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

The polygenic implication of clopidogrel responsiveness: Insights from platelet reactivity analysis and next-generation sequencing

Clopidogrel is widely used worldwide as an antiplatelet therapy in patients with acute coronary disease. Genetic factors influence interindividual variability in response. Some studies have explored the polygenic contributions in the drug response, generating pharmacogenomic risk scores (PgxPRS). Importantly, these factors are less explored in underrepresented populations, such as Latin-American countries. Identifying patients at risk of high-on-treatment platelet reactivity (HTPR) is highly valuable in translational medicine. In this study we used a custom next-generation sequencing (NGS) panel composed of 91 single nucleotide polymorphisms (SNPs) and 28 genes related to clopidogrel metabolism, to analyze 70 patients with platelet reactivity values, assessed through closure time (CT). Our results demonstrated the association of SNPs with HTPR and non-HTPR, revealing the strongest associations with rs2286823 (OR: 5,0; 95% CI: 1,02-24,48; p: 0,03), rs2032582 (OR: 4,41; 95% CI: 1,20-16,12; p: 0,019), and rs1045642 (OR: 3,38; 95% CI: 0,96-11,9; p: 0,05). Bivariate regression analysis demonstrated the significant association of several SNPs with the CT value, a "surrogate" biomarker of clopidogrel response. Exploratory results from the LASSO regression model showed a high discriminatory capacity between HTPR and non-HTPR patients (AUC: 0,955), and the generated PgxPRS demonstrated a significant negative association between the risk score, CT value, and the condition of HTPR and non-HTPR. To our knowledge, our study addresses for the first time the analysis of the polygenic contribution in platelet reactivity using NGS and establishes PgxPRS derived from the LASSO model. Our results demonstrate the polygenic implication of clopidogrel response and offer insights applicable to the translational medicine of antiplatelet therapy in an understudied population.

Optimizing warfarin dosing for patients with atrial fibrillation using machine learning

While novel oral anticoagulants are increasingly used to reduce risk of stroke in patients with atrial fibrillation, vitamin K antagonists such as warfarin continue to be used extensively for stroke prevention across the world. While effective in reducing the risk of strokes, the complex pharmacodynamics of warfarin make it difficult to use clinically, with many patients experiencing under- and/or over- anticoagulation. In this study we employed a novel implementation of deep reinforcement learning to provide clinical decision support to optimize time in therapeutic International Normalized Ratio (INR) range. We used a novel semi-Markov decision process formulation of the Batch-Constrained deep Q-learning algorithm to develop a reinforcement learning model to dynamically recommend optimal warfarin dosing to achieve INR of 2.0-3.0 for patients with atrial fibrillation. The model was developed using data from 22,502 patients in the warfarin treated groups of the pivotal randomized clinical trials of edoxaban (ENGAGE AF-TIMI 48), apixaban (ARISTOTLE) and rivaroxaban (ROCKET AF). The model was externally validated on data from 5730 warfarin-treated patients in a fourth trial of dabigatran (RE-LY) using multilevel regression models to estimate the relationship between center-level algorithm consistent dosing, time in therapeutic INR range (TTR), and a composite clinical outcome of stroke, systemic embolism or major hemorrhage. External validation showed a positive association between center-level algorithm-consistent dosing and TTR (R2 = 0.56). Each 10% increase in algorithm-consistent dosing at the center level independently predicted a 6.78% improvement in TTR (95% CI 6.29, 7.28; p < 0.001) and a 11% decrease in the composite clinical outcome (HR 0.89; 95% CI 0.81, 1.00; p = 0.015). These results were comparable to those of a rules-based clinical algorithm used for benchmarking, for which each 10% increase in algorithm-consistent dosing independently predicted a 6.10% increase in TTR (95% CI 5.67, 6.54, p < 0.001) and a 10% decrease in the composite outcome (HR 0.90; 95% CI 0.83, 0.98, p = 0.018). Our findings suggest that a deep reinforcement learning algorithm can optimize time in therapeutic range for patients taking warfarin. A digital clinical decision support system to promote algorithm-consistent warfarin dosing could optimize time in therapeutic range and improve clinical outcomes in atrial fibrillation globally.

Predicting quetiapine dose in patients with depression using machine learning techniques based on real-world evidence

BACKGROUND

Being one of the most widespread, pervasive, and troublesome illnesses in the world, depression causes dysfunction in various spheres of individual and social life. Regrettably, despite obtaining evidence-based antidepressant medication, up to 70% of people are going to continue to experience troublesome symptoms. Quetiapine, as one of the most commonly prescribed antipsychotic medication worldwide, has been reported as an effective augmentation strategy to antidepressants. The right quetiapine dose and personalized quetiapine treatment are frequently challenging for clinicians. This study aimed to identify important influencing variables for quetiapine dose by maximizing the use of data from real world, and develop a predictive model of quetiapine dose through machine learning techniques to support selections for treatment regimens.

METHODS

The study comprised 308 depressed patients who were medicated with quetiapine and hospitalized in the First Hospital of Hebei Medical University, from November 1, 2019, to August 31, 2022. To identify the important variables influencing the dose of quetiapine, a univariate analysis was applied. The prediction abilities of nine machine learning models (XGBoost, LightGBM, RF, GBDT, SVM, LR, ANN, DT) were compared. Algorithm with the optimal model performance was chosen to develop the prediction model.

