Evaluation of Potential Cancer Risks Associated with Chemical Mixtures Using Biologically Based Dose-Response Models: Sensitivity Analyses of Models Using Simulated Experiments. R. C. Lee, E. G. Luebeck, S. B. Curtis, S. M. Bartell, and E. M. Faustman, University of Washington Department of Environmental Health, Box 357234, Seattle, WA 98195; and Fred Hutchinson Cancer Research Center, Public Health Sciences Division, MP-665, Seattle, WA 98104
There has been much recent interest regarding methods of evaluating potential cancer risks associated with chemical-chemical and radionuclide-chemical mixtures. Stochastic biologically based dose-response (BBDR) models, along with physiologically based toxicokinetic (PPTK) models, offer possible means of quantifying these risks. PPTK models have been used to examine relative delivered doses of multiple compounds and metabolites, while BBDR models have been used to evaluate the combined effects of cancer initiators and promoters. The present effort examines the sensitivity of BBDR models to variations in model parameters. The BBDR model parameters tested include cell initiation rates, cell division rates, and net cell proliferation rates. Two methods are explored. The first varies specific model parameters individually over biologically reasonable ranges observed in previous experiments. The second uses maximum-likelihood estimation to fit parameters to simulated data sets consisting of different combinations of chemicals with similar and dissimilar carcinogenic mechanisms. In both instances, interaction effects on parameter values and model output are presented in 3-dimensional graphs. These analyses provide information regarding mechanistic uncertainties associated with chemical interactions and metabolites, as well as providing vital guidance for laboratory-based toxicological evaluations of mixtures.
Work partially supported by a cooperative agreement with the U.S. Department of Energy.