This project focuses on developing new statistical methods, and applying new and existing statistical techniques, to analyze data from laboratory animal studies. The endpoint of interest in a typical carcinogenicity experiment is the tumor incidence rate, which reflects the age-specific rate at which new tumors occur. Unfortunately, most tumors are not observable in live animals. Thus, simple estimates of the tumor incidence rate and tests for comparing incidence rates across groups are not available. Kaplan-Meier curves are typically displayed for overall survival, but analogous plots for tumor incidence generally require fairly elaborate model fitting. We developed a method for estimating tumor incidence as a function of more easily estimable components. A generalized additive model is assumed for one of the components, tumor prevalence, which leads to estimates that are more flexible than those derived under the usual parametric models. A multiplicative assumption about tumor lethality allows for straightforward incorporation of concomitant information, such as tumor size. Our approach requires only terminal sacrifice data, although additional sacrifice data are useful and can be easily accommodated. The method also yields a simple summary measure of tumor lethality, which can be useful in interpreting study results. We also developed a Bayesian method that focuses on tumor incidence and accommodates occult tumors without restricting tumor lethality, relying on cause of death data, or requiring interim sacrifices. We represent the underlying state of nature by a multistate stochastic process and assume general probit models for the time-specific transition rates. These models allow the incorporation of covariates, historical control data, and subjective prior information. The inherent flexibility of this approach facilitates the interpretation of results, particularly when the sample size is small or the data are sparse. We use a Gibbs sampler to estimate the relevant model parameters. - tumor, onset, incidence, lethality, sacrifice, survival, Bayesian, bioassay, carcinogenicity, prevalence
