BS34 Prediction and ranking of inotropic drug effects of checkpoint kinase inhibitors using an in silico cardiotoxicity screen and the primary cardiomyocyte work-loop assay

Introduction Checkpoint kinase (ChK) inhibitors represent an important class of cancer therapeutic agents, although several drugs within this class have failed in clinical trials due to adverse cardiovascular related toxicity. Unfortunately, adverse effect of drugs on the cardiovascular system (CVS) is a major cause for compounds failing in both non-clinical and clinical studies. Development of robust, predictive and reliable in-silico and in-vitro assay systems is important in order to establish better predictive triage systems for potential drug-induced cardiotoxicity in order to reduce attrition during drug development and adverse outcomes in a clinical setting. Developing drugs free from cardiotoxicity has been particularly challenging for pharmaceutical companies, due to the absence of effective assays deployable at the medicinal chemistry stage of pre-clinical research. Eight ChK inhibitors associated with varying degrees of drug-induced cardiotoxicity were investigated utilising an in-silico cardiotoxicity screen and a rat cardiomyocyte Work-Loop assay to understand the predictivity and associated concordance with human data for these two assays. Methods Cardiotoxicity in-silico modelling was performed with a calibrated structure-based algorithm comparing a range of ChK inhibitor structures (i.e. Chir-124, PF477736, AZD7762, Rabusertib, Prexasertib, SCH90076, SRA737 and SAR020106) to marketed drug structures. The number of structurally significant hits to (1) known cardiotoxic drugs and (2) known safe drugs were assessed and risk predicted from the number balance of hits. Work-Loop cardiomyocyte assay: primary rat cardiomyocytes were isolated and used to measure cardiac contractility (IonOptix) by clamping diastolic and systolic forces (pre- and after-load), mimicking the cardiac cycle. Contractility parameters included Emax (end-systolic elastance), Vmax (velocity of shortening) and Fmax (maximum isometric force). Stable cardiomyocyte Work-Loops were established and perfused in the absence or presence of a range of ChK inhibitors (1μM) or vehicle controls (n=4–8). Results The cardiotoxicity in-silico platform was able to predict the cardiotoxicity risk for all relevant ChK inhibitor compounds. The Work-Loop cardiomyocyte assay accurately predicted potential for rate and haemodynamic changes in all ChK inhibitor compounds where relevant. The Work-Loop assay provided multiple readouts for the negative inotropic responses (without system manipulation). For example, Chir-124 and Rabusertib elicited decreases in both Emax (55.5+/-15.3%; 27.8+/-7.2% respectively) and Fmax (64.4+/-19.1%; 35.2+/-18.9% respectively), whilst minimal differences in contractile performance with SRA737 were observed. The Work-Loop assay was able to detect electromechanical coupling alterations and relevant potency rank order for the ChK inhibitors. Discussion This study provides information on the use of a cardiotoxicity in-silico platform and Work-Loop assay. This alternative triage system usefully contains pre- and post-chemical synthesis components. This is the first study to demonstrate the utility of the in-silico platform and Work-Loop assay to predict cardiotoxicity risk for ChK inhibitors and provide insight into the intricate mechanisms implicated in the adverse effects of drug induced inotropy. It is also able to predict the relevant rank order of potency of compounds from the same class. Conclusion This study has demonstrated how these assays could be used to improve early detection of cardiac safety issues and support Safety Pharmacologists and Drug Discovery project teams avoid expensive downstream problems. Conflict of Interest n/a