Cardiotoxicity is one of the most common risks for drugs in development, often manifesting as prolongation of the QT interval on the ECG with the potential for fatal ventricular arrhythmias. Even the newest released heart drugs, like Ivabradine, block the hERG potassium channel (an off-target innocent bystander protein) to cause severe arrhythmias and potentially-lethal Torsades de Pointes. Frequently, the “off-target” binding site exists in proteins that are closely related to the therapeutic binding protein. Moreover, testing a single drug on a single cell is not the same as testing it in tissues or in animals.

The major factor plaguing drug development is the absence of a preclinical drug screen that can accurately predict unintended drug induced cardiac arrhythmias. The current approaches rely on substitute markers such as QT interval prolongation on the ECG. Unfortunately, QT prolongation is neither specific nor sensitive enough to indicate likelihood of arrhythmias. There is an urgent need to identify a new approach that can predict actual pro-arrhythmia rather than surrogate indicators. The challenge is that the myocardium is inherently a multi-scale organ, where microscopic events (drug binding to a cardiac ion channel) generate emergent electro-mechanical effects at the macroscopic level.

The fundamental mode of drug interaction derived from each drug’s unique structure activity relationship determines the resultant effects on cardiac electrical activity in cells and tissues and hence alters electro-mechanical coupling in the heart. By capturing these complex drug channel interactions in a computer simulations of key cardiac targets combined with experimental studies done in in-house developed cardiac myocyte cell lines, we will be able to predict drug safety or electro-toxicity in the heart. We strive to understand the receptor scale interactions of drugs with two key cardiac targets to functional scale predictions at the level of the channel, cell and tissues. Predictions from the atomic structure simulations can be used to inform the kinetic parameters of models that capture the complex dynamical interactions of drugs and ion channels.

The approach developed by Achlys will allow simulation and prediction of small molecule cardiotoxicity from its chemical structure, thus offering an indispensable tool for in silico pre-clinical and clinical trials. Experiments in mammalian cells and tissues are prospectively used to improve our models and validate predictions.