On-Chip Quasi-in vivo Predictive Cardiotoxicity Assay using Spatiotemporal Fluctuation Measurement on Human Cardiomyocyte Cell-Network

Abstract

Lethal arrhythmia is one of the major safety concerns for developing drug candidate compounds, nevertheless, conventional cardiotoxicity prediction methods like in vitro hERG/APD and in vivo QT assay cannot be eliminated the false negative/positive problem. To overcome this problem, we have developed an on-chip new in vitro predictive cardiotoxicity assay having a quality of clinical screening using constructive human cardiomyocyte networks derived from human stem cells (hES-CMs) employing a spatiotemporal measurement of fluctuation (short-term variability; STV) of temporal repolarization of cells and spatial cell-to-cell conduction, representing two origins of lethal arrhythmia. First, we examined the electrophysiological prolongation of field potential durations (FPDs) of human cardiomyocytes to compare the conventional in vivo QT prolongation and clinical results. The results showed (1) FPD prolongation has the ability to detect QT/pro-arrhythmic risks almost similar to the conventional in vitro APD measurements such as papillary muscle and Langendorff hearts. However, they did not show significant improvement of the false-positive/negative problems. Then we evaluated the new two spatiotemporal fluctuations, and found that (2) temporal fluctuations of FPDs (STVFPD) predicted ventricular arrhythmia risks more precisely in most of the representative compounds including false-positive/negative compounds except for terfenadine, and that (3) even the risk of those false-negative drugs including terfenadine was predicted by the fluctuation of spatial cell-to-cell conduction in the lined-up cardiomyocyte network. In conclusion, only the replacement of animal cells to human cardiomyocytes in conventional in vitro screening is not enough to get precise prediction, whereas, the combination of human cardiomyocytes with the new approach of two spatiotemporal fluctuation measurements of temporal STVFPD and spatial STVConductance gives us an potential of global precise predictive arrhythmic cardiotoxicity as a quasi-in vivo human cell network screening beyond conventional hERG and APD/QT assays.