Pharmacogenetic Studies Of Factors Affecting Warfarin Stable Dosage In Chinese Patients

Warfarin is the most widely used oral anticoagulant worldwide. It has a narrow therapeutic index, it is effective over wide dose variation and its stable dosage is influenced by a variety of factors. Close monitoring of prothrombin time (PT) standardized by the international normalized ratio (INR) is routinely carried out in clinical practice. Nevertheless, warfarin has been reported to be one of the ten drugs with the largest number of serious adverse events.Fixed dosing experience-based practices are common in the initiation of warfarin therapy, which may result in a high risk of bleeding or thrombosis, especially for patients who require extremely low or high dose of warfarin. In recent years, there have been efforts to use genotype information to create algorithms which could offer more precise estimates of safe and effective initial dosage for the individual patient. Most of these prediction algorithms are multiple linear regression equations with warfarin stable dosage the dependent variable and the genetic and non-genetic factors as independent variables. A very few prospective studies have confirmed that genotype-guided warfarin dosing algorithms can shorten the time to reach warfarin stable dosage and reduce outpatients’risk of hospitalization due to bleeding or thrombosis events caused by taking warfarin.Vitamin K epoxide reductase (VKOR) is the target of warfarin and CYP2C9is the main enzyme catalyzing warfarin. Numerous studies have shown that VKORC1-1639G>A and CYP2C9*2/*3genotypes are the major determinants of warfarin stable dosage. On average, VKORC1-1639G>A and CYP2C9*2/*3polymorphisms account respectively for27%and12%of variation in warfarin dose requirement. CYP4F2reduces hepatic-vitamin K1level via oxidation. However, the effect of CYP4F2(rs2108622) polymorphism on warfarin stable dosage is still controversial. POR is the only enzyme transporting electrons from NADPH to all CYP450enzyme families. POR activity is positively correlated with CYP2C9activity. POR polymorphisms have been shown to affect POR activity and influence CYP450-mediated drug metabolism. Therefore, POR polymorphisms may influence warfarin stable dosage.Chinese patients are known to have40%to50%lower wararin dose requirement than Caucasians. Also, Chinese patients have a lower target INR range compared to other ethnicities. Hence, warfarin dose prediction algorithms derived from data of Caucasians or from multiple populations may not be suitable for Chinese patients. Rheumatic heart disease affects at least2million adults in China, many of whom require HVR (heart valve replacement) surgery as their disease progresses. The target INR range for HVR patients is higher than that of those with other warfarin indications. Though numerous warfarin stable dosage prediction algorithms have been published, very few of them were specifically derived from data of HVR patients.Based on the above mentioned, we aimed to investigate the genetic and non-genetic determinants of wafarin stable dosage in Chinese patients undergoing HVR; to derive a warfarin stable dosage prediction algorithm from data of Chinese HVR patients on stable maintenance dose of warfarin; to validate and compare the performance of our derived algorithm with previously published warfarin dosing algorithms using the data of Chinese HVR patients; and to evaluate the impacts of CYP4F2C>T (rs2108622) and POR (rs72553971, rs1057868and rs2108622) polymorphisms on warfarin stable dosage in Chinese patients undergoing HVR.The findings in our studies were as below: 1. We confirmed that VKORC1-1639G>A and CYP2C9*3genotypes were the major determinants affecting warfarin stable dosage.2. We verified that the following non-genetic factors were minor determinants of warfarin stable dosage:body surface area, age, number of increasing INR drugs, smoking habit, preoperative stroke history and hypertension.3. We derived a warfarin dose prediction algorithm, designated XY algorithm, which included8factors:VKORC1-1639G>A, CYP2C9*3, body surface area, age, number of increasing INR drugs, smoking habit, preoperative stroke history and hypertension. Our XY algorithm accounted for56.4%of variation in the inter-patient warfarin stable doses.4. We evaluated the performance of up to14previously published warfarin dose prediction algorithms using the data of Chinese HVR patients. In comparison with the14previously published algorithms, our XY algorithm had the lowest MAE (-0.07mg/day) and the highest percentage of ideal prediction (62.8%) in the total validation cohort. Our XY algorithm generally performed well across the low, medium and high dose ranges.5. We confirmed that the CYP4F2C>T (rs2108622) genotype was a minor genetic determinant of warfarin stable dosage. The average warfarin stable dosage of patients harboring the wild CYP4F2CC genotype was about0.4mg/d lower than that of patients harboring at least one CYP4F2T allele. CYP4F2(rs2108622) polymorphism accounted for3.4%of variation in warfarin dose requirement in our study.6. We were the first to demonstrate that POR polymorphisms (rs72553971, rs1057868and rs2868177) did not significantly influence warfarin stable dosage in Chinese patients. Briefly, we systematically identified the genetic and non-genetic determinants of wafarin stable dosage in Chinese patients undergoing HVR. We believe the algorithm derived in our study is potentially useful in clinical practice. Although the CYP4F2(rs2108622) genotype is a minor genetic determinant of warfarin stable dosage, including this factor to develop a warfarin dosing algorithm should improve the algorithm precision.