While anticoagulation is life-saving in the management of patients with venous thromboembolic (VTE) disease, anticoagulation, most especially with warfarin, is associated with one of the highest rates of preventable adverse events as highlighted in the National Action Plan for Adverse Drug Event Prevention.

To improve safety, prescribing information for warfarin suggests that CYP2C9 and VKORC1 genotyping may help with initial dosing. Most resources that examine dosage alterations in the presence of drugs that interact with warfarin focus primarily on amiodarone, but what effect do other potentially interacting medications have on dosage adjustments needed to maintain the INR (international normalized ratio) in the therapeutic range long term?

To help answer this question, investigators in this retrospective cohort analysis examined long-term efficacy and safety measures for warfarin. These analyses involved correlating CYP2C9 (CYP2C9*2, CYP2C9*3, CYP2C9*4, CYP2C9*4) /VKORC1 (VKORC1 c.-`639 G>A) genotype data and the concomitant use of drugs known to interact with warfarin. Utilizing data from the NUgene biobank, which is owned by Northwestern Medicine hospital in Chicago, patients on warfarin who met inclusion criteria had their CYP2C9/VKORC1 genotype screened against a comprehensive list of medications known be inhibitors or inducers of CYP2C9.

This list was developed based on FDA labeling information and the Flockhart Table of Drug Interactions and included CYP2C9 inhibitors amiodarone, capecitabine, efavirenz, fenofibrate, fluconazole, fluvastatin, isoniazid, lovastatin, metronidazole, micronazole, paroxetine, probenecid, sertraline, sulfamethoxazole, tigecycline, voriconazole, and zafirlukast, as well as CYP2C9 inducers bosentan, carbamazepine, and rifampin. 

Of the 12,722 patients enrolled in the biobank, 302 met the inclusion criteria, which were Caucasian race, being prescribed warfarin for at least 1 year, having at least 20 international normalized ratio (INR) values from within the past year with at least 1 year between the first and last INR, not being involved in other pharmacogenomic studies, and having a sufficient amount of DNA testing sample.  

The mean age of the study population was 62.6 years, and 42% were prescribed warfarin for atrial fibrillation. Genotyping revealed 183 study participants were “normal responders.” The three most commonly interacting medications were amiodarone, sulfamethoxazole, and metronidazole. Thirty percent of warfarin users did not have a concomitant interacting drug whereas 40%, 20.2%, 5.6%, 3.3%, and 0.7% of patients had 1, 2, 3, 4, and 5 interacting medications, respectively. 

Overall, when time in therapeutic range (TTR) was analyzed, study participants were underanticoagulated 39% were overanticoagulated, and 15.6% were in the therapeutic range (2.0 < INR < 3.0) 45.3% of the time. When the association between INR variability, TTR and VKORC 1/CYP 2C9 genotypes were considered together, genotype impacted INR variability, but not TTR. However, when only CYP 2C9 genotype with examined, it was found that exposure to drugs interacting with this isozyme affected both INR variability as well as TTR. 

Interestingly, in the absence of interacting medications, CYP2C9/VKORC 1 genotype directly affected INR variability and TTR. However, this effect of genotype appeared to be masked when at least one interacting drug was present since there was no significant association with genotype and any outcome measures in the presence of an interacting medication. 

This study is useful to help pharmacists assess the potential effect of CYP2C9 inhibitors and inducers on maintaining TTR or affecting INR variability among patients receiving warfarin when their genotype status is known.

« Click here to return to Hematology Update.