Evolutionary synthetic minority oversampling technique with random forest for warfarin dose prediction in Chinese patients

SummaryPharmacogenetic (PG) dosing algorithms have been confirmed to predict warfarin therapeutic dose more accurately;however, most of them are based on standard intensity of warfarin anticoagulation, and their utility outside this range is limited. This study was designed to develop and validate a PG refinement algorithm in Chinese patients mainly under low-intensity warfarin anticoagulation. Consented Chinese-Han patients (n=310) under stable warfarin treatment were randomly divided into a derivation (n=207) and a validation cohort (n=103), with 83% and 80% of the patients under low-intensity anticoagulation, respectively. In the derivation cohort, a PG algorithm was constructed on the basis of genotypes (CYP2C9*3 and VKORC1–1639A/G) and clinical data. After integrating additional covariates of international normalised ratio (INR) values (INR on day 4 of therapy and target INR) and genotype of CYP4F2 (rs2108622), a PG refinement algorithm was established and explained 54% of warfarin dose variability. In the validation cohort, warfarin dose prediction was more accurate (p <0.01) with the PG refinement algorithm than with the PG algorithm and the fixed dose approach (3 mg/day). In the entire cohort, the PG refinement algorithm could accurately identify larger proportions of patients with lower dose requirement (≤2 mg/day) and higher dose requirement (≥4 mg/day) than did the PG algorithm. In conclusion, PG refinement algorithm integrating early INR response and three genotypes CYP2C9*3, VKORC1–1639A/G, CYP4F2 rs2108622) improves the accuracy of warfarin dose prediction in Chinese patients mainly under low-intensity anticoagulation.

Download Full-text

A post-hoc interpretable ensemble model to feature effect analysis in warfarin dose prediction for Chinese patients

IEEE Journal of Biomedical and Health Informatics ◽

10.1109/jbhi.2021.3092170 ◽

2021 ◽

pp. 1-1

Author(s):

Yuzhen Zhang ◽

Cheng Xie ◽

Ling Xue ◽

Yanyun Tao ◽

Guoqi Yue ◽

...

Keyword(s):

Warfarin Dose ◽

Chinese Patients ◽

Ensemble Model ◽

Effect Analysis ◽

Dose Prediction ◽

Post Hoc

Download Full-text

Accuracy assessment of pharmacogenetic algorithms for warfarin dose prediction in Chinese patients

American Journal of Hematology ◽

10.1002/ajh.23151 ◽

2012 ◽

Vol 87 (5) ◽

pp. 541-544 ◽

Cited By ~ 5

Author(s):

Xiaoying Lei ◽

Yanhai Guo ◽

Jianbin Sun ◽

Heping Zhou ◽

Yonglan Liu ◽

...

Keyword(s):

Accuracy Assessment ◽

Warfarin Dose ◽

Chinese Patients ◽

Dose Prediction

Download Full-text

An Ensemble Model With Clustering Assumption for Warfarin Dose Prediction in Chinese Patients

IEEE Journal of Biomedical and Health Informatics ◽

10.1109/jbhi.2019.2891164 ◽

2019 ◽

Vol 23 (6) ◽

pp. 2642-2654 ◽

Cited By ~ 3

Author(s):

Yanyun Tao ◽

Yenming J. Chen ◽

Ling Xue ◽

Cheng Xie ◽

Bin Jiang ◽

...

Keyword(s):

Warfarin Dose ◽

Chinese Patients ◽

Ensemble Model ◽

Dose Prediction

Download Full-text

The Negative Impact of Biological Variation in the Effect and Clearance of Warfarin on Methods for Prediction of Dose Requirements

Thrombosis and Haemostasis ◽

10.1055/s-0038-1661685 ◽

1986 ◽

Vol 56 (03) ◽

pp. 371-375 ◽

Cited By ~ 7

Author(s):

Peretz Weiss ◽

Hillel Halkin ◽

Shlomo Almog

Keyword(s):

Individual Variation ◽

Negative Impact ◽

Warfarin Dose ◽

Empirical Therapy ◽

Maintenance Phase ◽

Individual Variability ◽

Dose Prediction ◽

Group I ◽

Over Time ◽

Variance Explained

SummaryWithin-individual variation over time in the clearance (Cl) and effect (PT%) of warfarin, was measured in 25 inpatients (group I) studied after standard single or individualized split loading doses and 1-3 times (n = 16) 8-16 weeks later during maintenance. Mean Cl (2.5 α 0.9 ml/min) was similar in both phases but significant changes occurred in 6/16 patients, exceeding those expected from within-individual variation alone (defined by its 95% tolerance limits -24% to +62%). Initial PT% (21 α 5) was unaffected by dosing schedule, total or free plasma warfarin, varying between patients by only 18-24%. Mean initial and maintenance dose-PT% ratios (8.2 mg/d: 21% and 4.1 mg/d: 40%) were similar but significant changes in sensitivity to warfarin occurred in 4/16 patients. In group I and 64 other outpatients on maintenance therapy, between-individual variability was 36-52% for Cl and 49-56% for effect. PT% correlated best (r = 0.56) with free and total plasma warfarin but poorly with dose (r = 0.29), with only 30% of PT% variance explained at best, due to high between patient variability.Warfarin dose prediction whether based on extrapolation from initial effects to the maintenance phase, or on iterative methods not allowing for between- or within-patient variation in warfarin clearance or effect which may occur independently over time, have not improved on empirical therapy. This, due to the elements of biological variability as well as the intricacy of the warfarin - prothrombin complex interaction not captured by any kinetic-dynamic model used for prediction to date.

Download Full-text

Pharmacogenetic algorithm for predicting daily dose of warfarin in Caucasian patients of Czech origin

Drug Metabolism and Personalized Therapy ◽

10.1515/dmdi-2020-0171 ◽

2020 ◽

Vol 0 (0) ◽

Author(s):

Aleš Tomek ◽

Tereza Růžičková ◽

Vojtěch Kaplan ◽

Zuzana Lacinová ◽

Simona Kumstýřová ◽

...

Keyword(s):

Validation Cohort ◽

Warfarin Dose ◽

Therapeutic Index ◽

Coefficient Of Determination ◽

Dose Prediction ◽

Pharmacogenetic Algorithm ◽

Daily Dose ◽

Caucasian Patients ◽

Variant Alleles ◽

Initiation Of Therapy

Abstract Objectives Warfarin use is limited by a low therapeutic index and significant interindividual variability of the daily dose. The most important factor predicting daily warfarin dose is individual genotype, polymorphisms of genes CYP2C9 (warfarin metabolism) and VKORC1 (sensitivity for warfarin). Algorithms using clinical and genetic variables could predict the daily dose before the initiation of therapy. The aim of this study was to develop and validate an algorithm for the prediction of warfarin daily dose in Czech patients. Methods Detailed clinical data of patients with known and stable warfarin daily dose were collected. All patients were genotyped for polymorphisms in genes CYP2C9 and VKORC1. Results Included patients were divided into derivation (n=175) and validation (n=223) cohorts. The final algorithm includes the following variables: Age, height, weight, treatment with amiodarone and presence of variant alleles of genes CYP2C9 and VKORC1. The adjusted coefficient of determination is 72.4% in the derivation and 62.3% in the validation cohort (p<0.001). Conclusions Our validated algorithm for warfarin daily dose prediction in our Czech cohort had higher precision than other currently published algorithms. Pharmacogenetics of warfarin has the potential in the clinical practice in specialized centers.

Download Full-text