Abstract:Objectives Multiple warfarin pharmacogenetic algorithms have been published and been confirmed to predict warfarin dose more accurately than clinical algorithm or fixed-dose approach. However, their performance in Chinese patients has never been evaluated. We sought to compare the performance of published warfarin pharmacogenetic algorithms in a cohort of Chinese patients under warfarin anticoagulation. Methods The study population consisted of 282 unrelated Han-Chinese patients consecutively admitted to the General Hospital of Chinese People9s Liberation Army between January 2008 and July 2011. The patients had all undergone warfarin anticoagulation for various indications and were stably controlled with a target international normalisation ratio (INR) of 1.6–2.5. Genomic DNA extraction from each patient was followed by genotyping to detect the polymorphisms of CYP2C9*3 (rs1057910), VKORC1 −1639 G/A (rs9923231), and CYP4F2*3 (rs2108622) respectively, using standard techniques. A total of 8 eligible pharmacogenetic algorithms were selected for performance comparison. The performance of each algorithm was evaluated by calculating the percentage of patients whose predicted dose fell within 20% of their actual therapeutic dose (percentage within 20%), and the mean absolute error (MAE) between each predicted dose and actual stable dose. Results In the entire cohort, the pharmacogenetic algorithms could predict warfarin dose with the average MAE of 0.87±0.17 mg/day (0.73–1.17 mg/day), and the average percentage within 20% of 43.8%±8.1% (29.1–52.1%). By pair wise comparison, warfarin dose prediction was significantly more accurate with the algorithms derived from Asian patients (48.6–50.0%) than those from Caucasian patients (29.1–39.7%; OR 1.61–3.36, p≤0.02). Algorithms with additional covariates of INR values or CYP4F2*3 performed better than those without the covariates (adding INR: OR: 1.71 (1.08–2.72), p=0.029; adding CYP4F2*3: OR: 2.67 (1.41–5.05), p=0.004). When the patients were stratified according to the dose range, the algorithms from Caucasian and racially mixed populations tended to perform better in higher dose group (≥4.5 mg/day), and algorithms from Asian populations performed better in intermediate dose group (1.5–4.5 mg/day). None of the algorithms performed well in lower dose group (≤1.5 mg/day). Conclusions Our study indicated that none of the eligible pharmacogenetic algorithms could perform the best for all dosing ranges in the present cohort of Chinese patients under warfarin anticoagulation. It may be important to consider the ethnic and clinical specific characteristics when choosing the appropriate algorithm for a local service population. A more refinement pharmacogenetic algorithm combining 3 genotypes (CYP2C9, VKORC1 and CYP4F2) and clinical factors with INR response could potentially improve the warfarin dose prediction in Chinese patients.

Comparison of multivariate linear regression and a machine learning algorithm developed for prediction of precision warfarin dosing in a Korean population

Machine learning guided prediction of warfarin blood levels for personalized medicine based on clinical longitudinal data from cardiac surgery patients: a prospective observational study

Nonlinear Machine Learning in Warfarin Dose Prediction: Insights from Contemporary Modelling Studies

Building and analyzing machine learning-based warfarin dose prediction models using scikit-learn

Estimate the Warfarin Dose by Ensemble of Machine Learning Algorithms

An ensemble learning based framework to estimate warfarin maintenance dose with cross-over variables exploration on incomplete data set

Warfarin- A Natural Anticoagulant: A Review of Research Trends for Precision Medication

Comparative performance of warfarin pharmacogenetic algorithms in Chinese patients.

A Post-Hoc Interpretable Ensemble Model to Feature Effect Analysis in Warfarin Dose Prediction for Chinese Patients

Comparative Performance of Gene-Based Warfarin Dosing Algorithms in Chinese Patients

Accuracy Assessment of Pharmacogenetic Algorithms for Warfarin Dose Prediction in Chinese Patients

The application of a perceptron model to classify an individual's response to a proposed loading dose regimen of Warfarin

Stable warfarin dose prediction in sub-Saharan African patients: A machine-learning approach and external validation of a clinical dose-initiation algorithm

Long Short-Term Memory Network for Development and Simulation of Warfarin Dosing Model Based on Time Series Anticoagulant Data.

Development of A Novel Individualized Warfarin Dose Algorithm Based on A Population Pharmacokinetic Model with Improved Prediction Accuracy for Chinese Patients after Heart Valve Replacement

Warfarin dose estimation on multiple datasets with automated hyperparameter optimisation and a novel software framework

Development of a Machine Learning Model and Mobile Application to Aid in Predicting Dosage of Vitamin K Antagonists Among Indian Patients

A Computer-Aided System for Determining the Application Range of a Warfarin Clinical Dosing Algorithm Using Support Vector Machines with a Polynomial Kernel Function

Optimizing warfarin dosing for patients with atrial fibrillation using machine learning

Comparative performance of pharmacogenetics-based warfarin dosing algorithms in Chinese population: use of a pharmacokinetic/pharmacodynamic model to explore dosing regimen through clinical trial simulation

Improving Cardiovascular Risk Prediction Through Machine Learning Modelling of Irregularly Repeated Electronic Health Records