Video-electroencephalography in conscious non human primate using radiotelemetry and computerized analysis: refinement of a safety pharmacology model.

INTRODUCTION: Electroencephalography (EEG) investigations are occasionally required as follow-up studies for safety pharmacology core battery (S7A). Video-EEG monitoring is a standard diagnostic tool in humans but limited data is available on its use in telemetered freely moving macaque monkeys for safety pharmacology investigations. While proconvulsant risk evaluations are routinely conducted in rodents, pharmacological or pharmacokinetic considerations lead to the use of non human primates in toxicology and safety pharmacology in some cases. METHODS: Cynomolgus monkeys were instrumented with telemetry implants. Placement of EEG electrode was based on the 10-20 system using three derivations (C3-O1, Cz-Oz and C4-O2). EEG trace analysis was carried out using NeuroScore software. After 24 h of continuous video-EEG monitoring, animals received pentylenetetrazole (PTZ, 10 mg/kg/15 min) until convulsions were noted. Convulsions were immediately treated with diazepam (1.0 mg/kg). A seizure detection protocol with a dynamic spike train threshold was used for the entire EEG monitoring period (total of 44 h) including periods when PTZ was administered. Spectral analysis was done to quantify the absolute and relative amplitude of EEG frequency bands (delta, theta, alpha, sigma and beta waves). Sleep stages were quantified and EEGs during seizures were analyzed using fast Fourier transformation (FFT) to assess dominant frequencies. RESULTS: Spike trains were detected by computerized analysis in all animals presenting PTZ-induced seizures while paroxysmal activities were systematically predictive (at least 4-min prior to generalized seizures). Beta activity increased with visual stimulation using monkey treats. Characteristics of EEG for all sleep stages (I, II, III and IV) were present in all animals. Delta activity was predominant in normal awake EEG as well as in all sleep stages. Seizure peak frequency was 3-6 Hz on FFT, corresponding to the discharge of the underlying generator. DISCUSSION: EEG-video monitoring can be useful when using non human primates to characterize neurological adverse effects with unpredictable onset. Computerized video-EEG analysis was a valuable tool for safety pharmacology investigations including proconvulsant risk assessment, spectral analysis of frequency bands and sleep stage determination.

Conscious and anesthetized non-human primate safety pharmacology models: hemodynamic sensitivity comparison.

INTRODUCTION: Drug-induced cardiovascular effects identified in conscious cynomolgus monkeys equipped with tethers and prepared for radiotelemetry were compared with results from anesthetized non-human primate (cynomolgus and rhesus) models. METHODS: Remifentanil (4.0 microg/kg, bolus), esmolol (2.0 mg/kg, bolus) and dopamine (0.05 mg/kg/min, 30 min infusion) were given intravenously to all models. RESULTS: Remifentanil decreased heart rate (HR), systolic, mean and diastolic systemic arterial pressures (SAP) in anesthetized animals while conscious monkeys presented an increase in HR, systolic, mean and diastolic SAP, as seen in humans for the respective state of consciousness (conscious and anesthetized). Esmolol decreased HR, systolic, mean and diastolic SAP in anesthetized monkeys while only HR, systolic and mean SAP achieved a statistically significant decrease in the conscious model. The amplitude of SAP reduction was greater in anesthetized models, while the amplitude of HR reduction was greater in the conscious and anesthetized cynomolgus models than in the anesthetized rhesus model. Dopamine induced a significant increase in HR, systolic, mean and diastolic SAP in anesthetized models without any statistically significant effect on HR and SAP in the conscious model. DISCUSSION: The amplitude of hemodynamic and chronotropic alterations induced by positive control drugs was generally greater in anesthetized than in conscious models and statistical significance was achieved more often with the anesthetized models. These results suggest that an anesthetized model may be valuable as part of a drug screening program for cardiovascular safety evaluations in addition to a conscious model.