Carcinogenesis bioassays: study duration and biological relevance

Criticisms of the scientific value of rodent carcinogenicity bioassays have focused on the arguments that the studies are too long and that most organ-specific carcinogenic effects observed in experimental animals have little or no relevance to humans. For example, Davies et al. (Davies, T.S., Lynch, B.S., Monro, A.M., Munro, I.C., Nestmann, E.R., 2000. Rodent carcinogenicity tests need be no longer than 18 months: an analysis based on 210 chemicals in the IARC Monographs. Food and Chemical Toxicology 38, 219-235) concluded that the duration of rodent bioassays should be no more than 18 months, based on their analysis of 210 International Agency for Research on cancer (IARC) rodent carcinogens in which they report that most chemicals "tumorigenic effects;'' at or before 12 months. However, many of these "tumorigenic effects" reflect the occurrence of a single neoplasm, with most tumors occurring much later in the study. Reliance on a single tumor at an early time point as providing definitive evidence of rodent carcinogenicity is a dangerous practice that could produce both false positive and false negative outcomes. An extensive evaluation of the NTP database reveals that many rodent carcinogens produce later-appearing tumors that would not be detected as statistically significant in a 12-18 month study. Such a shortened duration study would be roughly equivalent to evaluating human cancer in subjects 30-50 years of age, which would, result in markedly reduced study sensitivity. In fact, many investigators recommend extending the duration of rodent studies to 30 months or to a true lifetime to increase study sensitivity. We also do not agree with the second conclusion of Davies et al. (2000) that the mode of action of rodent carcinogenesis is sufficiently well understood to justify discounting the majority of organ-specific carcinogenic effects found in these studies. The consequences of performing rodent carcinogenicity studies with inadequate sensitivity, and then discounting most of the carcinogenic effects that are observed will be that potential human carcinogens will not be detected, thus forcing near total reliance on human studies for this purpose. This is not prudent public health policy. Published by Elsevier Science Ltd. All rights reserved.