Evaluation of Uncertainty in Input Parameters to Pharmacokinetic Models and the Resulting Uncertainty in Output,

Physiologically based pharmacokinetic PB-PK models may be used to predict the concentrations of parent chemical or metabolites in tissues, resulting from specified chemical exposures. An important application of PB-PK modeling is in assessment of carcinogenic risks to humans, based on animal data. The parameters of a PB-PK model may include metabolic parameters, bloodair and tissueblood partition coefficients, and physiological parameters, such organ weights and blood flow rates. Uncertainty in estimates of these parameters results in uncertainty regarding tissue concentrations and resulting risks. Data are reviewed relevant to the quantification of these uncertainties, for a PB-PK model-based risk assessment for tetrachloroethylene. Probability distributions are developed to express uncertainty in model parameters, and uncertainties are propagated by a sequence of operations that simulates processes recognized as contributing to estimates of human risk. Distributions of PB-PK model output and human risk estimates are used to characterize uncertainty resulting from uncertainty in model parameters. Physiologically based pharmacokinetic models can be used in the animal-based assessment human cancer risks. A PB-PK model is assumed for rodents and humans with parameter values are possibly species-specific and dose surrogates are calculated on the basis of that model. A dose surrogate is a particular measure of chemical delivered to a putative target tissue.