Evaluation of cardiac pro-arrhythmic risks using the artificial neural network with ToR-ORd in silico model output.

Torsades de pointes (TdP) is a type of ventricular arrhythmia that can lead to sudden cardiac death. Drug-induced TdP has been an important concern for researchers and international regulatory boards. The Comprehensive in vitro Proarrhythmia Assay (CiPA) initiative was proposed that integrates in vitro testing and computational models of cardiac ion channels and human cardiomyocyte cells to evaluate the proarrhythmic risk of drugs. The TdP risk classification performance using only a single TdP metric may require some improvements because of information limitations and the instability of generalizing results. This study evaluates the performance of TdP metrics from the in silico simulations of the Tomek-O'Hara Rudy (ToR-ORd) ventricular cell model for classifying the TdP risk of drugs. We utilized these metrics as an input to an artificial neural network (ANN)-based classifier. The ANN model was optimized through hyperparameter tuning using the grid search (GS) method to find the optimal model. The study outcomes show an area under the curve (AUC) value of 0.979 for the high-risk category, 0.791 for the intermediate-risk category, and 0.937 for the low-risk category. Therefore, this study successfully demonstrates the capability of the ToR-ORd ventricular cell model in classifying the TdP risk into three risk categories, providing new insights into TdP risk prediction methods.

Machine learning approach to evaluate TdP risk of drugs using cardiac electrophysiological model including inter-individual variability.

Introduction: Predicting ventricular arrhythmia Torsade de Pointes (TdP) caused by drug-induced cardiotoxicity is essential in drug development. Several studies used single biomarkers such as qNet and Repolarization Abnormality (RA) in a single cardiac cell model to evaluate TdP risk. However, a single biomarker may not encompass the full range of factors contributing to TdP risk, leading to divergent TdP risk prediction outcomes, mainly when evaluated using unseen data. We addressed this issue by utilizing multi-in silico features from a population of human ventricular cell models that could capture a representation of the underlying mechanisms contributing to TdP risk to provide a more reliable assessment of drug-induced cardiotoxicity. Method: We generated a virtual population of human ventricular cell models using a modified O'Hara-Rudy model, allowing inter-individual variation. IC50 and Hill coefficients from 67 drugs were used as input to simulate drug effects on cardiac cells. Fourteen features (dVmdtrepol, dVmdtmax, Vmpeak, Vmresting, APDtri, APD90, APD50, Capeak, Cadiastole, Catri, CaD90, CaD50, qNet, qInward) could be generated from the simulation and used as input to several machine learning models, including k-nearest neighbor (KNN), Random Forest (RF), XGBoost, and Artificial Neural Networks (ANN). Optimization of the machine learning model was performed using a grid search to select the best parameter of the proposed model. We applied five-fold cross-validation while training the model with 42 drugs and evaluated the model's performance with test data from 25 drugs. Result: The proposed ANN model showed the highest performance in predicting the TdP risk of drugs by providing an accuracy of 0.923 (0.908-0.937), sensitivity of 0.926 (0.909-0.942), specificity of 0.921 (0.906-0.935), and AUC score of 0.964 (0.954-0.975). Discussion and conclusion: According to the performance results, combining the electrophysiological model including inter-individual variation and optimization of machine learning showed good generalization ability when evaluated using the unseen dataset and produced a reliable drug-induced TdP risk prediction system.

Application of Convolutional Neural Networks Using Action Potential Shape for In-Silico Proarrhythmic Risk Assessment.

This study proposes a convolutional neural network (CNN) model using action potential (AP) shapes as input for proarrhythmic risk assessment, considering the hypothesis that machine-learning features automatically extracted from AP shapes contain more meaningful information than do manually extracted indicators. We used 28 drugs listed in the comprehensive in vitro proarrhythmia assay (CiPA), consisting of eight high-risk, eleven intermediate-risk, and nine low-risk torsadogenic drugs. We performed drug simulations to generate AP shapes using experimental drug data, obtaining 2000 AP shapes per drug. The proposed CNN model was trained to classify the TdP risk into three levels, high-, intermediate-, and low-risk, based on in silico AP shapes generated using 12 drugs. We then evaluated the performance of the proposed model for 16 drugs. The classification accuracy of the proposed CNN model was excellent for high- and low-risk drugs, with AUCs of 0.914 and 0.951, respectively. The model performance for intermediate-risk drugs was good, at 0.814. Our proposed model can accurately assess the TdP risks of drugs from in silico AP shapes, reflecting the pharmacokinetics of ionic currents. We need to secure more drugs for future studies to improve the TdP-risk-assessment robustness.

