Default factors for interspecies differences in the major routes of xenobiotic elimination

For the risk to human health posed by chemicals that show threshold toxicit y there is an increasing need to move away from using the default approache s, which inherently incorporate uncertainty, towards more biologically defe nsible risk assessments. However, most chemical databases do not contain da ta of sufficient quantity or quality that can be used to replace either the interspecies or interindividual aspects of toxicokinetic and toxicodynamic uncertainty. The purpose of the current analysis was to evaluate the use o f alternative, species-specific, pathway-related, "categorical" default val ues to replace the current interspecies toxicokinetic default uncertainty f actor of 4.0. The extent of the difference in the internal dose of a compou nd, for each test species, could then be related to the specific route of m etabolism in humans. This refinement would allow for different categories o f defaults to be used, providing that the metabolic fate of a toxicant was known in humans. Interspecies differences in metabolism, excretion, and bioavailability have been compared for probe substrates for four different human xenobiotic-met abolizing enzymes: CYP1A2 (caffeine, paraxanthine, theobromine, and theophy lline), CYP3A4 (lidocaine), UDP-glucuronyltransferase (AZT), and esterases (aspirin). The results of this analysis showed that there are significant differences between humans and the four test species in the metabolic fate of the probe compounds, the enzymes involved, the route of excretion and oral bioavaila bility - all of which are factors that can influence the extent of the diff erence between humans and a test species in the internal dose of a toxicant . The wide variability between both compounds and the individual species su ggests that the categorical approach for species differences may be of limi ted use in refining the current default approach. However, future work to i ncorporate a wider database of compounds that are metabolized extensively b y any pathway in humans to provide more information on the extent to which the different test species are not covered by the default of 4.0. Ultimatel y this work supports the necessity to remove the uncertainty from the risk assessment process by the generation and use of compound-specific data.