In Situ Monitoring of Drug Induced QT Alterations and Pro-arrhythmic Effects of Compounds Within a 3D Beating Heart-on-Chip Platform

Abstract: Detecting drug cardiotoxicity at early stages of the drug development process is still critical. Hence the development of relevant in-vitro pre-clinical models resembling the behaviour of human heart and able to effectively predict the clinical outcomes is highly envisioned. Here we present uHeart, a human functional 3D cardiac model developed within a beating Organ-on-Chip (OoC) platform integrating fit-to-purpose assays for detecting drug-induced electrophysiological alterations in pre-clinical stages. The model was developed and qualified for cardiotoxicity screening by following the latest International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) S7b guidelines. 

uHeart platform encompasses i) uBeat® technology, that mechanically trains cardiac microtissues by providing them a physiological uniaxial cyclic strain (i.e., 10% stretching, 1 Hz) and ii) uSense technology, integrated in situ electrodes able to read electrophysiological cardiac signals (i.e., Field Potential FP). Human induced pluripotent stem cell derived cardiomyocytes (h-iPSC-CMs, iCell) and human dermal fibroblasts (h-DF), 75%-25% ratio, were embedded in fibrin hydrogel at 125*106 cells/ml and cultured for up to 10 days within uHeart. Upon achievement of functional cardiac microtissues synchronously and spontaneously beating, 11 drugs listed in the Comprehensive in vitro Proarrhytmia Assay (CiPA) and affecting single or multiple cardiac ion-channels were selected to qualify the model. Drug-induced alterations were evaluated at incremental doses and by analysing the FP morphology (i.e., beating period-BP, spike amplitude-AMP, field potential duration-FPD) and the onset of arrhythmic events. DMSO and Aspirin were used as vehicle and negative controls, respectively.

Microtissues beat synchronously after 5 days and FP signals showed the typical depolarization and re-polarization spikes. Ikr blockers (e.g., Dofetilide, Quinidine) prolonged the FPD at concentration near the Cmax, while ICaL blockers (e.g., Verapamil, Nifedipine) shorten it already at 5-50nM. Mexiletine, blocking INa, statistically decreased the FP amplitude at 10μM. Both Terfenadine and Dofetilide elicited arrhythmic events, matching FDA labels indications. Overall, the system showed 83.3% sensitivity, 100% specificity and 91.6% accuracy in detecting FPD prolongation. DMSO (up to 0.5% w/v) and Aspirin (up to 100µM) did not statistically alter the repolarization time.

uHeart generates functional 3D cardiac in-vitro models, predicting compound specific toxic effects at concentrations near the Cmax, resulting in a powerful tool for both detecting QT interval prolongation and pro-arrhythmic effects of drugs. Beyond this application, we are currently applying uHeart to investigate the acute effects of compounds on cardiac contractility (e.g. inotropic compounds), to study drug structural cardiotoxicity (e.g. effects after prolonged doxorubicin administration) and to model cardiac genetic diseases through integration of patient derived iPSC-CMs affected by specific mutations related to cardiac pathologies (e.g. laminopathies).