The whys and wherefores of statistics for the animal carcinogenicity experiments are inherent in the experimental design of the bioassay. The lifetime and high dose components of the bioassay cause the use of survival adjusted statistical methods and are the whys of statistics. The wherefores of statistics are the methods of survival adjustment and the assumptions needed for the statistical analyses. Some of the necessary statistical assumptions are not biologically tenable. In addition, univariate p-values computed by statistical methods, such as Peto's have to be adjusted for the multiplicity of organs/tissues examined for each animal.
FeronVJGriceHCGriesemerRPetoR Report 1, Basic requirements for long-term assays for carcinogenicity. In: Long-Term and Short-term Screening Assays for Carcinogens: A Critical Appraisal (IARC Monographs Supplement 2). Lyon: International Agency for Research on Cancer (IARC); 1980: 21–83.
2.
EdgingtonES. Randomization Tests. Second edition. New York: Marcel Dekker; 1987.
HoelDGWalburgHE. Statistical analysis of survival experiments. J Nat Cancer Inst.1972;49: 361–372.
5.
LagakosSWRyanLM. On the representativeness assumption in prevalence tests of carcinogenicity. Applied Statistics.1985;34: 54–62.
6.
PetoR.Guidelines on the analysis of tumor rates and death rates in experimental animals. British Journal of Cancer.1974;29: 101–105.
7.
GartJJKrewskiDLeePNTaroneREWahrendorfJ.Statistical Methods in Cancer Research, Vol. III. The Design and Analysis of Long-Term Animal Experiments (IARC Scientific Publication No. 79). Lyon: IARC; 1986: 13–15.
8.
GartJJKrewskiDLeePNTaroneREWahrendorfJ.Statistical Methods in Cancer Research, Vol. III. The Design and Analysis of Long-Term Animal Experiments (IARC Scientific Publication No. 79). Lyon: IARC; 1986: 101–105.
9.
PetoRPikeMCDayNEGrayRGLeePNParishSPetoJRichardsSWahrendorfJ.Guidelines for simple, sensitive significance tests for carcinogenic effects in long-term animal experiments. In: Long-Term and Short-Term Screening Assays for Carcinogens: A Critical Appraisal (IARC Monographs Supplement 2). Lyon: IARC; 1980: 311–426.
10.
KodellRLFarmerJHGaylorDWCameronAM. Influence of cause-of-death assignment on time-to-tumor analyses in animal carcinogenesis studies. J Nat Cancer Inst.1982;69: 596–664.
11.
McKnightBCrowleyJ.Tests for differences in tumor incidence based on animal carcinogenesis experiments. J Amer Statist Assoc.1984;79: 639–648.
12.
WestfallPHYoungSS. P-value adjustments for multiple tests in multivariate binomial models. J Amer Statist Assoc.1989;84: 780–786.
13.
SoperKWestfallPH. Monte Carlo estimation of significance levels for carcinogenicity tests using univariate and multivariate models. J Statist Comput Simul.1990;37: 189–209.
14.
HeyseJFRomD.Adjusting for multiplicity of statistical tests in the analysis of carcinogenicity studies. Biom J.1988;30: 883–896.
15.
FarrarDBCrumpKS. Exact statistical tests for any carcinogenic effect in animal bioassays. Fund Appl Toxicol.1988;11: 652–663.
16.
MillerRGJr.Simultaneous Statistical Inference, Second edition.New York: Springer-Verlag; 1981.
17.
TaroneRE. A modified Bonferroni method for discrete data. Biometrics.1990;46: 515–522.
18.
HollandBSCopenhaverMD. An improved sequentially rejective Bonferroni test procedure. Biometrics.1987;43: 417–424.
19.
SAS Institute, Inc.PROC. MULTTEST. In: SAS Technical Report P-229, Software changes and enhancements for SAS/STAT, Version 6.07. Cary, In: SAS Institute; 1991.
