Warfarin dosing algorithm using clinical, demographic and pharmacogenetic data from Chinese patients

CYP2C9 and VKORC1 genotypes could be used to predict warfarin requirement. The objective was to develop and validate a warfarin dosing algorithm using genetic, clinical and demographic data of Chinese patients from an anticoagulation clinic in Hong Kong. Blood samples were collected from 100 patients on stable maintenance dose of warfarin, recruited from an anticoagulation clinic, for genotyping CYP2C9 and VKORC1. Clinical and demographic data were obtained by face-to-face interview and medical chart review. Data of 80 patients (study cohort) were randomly selected for deriving a dosing algorithm. Comparison between predicted dose and actual stable doses was conducted in a validation cohort (n = 20). Sixty-nine (69%) of all 100 patients were homozygous for VKORC1 1173-TT, 25 (25%) were VKORC1 1173-CT heterozygotes and six (6%) were homozygous for VKORC1 1173-CC. 6 (6%) patients were CYP2C9 1*/3* and 94 (94%) were CYP2C9 1*/1*. CYP2C9 and VKORC1 genotype, age, weight and vitamin K intake were identified by stepwise regression modelling to produce the best model for estimating warfarin dose (R (2) = 68%, P < 0.001). In the validation cohort (n = 20), actual stable dose was significantly associated with predicted dose (R = 0.6, P = 0.005). Five of 11 (45.6%) and 5/9 (55.6%) patients whose mean warfarin requirements were <= 3 mg/day and > 3 mg/day, respectively, were within < 20% of actual doses. In conclusion, a genotype-guided dosing algorithm for warfarin therapy was developed for Chinese patients to explain 68% of dosage variation. The predicted doses differed from the actual doses by no more than 20% in 50% of patients.