Rat and human sensory evoked potentials and the predictability of human neurotoxicity from rat data.

The development of comprehensive quantitative models as alternatives to risk assessment based on uncertainty factors will require many steps, among them consideration of the relationships between the health endpoints which are measured in laboratory animals and humans. Sensory evoked potentials are measures of sensory function which can be recorded from many species, including humans, and as such provide an opportunity for examining the extrapolation of neurotoxicity data from laboratory animals to humans. Our research strategy for investigating how well laboratory rat data predict human neurotoxic risk involves comparing parametric stimulus manipulations and drug treatments in both species. Finally, we are comparing results in humans with neurodegenerative conditions, including those induced by neurotoxicant exposure, with animal models. To date, we have focused on pattern-elicited visual evoked potentials (VEPs) recorded from pigmented rats and humans. Parametric manipulations of spatial frequency, temporal frequency and stimulus contrast revealed parallel functions, displaced for differences in absolute sensitivity. Additionally, diazepam produced similar effects in rats and human volunteers. A quantitative cross-species map was developed to illustrate the prediction of human effects from rat data. Exposure to carbon disulfide produced changes in rat VEP-derived contrast sensitivity functions, which resembled psychophysically-measured loss of visual contrast sensitivity in human workers exposed to organic solvents. The results of these continuing efforts should help indicate how well animal electrophysiological measures predict human neurotoxicity.