Conclusions and Recommendations
Several decades ago, there was no standard bioassay for detecting chemical carcinogens. The current MTD bioassay in rodents closed that gap and became the standard assay in the United States. It is neither perfect nor unalterable, and by itself it is insufficient to produce data from which accurate human health risk assessments can be made. Nonetheless, the MTD bioassay does provide some useful information for hazard identification and risk assessment.
The committee began its deliberations by examining the question of agent specificity—i.e., whether carcinogenicity seen in animal bioassays at the MTD is a specific response to a chemical, rather than a response to general toxicity. Few studies have been conducted to address the issue directly by testing proposed mechanistic hypotheses in long-term animal studies. Recent studies have reported an inconsistent relationship between the sites and types of chronic toxic effects measured in conventional toxicicity studies and the development or lack of development of cancer. Although toxicity and carcinogenicity often are not detected simultaneously, nonspecific toxicity-induced carcinogenesis can occur, and increased rates of cell proliferation associated with carcinogenesis (when they occur) should be considered in risk assessments of genotoxic and nongenotoxic chemicals.
The committee concludes that the relationship between measures of general toxicity (e.g., the MTD) and carcinogenic potency can be expressed as follows: Potency of animal carcinogens is such that the increase in tumor incidence that is needed to show statistical significance
in standard bioassays occurs generally at doses near the MTD, but use of the MTD itself does not predict whether a material will elicit a carcinogenic response in a standard animal bioassay. The basis of the relationship is not clear. General toxicity and cancer induction have a number of steps in common. A material must be absorbed, possibly metabolized, and transported to a site of action. The sharing of those elements might be partially responsible for the observed relationship. In addition, the relationship is consistent with chemical-induced toxicity's being a mediator of carcinogenicity in some instances. However, the committee recognizes that other reasons for the relationship are also possible.
Because of the relationship between TD50s and MTDs of chemicals, the committee concludes that a preliminary upper bound on the potential carcinogenic potency of an untested chemical can be estimated from knowledge of its MTD. Such estimates can prove useful in setting priorities for carcinogenicity testing and in making preliminary upper-bound risk estimates of cancer risk whenever carcinogenicity bioassay results are not available. If such an upper-bound estimate predicts a low human risk, a chemical could be assigned a low priority for carcinogenicity testing.
The committee noted that, although testing at the MTD is currently included in most carcinogenicity bioassays, specific criteria for selecting the EMTD and evaluating the selection vary with the group conducting the study.
The committee recommends that, to facilitate interpretation, reports of bioassay results should include a clearly stated rationale for dose selection and a summary of the toxicity information important for evaluating the dose selection.
The committee concludes that the MTD bioassay as currently conducted in rodents is most useful as a qualitative screen to determine whether a chemical has the potential to induce cancer. It also provides information on the carcinogenic potential of a substance at high doses and some information about the dose-response curve. It does not provide (nor was it intended to provide) all the information useful for low dose human risk assessment. In most cases, additional information is likely to be needed to determine the extent to which the induction of cancer in rodents adequately predicts the human response and how the results of the
relatively high dose assay can best be used to make inferences about the expected effects at low doses. Some of the required information might be obtainable from study of tissues from animals subjected to long-term bioassay or from ancillary studies incorporated in the rodent bioassay. Future bioassays should be designed to reveal the overall toxic responses induced by the test chemical and not just the carcinogenic response. But in general, other information needs will require other studies.
The types of additional information that are needed will depend on the chemical under study, but several general subjects merit consideration. Studies should be conducted that can help to determine whether the mechanisms or metabolic processes involved in the production of cancer in rodents are relevant to humans. In addition, toxicokinetic studies are important to determine the effect of dose on the absorption and metabolic fate of the chemical. Physiologic responses induced by the chemical and the effect of dose on that response must be considered. Furthermore, chemically induced cell proliferation of target cells might play an important role in the induction of some tumors, and cell proliferation should be measured when that is appropriate. It is difficult to conceive how similar information could be gathered in humans.
The committee could not reach consensus on how additional information on mechanisms of carcinogenicity and other responses should be used in conjunction with the MTD bioassay, however. A majority of the committee believes that the assay identifies substances that do or do not increase the incidence of cancer under the conditions of the assay and provides an operational definition of animal noncarcinogens The assay also identifies target organs, demonstrates tumor types associated with exposure, provides a consistent basis for interspecies comparisons, and can serve as a guide in designing followup studies. The majority also believes that there are as yet no validated mechanisms of carcinogenicity that support lowering the MTD and that failing to use the MTD for carcinogen screening will reduce the sensitivity of the bioassay and diminish the opportunity to compare results among chemicals and species. As a result, the majority recommends implementation of Option 4B, described earlier in the report. This option is summarized as follows.
The MTD should continue to be one of the doses used in carcinogenicity bioassays. Other doses, ranging downward
from MTD/2 possibly to MTD/10 or less, should also be used. The capacity of the test animal to absorb and metabolize the test chemical should be taken into account in selection of doses below the MTD.
If a standard set of rodent bioassays that each include the MTD are negative, generally no additional tests related to carcinogenicity are required.
If a bioassay conducted with the MTD is positive, additional studies should be performed to reduce uncertainties in predicting human responses to the test material and to assist in performing quantitative risk assessments.
These additional studies should address mechanisms of cancer induction, toxicokinetics and metabolism of the material, and and physiologic responses induced by the material; they could also include validation of MTD bioassay results with epidemiologic studies.
Some committee members disagreed with those two recommendations and believe that a greater modification of the process for selecting doses to be used in carcinogenicity bioassays is required. The modification would emphasize specificity over sensitivity and would require that bioassay doses be selected after careful evaluation of subchronic studies conducted before the chronic bioassay, as discussed earlier in this report as Option 3. The minority recommends the following:
Bioassay doses should be selected by a panel of experts on the basis of careful evaluation of studies conducted before the bioassay is initiated. Information gathered before the bioassay is conducted would include information about mechanisms of toxicity in test animals and an elucidation of the dose-response relationship for such toxicity. The HDT should be chosen as the highest dose that can be expected to yield results relevant to humans, not the highest dose that can be administered to animals without causing early mortality from causes other than cancer.
The committee is aware that regulation of a chemical can take place
after any stage in data collection. Public health considerations can lead to cautious behavior, sometimes expressed as formal regulation early in the testing history of the chemical. Such regulation can be expected to use conservative (cautious) assumptions, which should be eased or otherwise modified as the accumulation of data makes estimation of the dose-response relationship more precise or improves knowledge of the pharmacokinetics of the material and its mode of action.