RESULTS

Four predictors were selected from 38 variables by the univariate analysis (p < 0.05), including quetiapine TDM value, age, mean corpuscular hemoglobin concentration, and total bile acid. Ultimately, the XGBoost algorithm was used to create a prediction model for quetiapine dose that had the greatest predictive performance (accuracy = 0.69) out of nine models. In the testing cohort (62 cases), a total of 43 cases were correctly predicted of the quetiapine dose regimen. In dose subgroup analysis, AUROC for patients with daily dose of 100 mg, 200 mg, 300 mg and 400 mg were 0.99, 0.75, 0.93 and 0.86, respectively.

CONCLUSIONS

In this work, machine learning techniques are used for the first time to estimate the dose of quetiapine for patients with depression, which is valuable for the clinical drug recommendations.

New insights into the role of VKORC1 polymorphisms for optimal warfarin dose selection in Caribbean Hispanic patients through an external validation of a population PK/PD model

Warfarin, an oral anticoagulant, has been used for decades to prevent thromboembolic events. The complex interplay between CYP2C9 and VKORC1 genotypes on warfarin PK and PD properties is not fully understood in special sub-groups of patients. This study aimed to externally validate a population pharmacokinetic/pharmacodynamic (PK/PD) model for the effect of warfarin on international normalized ratio (INR) and to evaluate optimal dosing strategies based on the selected covariates in Caribbean Hispanic patients. INR, and CYP2C9 and VKORC1 genotypes from 138 patients were used to develop a population PK/PD model in NONMEM. The structural definition of a previously published PD model for INR was implemented. A numerical evaluation of the parameter-covariate relationship was performed. Simulations were conducted to determine optimal dosing strategies for each genotype combinations, focusing on achieving therapeutic INR levels. Findings revealed elevated IC50 for G/G, G/A, and A/A VKORC1 haplotypes (11.76, 10.49, and 9.22 mg/L, respectively), in this population compared to previous reports. The model-guided dosing analysis recommended daily warfarin doses of 3-5 mg for most genotypes to maintain desired INR levels, although subjects with combination of CYP2C9 and VKORC1 genotypes * 2/* 2-, * 2/* 3- and * 2/* 5-A/A would require only 1 mg daily. This research underscores the potential of population PK/PD modeling to inform personalized warfarin dosing in populations typically underrepresented in clinical studies, potentially leading to improved treatment outcomes and patient safety. By integrating genetic factors and clinical data, this approach could pave the way for more effective and tailored anticoagulation therapy in diverse patient groups.

Evaluation of Machine Learning Approaches for Predicting Warfarin Discharge Dose in Cardiac Surgery Patients: Retrospective Algorithm Development and Validation Study

BACKGROUND

Warfarin dosing in cardiac surgery patients is complicated by a heightened sensitivity to the drug, predisposing patients to adverse events. Predictive algorithms are therefore needed to guide warfarin dosing in cardiac surgery patients.

OBJECTIVE

This study aimed to develop and validate an algorithm for predicting the warfarin dose needed to attain a therapeutic international normalized ratio (INR) at the time of discharge in cardiac surgery patients.

METHODS

We abstracted variables influencing warfarin dosage from the records of 1031 encounters initiating warfarin between April 1, 2011, and November 29, 2019, at St Michael's Hospital in Toronto, Ontario, Canada. We compared the performance of penalized linear regression, k-nearest neighbors, random forest regression, gradient boosting, multivariate adaptive regression splines, and an ensemble model combining the predictions of the 5 regression models. We developed and validated separate models for predicting the warfarin dose required for achieving a discharge INR of 2.0-3.0 in patients undergoing all forms of cardiac surgery except mechanical mitral valve replacement and a discharge INR of 2.5-3.5 in patients receiving a mechanical mitral valve replacement. For the former, we selected 80% of encounters (n=780) who had initiated warfarin during their hospital admission and had achieved a target INR of 2.0-3.0 at the time of discharge as the training cohort. Following 10-fold cross-validation, model accuracy was evaluated in a test cohort comprised solely of cardiac surgery patients. For patients requiring a target INR of 2.5-3.5 (n=165), we used leave-p-out cross-validation (p=3 observations) to estimate model performance. For each approach, we determined the mean absolute error (MAE) and the proportion of predictions within 20% of the true warfarin dose. We retrospectively evaluated the best-performing algorithm in clinical practice by comparing the proportion of cardiovascular surgery patients discharged with a therapeutic INR before (April 2011 and July 2019) and following (September 2021 and May 2, 2022) its implementation in routine care.

RESULTS

Random forest regression was the best-performing model for patients with a target INR of 2.0-3.0, an MAE of 1.13 mg, and 39.5% of predictions of falling within 20% of the actual therapeutic discharge dose. For patients with a target INR of 2.5-3.5, the ensemble model performed best, with an MAE of 1.11 mg and 43.6% of predictions being within 20% of the actual therapeutic discharge dose. The proportion of cardiovascular surgery patients discharged with a therapeutic INR before and following implementation of these algorithms in clinical practice was 47.5% (305/641) and 61.1% (11/18), respectively.

CONCLUSIONS

Machine learning algorithms based on routinely available clinical data can help guide initial warfarin dosing in cardiac surgery patients and optimize the postsurgical anticoagulation of these 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.

A framework for integrating artificial intelligence for clinical care with continuous therapeutic monitoring

The complex relationships between continuously monitored health signals and therapeutic regimens can be modelled via machine learning. However, the clinical implementation of the models will require changes to clinical workflows. Here we outline ClinAIOps ('clinical artificial-intelligence operations'), a framework that integrates continuous therapeutic monitoring and the development of artificial intelligence (AI) for clinical care. ClinAIOps leverages three feedback loops to enable the patient to make treatment adjustments using AI outputs, the clinician to oversee patient progress with AI assistance, and the AI developer to receive continuous feedback from both the patient and the clinician. We lay out the central challenges and opportunities in the deployment of ClinAIOps by means of examples of its application in the management of blood pressure, diabetes and Parkinson's disease. By enabling more frequent and accurate measurements of a patient's health and more timely adjustments to their treatment, ClinAIOps may substantially improve patient outcomes.

