Are tumor incidence rates from chronic bioassays telling us what we need to know about carcinogens?

Chronic bioassays for over 500 chemicals have been conducted under the auspices of the National Cancer Institute and/or the National Toxicology Program (NTP) to screen chemicals for carcinogenicity, providing a wealth of information about bioassays. Typically, chemicals are administered for two years to both sexes in each of one strain of rats and mice generally at the maximum tolerated dose (MTD), MTD/2, MTD/4 (in recent years), as well as unexposed control animals. In an attempt to ascertain the sensitivity of this bioassay to detect animal carcinogens tested at the MTD for the current experimental design, the false negative rate (failure to detect increased tumor rates) was investigated. This was accomplished by examining the tumor incidences from over 150 NTP bioassays and estimating the probability that a statistically significant (P0.01) dose-response trend would be obtained at one or more tissue sites in either sex of rats or mice if 200, rather than 50, animals were used per dose group. This provides an estimate of the proportion of chemicals that were not declared high-dose animal carcinogens due to the limited sample size of 50 animals per species-sex-dose group. In this series of chemicals tested, 97/156 (62%) were identified by the NTP to show some or clear evidence of carcinogenicity. With an increase of the number of animals per dose group from 50 to 200, it is estimated that 92% of these chemicals would show statistically significant (P0.01) dose-response trends at one or more tissue sites in either sex of rats or mice. Many of these chemicals are not genotoxic. Some chemicals had no structural alerts for carcinogenicity, but were tested because of potentially high human exposure. This analysis suggests that almost all of the chemicals selected would produce a statistically significant increase in tumor incidence at the MTD with larger sample sizes. Hence, this MTD bioassay screen is not distinguishing between true carcinogens and non-carcinogens. Rather, the screen is simply failing to detect the weaker carcinogens at the MTD. More than 30% of chemicals tested failed to detect statistically significant dose-response trends for tumors because of inadequate sample sizes of 50 animals per dose. Presumably, little or no action would have been taken to regulate exposures to these chemicals as potential carcinogens due to lack of a positive test result. This analysis does not suggest that most chemicals are carcinogenic at human exposure levels nor does it suggest that more than 50 animals should be tested per dose group. With an MTD that may produce a difference (up to 10%) in weight gain between treated and control animals, there quite possibly is cytotoxicity at the MTD. Increased carcinogenicity would be expected from increased opportunities for mutagenic activity during regenerative cell replication to compensate for cytotoxicity. Since it appears that almost all chemicals tested adequately at the MTD will demonstrate carcinogenicity, it is tempting to surmise that this is due in large part to one or more nearly universal modes of action, such as, regenerative cell replication at the MTD rather than due to some unique carcinogenic property of a chemical. That is, the current bioassay possibly is just primarily a screen for the more potent cytotoxins at the MTD, rather than a screen specifically for carcinogenicity. This issue should be examined and suggests that cytotoxicity and cell proliferation should be considered in setting the MTD, particularly for non-genotoxic (non-DNA reactive) chemicals. From a public health view, it is prudent to assume that most chemicals could demonstrate increased tumor incidence rates at the MTD in rodents. The current standard NTP bioassay provides sufficient data to estimate a benchmark dose associated with a specified low tumor incidence to be used as a point-of-departure for cancer risk assessments. The question that should be investigated by a bioassay is not whether a chemical is a carcinogen at the MTD, but what is the relationship between dose and cytotoxicity and/or other modes of action that could produce an excess of tumors?