REASSESSING BENZENE RISKS USING INTERNAL DOSES AND MONTE-CARLO UNCERTAINTY ANALYSIS

Human cancer risks from benzene have been estimated from epidemiologic al data, with supporting evidence from animal bioassay data. This arti cle reexamines the animal-based risk assessments using physiologically based pharmacokinetic (PBPK) models of benzene metabolism in animals and humans. Internal doses (total benzene metabolites) from oral gavag e experiments in mice are well predicted by the PBPK model. Both the d ata and the PBPK model outputs are also well described by a simple non linear (Michaelis-Menten) regression model, as previously used by Bail er and Heel [Metabolite-based internal doses used in risk assessment o f benzene. Environ Health Perspect 82:177-184 (1989)]. Refitting the m ultistage model family to internal doses changes the maximum-likelihoo d estimate (MLE) dose-response curve for mice from linear-quadratic to purely cubic, so that low-dose risk estimates are smaller than in pre vious risk assessments. In contrast to Bailer and Heel's findings usin g interspecies dose conversion, the use of internal dose estimates for humans from a PBPK model reduces estimated human risks at low doses. Sensitivity analyses suggest that the finding of a nonlinear MLE dose- response curve at low doses is robust to changes in internal dose defi nitions and more consistent with epidemiological data than earlier ris k models. A Monte-Carlo uncertainty analysis based on maximum-entropy probabilities and Bayesian conditioning is used to develop an entire p robability distribution for the true but unknown dose-response functio n. This allows the probability of a positive low-dose slope to be quan tified: it is about 10%. An upper 95% confidence limit on the low-dose slope of excess risk is also obtained directly from the posterior dis tribution and is similar to previous q(1) values. This approach sugge sts that the excess risk due to benzene exposure may be nonexistent (o r even negative) at sufficiently low doses. Two types of biological in formation about benzene effects-pharmacokinetic and hematotoxic-are ex amined to test the plausibility of this finding. A framework for incor porating causally relevant biological information into benzene risk as sessment is introduced, and it is shown that both pharmacokinetic and hematotoxic models appear to be consistent with the hypothesis that su fficiently low concentrations of inhaled benzene do not create an exce ss risk.