The paradoxical lack of interspecies correlation between plasma concentrations and chemical carcinogenicity

Several chemicals have been shown, within the limits of epidemiology, to be thus far not carcinogenic in humans at systemic exposure levels similar to those associated with carcinogenicity in rodents. The data are discussed in terms of mechanisms which appear to operate in rodents but not in humans. Pharmacokinetic methods for interspecies comparisons must always be validated by a complementary scrutiny of the pharmacodynamic processes involved, i.e., by considering the responsiveness of the species in question to a given systemic concentration of xenobiotic. For carcinogenesis this validation step is usually beyond reach, due to the time frame for the onset of cancer in humans. For these reasons it is argued that, in the absence of knowledge of mechanisms, there is usually no scientific basis for using the concentrations of a xenobiotic that occur in body fluids or tissues during a rodent carcinogenicity test to make a quantitative carcinogenic risk assessment in humans.