of action, and its intended use and estimated exposures (Carmichael et al. 2006). Those factors are used to refine testing priorities to focus first on areas of greatest concern in early tiers and then to move judiciously to advanced testing in later tiers as needed. In addition, an emphasis on pharmacokinetic studies in tiered approaches has been considered in recent discussions of improving toxicity testing of pesticides (Carmichael et al. 2006; Doe et al. 2006).
Tiered testing has been recommended in evaluating the toxicity of agricultural products (Doe et al. 2006), in screening for endocrine disruptors (Charles 2004), and in assessing developmental toxicity (Spielman 2005) and carcinogenicity (Stavanja et al. 2006) of chemicals and products. A tiered-testing approach also has the promise to include comparative genomic studies to help to identify genes, transcription-factor motifs, and other putative control regions that are involved in tissue responses (Ptacek and Sell 2005). The increasing complexity of biologic information—including genomic, proteomic, and cell-signaling information—has encouraged the use of a more systematic multilevel approach in toxicity screening (Yokota et al. 2004).
The systematic development of tiered, decision-tree selection of more limited suites of animal tests could conceivably provide toxicity-testing data nearly equivalent to those currently obtained but without the need to conduct tests for as many apical end points. The use of appropriately chosen computational models and in vitro screens might also permit sound risk-management decisions in some cases without the need for in vivo testing. Both types of tiered-testing strategies offer the potential of reducing animal use and toxicity-testing costs and allowing flexibility in testing based on risk-management information needs. Although the committee recognized the potential for incremental improvement in toxicity testing through a tiered approach, Option II still represents only a small step in improving coverage, reducing costs and animal use, and increasing mechanistic information in risk