Evaluation of supervised machine learning algorithms in predicting the poor anticoagulation control and stable weekly doses of warfarin

BACKGROUND

Machine learning algorithms (MLAs) carry a huge potential in identifying predicting factors and are being explored for their utility in the field of personalized medicine.

AIM

We aimed to investigate MLAs for identifying predictors (clinical and genetic) of poor anticoagulation status (ACS) and stable weekly warfarin dose (SWWD).

METHOD

Clinical factors, in addition to the CYP2C9, VKORC1, and CYP4F2 genotypes, were obtained for patients receiving warfarin for at least the previous six months. The C5.0 decision tree classification algorithm was used to predict poor ACS while classification and regression tree analysis (CART), in addition to the Chi-square automatic interaction detector (CHAID), was used to predict SWWD. The percentage of patients within 20% of the actual dose, root mean squared error (RMSE), and area under the receiver-operating characteristics curve (AUROC) were identified as performance indicators of the models.

RESULTS

In the C5.0 classification decision tree, the CYP4F2 genotype was the strongest predictor of ACS (AUROC = 0.53). In the CART analysis of SWWD, VKORC1 polymorphisms were the most significant predictor, followed by the CYP2C9 genotype (percentage of patients within 20% of the actual dose = 38.2%, RMSE = 13.6). For the CHAID algorithm, the percentage of patients within 20% of the actual dose was 49%, while the RMSE was found to be 13.4.

CONCLUSION

Genetic and non-genetic predictive factors were identified by the MLAs for ACS and SWWD. Further, the need to externally validate the MLAs in a prospective study was highlighted.

Individualized prediction of chronic kidney disease for the elderly in longevity areas in China: Machine learning approaches

Background

Chronic kidney disease (CKD) has become a major public health problem worldwide and has caused a huge social and economic burden, especially in developing countries. No previous study has used machine learning (ML) methods combined with longitudinal data to predict the risk of CKD development in 2 years amongst the elderly in China.

Methods

This study was based on the panel data of 925 elderly individuals in the 2012 baseline survey and 2014 follow-up survey of the Healthy Aging and Biomarkers Cohort Study (HABCS) database. Six ML models, logistic regression (LR), lasso regression, random forests (RF), gradient-boosted decision tree (GBDT), support vector machine (SVM), and deep neural network (DNN), were developed to predict the probability of CKD amongst the elderly in 2 years (the year of 2014). The decision curve analysis (DCA) provided a range of threshold probability of the outcome and the net benefit of each ML model.

Results

Amongst the 925 elderly in the HABCS 2014 survey, 289 (18.8%) had CKD. Compared with the other models, LR, lasso regression, RF, GBDT, and DNN had no statistical significance of the area under the receiver operating curve (AUC) value (>0.7), and SVM exhibited the lowest predictive performance (AUC = 0.633, p-value = 0.057). DNN had the highest positive predictive value (PPV) (0.328), whereas LR had the lowest (0.287). DCA results indicated that within the threshold ranges of ~0-0.03 and 0.37-0.40, the net benefit of GBDT was the largest. Within the threshold ranges of ~0.03-0.10 and 0.26-0.30, the net benefit of RF was the largest. Age was the most important predictor variable in the RF and GBDT models. Blood urea nitrogen, serum albumin, uric acid, body mass index (BMI), marital status, activities of daily living (ADL)/instrumental activities of daily living (IADL) and gender were crucial in predicting CKD in the elderly.

Conclusion

The ML model could successfully capture the linear and nonlinear relationships of risk factors for CKD in the elderly. The decision support system based on the predictive model in this research can help medical staff detect and intervene in the health of the elderly early.

Development and evaluation of uncertainty quantifying machine learning models to predict piperacillin plasma concentrations in critically ill patients

BACKGROUND

Beta-lactam antimicrobial concentrations are frequently suboptimal in critically ill patients. Population pharmacokinetic (PopPK) modeling is the golden standard to predict drug concentrations. However, currently available PopPK models often lack predictive accuracy, making them less suited to guide dosing regimen adaptations. Furthermore, many currently developed models for clinical applications often lack uncertainty quantification. We, therefore, aimed to develop machine learning (ML) models for the prediction of piperacillin plasma concentrations while also providing uncertainty quantification with the aim of clinical practice.

METHODS

Blood samples for piperacillin analysis were prospectively collected from critically ill patients receiving continuous infusion of piperacillin/tazobactam. Interpretable ML models for the prediction of piperacillin concentrations were designed using CatBoost and Gaussian processes. Distribution-based Uncertainty Quantification was added to the CatBoost model using a proposed Quantile Ensemble method, useable for any model optimizing a quantile function. These models are subsequently evaluated using the distribution coverage error, a proposed interpretable uncertainty quantification calibration metric. Development and internal evaluation of the ML models were performed on the Ghent University Hospital database (752 piperacillin concentrations from 282 patients). Ensuing, ML models were compared with a published PopPK model on a database from the University Medical Centre of Groningen where a different dosing regimen is used (46 piperacillin concentrations from 15 patients.).

RESULTS

The best performing model was the Catboost model with an RMSE and [Formula: see text] of 31.94-0.64 and 33.53-0.60 for internal evaluation with and without previous concentration. Furthermore, the results prove the added value of the proposed Quantile Ensemble model in providing clinically useful individualized uncertainty predictions and show the limits of homoscedastic methods like Gaussian Processes in clinical applications.

