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

Abstract

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.

Neural network emulation of the human ventricular cardiomyocyte action potential for more efficient computations in pharmacological studies

“…Second, timing offsets, for example between stimuli in the data to be fitted and the training data, can be accounted for using t 0 in (3) without retraining. Last but not least, the transmembrane potential gradient , recently highlighted in terms of proarrhythmic potential prediction ( Jeong et al, 2022 ), can be continuously derived and is not dependent on the temporal discretization.…”

“…Last but not least, the transmembrane potential gradient , recently highlighted in terms of proarrhythmic potential prediction [Jeong et al, 2022 ], can be continuously derived and is not dependent on the temporal discretization.…”

Neural network emulation of the human ventricular cardiomyocyte action potential: a tool for more efficient computation in pharmacological studies

“…between stimuli in the data to be fitted and the training data, can be accounted for using t 0 in (3) without retraining. Last but not least, the transmem-brane potential gradient ∂Vm ∂t , recently highlighted in terms of proarrhythmic potential prediction [Jeong et al, 2022], can be continuously derived and is not dependent on the temporal discretization.…”

Assessment of the proarrhythmic effects of repurposed antimalarials for COVID-19 treatment using a comprehensive in vitro proarrhythmia assay (CiPA)

“…To evaluate these risk factors, the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) has published S7B and E14 guidelines, which provide a framework for predicting cardiotoxicity assessment for drugs ( Cavero and Crumb, 2005 ). Unfortunately, the low specificity of ICH S7B/E14 leads to the undesirable regulation of drugs that are unlikely to cause cardiac arrhythmias ( Jeong et al, 2022 ). This is typically understood to result from focusing on only one of the many ion channels governing action potential (AP) formation ( Colatsky et al, 2016 ; Jeong et al, 2022 ).…”

“…Unfortunately, the low specificity of ICH S7B/E14 leads to the undesirable regulation of drugs that are unlikely to cause cardiac arrhythmias ( Jeong et al, 2022 ). This is typically understood to result from focusing on only one of the many ion channels governing action potential (AP) formation ( Colatsky et al, 2016 ; Jeong et al, 2022 ). To resolve this issue, the Comprehensive in vitro Proarrhythmia Assay (CiPA) project was proposed by the US Food and Drug Administration (FDA) and has grown into an international research project with the participation of the FDA and various researchers led by pharmaceutical companies.…”