less in patients with atrial fibrillation, for example, as they run a lower risk of VTE than do arthroplasty patients. The GIFT surgeries were all elective, so there was plenty of time to get back genotyping results before starting warfarin. That is a luxury not afforded to many patients in need of anticoagulation.

It’s possible, however, that the benefits of genotyping might be larger in the real world. GIFT was conducted at academic medical centers and used a clinical dosing algorithm as comparator. As a result, adverse event rates were likely lower in the comparison group than would be expected in other clinical settings with less-intense INR monitoring or empirically based initiation regimens.

Still, GIFT’s results are gaining global attention. Based on prepublication results of the GIFT trial, the Clinical Pharmacogenetics Implementation Consortium (CPIC), an international research network that develops consensus recommendations about the use of pharmacogenomic test results, recently published guidelines about genotype-guided dosing for warfarin. The group now recommends using genotype-guided warfarin dosing based on CYP2C9*2, CYP2C9*3, and VKORC1 variants for adult patients of non-African ancestry. It also recommends that patients with combinations of high-risk variants would benefit from an alternative anticoagulant strategy, because of likely greater risks of poor INR control and bleeding.

A single pharmacogenomic test covering many common variants relevant to multiple prescribing decisions over time is far more likely to be a cost-effective approach; however, there is no evidence for this proposition. Until then, it might be simpler and less expensive to use clinical dosing algorithms to reduce the risks of anticoagulation.

Jon D. Emery, PhD, is the Herman Professor of Primary Care Cancer Research at the University of Melbourne and Western Health, Melbourne. He made these remarks in an accompanying editorial (JAMA 2017;318;110-2 doi: 10.1001/jama.2017.11465 ).



Genotype-guided warfarin dosing reduced anticoagulation-related adverse events by almost 4% compared with a clinically based warfarin-dosing regimen in a randomized trial of 1,650 elderly patients undergoing hip or knee arthroplasty.

The difference in the composite endpoint (major bleeding within 30 days, international normalized ratio [INR] of 4 or greater within 30 days, venous thromboembolism within 60 days, or death within 30 days) in the Genetic Informatics Trial of Warfarin to Prevent Deep Vein Thrombosis (GIFT) trial was mainly driven by a significant difference in episodes of elevated INR, reported Brian F. Gage, MD, and his colleagues (JAMA 2017;318[12]:1115-1124. doi: 10.1001/jama.2017.11469).

A total of 1,597 patients completed the trial. Of 808 patients in the genotype-guided group, 10.8% met one of the endpoints. Of 789 in the clinically guided warfarin dosing group, 14.7% met at least 1 of the endpoints. There were no deaths in the study.

“Widespread use of genotype-guided dosing will depend on reimbursement, regulations, and logistics. Although several commercial platforms for warfarin-related genes have been approved by the Food and Drug Administration and the European Medicines Agency, routine genotyping is not yet recommended,” wrote Dr. Gage of Washington University, St. Louis, and his coauthors.

The Centers for Medicare and Medicaid Services used its Coverage with Evidence Development program to fund genotyping in this trial and will review the results to determine future coverage, the researchers added.

In GIFT, patients were randomized to an 11-day regimen of warfarin guided either by a clinical algorithm or by their individual genotype. The team tested for four polymorphisms known to affect warfarin metabolism: VKORC1-1639G>A, CYP2C9*2, CYP2C9*3, and CYP4F2 V433M. The treatment goal was an INR of 1.8-2. After 11 days, physicians could administer warfarin according to their own judgment.

The absolute difference of 3.9% in the composite endpoint was largely driven by a 2.8% absolute difference in the rate of an INR of 4 or greater. The rate difference between the two groups was 0.8% for major bleeding, and 0.7% for VTE.

About 41% of the cohort was considered to be at high risk of bleeding complications, and this group accrued the highest benefit from genotype-based dosing. Among them, the composite endpoint was 11.5% compared with 15.2% in the clinical algorithm group – an absolute difference of 3.76%.

The benefit was consistent among black patients, and those with CYP2C9.

By day 90, one VTE had occurred in each group. An intracranial hemorrhage occurred in one patient in the clinically guided group, 2 months after stopping warfarin.

The clinical benefit of genotype-based dosing influenced 90-day outcomes as well, with the composite endpoint occurring in 11% of the genotype group and 15% of the clinically guided group (absolute difference 3.9%).

Among the 1,588 patients who had their percentage of time in the therapeutic range (PTTR) calculated, genotyping improved PTTR time by 3.4% overall. The effect was especially strong from days 4 to 14, when it improved PTTR by 5.7% relative to clinical guidance.

Three other studies have examined the effect of a genotype-based warfarin dosing regimen, Dr. Gage and his coauthors noted: Two found no benefit, and a third found that such guidance improved INR control. GIFT has several advantages over those trials, which the authors said lend credence to its results.

“Compared with previous studies, this trial was larger, used genotype-guided dosing for a longer duration, and incorporated more genes into the dosing algorithm …The longer period of genotype-guided dosing likely prevented cases of supratherapeutic INR that were common in these trials,” they wrote.

Dr. Gage reported no financial disclosures, but several coauthors reported ties with pharmaceutical and imaging companies.

Next Article: