Comparison of the predictive abilities of pharmacogenetics-based warfarin dosing algorithms using seven mathematical models in Chinese patients

The Role of Vitamin D Metabolism Genes and Their Genomic Background in Shaping Cyclosporine A Dosage Parameters after Kidney Transplantation

Background: Kidney transplantation is followed by immunosuppressive therapy involving calcineurin inhibitors (CNIs) such as cyclosporin A. However, long-term high CNIs doses can lead to vitamin D deficiency, and genetic variations influencing vitamin D levels can indirectly impact the necessary CNIs dosage. This study investigates the impact of genetic variations of vitamin D binding protein (DBP) rs2282679 and CYP2R1 hydroxylase rs10741657 polymorphisms on the cyclosporin A dosage in kidney transplant recipients. Additional polymorphisims of genes that are predicted to influence the pharmacogenetic profile were included. Methods: Gene polymorphisms in 177 kidney transplant recipients were analyzed using data mining techniques, including the Random Forest algorithm and Classification and Regression Trees (C&RT). The relationship between the concentration/dose (C/D) ratio of cyclosporin A and genetic profiles was assessed to determine the predictive value of DBP rs2282679 and CYP2R1 rs10741657 polymorphisms. Results: Polymorphic variants of the DBP (rs2282679) demonstrated a strong predictive value for the cyclosporin A C/D ratio in post-kidney transplantation patients. By contrast, the CYP2R1 polymorphism (rs10741657) did not show predictive significance. Additionally, the immune response genes rs231775 CTLA4 and rs1800896 IL10 were identified as predictors of cyclosporin A response, though these did not result in statistically significant differences. Conclusions:DBP rs2282679 polymorphisms can significantly predict the cyclosporin A C/D ratio, potentially enhancing the accuracy of CNI dosing. This can help identify patient groups at risk of vitamin D deficiency, ultimately improving the management of kidney transplant recipients. Understanding these genetic influences allows for more personalized and effective treatment strategies, contributing to better long-term outcomes for patients.

Machine learning in medication prescription: A systematic review

BACKGROUND

Medication prescription is a complex process that could benefit from current research and development in machine learning through decision support systems. Particularly pediatricians are forced to prescribe medications "off-label" as children are still underrepresented in clinical studies, which leads to a high risk of an incorrect dose and adverse drug effects.

METHODS

PubMed, IEEE Xplore and PROSPERO were searched for relevant studies that developed and evaluated well-performing machine learning algorithms following the PRISMA statement. Quality assessment was conducted in accordance with the IJMEDI checklist. Identified studies were reviewed in detail, including the required variables for predicting the correct dose, especially of pediatric medication prescription.

RESULTS

The search identified 656 studies, of which 64 were reviewed in detail and 36 met the inclusion criteria. According to the IJMEDI checklist, five studies were considered to be of high quality. 19 of the 36 studies dealt with the active substance warfarin. Overall, machine learning algorithms based on decision trees or regression methods performed superior regarding their predictive power than algorithms based on neural networks, support vector machines or other methods. The use of ensemble methods like bagging or boosting generally enhanced the accuracy of the dose predictions. The required input and output variables of the algorithms were considerably heterogeneous and differ strongly among the respective substance.

CONCLUSIONS

By using machine learning algorithms, the prescription process could be simplified and dosing correctness could be enhanced. Despite the heterogenous results among the different substances and cases and the lack of pediatric use cases, the identified approaches and required variables can serve as an excellent starting point for further development of algorithms predicting drug doses, particularly for children. Especially the combination of physiologically-based pharmacokinetic models with machine learning algorithms represents a great opportunity to enhance the predictive power and accuracy of the developed algorithms.

Developing Chinese race-specific warfarin dose prediction algorithms

BACKGROUND

Numerous genotype-guided warfarin dosing algorithms have been developed to individualize warfarin doses, but they can only explain 47-52% of the variability.

AIM

This study aimed to develop new warfarin algorithms suitable to predict the stable warfarin dose for the Chinese population and to compare their prediction performance with those of the most commonly used algorithms.

METHOD

Multiple linear regression analysis with the warfarin optimal dose (WOD), logarithm (log) WOD, 1/WOD, and [Formula: see text], respectively, as the dependent variables were performed to deduce a new warfarin algorithm (NEW-Warfarin). WOD was the stable dose that maintained the international normalized ratio (INR) within the target range (2.0-3.0). Three major genotype-guided warfarin dosing algorithms were selected and compared against NEW-Warfarin predictive performance using the mean absolute error (MAE). Furthermore, patients were divided into five groups according to warfarin indications [atrial fibrillation (AF), pulmonary embolism (PE), cardiac-related disease (CRD), deep vein thrombosis (DVT), and other diseases (OD)]. Multiple linear regression analyses were also performed for each group.

RESULTS

The regression equation with [Formula: see text] as the dependent variable had the highest coefficient of determination (R² = 0.489). The NEW-Warfarin had the best predictive accuracy compared to the three algorithms selected. Group analysis, according to indications, showed that the R² of the five groups were PE (0.902) > DVT (0.608) > CRD (0.569) > OD (0.436) > AF (0.424).

CONCLUSION

Dosing algorithms based on warfarin indications are more suitable for predicting warfarin doses. Our research provides a novel strategy to develop indication-specific warfarin dosing algorithms to improve the efficacy and safety of warfarin prescribing.

A Real-World Exploration into Clinical Outcomes of Direct Oral Anticoagulant Dosing Regimens in Morbidly Obese Patients Using Data-Driven Approaches

INTRODUCTION

The clinical outcomes of direct oral anticoagulant (DOAC) dosage regimens in morbid obesity are uncertain due to limited clinical evidence. This study seeks to bridge this evidence gap by identifying the factors associated with clinical outcomes following the dosing of DOACs in morbidly obese patients.

METHOD

A data-driven observational study was carried out using supervised machine learning (ML) models with a dataset extracted from electronic health records and preprocessed. Following 70%:30% partitioning of the overall dataset via stratified sampling, the selected ML classifiers (e.g., random forest, decision trees, bootstrap aggregation) were applied to the training dataset (70%). The outcomes of the models were evaluated against the test dataset (30%). Multivariate regression analysis explored the association between DOAC regimens and clinical outcomes.

RESULTS

A sample of 4,275 morbidly obese patients was extracted and analysed. The decision trees, random forest, and bootstrap aggregation classifiers achieved acceptable (excellent) values of precision, recall, and F1 scores in terms of their contribution to clinical outcomes. The length of stay, treatment days, and age were ranked highest for relevance to mortality and stroke. Among DOAC regimens, apixaban 2.5 mg twice daily ranked highest for its association with mortality, increasing the mortality risk by 43% (odds ratio [OR] 1.430, 95% confidence interval [CI] 1.181-1.732, p = 0.001). On the other hand, apixaban 5 mg twice daily reduced the odds of mortality by 25% (OR 0.751, 95% CI 0.632-0.905, p = 0.003) but increased the odds of stroke events. No clinically relevant non-major bleeding events occurred in this group.

CONCLUSION

Data-driven approaches can identify key factors associated with clinical outcomes following the dosing of DOACs in morbidly obese patients. This will help design further studies to explore well tolerated and effective DOAC doses for morbidly obese patients.

Automated warfarin dose prediction for Asian, American, and Caucasian populations using a deep neural network

Existing warfarin dose prediction algorithms based on pharmacogenetics and clinical parameters have not been used clinically due to the absence of external validation, lack of assessment for clinical utility, and high risk of bias. Moreover, given the high degree of heterogeneity across different datasets used to develop these algorithms, it is unsurprising that prediction errors remain high, and dosing accuracy is dependent on specific ethnic populations. To circumvent these challenges, deep neural models are increasingly used to improve the precision and accuracy of warfarin dose predictions. Hence, this study sought to develop a deep learning-based model using a well-established curated dataset of over 6000 patients from the International Warfarin Pharmacogenomics Consortium (IWPC). Clinically-relevant input data such as physical attributes, medical conditions, concomitant medications, genotype status of functional warfarin genetic polymorphisms, and therapeutic INR were entered followed by applying a unique and robust training and validation method. The deep model yielded a low average mean absolute error (MAE) of 7.6 mg/week and a relatively low mean percentage of error of 40.9% in Asians, 14.2 mg/week MAE and 36.9% in African Americans, and 12.7 mg/week MAE and 45.4% mean percentage of error in White Caucasians. This model also resulted in 36.4% of all patients with a predicted dose within 20% of the administered dose. Hence, our proposed deep model provides an alternative to predicting warfarin dose in the clinical setting upon validation in ethnically-similar datasets.

Association of Cortical Superficial Siderosis with Post-Stroke Epilepsy

OBJECTIVE

To assess whether post-stroke epilepsy (PSE) is associated with neuroimaging findings of hemosiderin in a case-control study, and whether the addition of hemosiderin markers improves the risk stratification models of PSE.

METHODS

We performed a post-hoc analysis of the PROgnosis of POST-Stroke Epilepsy study enrolling PSE patients at National Cerebral and Cardiovascular Center, Osaka, Japan, from November 2014 to September 2019. PSE was diagnosed when one unprovoked seizure was experienced >7 days after the index stroke, as proposed by the International League Against Epilepsy. As controls, consecutive acute stroke patients with no history or absence of any late seizure or continuing antiseizure medications at least 3 months after stroke were retrospectively enrolled during the same study period. We examined cortical microbleeds and cortical superficial siderosis (cSS) using gradient-echo T2*-weighted images. A logistic regression model with ridge penalties was tuned using 10-fold cross-validation. We added the item of cSS to the existing models (SeLECT and CAVE) for predicting PSE and evaluated performance of new models.

RESULTS

The study included 180 patients with PSE (67 women; median age 74 years) and 1,183 controls (440 women; median age 74 years). The cSS frequency was higher in PSE than control groups (48.9% vs 5.7%, p < 0.0001). Compared with the existing models, the new models with cSS (SeLECT-S and CAVE-S) demonstrated significantly better predictive performance of PSE (net reclassification improvement 0.63 [p = 0.004] for SeLECT-S and 0.88 [p = 0.001] for CAVE-S at the testing data).

INTERPRETATION

Cortical superficial siderosis was associated with PSE, stratifying stroke survivors at high risk of PSE. ANN NEUROL 2023;93:357-370.

Optimizing the dynamic treatment regime of in-hospital warfarin anticoagulation in patients after surgical valve replacement using reinforcement learning

OBJECTIVE

Warfarin anticoagulation management requires sequential decision-making to adjust dosages based on patients' evolving states continuously. We aimed to leverage reinforcement learning (RL) to optimize the dynamic in-hospital warfarin dosing in patients after surgical valve replacement (SVR).

MATERIALS AND METHODS

10 408 SVR cases with warfarin dosage-response data were retrospectively collected to develop and test an RL algorithm that can continuously recommend daily warfarin doses based on patients' evolving multidimensional states. The RL algorithm was compared with clinicians' actual practice and other machine learning and clinical decision rule-based algorithms. The primary outcome was the ratio of patients without in-hospital INRs >3.0 and the INR at discharge within the target range (1.8-2.5) (excellent responders). The secondary outcomes were the safety responder ratio (no INRs >3.0) and the target responder ratio (the discharge INR within 1.8-2.5).

RESULTS

In the test set (n = 1260), the excellent responder ratio under clinicians' guidance was significantly lower than the RL algorithm: 41.6% versus 80.8% (relative risk [RR], 0.51; 95% confidence interval [CI], 0.48-0.55), also the safety responder ratio: 83.1% versus 99.5% (RR, 0.83; 95% CI, 0.81-0.86), and the target responder ratio: 49.7% versus 81.1% (RR, 0.61; 95% CI, 0.58-0.65). The RL algorithms performed significantly better than all the other algorithms. Compared with clinicians' actual practice, the RL-optimized INR trajectory reached and maintained within the target range significantly faster and longer.

DISCUSSION

RL could offer interactive, practical clinical decision support for sequential decision-making tasks and is potentially adaptable for varied clinical scenarios. Prospective validation is needed.

CONCLUSION

An RL algorithm significantly optimized the post-operation warfarin anticoagulation quality compared with clinicians' actual practice, suggesting its potential for challenging sequential decision-making tasks.

Warfarin sensitivity is associated with increased hospital mortality in critically Ill patients

Background

Warfarin is a widely used anticoagulant with a narrow therapeutic index and large interpatient variability in the therapeutic dose. Warfarin sensitivity has been reported to be associated with increased incidence of international normalized ratio (INR) > 5. However, whether warfarin sensitivity is a risk factor for adverse outcomes in critically ill patients remains unknown. In the present study, we aimed to evaluate the utility of different machine learning algorithms for the prediction of warfarin sensitivity and to determine the impact of warfarin sensitivity on outcomes in critically ill patients.

Methods

Nine different machine learning algorithms for the prediction of warfarin sensitivity were tested in the International Warfarin Pharmacogenetic Consortium cohort and Easton cohort. Furthermore, a total of 7,647 critically ill patients was analyzed for warfarin sensitivity on in-hospital mortality by multivariable regression. Covariates that potentially confound the association were further adjusted using propensity score matching or inverse probability of treatment weighting.

Results

We found that logistic regression (AUC = 0.879, 95% CI: 0.834–0.924) was indistinguishable from support vector machine with a linear kernel, neural network, AdaBoost and light gradient boosting trees, and significantly outperformed all the other machine learning algorithms. Furthermore, we found that warfarin sensitivity predicted by the logistic regression model was significantly associated with worse in-hospital mortality in critically ill patients with an odds ratio (OR) of 1.33 (95% CI, 1.01–1.77).

Conclusions

Our data suggest that the logistic regression model is the best model for the prediction of warfarin sensitivity clinically and that warfarin sensitivity is likely to be a risk factor for adverse outcomes in critically ill patients.

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

Objective: This study aimed to systematically assess the characteristics and risk of bias of previous studies that have investigated nonlinear machine learning algorithms for warfarin dose prediction. Methods: We systematically searched PubMed, Embase, Cochrane Library, Chinese National Knowledge Infrastructure (CNKI), China Biology Medicine (CBM), China Science and Technology Journal Database (VIP), and Wanfang Database up to March 2022. We assessed the general characteristics of the included studies with respect to the participants, predictors, model development, and model evaluation. The methodological quality of the studies was determined, and the risk of bias was evaluated using the Prediction model Risk of Bias Assessment Tool (PROBAST). Results: From a total of 8996 studies, 23 were assessed in this study, of which 23 (100%) were retrospective, and 11 studies focused on the Asian population. The most common demographic and clinical predictors were age (21/23, 91%), weight (17/23, 74%), height (12/23, 52%), and amiodarone combination (11/23, 48%), while CYP2C9 (14/23, 61%), VKORC1 (14/23, 61%), and CYP4F2 (5/23, 22%) were the most common genetic predictors. Of the included studies, the MAE ranged from 1.47 to 10.86 mg/week in model development studies, from 2.42 to 5.18 mg/week in model development with external validation (same data) studies, from 12.07 to 17.59 mg/week in model development with external validation (another data) studies, and from 4.40 to 4.84 mg/week in model external validation studies. All studies were evaluated as having a high risk of bias. Factors contributing to the risk of bias include inappropriate exclusion of participants (10/23, 43%), small sample size (15/23, 65%), poor handling of missing data (20/23, 87%), and incorrect method of selecting predictors (8/23, 35%). Conclusions: Most studies on nonlinear-machine-learning-based warfarin prediction models show poor methodological quality and have a high risk of bias. The analysis domain is the major contributor to the overall high risk of bias. External validity and model reproducibility are lacking in most studies. Future studies should focus on external validity, diminish risk of bias, and enhance real-world clinical relevance.

Determining the adjusted initial treatment dose of warfarin anticoagulant medicine using kernel-based support vector regression

BACKGROUND AND OBJECTIVE

A novel research field in bioinformatics is pharmacogenomics and the corresponding applications of artificial intelligence tools. Pharmacogenomics is the study of the relationship between genotype and responses to medical measures such as drug use. One of the most effective drugs is warfarin anticoagulant, but determining its initial treatment dose is challenging. Mistakes in the determination of the initial treatment dose can result directly in patient death.

METHODS

Some of the most successful techniques for estimating the initial treatment dose are kernel-based methods. However, all the available studies use pre-defined and constant kernels that might not necessarily address the problem's intended requirements. The present study seeks to define and present a new computational kernel extracted from a data set. This process aims to utilize all the data-related statistical features to generate a dose determination tool proportional to the data set with minimum error rate. The kernel-based version of the least square support vector regression estimator was defined. Through this method, a more appropriate approach was proposed for predicting the adjusted dose of warfarin.

RESULTS AND CONCLUSION

This paper benefits from the International Warfarin Pharmacogenomics Consortium (IWPC) Database. The results obtained in this study demonstrate that the support vector regression with the proposed new kernel can successfully estimate the ideal dosage of warfarin for approximately 68% of patients.

Machine Learning: An Overview and Applications in Pharmacogenetics

This narrative review aims to provide an overview of the main Machine Learning (ML) techniques and their applications in pharmacogenetics (such as antidepressant, anti-cancer and warfarin drugs) over the past 10 years. ML deals with the study, the design and the development of algorithms that give computers capability to learn without being explicitly programmed. ML is a sub-field of artificial intelligence, and to date, it has demonstrated satisfactory performance on a wide range of tasks in biomedicine. According to the final goal, ML can be defined as Supervised (SML) or as Unsupervised (UML). SML techniques are applied when prediction is the focus of the research. On the other hand, UML techniques are used when the outcome is not known, and the goal of the research is unveiling the underlying structure of the data. The increasing use of sophisticated ML algorithms will likely be instrumental in improving knowledge in pharmacogenetics.

