Predictivity of in vitro non-clinical cardiac contractility assays for inotropic effects in humans--A literature search

Adverse drug effects on the cardiovascular system are a major cause of compound attrition throughout compound discovery and development. There are many ways by which drugs can affect the cardiovascular system, including effects on the electrocardiogram, vascular resistance, heart rate and the force of contraction of the heart (inotropy). Compounds that increase the force of contraction of the heart can be harmful in patients with ischemic heart disease, whilst negative inotropes can induce symptoms of heart failure. There is a range of non-clinical in vitro and in vivo assays used to detect inotropic effects of drugs. We have conducted a literature review of the in vitro assays and compared the findings from these with known effects on cardiac contractility in man. There was a wide variety of assays used, ranging from perfuse whole hearts to isolated regions of the heart (papillary muscle, ventricle and atria), which were removed from a number of species (cat, guinea pig, rabbit and rat). We conducted two analyses. The first was investigating the concordance of the findings from the in vitro assays at any concentration with those observed in man (an assessment of hazard identification) and the second was the concordance of the in vitro findings at concentrations tested up to 10-fold higher than those tested in the clinic. We found that when used as a hazard identification tool, the available assays had good sensitivity (88%), although the specificity was not so good (60%), but when used as a risk management tool the sensitivity was considerably reduced (sensitivity 58-70% and specificity 60%). These data would suggest that the available in vitro assays can be used as hazard identification tools for adverse drug effects on cardiac contractility, but there is a need for new assays to better predict the exposures in man that may cause a change in cardiac contractility and therefore better predict the likely therapeutic index of compounds prior to nomination of compounds for clinical development.