In silico assessment on TdP risks of drug combinations under CiPA paradigm.

Researchers have recently proposed the Comprehensive In-vitro Proarrhythmia Assay (CiPA) to analyze medicines' TdP risks. Using the TdP metric known as qNet, numerous single-drug effects have been studied to classify the medications as low, intermediate, and high-risk. Furthermore, multiple medication therapies are recognized as a potential method for curing patients, mainly when limited drugs are available. This work expands the TdP risk assessment of drugs by introducing a CiPA-based in silico analysis of the TdP risk of combined drugs. The cardiac cell model was simulated using the population of models approach incorporating drug-drug interactions (DDIs) models on several ion channels for various drug pairs. Action potential duration (APD90), qNet, and calcium duration (CaD90) were computed and analyzed as biomarker features. The drug combination maps were also used to illustrate combined medicines' TdP risk. We found that the combined drugs alter cell responses in terms of biomarkers such as APD90, qNet, and CaD90 in a highly nonlinear manner. The results also revealed that combinations of high-risk with low-risk and intermediate-risk with low-risk drugs could result in compounds with varying TdP risks depending on the drug concentrations.

Sensitivity Analysis of Cardiac Alternans and Tachyarrhythmia to Ion Channel Conductance Using Population Modeling.

Action potential duration (APD) alternans, an alternating phenomenon between action potentials in cardiomyocytes, causes heart arrhythmia when the heart rate is high. However, some of the APD alternans observed in clinical trials occurs under slow heart rate conditions of 100 to 120 bpm, increasing the likelihood of heart arrhythmias such as atrial fibrillation. Advanced studies have identified the occurrence of this type of APD alternans in terms of electrophysiological ion channel currents in cells. However, they only identified physiological phenomena, such as action potential due to random changes in a particular ion channel's conductivity through ion models specializing in specific ion channel currents. In this study, we performed parameter sensitivity analysis via population modeling using a validated human ventricular physiology model to check the sensitivity of APD alternans to ion channel conductances. Through population modeling, we expressed the changes in alternans onset cycle length (AOCL) and mean APD in AOCL (AO meanAPD) according to the variations in ion channel conductance. Finally, we identified the ion channel that maximally affected the occurrence of APD alternans. AOCL and AO meanAPD were sensitive to changes in the plateau Ca2+ current. Accordingly, it was expected that APD alternans would be vulnerable to changes in intracellular calcium concentration.

Validation of in silico biomarkers for drug screening through ordinal logistic regression.

Since the Comprehensive in vitro Proarrhythmia Assay (CiPA) initiation, many studies have suggested various in silico features based on ionic charges, action potentials (AP), or intracellular calcium (Ca) to assess proarrhythmic risk. These in silico features are computed through electrophysiological simulations using in vitro experimental datasets as input, therefore changing with the quality of in vitro experimental data; however, research to validate the robustness of in silico features for proarrhythmic risk assessment of drugs depending on in vitro datasets has not been conducted. This study aims to verify the availability of in silico features commonly used in assessing the cardiac toxicity of drugs through an ordinal logistic regression model and three in vitro datasets measured under different experimental environments and with different purposes. We performed in silico drug simulations using the Tomek-Ohara Rudy (ToR-ORD) ventricular myocyte model and computed 12 in silico features comprising six AP features, four Ca features, and two ion charge features, which reflected the effect and characteristics of each in vitro data for CiPA 28 drugs. We then compared the classific performances of ordinal logistic regressions according to these 12 in silico features and used in vitro datasets to validate which in silico feature is the best for assessing the proarrhythmic risk of drugs at high, intermediate, and low levels. All 12 in silico features helped determine high-risky torsadogenic drugs, regardless of the in vitro datasets used in the in silico simulation as input. In the three types of in silico features, AP features were the most reliable for determining the three Torsade de Pointes (TdP) risk standards. Among AP features, AP duration at 50% repolarization (APD50) was the best when individually using in silico features per in vitro dataset. In contrast, the AP repolarization velocity (dVm/dtMax_repol) was the best when merging all in silico features computed through three in vitro datasets.

