Phenytoin Metabolic Ratio, a Marker of CYP2C9 Activity, is Superior to the CYP2C9 Genotype as a Predictor of (S)-Warfarin Clearance

Background CYP2C9 is a member of the cytochrome P450 (CYP) superfamily responsible for the metabolism of 16% of drugs that undergo oxidative metabolism. The activity of CYP2C9 exhibits marked inter-individual variability, which translates into prominent differences in the pharmacokinetics of CYP2C9 substrates, some of which are characterized by a narrow therapeutic window. Genetic polymorphisms in the gene encoding for CYP2C9 account for a fraction of the variability in CYP2C9 activity. The phenytoin metabolic ratio (PMR) is a marker of CYP2C9 activity in vivo, which correlates with CYP2C9 genetic polymorphisms. Objective The purpose of the current study was to evaluate the ability of the PMR to predict the oral clearance of (S)-warfarin (SWOCL) and its formation clearance towards its CYP2C9-mediated metabolites (SWCLf) [i.e., 6- and 7-hydroxy-(S)-warfarin]. Methods The study was conducted in 150 healthy non-smoker subjects (segment 1) and 60 patients treated with warfarin (segment 2). In the first segment, the participants received on two separate occasions a single 300-mg dose of phenytoin and at least 7 days later a single dose of warfarin (5 or 10 mg). The same PMR procedure was performed in the second segment, except that it was performed either before warfarin initiation or after the patients had reached stable anticoagulation. The PMR was derived from the ratio of 5-(4-hydroxyphenyl)-5-phenyl-hydantoin content in a 24-hour urine collection to plasma phenytoin concentration 12- (PMR24/12) or 24- (PMR24/24) post-dosing. In segment 1, SWOCL was calculated from the ratio of (S)-warfarin dose to the warfarin area under the plasma concentration-time curve extrapolated to infinity and the SWCLf from the ratio of urine content of 6- and 7-hydroxy-(S)-warfarin to (S)-warfarin area under the (S)-warfarin plasma concentration-time curve until the last measured timepoint. In segment 2, estimated SWOCL was derived from the ratio of (S)-warfarin dose to the mid-interval plasma concentration of (S)-warfarin. Results The PMR, SWOCL, and SWCLf varied significantly between carriers of different CYP2C9 genotypes in both healthy subjects (p < 0.001) and patients (p < 0.005). However, PMR and SWOCL values exhibited substantial intra-genotypic variability. PMR24/12 and PMR24/24 were significantly correlated with SWOCL both in healthy subjects (r = 0.62 and r = 0.67, respectively, p < 0.001) and in patients (r = 0.57 and r = 0.61, respectively, p < 0.001). In a multiple regression model that included all variables that correlated with SWOCL, PMR was the strongest predictor, explaining 44% and 38% of the variability in SWOCL among healthy subjects and patients, respectively, and accounting for 95.7% (44%/46%) and 90.5% (38%/42%) of the total explained variability in SWOCL among healthy subjects and patients, respectively. Conclusions The PMR is the strongest predictor of SWOCL, and as such, it exhibits a significant advantage over the CYP2C9 genotype. The inclusion of PMR in future dosing algorithms of CYP2C9 substrates characterized by a narrow therapeutic window should be encouraged and further investigated.