Benzene dosimetry in experimental animals: relevance for risk assessment.

The findings of the studies summarized in this report provide some understanding of the possible role of dosimetry in the different response of the rats and mice to benzene in the long-term bioassay studies. The more sensitive species, the mice, definitely has a higher capacity to metabolize benzene and to metabolize it to more of the putative toxic metabolites than do rats. A major finding of these studies is that in three different animal species, from mice to monkeys, the metabolic pathways leading to production of the putative toxic metabolites appear to be low-capacity, high-affinity pathways that are saturated at relatively low-exposure concentrations. This does not prove, but suggests, that the same may be true in humans. If the total formation of the putative toxic metabolites is predictive of the toxicity of benzene, then the animal studies suggest that calculations of the risk associated with low dose exposures based on the results of animal studies conducted at high doses would underestimate the toxicity of benzene. The current report concerns only dosimetry. Another problem in assessing the risk to humans from benzene exposure is the fact that the animal models do not respond to benzene in the same way as humans. The major concern for humans exposed to benzene, based on epidemiology studies, is the risk of developing acute myelogenous leukemia (Rinksy, 1987). The cancers developed by the rodents on the long-term bioassay studies were at other sites (liver, lung, Zymbal's gland, lymph tissue, ovaries, and mammary gland). There is as yet no good animal model for benzene-induced leukemia. However, it has been suggested that benzene may also increase the incidence of Hodgkin's disease, malignant lymphoma, multiple myeloma and lung cancer in humans, although a statistical basis for this is lacking (Askoy, 1985). It is not unreasonable to assume that whatever form of cancer is induced, the induction is most likely through the reactive metabolites produced from benzene. Therefore, the dosimetry of these metabolites is pertinent. Our studies indicate that benzene metabolite dosimetry data obtained in animals provides data relevant to the estimation of human risks.