Statistical methods for the analysis of tumor multiplicity data.

The fit of tumor multiplicity data from 93 mouse skin, lung, and liver carcinogenicity experiments to Poisson, negative binomial, and normal distributions was studied. The data were fitted well by the negative binomial distribution. This distribution has two parameters, the mean tumor multiplicity and an exponent determined by the interanimal homogeneity of tumor response. The value of the latter parameter was related to animal strain and the target tissue studied in the carcinogenicity experiments. The null distribution of the two-sample likelihood ratio test based on the negative binomial with common exponent model for tumor multiplicity data was shown by simulation studies to be approximately chi 2 with 1 d.f. Simulation also indicated that the likelihood ratio test has sufficiently better performance when the negative binomial model is valid to make its use more attractive than the more commonly used Wilcoxon test or Student t test. Charts for estimating the number of animals per group that are required to detect specified differences in tumor multiplicities are provided for several commonly used assays.

[1]  P. Emmelot,et al.  Multi-hit kinetics of tumor formation, with special reference to experimental liver and human lung carcinogenesis and some gneral conclusions. , 1977, Cancer research.

[2]  R. Fisher.  250: Note on the Efficient Fitting of the Negative Binomial. , 1953 .

[3]  J. Miller,et al.  Vinyl carbamate as a promutagen and a more carcinogenic analog of ethyl carbamate. , 1978, Cancer research.

[4]  Laura Rountree Smith.  January , 1890, The Hospital.

[5]  P. Fialkow.  Use of genetic markers to study cellular origin and development of tumors in human females. , 1972, Advances in cancer research.

[6]  J. Reeds,et al.  Compound Multinomial Likelihood Functions are Unimodal: Proof of a Conjecture of I. J. Good , 1977 .

[7]  J. Miller,et al.  The N-hydroxy metabolites of N-methyl-4-aminoazobenzene and related dyes as proximate carcinogens in the rat and mouse. , 1979, Cancer research.

[8]  C. I. Bliss,et al.  FITTING THE NEGATIVE BINOMIAL DISTRIBUTION TO BIOLOGICAL DATA AND NOTE ON THE EFFICIENT FITTING OF THE NEGATIVE BINOMIAL , 1953 .

[9]  W. Baird,et al.  Tumor-promoting activity of phorbol and four diesters of phorbol in mouse skin. , 1971, Cancer research.

[10]  Z. Govindarajulu.  TABLES OF THE CUMULATIVE NON-CENTRAL CHI-SQUARE DISTRIBUTION, , 1962 .

[11]  R. Fisher.  The Advanced Theory of Statistics , 1943, Nature.

[12]  F. J. Anscombe,et al.  Sampling theory of the negative binomial and logarithmic series distributions. , 1950, Biometrika.

[13]  M. Fox.  Naturally Occurring Carcinogens, Mutagens and Modulators of Carcinogenesis , 1980, British Journal of Cancer.

[14]  E C Miller,et al.  Hepatocarcinogenicity of estragole (1-allyl-4-methoxybenzene) and 1'-hydroxyestragole in the mouse and mutagenicity of 1'-acetoxyestragole in bacteria. , 1976, Journal of the National Cancer Institute.

[15]  D. B. Owen,et al.  The Power of Student's t-test , 1965 .

[16]  B. Hoadley.  Asymptotic Properties of Maximum Likelihood Estimators for the Independent Not Identically Distributed Case , 1971 .

[17]  F. J. Anscombe,et al.  THE TRANSFORMATION OF POISSON, BINOMIAL AND NEGATIVE-BINOMIAL DATA , 1948 .

[18]  Anscombe Fj.  The statistical analysis of insect counts based on the negative binomial distribution. , 1949 .

[19]  D. Falconer,et al.  A Genetic Study of Induced Lung-Tumours in Mice , 1962, British Journal of Cancer.

[20]  W. E. Lever,et al.  THE LIMITING DISTRIBUTION OF THE LIKELIHOOD RATIO STATISTIC UNDER A CLASS OF LOCAL ALTERNATIVES. , 1967 .

[21]  G. Bateman.  The power of the X2 index of dispersion test when Neyman's contagious distribution is the alternate hypothesis. , 1950, Biometrika.

[22]  F. J. Anscombe,et al.  The statistical analysis of insect counts based on the negative binomial distribution. , 1949, Biometrics.

[23]  Paul G. Hoel,et al.  On Indices of Dispersion , 1943 .