Literature-based evaluation of four 'hard endpoint' models for assessing drug-induced torsades de pointes liability.

In safety pharmacology, a number of preclinical models for detecting drug-induced proarrhythmia liability have been recently introduced that utilize hard endpoints: early after depolarziations (EADs), torsades de pointes (TdP) or both as the principal biomarker. To explore the validity of four of the most common of these models, (the isolated canine/rabbit left ventricular wedge preparation, the isolated rabbit heart, the methoxamine-pretreated anaesthetized rabbit and the complete, chronic AV-blocked (CAVB) dog (conscious and anaesthetized), the present article reviews published data sets for three drugs with recognized and different human TdP liabilities (cisparide, terfenadine and moxifloxacinin). Finally, this review considers the value of inclusion of analysis of beat-to-beat variability of repolarization (BVR) in TdP liability testing to improve sensitivity and specificity.