| Autor: |
Serrano, Ricardo, Feyen, Dries A, Bruyneel, Arne A, Hnatiuk Hnatiuk, Anna P, Vu, Michelle M, Amatya, Prashila L, Perea-Gil, Isaac, Prado, Maricela, Seeger, Timon, Wu, Joseph C, Karakikes, Ioannis, Mercola, Mark |
| Zdroj: |
Circulation (Ovid); November 2021, Vol. 144 Issue: Supplement 1 pA14393-A14393, 1p |
| Abstrakt: |
Introduction:Drug safety initiatives globally have endorsed human iPSC-derived cardiomyocytes (hiPSC-CMs) as an in vitro model for predicting drug-induced cardiac arrhythmia. However, the extent to which human-defined features of in vitro arrhythmia, such as action potential prolongation and after depolarizations, predict actual clinical risk has been much debated.Hypothesis:Deep learning can be used to discriminate features of in vitro action potential features that associate with dangerous drugs and distinguish them from features evoked by safe drugs. Furthermore, the deep learning algorithm should be able to discern the influence of myopathic gene variants on drug-induced arrhythmia.Methods:We trained a deep neural network (DNN) to detect in vitrofeatures of electrophysiological recordings induced by drugs with established clinical risk. The 40 drugs used in the study comprised high, intermediate, and low risk. Action potentials were optically recorded for each drug at 8 concentrations in iPSC-CMs from 3 healthy donors and in isogenic iPSC-CMs carrying 5 gene variants that cause dilated and hypertrophic cardiomyopathies. Multiple differentiation batches were evaluated for each.Results:The trained DNN accurately classified reference drugs according to high, intermediate, and low arrhythmic risk in people. The risk profile of the test drugs was similar across hiPSC-CMs from different healthy donors. The introduction of gene variants associated with arrhythmogenic cardiomyopathies increased the propensity of certain intermediate and high-risk drugs to induce arrhythmia.Conclusions:Deep learning can identify in vitro arrhythmic features that correlate with clinical arrhythmia and discern the influence of pathological gene variants on drug-induced arrhythmia. |
| Databáze: |
Supplemental Index |
| Externí odkaz: |
|
