Measuring the risk of torsades de pointes: electrocardiographic evaluation of PNU-142093 in conscious cynomolgus non-human primates using restraint and non-restraint procedures.

OBJECTIVE: Drug-induced torsades de pointes (TdP) arrhythmia is a serious public health concern that has significantly slowed the advancement of promising new therapeutic agents to the marketplace. Modeling for the potential to produce TdP has relied in part on the surrogate biomarker QT interval prolongation, measured in vivo in animals and in the clinic in man. This study was a comparison of the effects of PNU-142093, a selective 5HT1D-serotonin receptor agonist, on QT interval prolongation under restraint and non-restraint conditions in conscious cynomolgus non-human primates. METHODS: Lead II electrocardiograms (ECG) were collected following an oral single-dose (non-restraint conditions using radio-telemetry) and single- and multiple-doses for 14 days (restraint conditions using electrodes applied to the surface) at doses of 0, 5, 15, and 25 mg/kg. ECG were collected from non-restrained animals predose and for up to 5 hrs, and again at 7 hrs, postdose on 4 different days in a Latin-square crossover design; N=4/sex/dose level. ECG were collected from restrained animals on days 1, 7, and 13, predose and at approximately 4 hrs postdose; N=2/sex/group. RESULTS: Non-restrained animal heart rate ranged from 159+/-22.1 to 168+/-21.4 beats/minute when compared to restrained animal heart rate (ranging from 242+/-17.2 to 246+/-11.5 beats/minute), suggesting that non-restrained animals were under less stress. In non-restrained animals, PNU-142093 produced a non-dose related decrease in heart rate, associated with a dose-related increase in QT and QTc (QT interval corrected for changes in heart rate) intervals, which was accompanied by alterations in T-wave morphology (e.g., widening and notching of the T wave). In restrained non-human primates, PNU-142093 had no effect on heart rate or ECG morphology on any day of dosing and no statistically significant effect on QT or QTc intervals on days 1 or 7 of dosing. By day 13 there were statistically significant increases in QT and QTc intervals at 15 and 25 mg/kg. The increase in QTc interval in restrained animals on day 13 was 29+/-12 and 30+/-19 msec at 15 and 25 mg/kg/day, respectively, and that in non-restrained animals was 65+/-23 and 73+/-28 msec. DISCUSSION: These data demonstrate an ability to detect problematic drugs in conscious cynomolgus non-human primates using both restraint and non-restraint procedures. They further show that the sensitivity of these assays to identify this signal of cardiac risk is significantly improved under the condition of non-restraint.

ILSI-HESI cardiovascular safety subcommittee initiative: evaluation of three non-clinical models of QT prolongation.

INTRODUCTION: Drugs that delay cardiac repolarization pose potential safety risks to patients and cause serious regulatory concern because of the link between QT interval prolongation and the potentially fatal arrhythmia torsades de pointes (TdP). Predicting which drugs will cause TdP is an inexact and difficult science. The utility of non-clinical assays was not well understood due in part to variability in methods, species, and consistency in the assays reported in the literature. The Health and Environmental Sciences Institute of the International Life Sciences Institute (ILSI/HESI) outlined a set of studies to determine how well selected commonly used non-clinical assays identified compounds known to cause TdP and prolong QT interval in humans. METHODS: Compounds known to prolong ventricular repolarization and compounds considered safe by years of clinical use were tested in three assays: HERG ionic current, Purkinje fiber repolarization, and in vivo QT studies in conscious telemeterized dogs. RESULTS: The data from each of these assays demonstrate that compounds that may pose a proarrhythmia risk for patients can be distinguished from those that are considered safe. DISCUSSION: Taken collectively, the in-vitro and in-vivo preclinical results can be integrated to develop an accurate preclinical risk assessment to support clinical safety.