Performance Evaluation of Warfarin Dose Prediction Algorithms and Effects of Clinical Factors on Warfarin Dose in Chinese Patients

BACKGROUND

The clinical utility of warfarin dose prediction algorithms remains controversial, our purpose is to evaluate the performance of warfarin dose prediction algorithms and the effects of clinical factors on warfarin dose in Chinese patients.

METHODS

Clinical data of 217 patients who received warfarin treatment were used to assess 6 warfarin dose prediction algorithms (OHNO, IWPC [International Warfarin Pharmacogenetics Consortium], HUANG, KIM, BRESS, and MIAO). The predicted dose (PD) was compared with the warfarin optimal dose (WOD, defined as the dose that maintains the international normalized ratio within the target range of 2.0-3.0). A multiple regression analysis with WOD as the dependent variable was performed to evaluate the effects of clinical factors on warfarin dose.

RESULTS

The mean absolute error analysis ranked the predictive accuracies of the algorithms as OHNO > IWPC > HUANG > KIM > BRESS > MIAO. Stratified analysis indicated that HUANG most accurately predicted that patients required lower WODs (≤3 mg/d), whereas OHNO was the most effective in predicting medium WODs (3-5 mg/d). KIM was effective in predicting high WODs (>5 mg/d). Multiple linear regression analysis showed that VKORC1 (rs9923231) and body mass index were significantly positively correlated with WOD, whereas concurrent atrial fibrillation status, CYP2C9*3 (rs1057910), and sex were significantly negatively correlated with WOD.

CONCLUSIONS

In Chinese patients, OHNO should be given priority during the prediction and selection of warfarin dose. When using OHNO to predict warfarin dose (≤3 mg/d or >5 mg/d), HUANG or KIM algorithms can provide precise predictions. At the same time, physicians should pay close attention to clinical factors, such as VKORC1 (rs9923231), concurrent atrial fibrillation status, CYP2C9*3 (rs1057910), body mass index, and sex, to improve warfarin dose adjustment strategies in Chinese patients.

Development of a system to support warfarin dose decisions using deep neural networks

The first aim of this study was to develop a prothrombin time international normalized ratio (PT INR) prediction model. The second aim was to develop a warfarin maintenance dose decision support system as a precise warfarin dosing platform. Data of 19,719 inpatients from three institutions was analyzed. The PT INR prediction algorithm included dense and recurrent neural networks, and was designed to predict the 5th-day PT INR from data of days 1-4. Data from patients in one hospital (n = 22,314) was used to train the algorithm which was tested with the datasets from the other two hospitals (n = 12,673). The performance of 5th-day PT INR prediction was compared with 2000 predictions made by 10 expert physicians. A generator of individualized warfarin dose-PT INR tables which simulated the repeated administration of varying doses of warfarin was developed based on the prediction model. The algorithm outperformed humans with accuracy terms of within ± 0.3 of the actual value (machine learning algorithm: 10,650/12,673 cases (84.0%), expert physicians: 1647/2000 cases (81.9%), P = 0.014). In the individualized warfarin dose-PT INR tables generated by the algorithm, the 8th-day PT INR predictions were within 0.3 of actual value in 450/842 cases (53.4%). An artificial intelligence-based warfarin dosing algorithm using a recurrent neural network outperformed expert physicians in predicting future PT INRs. An individualized warfarin dose-PT INR table generator which was constructed based on this algorithm was acceptable.

Warfarin maintenance dose prediction for Chinese after heart valve replacement by a feedforward neural network with equal stratified sampling

Patients requiring low-dose warfarin are more likely to suffer bleeding due to overdose. The goal of this work is to improve the feedforward neural network model's precision in predicting the low maintenance dose for Chinese in the aspect of training data construction. We built the model from a resampled dataset created by equal stratified sampling (maintaining the same sample number in three dose-groups with a total of 3639) and performed internal and external validations. Comparing to the model trained from the raw dataset of 19,060 eligible cases, we improved the low-dose group's ideal prediction percentage from 0.7 to 9.6% and maintained the overall performance (76.4% vs. 75.6%) in external validation. We further built neural network models on single-dose subsets to invest whether the subsets samples were sufficient and whether the selected factors were appropriate. The training set sizes were 1340 and 1478 for the low and high dose subsets; the corresponding ideal prediction percentages were 70.2% and 75.1%. The training set size for the intermediate dose varied and was 1553, 6214, and 12,429; the corresponding ideal prediction percentages were 95.6, 95.1%, and 95.3%. Our conclusion is that equal stratified sampling can be a considerable alternative approach in training data construction to build drug dosing models in the clinic.

An Adapted Neural-Fuzzy Inference System Model Using Preprocessed Balance Data to Improve the Predictive Accuracy of Warfarin Maintenance Dosing in Patients After Heart Valve Replacement

BACKGROUND

Tailoring warfarin use poses a challenge for physicians and pharmacists due to its narrow therapeutic window and substantial inter-individual variability. This study aimed to create an adapted neural-fuzzy inference system (ANFIS) model using preprocessed balance data to improve the predictive accuracy of warfarin maintenance dosing in Chinese patients undergoing heart valve replacement (HVR).

METHODS

This retrospective study enrolled patients who underwent HVR between June 1, 2012, and June 1, 2016, from 35 centers in China. The primary outcomes were the mean difference between predicted warfarin dose by ANFIS models and actual dose and the models' predictive accuracy, including the ideal predicted percentage, the mean absolute error (MAE), and the mean squared error (MSE). The eligible cases were divided into training, internal validation, and external validation groups. We explored input variables by univariate analysis of a general linear model and created two ANFIS models using imbalanced and balanced training sets. We finally compared the primary outcomes between the imbalanced and balanced ANFIS models in both internal and external validation sets. Stratified analyses were conducted across warfarin doses (low, medium, and high doses).

