Objectives:To investigate the relationships between CYP2C9 and VKORC1 polymorphism and the steady-state dose of warfarin in Chinese Han patients with nonvalvular atrial fibrillation.Methods:A total 544 Chinese Han patients with atrial fibrillation who received warfarin anticoagulant therapy in Department of Cardiology of Beijing Hospital, Tongren Hospital, Xuanwu Hospital, Anzhen Hospital and Tiantan Hospital were enrolled from January 2016 to may 2019. The genotype and allele frequency in exon 1, 2, 3, 7 of CYP2C9 gene and 1639 site of VKORC1 gene were analyzed; and the general information, warfarin steady-state dose and concomitant medication of patients were recorded.Results:There were four genotypes CYP2C9:CYP2C9*1*1 (93.57%, 509/544), CYP2C9*1 *2 (0.18%, 1/544), CYP2C9*1 *3 (5.88%, 32/544) and CYP2C9*1 *60 (0.37%, 2/544); while VKORC1 had three genotypes: AA (82.72%, 450/544), GA (15.99%, 87/544) and GG (1.29%, 7/544). After reaching the anticoagulation index (INR 2.0-3.0), the steady-state dose of warfarin was the highest in patients with CYP2C9 *1/*1 and VKORC1 GA/GG genotypes, reaching (3.70±1.34) mg/d. The lowest steady-state dose of warfarin was (2.17±0.29)mg/d in patients with both new mutations ( F=22.09, P<0.001). Multiple linear regression analysis showed that body surface area, use of amiodarone, CYP2C9 and VKORC1 genotypes were the independent influencing factors of warfarin steady-state dose ( t=4.44, -2.90, -6.96, 2.14; P<0.05) and the steady-state dose prediction model of warfarin was established. Conclusion:Body weight, height, body surface area, gender, smoking, and combination of amiodarone may significantly affect the steady-state dose of warfarin in patients. CYP2C9 and VKORC1 mutant genotypes were significantly related to the steady-state dose of warfarin. The prediction model based on genetic factors and other influencing factors may effectively predict the steady-state dose of warfarin in Han patients with atrial fibrillation.

Objective:To verify the accuracy of the International Warfarin Pharmacogenetics Consortium(IWPC)model, identify the effects of genetic and clinical factors on steady-state doses of Warfarin, and establish a Warfarin dose prediction model for the Han-Chinese population aged 75 years and over under the guidance of pharmacogenetics.Methods:A total of 544 Han-Chinese patients receiving Warfarin therapy for atrial fibrillation were divided into two groups: those aged 75 years and over(n=164)and those aged below 75 years(n=380). Data for the whole population and the two age groups were each substituted into the IWPC prediction model for accuracy verification.Demographic and clinical characteristics of 164 patients aged 75 years and over were recorded, and the genotypes of CYP2 C9 and VKORC1- G1639 A were detected by polymerase chain reaction.A new pharmacogenetic-guided dosing algorithm for the elderly was obtained by stepwise multiple linear regression.The accuracy of the new model was compared with that of the IWPC model. Results:The predictive accuracy of IWPC for steady-state dosing of warfarin was 35.47% in all subjects, 33.75% in 164 subjects aged below 75 years, and only 28.70% in subjects aged 75 years and over, respectively.In 164 subjects aged 75 years and over, three genotypes of *1/*1, *1/*3 and *1/*2 were detected in CYP2 C9 polymorphism, and the CYP2 C9*1/*1 genotype was the most common one, with a frequency of 87.80%(144/164), followed by the CYP2 C9*1/*3 genotype, at 11.59%(19/164). GG, GA and AA genotypes were detected in VKORC1 polymorphism, among which the AA genotype accounted for 82.32%(135/164)and the GG genotype accounted for only 1.83%(3/164). The steady state dose for Warfarin in patients with the wild-type CYP2 C9*1/*1 was higher than in those with the heterozygote CYP2 C9*1/*3 and *1/*2(3.18±0.86 mg/d vs.2.27±0.51 mg/d, t=5.637, P<0.05). Patients with a mutant homozygotic AA genotype of VKORC1 required lower maintenance doses than those with the heterozygotic GA and GG genotypes(2.96±0.66 mg/d vs. 3.59±1.43 mg/d, t=-2.092, P<0.05). The steady-state dose for Warfarin in subjects carrying CYP2 C9 (*1/*2 or *3)and VKORC1 (GA and GG)was(2.00±0.63)mg/d, lower than in those carrying other genotype combinations( P<0.05). We established a new Warfarin dosing algorithm for elderly subjects aged 75 years and over containing height, creatinine, amiodarone usage, CYP2 C9 and VKORC1 mutants, and the accuracy of the new model was 56.0%, which could explain 56.0% of individual variability, and the accuracy was higher than that of the IWPC algorithm(56.0% vs. 45.8%, P<0.05). Conclusions:Polymorphisms of CYP2 C9 and VKORC1 clearly affect the steady-state dose for Warfarin in the elderly Han-Chinese population aged 75 years and over.A combination of pharmacogenomics with clinical factors can better guide warfarin medication in Han-Chinese people aged 75 years and over.