摘要：Benchmark dose (BMD) method have been used in the toxicological assessment of chemical substances so that the point of departure can be derived, as an alternative to the use of no observable adverse effect level (NOAEL), and the method is often applied to the incidence data of histopathological findings in the toxicity studies. In the present study, the BMD method was applied to various patterns of incidence data derived from some toxicity studies as case studies, and the validity of each application was discussed. Five independent applications including toxicity studies of madder color or semicarbazide hydrochloride were prepared and model averaging over the three models with the lowest three AIC (Akaike information criteria) values (MA-3), a recently proposed model averaging method, was employed. The series of case studies indicated, for the better application of the BMD method to histopathological findings, the following points:(i) If there are incidence data with severity grading of pathologically significant lesions, we must discuss whether the BMD method should be applied to the total incidence data or the incidence data above certain grade with or without data aggregation.(ii) If a lesion of interest had higher toxicological significance rather than the secondary lesions with higher severity, the BMD method should be applied to the incidence data of the lesion of interest.(iii) If it is highly necessary to apply the BMD method to obtained incidence data without toxicological and statistical validity, toxicological pathologists are advised to review individual datasets of histopathology and associated data, and provide new incidence data of comprehensive findings (diagnostic name) such as hepatocellular injury or nephropathy, if possible. In all cases, toxicological significance and mechanism of a lesion of interest need to be considered in light of the dose-dependence. In view of both toxicology and statistics, sufficient discussions must be made on the validity of applying BMD method and its estimate.