RESULTS

A total of 15,108 patients were included and grouped as follows: 12,086 in the imbalanced training set; 2820 in the balanced training set; 1511 in the internal validation set; and 1511 in the external validation set. Eight variables were explored as predictors related to warfarin maintenance doses, and imbalanced and balanced ANFIS models with multi-fuzzy rules were developed. The results showed a low mean difference between predicted and actual doses (< 0.3 mg/d for each model) and an accurate prediction property in both the imbalanced model (ideal prediction percentage, 74.39-78.16%; MAE, 0.37 mg/daily; MSE, 0.39 mg/daily) and the balanced model (ideal prediction percentage, 73.46-75.31%; MAE, 0.42 mg/daily; MSE, 0.43 mg/daily). Compared to the imbalanced model, the balanced model had a significantly higher prediction accuracy in the low-dose (14.46% vs. 3.01%; P < 0.001) and the high-dose warfarin groups (34.71% vs. 23.14%; P = 0.047). The results from the external validation cohort confirmed this finding.

CONCLUSIONS

The ANFIS model can accurately predict the warfarin maintenance dose in patients after HVR. Through data preprocessing, the balanced model contributed to improved prediction ability in the low- and high-dose warfarin groups.

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

MOTIVATION

Warfarin is a widely used oral anticoagulant, but it is challenging to select the optimal maintenance dose due to its narrow therapeutic window and complex individual factor relationships. In recent years, machine learning techniques have been widely applied for warfarin dose prediction. However, the model performance always meets the upper limit due to the ignoration of exploring the variable interactions sufficiently. More importantly, there is no efficient way to resolve missing values when predicting the optimal warfarin maintenance dose.

METHODS

Using an observational cohort from the Xinhua Hospital affiliated to Shanghai Jiaotong University School of Medicine, we propose a novel method for warfarin maintenance dose prediction, which is capable of assessing variable interactions and dealing with missing values naturally. Specifically, we examine single variables by univariate analysis initially, and only statistically significant variables are included. We then propose a novel feature engineering method on them to generate the cross-over variables automatically. Their impacts are evaluated by stepwise regression, and only the significant ones are selected. Lastly, we implement an ensemble learning based approach, LightGBM, to learn from incomplete data directly on the selected single and cross-over variables for dosing prediction.

RESULTS

377 unique patients with eligible and time-independent 1173 warfarin order events are included in this study. Through the comprehensive experimental results in 5-fold cross-validation, our proposed method demonstrates the efficiency of exploring the variable interactions and modeling on incomplete data. The R² can achieve 75.0% on average. Moreover, the subgroup analysis results reveal that our method performs much better than other baseline methods, especially in the medium-dose and high-dose subgroups. Lastly, the IWPC dosing prediction model is used for further comparison, and our approach outperforms it by a significant margin.

CONCLUSION

In summary, our proposed method is capable of exploring the variable interactions and learning from incomplete data directly for warfarin maintenance dose prediction, which has a great premise and is worthy of further research.

Artificial Intelligence in Pharmacoepidemiology: A Systematic Review. Part 2-Comparison of the Performance of Artificial Intelligence and Traditional Pharmacoepidemiological Techniques

Aim: To summarize the evidence on the performance of artificial intelligence vs. traditional pharmacoepidemiological techniques.

Methods: Ovid MEDLINE (01/1950 to 05/2019) was searched to identify observational studies, meta-analyses, and clinical trials using artificial intelligence techniques having a drug as the exposure or the outcome of the study. Only studies with an available full text in the English language were evaluated.

Results: In all, 72 original articles and five reviews were identified via Ovid MEDLINE of which 19 (26.4%) compared the performance of artificial intelligence techniques with traditional pharmacoepidemiological methods. In total, 44 comparisons have been performed in articles that aimed at 1) predicting the needed dosage given the patient’s characteristics (31.8%), 2) predicting the clinical response following a pharmacological treatment (29.5%), 3) predicting the occurrence/severity of adverse drug reactions (20.5%), 4) predicting the propensity score (9.1%), 5) identifying subpopulation more at risk of drug inefficacy (4.5%), 6) predicting drug consumption (2.3%), and 7) predicting drug-induced lengths of stay in hospital (2.3%). In 22 out of 44 (50.0%) comparisons, artificial intelligence performed better than traditional pharmacoepidemiological techniques. Random forest (seven out of 11 comparisons; 63.6%) and artificial neural network (six out of 10 comparisons; 60.0%) were the techniques that in most of the comparisons outperformed traditional pharmacoepidemiological methods.

Conclusion: Only a small fraction of articles compared the performance of artificial intelligence techniques with traditional pharmacoepidemiological methods and not all artificial intelligence techniques have been compared in a Pharmacoepidemiological setting. However, in 50% of comparisons, artificial intelligence performed better than pharmacoepidemiological techniques.

LRP1 polymorphisms associated with warfarin stable dose in Chinese patients: a stepwise conditional analysis

Aim: The aim of this study was to investigate whether variability in warfarin stable dose (WSD) could be influenced by vitamin K-related polymorphisms in patients with heart valve replacement. Patients & methods: Twenty-nine vitamin K-related SNPs in 208 patients who initially took warfarin and achieved WSD were genotyped. Results: After conducting conditional analysis for both VKORC1 -1639G>A and CYP2C9*3, LRP1 rs1800139 and LRP1 rs1800154 were significantly associated with WSD (p = 0.007 and p = 0.015, respectively). Multivariate analysis showed that LRP1 rs1800139 accounted for 5.9% WSD variability. Conclusion: Our results suggest that a novel vitamin K-related gene, LRP1, exerts a relevant influence on WSD, independent of VKORC1 -1639G>A and CYP2C9*3.

Comparison of the prediction performance of different warfarin dosing algorithms based on Chinese patients

Aim: To compare the prediction performance of different warfarin dosing algorithms based on Chinese patients. Materials & methods: A total of 18 algorithms were tested in 325 patients. The predictive efficacy of selected algorithms was evaluated by calculating the percentage of patients whose predicted dose fell within ±20% of their actual stable warfarin dose and the mean absolute error. Results: The percentage within ± 20% and the mean absolute error of the algorithms ranged from 11.9 to 41.2% and -0.20 (-0.29 to -0.11) mg/d to -1.63 (-1.75 to -1.50) mg/d. The algorithms established by Miao et al. and Wei et al. had optimal predictive performance. Conclusion: Algorithms based on geographical populations might be more suitable for the prediction of stable warfarin doses in local patients.