Verification of the Efficacy of Mexiletine Treatment for the A1656D Mutation on Downgrading Reentrant Tachycardia Using a 3D Cardiac Electrophysiological Model.

The SCN5A mutations have been long associated with long QT variant 3 (LQT3). Recent experimental and computation studies have reported that mexiletine effectively treats LQT3 patients associated with the A1656D mutation. However, they have primarily focused on cellular level evaluations and have only looked at the effects of mexiletine on action potential duration (APD) or QT interval reduction. We further investigated mexiletine's effects on cardiac cells through simulations of single-cell (behavior of alternant occurrence) and 3D (with and without mexiletine). We discovered that mexiletine could shorten the cell's APD and change the alternant's occurrence to a shorter basic cycle length (BCL) between 350 and 420 ms. The alternant also appeared at a normal heart rate under the A1656D mutation. Furthermore, the 3D ventricle simulations revealed that mexiletine could reduce the likelihood of a greater spiral wave breakup in the A1656D mutant condition by minimizing the appearance of rotors. In conclusion, we found that mexiletine could provide extra safety features during therapy for LQT3 patients because it can change the alternant occurrence from a normal to a faster heart rate, and it reduces the chance of a spiral wave breakup. Therefore, these findings emphasize the promising efficacy of mexiletine in treating LQT3 patients under the A1656D mutation.

Proarrhythmic risk assessment of drugs by dVm /dt shapes using the convolutional neural network.

Comprehensive in vitro Proarrhythmia Assay (CiPA) projects for assessing proarrhythmic drugs suggested a logistic regression model using qNet as the Torsades de Pointes (TdP) risk assessment biomarker, obtained from in silico simulation. However, using a single in silico feature, such as qNet, cannot reflect whole characteristics related to TdP in the entire action potential (AP) shape. Thus, this study proposed a deep convolutional neural network (CNN) model using differential action potential shapes to classify three proarrhythmic risk levels: high, intermediate, and low, considering both characteristics related to TdP not only in the depolarization phase but also the repolarization phase of AP shape. We performed an in silico simulation and got AP shapes with drug effects using half-maximal inhibitory concentration and Hill coefficients of 28 drugs released by CiPA groups. Then, we trained the deep CNN model with the differential AP shapes of 12 drugs and tested it with those of 16 drugs. Our model had a better performance for classifying the proarrhythmic risk of drugs than the traditional logistic regression model using qNet. The classification accuracy was 98% for high-risk level drugs, 94% for intermediate-risk level drugs, and 89% for low-risk level drugs.

qInward variability-based in-silico proarrhythmic risk assessment of drugs using deep learning model.

Many researchers have suggested evaluation methods and Torsades de Pointes (TdP) metrics to assess the proarrhythmic risk of a drug based on the in silico simulation, as part of the Comprehensive in-vitro Proarrhythmia Assay (CiPA) project. In the previous study, we validated the robustness of 12 in silico features using the ordinal logistic regression (OLR) model by comparing the classification performances of metrics according to the in-vitro experimental datasets used; however, the OLR model using 12 in silico features did not provide desirable results. This study proposed a convolutional neural network (CNN) model using the variability of promising in silico TdP metrics hypothesizing that the variability of in silico features based on beats has more information than the single value of in silico features. We performed the action potential (AP) simulation using a human ventricular myocyte model to calculate seven in silico features representing the electrophysiological cell states of drug effects over 1,000 beats: qNet, qInward, intracellular calcium duration at returning to 50% baseline (CaD50) and 90% baseline (CaD90), AP duration at 50% repolarization (APD50) and 90% repolarization (APD90), and dVm/dtMax_repol. The proposed CNN classifier was trained using 12 train drugs and tested using 16 test drugs among CiPA drugs. The torsadogenic risk of drugs was classified as high, intermediate, and low risks. We determined the CNN classifier by comparing the classification performance according to the variabilities of seven in silico biomarkers computed from the in silico drug simulation using the Chantest dataset. The proposed CNN classifier performed the best when using qInward variability to classify the TdP-risk drugs with 0.94 AUC for high risk and 0.93 AUC for low risk. In addition, the final CNN classifier was validated using the qInward variability obtained after merging three in-vitro datasets, but the model performance decreased to a moderate level of 0.75 and 0.78 AUC. These results suggest the need for the proposed CNN model to be trained and tested using various types of drugs.

