Tumor incidence data

In spite of its current popularity in the pharmaceutical industry, the use of two control groups is opposed by some statisticians on the grounds that a significant difference between the two groups may indicate that the study was compromised by excessive, uncontrolled variation. Haseman et al. (1986), however, analyzed tumor incidence data from 18 color additives tested in rats and mice and found that the frequency of significant pairwise differences between the two concurrent control groups did not exceed that which would be expected by chance alone. 

In 1985, the United States Federal Register recommended that the analysis of tumor incidence data be carried out with a Cochran-Armitage (Armitage, 1955 Cochran, 1954) trend test. The test statistic of the Cochran-Armitage test is defined as this term ... 

In the absence of sufficient data to develop a robust, biologically based model for quantitative risk assessment, a single curve-fitting model for each type of data set is preferred. It is noted that many different curve-fitting models have been developed, and those that fit the observed data reasonably well may lead to several-fold differences in estimated risk at the lower end of the observed range. Therefore, the US-EPA uses a standard curve-fitting procedure for tumor incidence data. 

A nonlinear approach should be selected when there are sufficient data to ascertain the mode of action and to conclude that it is not linear at low doses, and the agent does not demonstrate mutagenic or other activity consistent with linearity at low doses. The POD is in this case generally a BMDL when incidence data are modeled. A sufficient basis to support this nonlinear procedure is likely to include data on responses that are key events in the carcinogenic process. This means that the POD may be based on these precursor response data, for example hormone levels or mitogenic effects, rather than tumor incidence data. A nonlinear approach can be used to develop an RfD or an RfC. This approach expands such reference values, previously reserved for threshold effects, to include carcinogenic effects determined to have a nonlinear mode of action. A nonlinear approach should generally not be used in cases where the mode of action has not been ascertained. 

The animal tests are designed to provide a certain level of sensitivity in tumor detection, and the tumor incidence data are used to develop cancer potency factors... 

TABLE 29.5. Carcinogenic Potency Estimates Using Dichotomous Tumor Incidence Data for Mice Exposed by Inhalation to 1,3-Butadiene... 

Another problem concerns the identiflability of carcinogenic mechanism using tumor incidence data. Using a simpler model than that shown in Figure 3, Fortier (1987) examined the ability of tumor incidence data to accurately differentiate between a low-dose linear chemically induced increase in the rate of mutations from normal cells to initiated cells and a low-dose nonlinear chemically induced increase in the birth rate of initiated cells. It was shown that the probability of incorrectly classifying one mechanism as the other was quite high, exceeding 50% in some cases. 

---