Artificial Intelligence in Pharmacoepidemiology: A Systematic Review. Part 1-Overview of Knowledge Discovery Techniques in Artificial Intelligence

Aim

To perform a systematic review on the application of artificial intelligence (AI) based knowledge discovery techniques in pharmacoepidemiology.

Study Eligibility Criteria

Clinical trials, meta-analyses, narrative/systematic review, and observational studies using (or mentioning articles using) artificial intelligence techniques were eligible. Articles without a full text available in the English language were excluded.

Data Sources

Articles recorded from 1950/01/01 to 2019/05/06 in Ovid MEDLINE were screened.

Participants

Studies including humans (real or simulated) exposed to a drug.

Results

In total, 72 original articles and 5 reviews were identified via Ovid MEDLINE. Twenty different knowledge discovery methods were identified, mainly from the area of machine learning (66/72; 91.7%). Classification/regression (44/72; 61.1%), classification/regression + model optimization (13/72; 18.0%), and classification/regression + features selection (12/72; 16.7%) were the three most frequent tasks in reviewed literature that machine learning methods has been applied to solve. The top three used techniques were artificial neural networks, random forest, and support vector machines models.

Conclusions

The use of knowledge discovery techniques of artificial intelligence techniques has increased exponentially over the years covering numerous sub-topics of pharmacoepidemiology.

Systematic Review Registration

Systematic review registration number in PROSPERO: CRD42019136552.

Machine Learning Algorithm for Predicting Warfarin Dose in Caribbean Hispanics Using Pharmacogenetic Data

Despite some previous examples of successful application to the field of pharmacogenomics, the utility of machine learning (ML) techniques for warfarin dose predictions in Caribbean Hispanic patients has yet to be fully evaluated. This study compares seven ML methods to predict warfarin dosing in Caribbean Hispanics. This is a secondary analysis of genetic and non-genetic clinical data from 190 cardiovascular Hispanic patients. Seven ML algorithms were applied to the data. Data was divided into 80 and 20% to be used as training and test sets. ML algorithms were trained with the training set to obtain the models. Model performance was determined by computing the corresponding mean absolute error (MAE) and % patients whose predicted optimal dose were within ±20% of the actual stabilization dose, and then compared between groups of patients with “normal” (i.e., > 21 but <49 mg/week), low (i.e., ≤21 mg/week, “sensitive”), and high (i.e., ≥49 mg/week, “resistant”) dose requirements. Random forest regression (RFR) significantly outperform all other methods, with a MAE of 4.73 mg/week and 80.56% of cases within ±20% of the actual stabilization dose. Among those with “normal” dose requirements, RFR performance is also better than the rest of models (MAE = 2.91 mg/week). In the “sensitive” group, support vector regression (SVR) shows superiority over the others with lower MAE of 4.79 mg/week. Finally, multivariate adaptive splines (MARS) shows the best performance in the resistant group (MAE = 7.22 mg/week) and 66.7% of predictions within ±20%. Models generated by using RFR, MARS, and SVR algorithms showed significantly better predictions of weekly warfarin dosing in the studied cohorts than other algorithms. Better performance of the ML models for patients with “normal,” “sensitive,” and “resistant” to warfarin were obtained when compared to other populations and previous statistical models.

Precision dosing of warfarin: open questions and strategies

Warfarin has a very narrow therapeutic window and obvious interindividual variability in its effects, with many factors contributing to the body's response. Algorithms incorporating multiple genetic, environment and clinical factors have been established to select a precision dose for each patient. A number of randomized controlled trials (RCTs) were conducted to explore whether patients could benefit from these algorithms; however, the results were inconsistent. Some questions remain to be resolved. Recently, new genetic and non-genetic factors have been discovered to contribute to variability in optimal warfarin doses. The results of further RCTs have been unveiled, and guidelines for pharmacogenetically guided warfarin dosing have been updated. Based on these most recent advancements, we summarize some open questions in this field and try to propose possible strategies to resolve them.

Evaluation of the predictive performance of Bayesian dosing for warfarin in Chinese patients

Aim: To evaluate the accuracy and predictive performance of Bayesian dosing for warfarin in Chinese patients. Materials & methods: Six multiple linear regression algorithms (Wei, Lou, Miao, Huang, Gage and IWPC) and a Bayesian method implemented in Warfarin Dose Calculator were compared with each other. Results: Six multiple linear regression warfarin dosing algorithms had similar predictive ability, except Miao and Lou. The mean prediction error of Bayesian priori and posteriori method were 0.01 mg/day (95% CI: -0.18 to 0.19) and 0.17 mg/day (95% CI: -0.05 to 0.29), respectively, and Bayesian posteriori method demonstrated better performance in all dose ranges. Conclusion: The Bayesian method showed a good potential for warfarin maintenance dose prediction in Chinese patients requiring less than 6 mg/day.

A factor VII-based method for the prediction of anticoagulant response to warfarin

Warfarin dosing methods based on existing models for warfarin and the international normalised ratio (INR) give biased maintenance dose predictions at the upper and lower quantiles of dose requirements. The aim of this work is to propose a conceptually different approach to predict INR after warfarin dosing. Factor VII concentration was proposed as the principal driving force for the INR. The time to steady-state INR (t_SS,INR) was determined based on the INR response to changes in factor VII concentrations following warfarin initiation, and from this the steady-state INR (INR_SS) was derived. The proposed method requires timed, paired blood samples of INR and factor VII. At different simulated warfarin dose rates, the prediction error associated with the proposed method was shown to be within clinically acceptable limits for both the t_SS,INR (±2 days) and INR_SS (±0.2). The use of the method was demonstrated in two patients who were initiated with 5 mg of warfarin daily. The difference in predicted versus actual steady-state INR were 0.0 and −0.4. The proposed method represents a unique approach to predict the INR. It considers factor VII as the main driver for INR and provides valuable information about the time to steady state INR.