CONCLUSIONS

Our results show that ML models can consistently estimate piperacillin concentrations with acceptable and high predictive accuracy when identical dosing regimens as in the training data are used while providing highly relevant uncertainty predictions. However, generalization capabilities to other dosing schemes are limited. Notwithstanding, incorporating ML models in therapeutic drug monitoring programs seems definitely promising and the current work provides a basis for validating the model in clinical practice.

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.

Deep compartment models: A deep learning approach for the reliable prediction of time-series data in pharmacokinetic modeling

Nonlinear mixed effect (NLME) models are the gold standard for the analysis of patient response following drug exposure. However, these types of models are complex and time-consuming to develop. There is great interest in the adoption of machine-learning methods, but most implementations cannot be reliably extrapolated to treatment strategies outside of the training data. In order to solve this problem, we propose the deep compartment model (DCM), a combination of neural networks and ordinary differential equations. Using simulated datasets of different sizes, we show that our model remains accurate when training on small data sets. Furthermore, using a real-world data set of patients with hemophilia A receiving factor VIII concentrate while undergoing surgery, we show that our model more accurately predicts a priori drug concentrations compared to a previous NLME model. In addition, we show that our model correctly describes the changing drug concentration over time. By adopting pharmacokinetic principles, the DCM allows for simulation of different treatment strategies and enables therapeutic drug monitoring.

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.

Machine learning-based method for tacrolimus dose predictions in Chinese kidney transplant perioperative patients

WHAT IS KNOWN AND OBJECTIVES

Tacrolimus (TAC), a first-line immunosuppressant in solid-organ transplant, has a narrow therapeutic window and large inter-individual variability, which affects its use in clinical practice. Successful predictions using machine learning algorithms have been reported in several fields. However, a comparison of 10 machine learning model-based TAC pharmacogenetic and pharmacokinetic dosing algorithms for kidney transplant perioperative patients of Chinese descent has not been reported. The objective of this study was to screen and establish an appropriate machine learning method to predict the individualized dosages of TAC for perioperative kidney transplant patients.

METHODS

The records of 2551 patients were collected from three transplant centres, 80% of which were randomly selected as a 'derivation cohort' to develop the dose prediction algorithm, while the remaining 20% constituted a 'validation cohort' to validate the final algorithm selected. Important features were screened according to our previously established population pharmacokinetic model of tacrolimus. The performances of the algorithms were evaluated and compared using R-squared and the mean percentage in the remaining 20% of patients.

RESULTS AND DISCUSSION

This study identified several factors influencing TAC dosage, including CYP3A5 rs776746, CYP3A4 rs4646437, haematocrit, Wuzhi capsules, TAC daily dose, age, height, weight, post-operative time, nifedipine and the medication history of the patient. According to our results, among the 10 machine learning models, the extra trees regressor (ETR) algorithm showed the best performance in the training set (R-squared: 1, mean percentage within 20%: 100%) and test set (R-squared: 0.85, mean percentage within 20%: 92.77%) of the derivation cohort. The ETR model successfully predicted the ideal TAC dosage in 97.73% of patients, especially in the intermediate dosage range (>5 mg/day to <8 mg/day), whereby the ideal TAC dosage could be successfully predicted in 99% of the patients.

WHAT IS NEW AND CONCLUSION

The results indicated that the ETR algorithm, which was chosen to establish the dose prediction model, performed better than the other nine machine learning models. This study is the first to establish ETR algorithms to predict TAC dosage. This study will further promote the individualized medication of TAC in kidney transplant patients in the future, which has great significance in ensuring the safety and effectiveness of drug use.

Machine Learning for Prediction of Stable Warfarin Dose in US Latinos and Latin Americans

Populations used to create warfarin dose prediction algorithms largely lacked participants reporting Hispanic or Latino ethnicity. While previous research suggests nonlinear modeling improves warfarin dose prediction, this research has mainly focused on populations with primarily European ancestry. We compare the accuracy of stable warfarin dose prediction using linear and nonlinear machine learning models in a large cohort enriched for US Latinos and Latin Americans (ULLA). Each model was tested using the same variables as published by the International Warfarin Pharmacogenetics Consortium (IWPC) and using an expanded set of variables including ethnicity and warfarin indication. We utilized a multiple linear regression model and three nonlinear regression models: Bayesian Additive Regression Trees, Multivariate Adaptive Regression Splines, and Support Vector Regression. We compared each model’s ability to predict stable warfarin dose within 20% of actual stable dose, confirming trained models in a 30% testing dataset with 100 rounds of resampling. In all patients (n = 7,030), inclusion of additional predictor variables led to a small but significant improvement in prediction of dose relative to the IWPC algorithm (47.8 versus 46.7% in IWPC, p = 1.43 × 10⁻¹⁵). Nonlinear models using IWPC variables did not significantly improve prediction of dose over the linear IWPC algorithm. In ULLA patients alone (n = 1,734), IWPC performed similarly to all other linear and nonlinear pharmacogenetic algorithms. Our results reinforce the validity of IWPC in a large, ethnically diverse population and suggest that additional variables that capture warfarin dose variability may improve warfarin dose prediction algorithms.

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.

Robust Performance of Potentially Functional SNPs in Machine Learning Models for the Prediction of Atorvastatin-Induced Myalgia

Statins can cause muscle symptoms resulting in poor adherence to therapy and increased cardiovascular risk. We hypothesize that combinations of potentially functional SNPs (pfSNPs), rather than individual SNPs, better predict myalgia in patients on atorvastatin. This study assesses the value of potentially functional single nucleotide polymorphisms (pfSNPs) and employs six machine learning algorithms to identify the combination of SNPs that best predict myalgia.

