Experimental and mathematical modeling methods for the investigation of toxicological interactions

While procedures have been developed and used for many years to assess risk and determine acceptable exposure levels to individual chemicals, most cases of environmental contamination can result in concurrent or sequential exposure to more than one chemical. Toxicological predictions of such combinations must be based on an understanding of the mechanisms of action and interaction of the components of the mixtures. Statistical and experimental methods test the existence of toxicological interactions in a mixture. However, these methods are limited to experimental data ranges for which they are derived, in addition to limitations caused by response differences from experimental animals to humans. Empirical methods such as isobolograms, median-effect principle and response surface methodology (RSM) are based on statistical experimental design and regression of data. For that reason, the predicted response surfaces can be used for extrapolation across dose regions where interaction mechanisms are not anticipated to change. In general, using these methods for predictions can be problematic without including biologically based mechanistic descriptions that can account for dose and species differences. Mechanistically based models, such as physiologically based pharmacokinetic/pharmacodynamic (PBPK/PD) models, include explicit descriptions of interaction mechanisms which are related to target tissues levels. These models include dose-dependent mechanistic hypotheses of toxicological interactions which can be tested by model-directed experimental design and used to identify dose regions where interactions are not significant. Published by Elsevier Inc.