Abstract
In the 2022-Risk Assessment I (RA I) under the Chemical Substances Control Law, hazard assessment values (HAVs: carcinogenic 10-5 risk levels) for 6 “no-threshold” carcinogens were derived from CPDB representative TD50 (rTD50). The validity of rTD50 was, however, low. We herein investigated alternative HAVs derivation for 3 substances (A: p-chloronitrobenzene, B: 2,2'-dichloro-4,4'-methylenedianiline, C: vinyl acetate) instead of applying rTD50. Initially, tumor incidence in studies done with multiple doses were selected, and candidate HAVs were derived from TD50 of individual tumor(s) (iTD50) as well as from the NOAEL. The validity of such HAVs was evaluated referring to the HAV derived from the BMDL10 based on the incidence of the respective tumor. As a result, the lowest iTD50-derived HAV was close to the BMDL10-derived HAV in the case of B, which induced tumors at the lowest dose and thus had no NOAEL. For A, the NOAEL-based HAV was considered appropriate, because TD50 was not available for the best study showing clear dose-response in the tumor incidence. The same was true for C, which showed a low incidence of tumors at the highest dose and an adverse effect at an intermediate dose in a study done with suitable dose range for carcinogenicity detection. As shown above, it was clarified that, instead of applying TD50, selection of the method for deriving the best carcinogenic HAV for “no-threshold” substances in RA I is needed depending on the study conditions, tumor incidence and its dose-response. (Please note that this assessment is based on our own considerations and is not directly related to any administrative decisions made by the Ministry of Health, Labour and Welfare)
