Cover Image

PAPERBACK
$49.00



View/Hide Left Panel

regressions were performed using the one-stage, multistage, and Weibull models to determine the TD50.

As indicated in Table 3, the estimate of the slope is near unity for all three models. The error variance s2 is greatest for the Weibull model, and least for the one-stage model. An approximate 95% prediction interval for an individual log10(TD50) at a given value of the MTD is given by a+blog10(MTD)+2s. Since b ˜ 1, this corresponds to the interval 10±2s × MTD for the TD50. For the one-stage model, for example, the 95% prediction interval encompasses a range of 7 × 7 = 49-fold about the MDT, comparable to the 32-fold range derived by Bernstein et al. (1985).

4.2 Predictions Based on Mutagenicity and Acute Toxicity

The preceding results indicate that the MTD, which is essentially a measure of chronic toxicity, is a fair predictor of the TD50. Determination of the MTD is normally based on the results of subchronic toxicity tests lasting about three-months. This observation raises the question

TABLE 3 Regression of Carcinogenic Potency on the Maximum Tolerated Dose

Regression Parameter

Model

One-Stage

Multistage

Weibull

Intercept ± SE

- 0.07 ± 0.05

- 0.10 ± 0.04

0.15 ± 0.06

Slope ± SE

1.04 ± 0.02

1.04 ± 0.02

0.99 ± 0.03

Correlation

0.952

0.964

0.903

Root Mean Square (s)

0.423

0.362

0.592

Factor 102s for 95% Prediction Intervalb

7.0

5.3

15.3

aBased on simple linear regression of log TD50 on log MDT

bUpper limit is 102s x MDT; lower limit is 10-2s x MDT.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 132
APPENDIX F 132 original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution. regressions were performed using the one-stage, multistage, and Weibull models to determine the TD50. As indicated in Table 3, the estimate of the slope is near unity for all three models. The error variance σ2 is greatest for the Weibull model, and least for the one-stage model. An approximate 95% prediction interval for an individual log10(TD50) at a given value of the MTD is given by a+blog10(MTD)+2σ. Since b ≈ 1, this corresponds to the interval 10±2σ × MTD for the TD50. For the one- stage model, for example, the 95% prediction interval encompasses a range of 7 × 7 = 49-fold about the MDT, comparable to the 32-fold range derived by Bernstein et al. (1985). 4.2 Predictions Based on Mutagenicity and Acute Toxicity The preceding results indicate that the MTD, which is essentially a measure of chronic toxicity, is a fair predictor of the TD50. Determination of the MTD is normally based on the results of subchronic toxicity tests lasting about three-months. This observation raises the question TABLE 3 Regression of Carcinogenic Potency on the Maximum Tolerated Dose Regression Parameter Model One-Stage Multistage Weibull Intercept ± SE - 0.07 ± 0.05 - 0.10 ± 0.04 0.15 ± 0.06 Slope ± SE 1.04 ± 0.02 1.04 ± 0.02 0.99 ± 0.03 Correlation 0.952 0.964 0.903 Root Mean Square (σ) 0.423 0.362 0.592 102σ Factor for 95% Prediction 7.0 5.3 15.3 Intervalb aBased on simple linear regression of log TD50 on log MDT bUpper limit is 102σ x MDT; lower limit is 10-2σ x MDT.