Methods: Whole genome sequencing of 183 Chinese, Malay and Indian patients from Singapore was conducted to identify genetic variants associated with atorvastatin induced myalgia. To adjust for confounding factors, demographic and clinical characteristics were also examined for their association with myalgia. The top factor, sex, was then used as a covariate in the whole genome association analyses. Variants that were highly associated with myalgia from this and previous studies were extracted, assessed for potential functionality (pfSNPs) and incorporated into six machine learning models. Predictive performance of a combination of different models and inputs were compared using the average cross validation area under ROC curve (AUC). The minimum combination of SNPs to achieve maximum sensitivity and specificity as determined by AUC, that predict atorvastatin-induced myalgia in most, if not all the six machine learning models was determined.

Results: Through whole genome association analyses using sex as a covariate, a larger proportion of pfSNPs compared to non-pf SNPs were found to be highly associated with myalgia. Although none of the individual SNPs achieved genome wide significance in univariate analyses, machine learning models identified a combination of 15 SNPs that predict myalgia with good predictive performance (AUC >0.9). SNPs within genes identified in this study significantly outperformed SNPs within genes previously reported to be associated with myalgia. pfSNPs were found to be more robust in predicting myalgia, outperforming non-pf SNPs in the majority of machine learning models tested.

Conclusion: Combinations of pfSNPs that were consistently identified by different machine learning models to have high predictive performance have good potential to be clinically useful for predicting atorvastatin-induced myalgia once validated against an independent cohort of patients.

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.

A machine learning approach to personalized dose adjustment of lamotrigine using noninvasive clinical parameters

The pharmacokinetic variability of lamotrigine (LTG) plays a significant role in its dosing requirements. Our goal here was to use noninvasive clinical parameters to predict the dose-adjusted concentrations (C/D ratio) of LTG based on machine learning (ML) algorithms. A total of 1141 therapeutic drug-monitoring measurements were used, 80% of which were randomly selected as the "derivation cohort" to develop the prediction algorithm, and the remaining 20% constituted the "validation cohort" to test the finally selected model. Fifteen ML models were optimized and evaluated by tenfold cross-validation on the "derivation cohort,” and were filtered by the mean absolute error (MAE). On the whole, the nonlinear models outperformed the linear models. The extra-trees’ regression algorithm delivered good performance, and was chosen to establish the predictive model. The important features were then analyzed and parameters of the model adjusted to develop the best prediction model, which accurately described the C/D ratio of LTG, especially in the intermediate-to-high range (≥ 22.1 μg mL⁻¹ g⁻¹ day), as illustrated by a minimal bias (mean relative error (%) =  + 3%), good precision (MAE = 8.7 μg mL⁻¹ g⁻¹ day), and a high percentage of predictions within ± 20% of the empirical values (60.47%). This is the first study, to the best of our knowledge, to use ML algorithms to predict the C/D ratio of LTG. The results here can help clinicians adjust doses of LTG administered to patients to minimize adverse reactions.

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.

Developing and Validating a Clinical Warfarin Dose-Initiation Model for Black-African Patients in South Africa and Uganda

Warfarin remains the oral anticoagulant of choice in sub-Saharan Africa. However, dosing is challenging due to a highly variable clinical response for a given dose. This study aimed to develop and validate a clinical warfarin dose-initiation model in sub-Saharan Black-African patients. For the development cohort, we used data from 364 patients who were recruited from 8 outpatient clinics and hospital departments in Uganda and South Africa (June 2018-July 2019). Validation was undertaken using the International Warfarin Pharmacogenetics Consortium (IWPC) dataset (690 black patients). Four predictors (age, weight, target International Normalized Ratio range, and HIV status) were included in the final model, which achieved mean absolute errors (MAEs; mean of absolute differences between true dose and dose predicted by the model) of 11.6 (95% confidence interval (CI) 10.4-12.8) and 12.5 (95% CI 11.6-13.4) mg/week in the development and validation cohorts, respectively. Two other clinical models, IWPC and Gage, respectively, obtained MAEs of 12.5 (95% CI 11.3-13.7) and 12.7 (95% CI 11.5-13.8) mg/week in the development cohort, and 12.1 (95% CI 11.2-13.0) and 12.2 (95% CI 11.4-13.1) mg/week in the validation cohort. Compared with fixed dose-initiation, our model decreased the percentage of patients at high risk of suboptimal anticoagulation by 7.5% (1.5-13.7%) and 11.9% (7.1-16.8%) in the development and validation cohorts, respectively. The clinical utility of this model will be tested in a prospective study. Study Highlights WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC? ☑ Warfarin dosing remains challenging due to a highly variable clinical response for a given dose. WHAT QUESTION DID THIS STUDY ADDRESS? ☑ Can a clinical dose-initiation model be developed and validated for sub-Saharan Black-African patients? WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE? ☑ We have developed the first warfarin dose-initiation clinical model for Black-African patients in Uganda and South Africa. HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE? ☑ We will be implementing and validating this model in a prospective cohort to inform future large-scale implementation. More optimized dosing should improve the quality of warfarin anticoagulation in these two developing countries.

Evaluating warfarin dosing models on multiple datasets with a novel software framework and evolutionary optimisation

Warfarin is an effective preventative treatment for arterial and venous thromboembolism, but requires individualised dosing due to its narrow therapeutic range and high individual variation. Many machine learning techniques have been demonstrated in this domain. This study evaluated the accuracy of the most promising algorithms on the International Warfarin Pharmacogenetics Consortium dataset and a novel clinical dataset of South African patients. Support vectors and linear regression were amongst the top performers in both datasets and performed comparably to recent stacked ensemble approaches, whilst neural networks were one of the worst performers in both datasets. We also introduced genetic programming to automatically optimise model architectures and hyperparameters without human guidance. Remarkably, the generated models were found to match the performance of the best models hand-crafted by human experts. Finally, we present a novel software framework (Warfit-learn) for warfarin dosing research. It leverages the most successful techniques in preprocessing, imputation, and parallel evaluation, with the goal of accelerating research and making results in this domain more reproducible.

Computationally Driven Discovery in Coagulation

Bleeding frequency and severity within clinical categories of hemophilia A are highly variable and the origin of this variation is unknown. Solving this mystery in coagulation requires the generation and analysis of large data sets comprised of experimental outputs or patient samples, both of which are subject to limited availability. In this review, we describe how a computationally driven approach bypasses such limitations by generating large synthetic patient data sets. These data sets were created with a mechanistic mathematical model, by varying the model inputs, clotting factor, and inhibitor concentrations, within normal physiological ranges. Specific mathematical metrics were chosen from the model output, used as a surrogate measure for bleeding severity, and statistically analyzed for further exploration and hypothesis generation. We highlight results from our recent study that employed this computationally driven approach to identify FV (factor V) as a key modifier of thrombin generation in mild to moderate hemophilia A, which was confirmed with complementary experimental assays. The mathematical model was used further to propose a potential mechanism for these observations whereby thrombin generation is rescued in FVIII-deficient plasma due to reduced substrate competition between FV and FVIII for FXa (activated factor X).

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.

Evaluation of the impact of body mass index on venous thromboembolism risk factors

In this paper, we investigate the interaction impacts of body mass index (BMI) on the other important risk factors for venous thromboembolism (VTE), using deep venous thrombosis (DVT) patient data from the International Warfarin Pharmacogenetics Consortium (IWPC). We apply eight machine learning techniques, including naive Bayes classifier (NB), support vector machine (SVM), elastic net regression (ENET), logistic regression (LR), lasso regression (LAR), multivariate adaptive regression splines (MARS), boosted regression tree (BRT) and random forest model (RF). The RF method is selected as the best model for classification. Out of 33 features considered in this study, we identify 12 variables as relatively important risk factors for VTE. Finally, we examine the interaction impacts of BMI on these important VTE risk factors. We conclude that the impacts of risk factors on VTE incidence are varying across different BMI groups, and the variations are different for different risk factors. Therefore the interaction impacts of BMI on the other risk factors have to be taken into account in order to better understand the incidence of VTE.

Therapeutic drug monitoring of β-lactam antibiotics in the ICU

INTRODUCTION

Individualizing antibiotic therapy is paramount to improve clinical outcomes while minimizing the risk of toxicity and antimicrobial therapy. β-lactam antibiotics are amongst the drugs most commonly prescribed in the Intensive Care Unit (ICU). The pharmacokinetics of β-lactam antibiotics are profoundly altered in critically ill patients, leading to the failure of standard drug dosing regimens to result in adequate drug concentrations. Therapeutic Drug Monitoring (TDM) of β-lactam antibiotics is a promising tool to help optimize β-lactam antibiotic therapy.

AREAS COVERED

The rationale behind TDM for β-lactam antibiotics is explained, as well as some more practical aspects such as when to sample, what concentrations to strive for and how to use it in clinical practice. We also discuss microbiological and analytical considerations, knowledge gaps, and future perspectives of β-lactam antibiotics TDM in ICU patients.

EXPERT OPINION

TDM of β-lactam antibiotics has been studied intensively in recent years. While TDM may not yet be widely available, and targets need to be further refined, TDM of β-lactam antibiotics will help to optimize antibiotic therapy in the critically ill patient, as an integrated part of an antimicrobial stewardship program.

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.

Implications of genetic variation of common Drug Metabolizing Enzymes and ABC Transporters among the Pakistani Population

Genetic polymorphism of drug metabolizing enzymes and transporters may influence drug response. The frequency varies substantially between ethnicities thus having implications on appropriate selection and dosage of various drugs in different populations. The distribution of genetic polymorphisms in healthy Pakistanis has so far not been described. In this study, 155 healthy adults (98 females) were included from all districts of Karachi. DNA was extracted from saliva and genotyped for relevant SNVs in CYP1A1, CYP2B6, CYP2C9, CYP2C19, CYP2D6, CYP3A4 and CYP3A5 as well as ALDH3A1, GSTA1, ABCB1 and ABCC2. About 64% of the participants were born to parents who were unrelated to each other. There was generally a higher prevalence (p < 0.05) of variant alleles of CYP450 1A2, 2B6, 2C19, 3A5, ALDH3A1, GSTM1 as well as ABCB1 and ABCC2 in this study cohort than in other ethnicities reported in the HapMap database. In contrast, the prevalence of variant alleles was lower in GSTA1. Therefore, in the Pakistani population sample from Karachi a significantly different prevalence of variant drug metabolizing enzymes and ABC transporters was observed as compared to other ethnicities, which could have putative clinical consequences on drug efficacy and safety.

Prediction of cardiovascular outcomes with machine learning techniques: application to the Cardiovascular Outcomes in Renal Atherosclerotic Lesions (CORAL) study

Background

Data derived from the Cardiovascular Outcomes in Renal Atherosclerotic Lesions (CORAL) study were analyzed in an effort to employ machine learning methods to predict the composite endpoint described in the original study.