Assessment of Drug Proarrhythmicity Using Artificial Neural Networks With in silico Deterministic Model Outputs.

As part of the Comprehensive in vitro Proarrhythmia Assay initiative, methodologies for predicting the occurrence of drug-induced torsade de pointes via computer simulations have been developed and verified recently. However, their predictive performance still requires improvement. Herein, we propose an artificial neural networks (ANN) model that uses nine multiple input features, considering the action potential morphology, calcium transient morphology, and charge features to further improve the performance of drug toxicity evaluation. The voltage clamp experimental data for 28 drugs were augmented to 2,000 data entries using an uncertainty quantification technique. By applying these data to the modified O'Hara Rudy in silico model, nine features (dVm/dtmax, APresting, APD90, APD50, Caresting, CaD90, CaD50, qNet, and qInward) were calculated. These nine features were used as inputs to an ANN model to classify drug toxicity into high-risk, intermediate-risk, and low-risk groups. The model was trained with data from 12 drugs and tested using the data of the remaining 16 drugs. The proposed ANN model demonstrated an AUC of 0.92 in the high-risk group, 0.83 in the intermediate-risk group, and 0.98 in the low-risk group. This was higher than the classification performance of the method proposed in previous studies.

Sensitivity Analysis of Ion Channel Conductance on Myocardial Electromechanical Delay: Computational Study.

It is well known that cardiac electromechanical delay (EMD) can cause dyssynchronous heart failure (DHF), a prominent cardiovascular disease (CVD). This work computationally assesses the conductance variation of every ion channel on the cardiac cell to give rise to EMD prolongation. The electrical and mechanical models of human ventricular tissue were simulated, using a population approach with four conductance reductions for each ion channel. Then, EMD was calculated by determining the difference between the onset of action potential and the start of cell shortening. Finally, EMD data were put into the optimized conductance dimensional stacking to show which ion channel has the most influence in elongating the EMD. We found that major ion channels, such as L-type calcium (CaL), slow-delayed rectifier potassium (Ks), rapid-delayed rectifier potassium (Kr), and inward rectifier potassium (K1), can significantly extend the action potential duration (APD) up to 580 ms. Additionally, the maximum intracellular calcium (Cai) concentration is greatly affected by the reduction in channel CaL, Ks, background calcium, and Kr. However, among the aforementioned major ion channels, only the CaL channel can play a superior role in prolonging the EMD up to 83 ms. Furthermore, ventricular cells with long EMD have been shown to inherit insignificant mechanical response (in terms of how strong the tension can grow and how far length shortening can go) compared with that in normal cells. In conclusion, despite all variations in every ion channel conductance, only the CaL channel can play a significant role in extending EMD. In addition, cardiac cells with long EMD tend to have inferior mechanical responses due to a lack of Cai compared with normal conditions, which are highly likely to result in a compromised pump function of the heart.

Effect of KCNQ1 G229D mutation on cardiac pumping efficacy and reentrant dynamics in ventricles: Computational study.

There is growing interest in genetic arrhythmia since mutations in gene which encodes the ion channel underlie numerous arrhythmias. Hasegawa et al reported that G229D mutation in KCNQ1 underlies atrial fibrillation due to significant shortening of action potential duration (APD) in atrial cells. Here, we predicted whether KCNQ1 G229D mutation affects ventricular fibrillation generation, although it shortens APD slightly compared with the atrial cell. We analyzed the effects of G229D mutation on electrical and mechanical ventricle behaviors (not considered in previous studies). We compared action potential shapes under wild-type and mutant conditions. Electrical wave propagations through ventricles were analyzed during sinus rhythm and reentrant conditions. IKs enhancement due to G229D mutation shortened the APD in the ventricular cells (6%, 0.3%, and 8% for endo, M, and epi-cells, respectively). The shortened APD contributed to 7% shortening of QT intervals, 29% shortening of wavelengths, 20% decrease in intraventricular pressure, and increase in end-systolic volume 17%, end-diastolic volume 7%, and end-diastolic pressure 11%, which further resulted in reduction in stroke volume as well as cardiac output (28%), ejection fraction 33% stroke work 44%, and ATP consumption 28%. In short, using computational model of the ventricle, we predicted that G229D mutation decreased cardiac pumping efficacy and increased the vulnerability of ventricular fibrillation.