Biologically based risk assessment models for developmental toxicity.

Abstract

It is obvious that the task of incorporating mechanistic information into dose-response assessment for developmental toxicity is, by and large, still at its conceptual stage. Our immature understanding of embryogenesis and teratogenesis forces us to make simplifying biological assumptions that may turn out to be erroneous; therefore, these mechanistically based models should be constructed so as to be easily modified as new information becomes available. The data-intensive (and costly) nature of these modeling efforts may also limit their practice to extraordinary situations where, for instance, large segments of the human population are exposed to low levels of a compound and the determination of a safe level of exposure is of utmost importance, or where compounds are of such immense economic or therapeutic value that their use would warrant a concerted effort to minimize the uncertainties inherent in the current methods of extrapolation. Nevertheless, this chapter has presented several attempts to translate the concept of BBDR into practice. When these applications become successful, these models should provide risk assessors with more reliable response indicators at low doses (where human exposures are realistic) and more accurate cross-strain/cross-species comparisons as well as extrapolations across exposure conditions. Although the BBDR models do not necessarily redefine the current practice of risk assessment using either the NOAEL or the benchmark approach, information derived from these efforts should provide a credible scientific basis for the estimates of RfD.