Warfarin maintenance dose Prediction for Patients undergoing heart valve replacement- a hybrid model with genetic algorithm and Back-Propagation neural network

Warfarin is the most recommended anticoagulant drug for patients undergoing heart valve replacement. However, due to the narrow therapeutic window and individual dose, the use of warfarin needs more advanced technology. We used the data collected from a multi-central registered clinical system all over China about the patients who have undergone heart valve replacement, subsequently divided into three groups (training group: 10673 cases; internal validation group: 3558 cases; external validation group: 1463 cases) in order to construct a hybrid model with genetic algorithm and Back-Propagation neural network (BP-GA), For testing the model's prediction accuracy, we used Mean absolute error (MAE), Root mean squared error (RMSE) and the ideal predicted percentage of total and dose subgroups. In results, whether in internal or in external validation group, the total ideal predicted percentage was over 58% while the intermediate dose subgroup manifested the best. Moreover, it showed higher prediction accuracy, lower MAE value and lower RMSE value in the external validation group than that in the internal validation group (p < 0.05). In conclusion, BP-GA model is promising to predict warfarin maintenance dose.

Clinical verification of Lou type warfarin pharmacokinetic dosing algorithms equation

Warfarin is the most commonly used oral anti-coagulant in clinic practice. However, it is difficult to recommend the correct dosage due to its narrow therapeutic window. The aim of the present study was to verify the clinical value of the Lou type equation, using pharmacogenetics‑based warfarin dosing algorithms to appropriately predict the actual maintenance dose. A total of 87 Chinese Han patients who required treatment with warfarin were enrolled and randomly divided into the experimental and control groups. In the experimental group, the first 3 doses of warfarin were calculated according to the Lou type equation. While in the control group, these 3 treatments were performed following the doctors' recommendations. Then the dose of warfarin was gradually adjusted to the stable dose according to the changes in the international standardized ratio. At the end of the 50 day experimental period, there were a greater number of patients in the experimental group who exhibited a stable blood concentration of warfarin than those in the control group (83.35 and 64.4%, respectively). In addition, the mean and median times for patients to obtain a stable dose in the experimental group were significantly shorter than those in the control group (mean, 18.2±1.7 and 27.3±2.0 days; and median, 11.7±1.1 and 20.5±1.8 days, respectively). The adverse reaction rate of the experimental group (9.5%) was markedly lower than that of the control group (26.7%). The occurrence of adverse reactions in the experimental group was also significantly later when compared with the control group (43.9±1.6 and 38.6±1.5 days, respectively). Furthermore, there was no significant difference between the average predicted dose (3.4±1.1 mg/day) and the average actual dose (3.5±1.4 mg/day; P=0.313). In conclusion, using the Lou type warfarin pharmacokinetic dosing algorithm equation to administer warfarin markedly shortened the adjustment time of warfarin to reach a stable dose and reduced the adverse reactions rate, thus supporting clinical feasibility.

A prediction study of warfarin individual stable dose after mechanical heart valve replacement: adaptive neural-fuzzy inference system prediction

Background

It’s difficult but urgent to achieve the individualized rational medication of the warfarin, we aim to predict the individualized warfarin stable dose though the artificial intelligent Adaptive neural-fuzzy inference system (ANFIS).

Methods

Our retrospective analysis based on a clinical database, involving 21,863 patients from 15 Chinese provinces who receive oral warfarin after the heart valve replacement. They were allocated into four groups: the external validation group (A group), the internal validation group (B group), training group (C group) and stratified training group (D group). We used a univariate analysis of general linear models(GLM-univariate) to select the input variables and construct two prediction models by the ANFIS with the training and stratified training group, and then verify models with two validation groups by the mean squared error(MSE), mean absolute error(MAE) and the ideal predicted percentage.

Results

A total of 13,639 eligible patients were selected, including 1639 in A group, 3000 in B group, 9000 in C group, and 3192 in D group. Nine input variables were selected out and two five-layered ANFIS models were built. ANFIS model achieved the highest total ideal predicted percentage 63.7%. In the dose subgroups, all the models performed best in the intermediate-dose group with the ideal predicted percentage 82.4~ 86.4%, and the use of the stratified training group slightly increased the prediction accuracy in low-dose group by 8.8 and 5.2%, respectively.

Conclusion

As a preliminary attempt, ANFIS model predicted the warfarin stable dose properly after heart valve surgery among Chinese, and also proved that Chinese need lower anticoagulation intensity INR (1.5–2.5) to warfarin by reference to the recommended INR (2.5–3.5) in the developed countries.

Impact of GGCX polymorphisms on warfarin dose requirements in atrial fibrillation patients

Background/aim: Warfarin is a common anticoagulant with large interindividual differences and a narrow therapeutic range. The polymorphisms of gamma-glutamyl carboxylase (GGCX) are important genetic factors for warfarin dose requirements. Materials and methods: Polymerase chain reaction and direct sequencing methods were used to detect the GGCX rs699664 genotype in 215 atrial fibrillation (AF) patients with warfarin administration. The effects on warfarin dose by different genotypes were analyzed. A warfarin dosing algorithm was developed based on age, height, CYP2C9, VKORC1, and GGCX genotype. Results: In 215 AF patients, there were 104 cases of wild-type GG genotype (48.4%), 92 cases of GA genotype (42.8%), and 19 cases of AA genotype (8.8%). Patients with the GGCX rs699664 A allele (GA or AA genotypes) needed higher warfarin doses than those with the GG genotype (P < 0.05). A warfarin dosing algorithm showed that age, height, CYP2C9, VKORC1, and GGCX genotype were the best variables for estimating warfarin dose (R2 = 41.2%). Another independent cohort of 60 AF patients showed a significant linear correlation between predicted warfarin maintenance dose and actual dose (R = 0.660, P < 0.01). Conclusion: AF patients with the GA and AA genotypes in GGCX rs699664 required significantly higher warfarin doses. GGCX rs699664 is a potential predictor for the warfarin dose of AF patients.

Warfarin Dosing Algorithms Underpredict Dose Requirements in Patients Requiring ≥7 mg Daily: A Systematic Review and Meta-analysis

There is preliminary evidence to suggest that some published warfarin dosing algorithms produce biased maintenance dose predictions in patients who require higher than average doses. We conducted a meta-analysis of warfarin dosing algorithms to determine if there exists a systematic under- or overprediction of dose requirements for patients requiring ≥7 mg/day across published algorithms. Medline and Embase databases were searched up to September 2015. We quantified the proportion of over- and underpredicted doses in patients whose observed maintenance dose was ≥7 mg/day. The meta-analysis included 47 evaluations of 22 different warfarin dosing algorithms from 16 studies. The meta-analysis included data from 1,492 patients who required warfarin doses of ≥7 mg/day. All 22 algorithms were found to underpredict warfarin dosing requirements in patients who required ≥7 mg/day by an average of 2.3 mg/day with a pooled estimate of underpredicted doses of 92.3% (95% confidence interval 90.3-94.1, I² = 24%).

Clinical application of a new warfarin-dosing regimen based on the CYP2C9 and VKORC1 genotypes in atrial fibrillation patients

The polymorphisms of cytochrome P450 2C9 (CYP2C9) and vitamin K epoxide reductase complex 1 (VKORC1) are important genetic factors for warfarin dose determinations. The present study aimed to investigate the contribution of the CYP2C9 and VKORC1 genotypes to warfarin dose requirement in atrial fibrillation (AF) patients, and to evaluate the clinical application of a warfarin-dosing algorithm. A total of 122 AF patients with a target international normalized ratio of 2.0 to 3.0 were included to determine the genotypes of CYP2C9 (rs1057910) and VKORC1 (rs9923231). A warfarin-dosing algorithm was developed based on age, height, and the CYP2C9 and VKORC1 genotypes of AF patients. The results indicated that the mean warfarin daily dose requirement was lower in the CYP2C9*1/*3 genotype compared with those in the homozygous wild-type CYP2C9*1/*1 patients (P<0.05), and was higher in patients with the VKORC1 AG and GG genotypes compared with those with the AA genotype (P<0.05). The multivariate regression model showed that age, height, and the CYP2C9 and VKORC1 genotypes were the best variables for estimating warfarin dose (R2=56.4%). A new warfarin-dosing algorithm was developed and its validity was confirmed in a second cohort of AF patients. During the 50-day follow-up, 63.3% (19/30) of control group patients and 86.7% (26/30) of patients in the experimental group acquired the warfarin maintenance dose. Among all the patients who acquired the warfarin maintenance dose, the mean time elapse from initiation until warfarin maintenance dose was significantly less in the experimental group (25.8±1.7 day) compared to the control group (33.1±1.9 day) (P<0.05). There was significant linear correlation between predicted warfarin maintenance dose and actual dose (r=0.822, P<0.01). In conclusion, a new warfarin-dosing algorithm was developed based on the CYP2C9 and VKORC1 genotypes, and it can shorten the time elapse from initiation until warfarin maintenance dose in AF patients with warfarin therapy.

Comparison of Nine Statistical Model Based Warfarin Pharmacogenetic Dosing Algorithms Using the Racially Diverse International Warfarin Pharmacogenetic Consortium Cohort Database

Objective

Multiple linear regression (MLR) and machine learning techniques in pharmacogenetic algorithm-based warfarin dosing have been reported. However, performances of these algorithms in racially diverse group have never been objectively evaluated and compared. In this literature-based study, we compared the performances of eight machine learning techniques with those of MLR in a large, racially-diverse cohort.

Methods

MLR, artificial neural network (ANN), regression tree (RT), multivariate adaptive regression splines (MARS), boosted regression tree (BRT), support vector regression (SVR), random forest regression (RFR), lasso regression (LAR) and Bayesian additive regression trees (BART) were applied in warfarin dose algorithms in a cohort from the International Warfarin Pharmacogenetics Consortium database. Covariates obtained by stepwise regression from 80% of randomly selected patients were used to develop algorithms. To compare the performances of these algorithms, the mean percentage of patients whose predicted dose fell within 20% of the actual dose (mean percentage within 20%) and the mean absolute error (MAE) were calculated in the remaining 20% of patients. The performances of these techniques in different races, as well as the dose ranges of therapeutic warfarin were compared. Robust results were obtained after 100 rounds of resampling.

Results

BART, MARS and SVR were statistically indistinguishable and significantly out performed all the other approaches in the whole cohort (MAE: 8.84–8.96 mg/week, mean percentage within 20%: 45.88%–46.35%). In the White population, MARS and BART showed higher mean percentage within 20% and lower mean MAE than those of MLR (all p values < 0.05). In the Asian population, SVR, BART, MARS and LAR performed the same as MLR. MLR and LAR optimally performed among the Black population. When patients were grouped in terms of warfarin dose range, all machine learning techniques except ANN and LAR showed significantly higher mean percentage within 20%, and lower MAE (all p values < 0.05) than MLR in the low- and high- dose ranges.

Conclusion

Overall, machine learning-based techniques, BART, MARS and SVR performed superior than MLR in warfarin pharmacogenetic dosing. Differences of algorithms’ performances exist among the races. Moreover, machine learning-based algorithms tended to perform better in the low- and high- dose ranges than MLR.

Validation of warfarin pharmacogenetic algorithms in 586 Han Chinese patients

AIM

To compare the accuracy of the nine selected algorithms for predicting warfarin dose with 586 Han Chinese patients.

MATERIALS & METHODS

Genotyping of VKORC1 1639G>A, CYP2C9*2 and CYP2C9*3 variants was performed. Both the mean absolute error and ideal estimation value were used for comparison.

RESULTS

The top three performers were from East Asians. The algorithms from Caucasians generally performed better in the medium-dose subgroup (>3 and <7 mg/day), while the algorithms from East Asians generally performed better in the low-dose subgroup (≤ 3 mg/day). None of the algorithms performed well in the high-dose subgroup (≥ 7 mg/day).

CONCLUSION

Algorithms built for specific ethnic groups and preassigned-dose groups are suggested for better prediction.