PHYSIOLOGICALLY-BASED PHARMACOKINETIC MODELING OF A TERNARY MIXTURE OF ALKYL BENZENES IN RATS AND HUMANS

The objective of the present study was to develop a physiologically ba sed pharmacokinetic (PBPK) model for a ternary mixture of alkyl benzen es [toluene (TOL), m-xylene (XYL), and ethylbenzene (EBZ)] in rats and humans. The approach involved the development of the mixture PBPK mod el in the rat and extrapolation to humans by substituting rat physiolo gical parameters and blood:air partition coefficients in the model wit h those of humans, scaling maximal velocity for metabolism on the basi s of body weight(0.75) and keeping all other model parameters species- invariant. The development of the PBPK model for the ternary mixture i n the rat was accomplished by initially validating or refining the exi sting PBPK models for TOL, XYL, and EEZ and linking the individual che mical models via the hepatic metabolism term. accordingly, the Michael is-Menten equation for each solvent was modified to test four possible mechanisms of metabolic interaction (i.e., no interaction, competitiv e inhibition, noncompetitive inhibition, and uncompetitive inhibition) . The metabolic inhibition constant (K-i) for each binary pair of alky l benzenes was estimated by fitting the binary chemical PBPK model sim ulations to previously published data on blood concentrations of TOL, XYL, and EBZ in rats exposed for 4 hr to a binary combination of 100 o r 200 ppm of each of these solvents. Competitive metabolic inhibition appeared to be the most plausible mechanism of interaction at relevant exposure concentrations for all binary mixtures of alkyl benzenes in the rat K-i,K-TOL-XYL = 0.33; K-i,K-EBZ-XYL = 0.23 mg/L). Incorporatin g the K-i values obtained with the binary chemical mixtures, the PBPK model for the ternary mixture simulated adequately the time course of the venous blood concentrations of TOL, XYL, and EBZ in rats exposed t o a mixture containing 100 ppm each of these solvents. Following the v alidation of the ternary mixture model in the rat, it was scaled to pr edict the kinetics of TOL, XYL, and EBZ in blood and alveolar air of h uman volunteers exposed for 7 hr to a combination of 17, 33, and 33 pp m, respectively, of these solvents. Model simulations and experimental data obtained in humans indicated that exposure to atmospheric concen trations of TOL, XYL, and EBZ that remain within the permissible conce ntrations for a mixture would not result in biologically significant m odifications of their pharmacokinetics. Overall, this study demonstrat es the utility of PBPK models in the prediction of the kinetics of com ponents of chemical mixtures, by accounting for mechanisms of binary c hemical interactions. (C) 1997 Academic Press.