If the numbers of animals in a test were not substantially increased, agents that were formerly found to be carcinogenic only at the MTD would not be identified as positive in this procedure. Thus, this procedure would focus attention on agents that had high carcinogenic potency relative to their subacute toxicity (Region C carcinogens in Figure 1).
An advantage of this procedure is a flexibility that reflects differences between potentially hazardous substances themselves or the expected exposures of humans to them. Most such chemicals will never be tested in chronic animal bioassays; current evidence suggests that, if they were, about half would be identified as carcinogens and almost all, by definition, would be in Region B. It would clearly pose a dilemma for regulators to be faced with decisions on a multitude of chemicals that would be identified as carcinogens under today's regulatory standards. However, substances with low toxicity but high carcinogenic potency (Region C carcinogens) might well present an unusually high cancer risk to human populations but not produce toxicity in the bioassay that serves to warn people of a hazard; that is apparently what happened with vinyl chloride workers (Fox and Collier, 1977; IARC, 1979).
A disadvantage of this option is that it would decrease the sensitivity of the assay, thus reducing its usefulness as a means of hazard identification. This disadvantage can be compensated for by increasing the numbers of animals in test groups; however, the expense of increasing the number of animals to a point necessary to retain the same power would probably be prohibitive. Furthermore, the choice of a lower dose, such as MTD/3, as the highest dose is arbitrary. Although implementation of this option would identify primarily Region C carcinogens, there is little evidence that Region C carcinogens contribute a predominant portion of chemically induced cancer risk to human populations. Future uses of a chemical cannot always be anticipated. If the HDT were based on current uses, a future use that entailed high human exposures could necessitate a new bioassay.
OPTION 3
Base the HDT on preliminary studies that determined the dose dependence of physiologic effects induced by the chemical and the dose dependence of its metabolism.
In this option, a comprehensive series of tests would be conducted before the bioassay were initiated. The tests would be designed to provide information about mechanisms of toxicity, as well as the dose-response curve for such toxicity. Microscopic examination of tissues would continue to be a part of studies, as would clinical chemistry (e.g., serum enzyme measurements and urinalysis.) However, the studies would be expanded to include measurement of physiologic and biochemical effects (e.g., alterations in hormone status) and quantitative measurements of cell proliferation. In addition, extensive pharmacokinetic studies (quantitative measurements of uptake, distribution, metabolism, and elimination) would be carried out.
An expert panel would be convened to evaluate preliminary data before doses for the bioassay were selected. The panel would select the HDT and lower doses on the basis of evaluation of the preliminary studies. The objective would be to design a study that could be expected to yield results that would be useful for human risk assessment and not simply to administer as much chemical as possible without causing early mortality from causes other than cancer. This approach constitutes a change in emphasis of the bioassay. Studies that use the MTD as currently defined are designed to maximize the sensitivity of the bioassay (i.e., to prevent false-negative results). The objective of the new approach would be to increase the selectivity of the bioassay (i.e., to decrease the number of false-positive results).
In some cases, adoption of this option would not change the selection of the MTD as the HDT. For example, if human populations were reasonably expected to encounter high concentrations of the test substance, the MTD would continue to be used. In many cases, however, the HDT would be lower than the MTD as currently defined, and the spacing of doses could be much wider than commonly adopted by programs such as the National Toxicology Program (NTP).
An advantage of this modification is that the mechanisms underlying any observed carcinogenic response would be more likely to be qualitatively and quantitatively similar to those operating at lower doses than mechanisms underlying responses observed at the MTD.
A disadvantage of this modification might be that doses that caused a physiologic change in one organ might not cause physiologic changes in other organs. Without knowledge of the target sites for carcinogenicity, it would be unclear whether a physiologic change that is being avoided