ABSTRACT
Historically and still today, genotoxicity tests are conducted at high doses and interpreted in a binary fashion, either as positive or not. The basis for this approach is no longer relevant, given what is now understood about general lack of relevance of high dose effects to low dose exposure scenarios and modes of action for genotoxicity. Thus, dose selection for genotoxicity tests, previously driven by maximum tolerated dose concepts, should be revisited. A new, more pragmatic approach is needed, one that encompasses the entire dose–response and encourages integration of all available information including information on human exposure, internal dose, and dose to the critical targets. We believe that integrating actual/predicted human exposure information into dose selection will result in data that are more useful to inform assessment of risks to human health.
ACKNOWLEDGMENTS
The authors acknowledge valuable discussions of these concepts with Drs. Dan Wilson and Jim Bus.