Lung cancer rates in carbon-black workers are discordant with predictions from rat bioassay data.

The recent demonstration of lung tumors in rats chronically exposed to airborne carbon-black particles has prompted reevaluation of the carcinogenicity of carbon black. However, accumulating evidence suggests that rat responses to inhaled particles are unique and are mediated by the sequelae to lung overload conditions. To test the predictive value of the rat inhalation bioassay for human lung cancer risk, we examined epidemiologic studies of workers exposed to carbon black during its manufacture or use. Industries in which significant airborne carbon-black exposure has occurred were not associated with increased lung cancer risk. We also predicted the number of occupational lung cancer cases that one would expect in exposed workers, based on the cancer potency of carbon black derived from rat studies. Our quantitative comparison of the tumorigenicity of carbon black predicted from rat studies to the lung cancer rate in carbon-black workers showed a marked discrepancy between the lung cancers predicted and those actually observed. We found that far more lung cancers are predicted from the rat bioassay than can be demonstrated in workers. We determined that it is highly unlikely that this discrepancy is due to chance. We conclude that extrapolating the incidence of lung tumors in rats inhaling inert, insoluble particles, such as carbon black, to humans must be seriously questioned. Using rat inhalation bioassay data for carbon black to estimate lung tumor risk in humans must be seriously questioned unless the mechanisms of the rat's unique response are shown to be relevant to humans.