Commentary on ‘‘Toxicity Testing in the 21st Century: A vision and a Strategy’’

Abstract

The vision of the NAS Committee represents a transition in toxicology from an emphasis on the study of apical endpoints in whole animals to a focus on the study of pathway perturbations, based primarily on in vitro data, from which dose-response models will ultimately be developed. It should be emphasized that this vision does not seek to eliminate whole-animal testing altogether; however, it does aim to eventually use animal testing to a far less extent than in our current system. Such a change to the current method of toxicity testing represents an important goal for many reasons. For example, toxicity testing needs to incorporate the knowledge that has been developed through the most recent advances in research, particularly at the molecular level. Through the application of mechanistic information resulting from the testing of chemicals by the proposed in vitro methods, it may be possible to predict the toxicity of similar chemicals that have not yet been tested in whole animals. The number of chemicals in commercial use for which we have adequate toxicity information is limited, and higherthroughput assays have the potential to increase this number. While the vision’s approach would result in a loss of information from a reduction in testing at high doses, the NAS Committee has proposed that this lack of information would be compensated by a greater understanding of effects at low doses, and that such an understanding can eventually be applied in a practical context. Clearly, however, there are multiple challenges to the actual implementation of the vision. First, we need to consider the vast biological complexity of whole organisms. Will in vitro assays for pathway perturbations, in the absence of some understanding from in vivo studies, be able to identify all of the possible toxic endpoints of a particular chemical? The application of in vitro methods may be a slow and uncertain process without some initial hypotheses about likely critical apical endpoints for a particular chemical. In contrast, with animal bioassays, we can identify toxic endpoints that we may not have originally thought to look for, through the analysis of different target organs. An understanding of how pathway activation and the corresponding dose-response relationships change with differences in the duration of exposure will be necessary, and yet it may be difficult to distinguish between acute, subchronic, and chronic effects at the pathway level. In addition, in vitro assays cannot mirror the metabolism of a whole animal, and the NAS Committee has agreed that much research will be needed to ensure that the new testing methods associated with their vision fully evaluate the effects of both chemicals and their metabolites. The development of ‘-omic’ technologies (such as genomics, proteomics, and metabolomics) is a key element for advancing the vision of the NAS Committee; however, there are challenges to the use of these approaches for the prediction of in vivo toxicity. Rhomberg et al. state that: