The changing landscape of cardiac safety testing
Rationale for more predictive cardiac safety assays in drug discovery
Drug discovery is an expensive and complicated process, with recent estimates of the cost of bringing each new drug successfully to market of $1-$2 billion.
Implicit in these calculations are the cost of failed drug candidates, which, in industry surveys over the last two decades (Arrowsmith & Miller 2013; Hay et al., 2014; Cook et al., 2014), is attributed firstly to lack of efficacy (~50% of failed compounds) followed by safety issues (~30% of failed compounds). Significantly, cardiac toxicity remains the leading cause of new drug safety side effects (Valentin & Redfern, 2017). Drug attrition due to cardiotoxicity occurs in all phases of drug development (Fig. 1), and the later it occurs the greater the out-of-pocket cost. What is apparent from these industry surveys is that existing pre-clinical cardiac safety assays that rely on in vitro, ex vivo and in vivo animal models are not sufficiently reliable in predicting the human cardiac
risk of new compounds being tested in clinical trials or approved for marketing (red arrow, Fig. 1). Given the ongoing poor predictability of existing cardiac safety testing regimes, industry and regulatory groups across the globe have initiated efforts to improve the translational potential of cardiac safety assays. One key aspect of this is to reduce reliance on animal tissues and models derived from different species. This is because there are key species differences in
cardiac physiology, as well as a growing commitment in industry to meet recommended guidelines for reducing animal testing. Three general approaches
have been proposed to improve the predictability of future cardiac safety assays, all of which have been successfully implemented at Metrion Biosciences:
i Include additional human cardiac ion channels for in vitro screening panels to capture the full cardiac risk profile of test compounds;
ii Use high quality in vitro data in sophisticated in silico models of the human cardiac action potential (AP) to predict arrhythmias, and;
iii Test compounds empirically, and confirm the arrhythmia predictions of in silico models, using translational models employing human induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs). Please download the full pdf of the whitepaper using the tab below.