| Warfarin is a commonly prescribed oral anticoagulant for the treatment and prevention of thrombotic diseases, including venous thrombosis, myocardial infarction, ischemic stroke, and following heart valve replacement and atrial fibrillation. Warfarin is available as a racemic mixture of S- and R- enantiomer, and S- warfarin is 2.7 to 3.8 times more potent than R- warfarin. Cytochrome P4502C9 (CYP2C9) accounts for nearly complete hepatic metabolism of S- warfarin. Recent studies have shown that CYP2C9*2 and CYP2C9*3-the two important SNPs of CYP2C9, are associated with low maintenance dosage requirement in warfarin therapy and increasing the risk of major bleeding events. Our previous study also supported this finding. In addition, warfarin inhibits the action of vitamin K epoxide reductase, whose gene (VKORC1) was discovered in 2004 by two independent teams. This complex recycles reduced vitamin K, which is essential for the post-translationalγ-carboxylation of vitamin K-dependent clotting factors (Ⅱ,Ⅶ,ⅨandⅩ). Several recent studies have found that genetic variants of VKORC1 have a significant effect on warfarin dose and response. Yet there have been few extensive and thorough studies in this field in China so far. In this study, the distribution of CYP2C9 and VKORC1 of Han nationality in China was investigated, and the correlation between dosage requirement and CYP2C9 & VKORC1 genotypes and other factors was analyzed, finally an individualized warfarin therapeutic regimen was established in Chinese population.In this study, a PCR-RFLP method was firstly developed to determine CYP2C9 and VKORC1 genotype distribution in Chinese healthy subjects and warfarin-treated patients, and the result was verified by direct sequencing analysis. C YP2C9 genotyping showed that in healthy subjects the frequency of*1/*1,*1/*2,*1/*3 was 90.5%(200/221), 0.5%(1/221), and 9.0%(20/221), respectively, which for patients was 91.6%(174/190), 0%(0/190) and 8.4%(16/190), respectively. VKORC1 genotyping showed that in healthy subjects the frequency of GG, GA, AA was 0%(0/221),17.6%(39/221) and 82.4%(182/221), respectively, which for patients was 1.1%(2/190),13.2%(25/190) and 85.8%(163/190), respectively. The distribution of CYP2C9 and VKORC1 in warfarin-treated patients was in accordance with that in healthy subjects. In addition, a CYP2C9*2 variant allele was found in one healthy subject in this study, which confirmed that there are CYP2C9*2 alleles in Chinese.Secondly, the correlation between warfarin maintenance dose requirement and CYP2C9 & VKORC1 polymorphisms was investigated. On the basis of the above methods,190 patients were recruited. Patients' age, sex, body weight, height, serum albumin and warfarin maintenance dose were recorded, and venous blood was taken for the determination of CYP2C9 & VKORC1 genotypes as well as PT-INR. The statistical analysis showed that warfarin dose was significantly different among the groups of genotypes. Patients with VKORC1(1639G>A) AA genotype received significantly lower doses of warfarin than those with GA or GG genotype [(2.61±0.84)mg/d vs. (4.44±0.87)mg/d, P<0.0005], and patients with CYP2C9*1/*3 genotype also received significantly lower doses of warfarin than those with CYP2C9*1/*1 genotype [(2.29±0.83)mg/d vs. (2.98±1.09)mg/d, P=0.014]. Warfarin dose was significantly negatively correlated with age (r=-0.253, P<0.0005) and significantly positively correlated with body weight (r=0.306, P<0.0005) and height (r=0.216, P=0.023). There was no significantly correlation between dose and serum albumin or stable INR. The multiple linear regression analysis showed that age, body weight, genotype had independent and significant contributions to the overall variability in warfarin dose, while height showed no significant effect on warfarin dose. The multivariate regression model including the variables age, body weight, CYP2C9 and VKORCl genotype produce the best model for estimating individual warfarin dose (r2=0.550, P<0.0005), the algorithm of warfarin dose was as follows:Dose(Ygenetic)=2.791-0.0127 X age(y)+0.0239 X body weight(kg)+0.878×CYP2C9(input 1 for*1/*1, and input 0 for*1/*3)-1.792 X VKORC1(input 1 forAA, and input 0 for GA or GG).And the algorithm that only including the clinical factors could only explain 12.8% of the variability in response to warfarin (r2=0.128, P<0.0005). the algorithm was as follows:Dose(YcIinicaI)=1.757-0.012 X age(y)+0.028 X body weight(kg).Genetic model accurately identified larger proportions of patients who required lower dose (<3mg/d) and of those who required higher dose (>4.5mg/d) to achieve the target INR than did the clinical model (43.5% vs.30.4%, among patients requiring <3mg/d, and 71.4% vs.0%, among those requiring>4.5mg/d). Thirdly, genetic model was validated by randomly dividing the whole patients data (156 patients whose data was complete) into 3 groups containing 58,51 and 47 samples, respectively. Each two of the three groups was combined to build a new model, and the remaining group constituted the validation data, which was used for testing the new model. The final model was still base on the whole data. The results showed that the percentage of patients whose predicted dose of warfarin was within 20% of the actual stable therapeutic dose was 59.6%(28/47),47.1%(24/51) and 55.2(32/58), mean value was 54.0%, just in accordance with that in the whole model (57.1%,89/156).In addition, to explore an appropriate clinical practice method on anticoagulation patients, we cooperated with clinical department, built a database containing the complete information of patients receiving warfarin therapy, and established an anticoagulation medication record and provided education for warfarin-treated patients. Moreover, we validated the genetic algorithm with several new patients, the result was also satisfying. Clinical pharmacy practices could improve patients'compliance and make them trust pharmacists more.It can be concluded that 1) The distribution of CYP2C9 and VKORC1 genotypes in Han nationality of China is significantly different from those in other nationalities of China as well as those in America-European or African population, but similar to those of Asian population.2) The distribution of CYP2C9 and VKORC1 in warfarin-treated patients was in accordance with that in healthy subjects. The CYP2C9 and VKORC1 genotypes were not associated with altered risk of thromboembolic disease.3) Warfarin maintenance dose was highly significantly correlated with age, body weight, CYP2C9 and VKORC1 polymorphisms, and the use of a pharmacogenetic algorithm could predict warfarin dose more accurately than those derived from a clinical algorithm.4) The clinical pharmacy practices on anticoagulation performed by clinical pharmacists could significantly improve patients'compliance, which made patients trust pharmacists more. 5) Detection of CYP2C9 and VKORC1 genotypes prior to warfarin medication, combined with pharmaceutical services on anticoagulation, has the potential to estimate the maintenance warfarin dose and shorten the time to achive the maintanence warfarin dose. |
PDF Full Text Request