OCR for page 132
APPENDIX F 133 original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution. as to whether there exist other variables that are highly correlated with cancer potency that may be considered as possible predictors of the TD50, and which may be determined with less effort than through a subchronic study. The two-stage initiation-promotion-progression model of carcinogenesis described by Moolgavkar & Luebeck (1990) is based on the concept that mutation and cell proliferation are the two most important determinants of neoplastic change. Meselson & Russell (1977) reported a near-perfect linear relationship between the mutagenic potency of 14 chemical carcinogens, as measured by the dose inducing a mutation rate of 100 revertant colonies in the Ames Salmonella/microsome assay, and carcinogenic potency, as measured by the TD50. McCann et al (1988) found a correlation of r = 0.41 between the mutagenic and carcinogenic potencies of 80 chemicals drawn from both the general literature and the U.S. National Toxicology Program. More recently, Piegorsch & Hoel (1988) examined the correlation between mutagenic and carcinogenic potency using 97 chemicals tested in the U.S. National Toxicology Program. In this analysis, mutagenic potency was measured by the slope of the initial linear portion of the dose-response curve (cf. Krewski et al., 1992, 1993). Although a significant positive correlation of r=0.48 between mutagenic and carcinogenic potency was apparent, the overall scatter was considered to be to sufficiently great to preclude the use of Ames test data as a quantitative predictor of carcinogenic potency. Parodi et al. (1990) reviewed studies of the correlation between mutagenic and carcinogenic potency conducted between 1976 and 1988. In addition to using the Salmonella assay to measure mutagenic potency, mutation in L5178Y mouse lymphoma cells, in vivo DNA adducts in rodent liver, and in vitro DNA repair in rodent liver were also considered. This investigation suggests that the correlation between carcinogenic potency and mutagenic potency based on each of these short-term tests is moderate, with correlation coefficients in the neighborhood of r=0.4. The relationship between acute toxicity, which may provide some indication of the ability of a chemical to induce cellular proliferation, and carcinogenic potency has been the subject of several investigations. Parodi et al. (1982a) found a significant correlation (r = 0.49) between carcinogenic potency and acute toxicity. Parodi et al. (1990) suggested that this association may be due in large part to the fact that acute and chronic toxicity are correlated, with chronic toxicity (as measured by the

OCR for page 132
APPENDIX F 134 original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution. MTD) in turn being highly correlated with carcinogenic potency. Zeise et al. (1986) reported a high correlation between acute toxicity as measured by the LD50, and carcinogenic potency, as measured by the parameter β in the one-hit model. Although correlation coefficients as high as r=-0.93 were found with both variables expressed on a logarithmic scale, some exceptions were noted. For example, the carcinogenic potency of 7,12-dimethyl benz(a)anthracene (DMBA) is about 5,000-fold greater than would be predicted on the basis of its LD50. Nonetheless, Zeise et al. (1984) suggested that this relationship between acute toxicity and carcinogenicity could be used to substantially narrow the uncertainty in the TD50 values of untested carcinogens, which, as shown previously, vary over some seven orders of magnitude. More recently, Metzger et al. (1987) reported somewhat lower correlations (r = 0.6) between the LD50 and TD50 for 264 carcinogens selected from the CPDB, with an average TD50/LD50 ratio of 0.06. Travis et al. (1990a) argued that since both mutation and cell proliferation are important determinants of carcinogenesis, attempts to correlate carcinogenic potency with mutagenicity or acute toxicity alone are inadequate. Thus, Travis et al. (1990ab, 1991) investigated the correlation between the TD50 and composite indices based on mutation, toxicity, reproductive anomalies, and tumorigenicity data derived from the Registry of Toxic Effects of Chemical Substances (RTECS) (Sweet, 1987). In general, this analysis confirmed the previously reported correlation of r = 0.4 between mutagenic potency in the Ames assay and the minimum TD50 observed in rodent carcinogenicity studies (McCann et al., 1988; Piegorsch & Hoel, 1988); the correlation of r = 0.7 between LD50 and TD50 was also somewhat greater than that reported by Metzger et al. (1987). In addition to confirming previous findings, Travis et al. (1990ab, 1991) investigated the correlation between composite predictors of carcinogenic potency based on results from two or more of 870 different assays for mutagenicity or toxicity, including 20 assays for mutation reported in RTECS. For each assay, a relative potency index was established in terms of weighted average of the potency relative to 20 reference compounds; a geometric mean of all available assays was then used to obtain an overall predictor of carcinogenicity. For some chemicals, the relative potency values based on different assays varied by as much as five orders of magnitude.