Methods

We identified 573 CORAL subjects with complete baseline data and the presence or absence of a composite endpoint for the study. These data were subjected to several models including a generalized linear (logistic-linear) model, support vector machine, decision tree, feed-forward neural network, and random forest, in an effort to attempt to predict the composite endpoint. The subjects were arbitrarily divided into training and testing subsets according to an 80%:20% distribution with various seeds. Prediction models were optimized within the CARET package of R.

Results

The best performance of the different machine learning techniques was that of the random forest method which yielded a receiver operator curve (ROC) area of 68.1%±4.2% (mean ± SD) on the testing subset with ten different seed values used to separate training and testing subsets. The four most important variables in the random forest method were SBP, serum creatinine, glycosylated hemoglobin, and DBP. Each of these variables was also important in at least some of the other methods. The treatment assignment group was not consistently an important determinant in any of the models.

Conclusion

Prediction of a composite cardiovascular outcome was difficult in the CORAL population, even when employing machine learning methods. Assignment to either the stenting or best medical therapy group did not serve as an important predictor of composite outcome.

Clinical Trial Registration

ClinicalTrials.gov, NCT00081731

Ensemble of machine learning algorithms using the stacked generalization approach to estimate the warfarin dose

Warfarin dosing remains challenging due to narrow therapeutic index and highly individual variability. Incorrect warfarin dosing is associated with devastating adverse events. Remarkable efforts have been made to develop the machine learning based warfarin dosing algorithms incorporating clinical factors and genetic variants such as polymorphisms in CYP2C9 and VKORC1. The most widely validated pharmacogenetic algorithm is the IWPC algorithm based on multivariate linear regression (MLR). However, with only a single algorithm, the prediction performance may reach an upper limit even with optimal parameters. Here, we present novel algorithms using stacked generalization frameworks to estimate the warfarin dose, within which different types of machine learning algorithms function together through a meta-machine learning model to maximize the prediction accuracy. Compared to the IWPC-derived MLR algorithm, Stack 1 and 2 based on stacked generalization frameworks performed significantly better overall. Subgroup analysis revealed that the mean of the percentage of patients whose predicted dose of warfarin within 20% of the actual stable therapeutic dose (mean percentage within 20%) for Stack 1 was improved by 12.7% (from 42.47% to 47.86%) in Asians and by 13.5% (from 22.08% to 25.05%) in the low-dose group compared to that for MLR, respectively. These data suggest that our algorithms would especially benefit patients requiring low warfarin maintenance dose, as subtle changes in warfarin dose could lead to adverse clinical events (thrombosis or bleeding) in patients with low dose. Our study offers novel pharmacogenetic algorithms for clinical trials and practice.

Support Vector Regression Approach to Predict the Design Space for the Extraction Process of Pueraria lobata

A support vector regression (SVR) method was introduced to improve the robustness and predictability of the design space in the implementation of quality by design (QbD), taking the extraction process of Pueraria lobata as a case study. In this paper, extraction time, number of extraction cycles, and liquid–solid ratio were identified as critical process parameters (CPPs), and the yield of puerarin, total isoflavonoids, and extracta sicca were the critical quality attributes (CQAs). Models between CQAs and CPPs were constructed using both a conventional quadratic polynomial model (QPM) and the SVR algorithm. The results of the two models indicated that the SVR model had better performance, with a higher R² and lower root-mean-square error (RMSE) and mean absolute deviation (MAD) than those of the QPM. Furthermore, the design space was predicted using a grid search technique. The operational range was extraction time, 24–51 min; number of extraction cycles, 3; and liquid–solid ratio, 14–18 mL/g. This study is the first reported work optimizing the design space of the extraction process of P. lobata based on an SVR model. SVR modeling, with its better prediction accuracy and generalization ability, could be a reliable tool for predicting the design space and shows great potential for the quality control of QbD.

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.

Warfarin Pharmacogenomics in Diverse Populations

Genotype-guided warfarin dosing algorithms are a rational approach to optimize warfarin dosing and potentially reduce adverse drug events. Diverse populations, such as African Americans and Latinos, have greater variability in warfarin dose requirements and are at greater risk for experiencing warfarin-related adverse events compared with individuals of European ancestry. Although these data suggest that patients of diverse populations may benefit from improved warfarin dose estimation, the vast majority of literature on genotype-guided warfarin dosing, including data from prospective randomized trials, is in populations of European ancestry. Despite differing frequencies of variants by race/ethnicity, most evidence in diverse populations evaluates variants that are most common in populations of European ancestry. Algorithms that do not include variants important across race/ethnic groups are unlikely to benefit diverse populations. In some race/ethnic groups, development of race-specific or admixture-based algorithms may facilitate improved genotype-guided warfarin dosing algorithms above and beyond that seen in individuals of European ancestry. These observations should be considered in the interpretation of literature evaluating the clinical utility of genotype-guided warfarin dosing. Careful consideration of race/ethnicity and additional evidence focused on improving warfarin dosing algorithms across race/ethnic groups will be necessary for successful clinical implementation of warfarin pharmacogenomics. The evidence for warfarin pharmacogenomics has a broad significance for pharmacogenomic testing, emphasizing the consideration of race/ethnicity in discovery of gene-drug pairs and development of clinical recommendations for pharmacogenetic testing.

