Resumen de Empirical analysis of the variability of the maximum likelihood estimates of benzene cancer risks

Paolo F. Ricci, Louis A. Cox Jr.

• This paper presents results from an on-going study of the carcinogenicity of benzene at low doses. It focuses on the statistical variability of the parametric maximum likelihood estimator (MLE) of the parameters of the linearized multistage (LMS) cancer dose-response model. This paper describes: 1) how pharmacokinetic conversions from animal to human were made to convert administered benzene to estimated internal doses of its metabolites; 2) the exact form of the LMS; 3) likelihood surfaces for fitting the exact form to data; and 4) results of bootstrap and Monte Carlo simulations that assess the variability of the parameters of the dose-response model. This paper suggests that the cubic form of the LMS dose-response for animal tumors is robust to sampling variability. For environmental policy, the hypothesis that linearity governs low dose benzene carcinogenicity is probably incorrect.