Moving towards better predictors of drug-induced Torsade de Pointes

The Health and Environmental Sciences Institute (HESI) convened a two-day workshop to explore the biological mechanisms responsible for inherited, acquired and induced long QT interval syndromes. Investigational areas included QT dynamics, cell biology, nonclinical models of Torsade de Pointes (TdP) arrhythmia as well as how to ascertain the predictability of such models for human outcome were discussed. Magisterial lectures from academia covered state-of-science knowledge on these domains, whereas industry and regulatory authority representatives dealt with the advantages, disadvantages and desirable requisites of nonclinical assays to assess drug-associated cardiac safety. Proposals for future research projects were not sufficiently factual to allow discrimination between pragmatic and ideal solutions. A key objective of the workshop was to foster initiatives addressing development of nonclinical proarrhythmic models for identifying without failure drug candidates with the potential to cause threshold QT interval increases of regulatory concern (5-10 ms) in healthy volunteers and a TdP event in 1/10(5)-10(7) patients. In the authors' opinion, fulfilling this goal (an FDA prerequisite for abrogating E14 'thorough QT study' requirement) pertains more to a Faustian quest than to a realistic, nonclinical safety pharmacology assignment. Indeed, biological assays are, by nature, characterised by an implicit degree of uncertainty, contradicting the precautionary principle of zero error expectancy. For the moment, relatively successful strategies should rely on expertly designed and executed S7B core assays, complemented as needed, by reliable proarrhythmia tests. These studies should be preceded, when available, by application of powerful in silico fingerprint technology mining databases containing pertinent public and proprietary (liberally released) cardiac safety information on reference, marketed, withdrawn and failed drugs.