Intuitive pharmacogenetic dosing of risperidone according to CYP2D6 phenotype extrapolated from genotype in a cohort of first episode psychosis patients

Risperidone (R) is the most prescribed antipsychotic drug for patients with a first episode of psychosis (FEP). In a naturalistic cohort of chronic psychiatric inpatients, we demonstrated that clinicians adjust R dosage by CYP2D6 activity, despite being blinded to the genotype, which we described as an "intuitive pharmacogenetic" process. The aim of the present study is to replicate our previous findings of intuitive pharmacogenetic in a cohort of FEP patients using CYP2D6 phenotype extrapolated from genotypes. 70 FEP patients, under baseline treatment with R monotherapy were genotyped using the iPLEX® ADME PGx multiplex panel and TaqMan® Genotyping and Copy Number Assays. Plasma concentrations of R and its metabolite, 9-hydroxyrisperidone (9-OH), were determined. The predictive properties of those variables associated with R dosage were tested using a multiple linear regression model as well as regression trees. Significant differences in the mean daily dosage of R among CYP2D6 phenotypes were observed (Kruskal-Wallis test p=0.02): PM (4.00±2.3mg/mL), IM (4.56±2.44), EM (6.22±4.0mg/day) and UM (10.20±4.91mg/day). However, non-significant differences were observed in the R/9-OH ratio or in the Concentration/Dose ratio. Regression tree provided better estimations of R dosage than the multiple linear regression model (MAE=0.958 and R²=0.871). We confirm the "intuitive pharmacogenetic" dosing of R according to the CYP2D6 phenotype in a FEP cohort. The results presented provides a rationale for the clinical use of CYP2D6 genotyping in personalized medicine.

Application of Machine-Learning Models to Predict Tacrolimus Stable Dose in Renal Transplant Recipients

Tacrolimus has a narrow therapeutic window and considerable variability in clinical use. Our goal was to compare the performance of multiple linear regression (MLR) and eight machine learning techniques in pharmacogenetic algorithm-based prediction of tacrolimus stable dose (TSD) in a large Chinese cohort. A total of 1,045 renal transplant patients were recruited, 80% of which were randomly selected as the “derivation cohort” to develop dose-prediction algorithm, while the remaining 20% constituted the “validation cohort” to test the final selected algorithm. 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 and their performances were compared in this work. Among all the machine learning models, RT performed best in both derivation [0.71 (0.67–0.76)] and validation cohorts [0.73 (0.63–0.82)]. In addition, the ideal rate of RT was 4% higher than that of MLR. To our knowledge, this is the first study to use machine learning models to predict TSD, which will further facilitate personalized medicine in tacrolimus administration in the future.

Formation of δ-Lactones by Cerium-Catalyzed, Baeyer-Villiger-Type Coupling of β-Oxoesters, Enol Acetates, and Dioxygen

Formation of δ-lactones is observed when cyclopentanone-2-carboxylates are converted in a cerium-catalyzed reaction with α-aryl vinyl acetates under oxidative conditions. The products of this transformation possess a 1,4-dicarbonyl constitution together with a quaternary carbon center. Atmospheric oxygen is the oxidant in this process, which can be regarded as ideal from economic and ecological points of view. Further advantages of this new C-C coupling and oxidation reaction are its operational simplicity and the application of nontoxic and inexpensive CeCl3·7 H2O as precatalyst. This so far unprecedented reaction is proposed to proceed via 1,2-dioxane derivatives, which decompose under formation of an oxycarbenium cation in a Baeyer-Villiger-type pathway. This mechanistic picture is supported by the observation that electron-rich (donor substituted or heteroaromatic) enol esters give higher yields than electron deficient congeners. Apart from 1,4-diketones and α-hydroxylated β-oxoesters formed as byproducts, the yields of δ-lactones range from moderate to good (up to 74%).

The Non-Linear Child: Ontogeny, Isoniazid Concentration, and NAT2 Genotype Modulate Enzyme Reaction Kinetics and Metabolism

N-acetyltransferase 2 (NAT2) catalyzes the acetylation of isoniazid to N-acetylisoniazid. NAT2 polymorphism explains 88% of isoniazid clearance variability in adults. We examined the effects of clinical and genetic factors on Michaelis-Menten reaction kinetic constants of maximum velocity (V_max) and affinity (K_m) in children 0–10 years old. We measured the rates of isoniazid elimination and N-acetylisoniazid production in the blood of 30 children. Since maturation effects could be non-linear, we utilized a pharmacometric approach and the artificial intelligence method, multivariate adaptive regression splines (MARS), to identify factors predicting NAT2 V_max and K_m by examining clinical, genetic, and laboratory factors in toto. Isoniazid concentration predicted both V_max and K_m and superseded the contribution of NAT2 genotype. Age non-linearly modified the NAT2 genotype contribution until maturation at ≥ 5.3 years. Thus, enzyme efficiency was constrained by substrate concentration, genes, and age. Since MARS output is in the form of basis functions and equations, it allows multiscale systems modeling from the level of cellular chemical reactions to whole body physiological parameters, by automatic selection of significant predictors by the algorithm.

•

We identified the NAT2 K_m and V_max in children treated with isoniazid.

•

Artificial intelligence (AI) algorithms were used to find predictors of K_m and V_max.

•

Isoniazid concentration affected V_max and K_m, and superseded NAT2 genotype.

•

Age non-linearly modified NAT2 genotype contribution until maturation at ≥ 5.3 years.

•

AI output is in the form of equations that allow multiscale systems modeling.

The effects of maturation on drug metabolism have not been studied for the type phase II enzymes such as NAT2, which metabolizes the drug isoniazid. Genes have been found to control speed of isoniazid metabolism. Studies to characterize affinity and maximum velocity for isoniazid metabolism in people were last performed in two individuals' livers in the 1960s. We identified NAT2 affinity and maximum velocity in 30 tuberculosis children treated with isoniazid. Artificial intelligence methods found that metabolism was affected by the drug's concentration more than by genes, which were affected by age up to 5.3 years.