Related Experiment Videos
Analysis of animal carcinogenicity experiments with multiple tumor types
1Department of Epidemiology and Public Health, University of Miami School of Medicine, Florida 33101, USA.
Biometrics
|March 1, 1995
Summary
New statistical methods allow simultaneous comparison of tumor incidence rates in animal carcinogenicity studies. These approaches are validated using data from a large mouse study on 2-acetylaminofluorene (2-AAF) and simulations.
Area of Science:
- Toxicology
- Biostatistics
- Cancer Research
Background:
- Tumor development time is often unobservable in chemical carcinogenicity studies.
- Accurate statistical methods are needed to analyze tumor incidence rates.
Purpose of the Study:
- To propose novel statistical test approaches for simultaneous comparison of marginal tumor incidence rates across multiple tumor types.
- To develop a model for tumor development and assess parameter identifiability.
Main Methods:
- A model for tumor development was established.
- Tests for individual marginal tumor incidence rates were combined.
- Simultaneous tests for all marginal tumor incidences were developed.
Main Results:
- The proposed methods were applied to data from the ED01 study, involving approximately 24,000 mice exposed to 2-acetylaminofluorene (2-AAF).
- Simulation results demonstrated the utility of simultaneous test methods for smaller experiments (100 mice per dose group).
Conclusions:
- The developed statistical methods provide a robust framework for analyzing tumor incidence in chemical carcinogenicity studies.
- These methods are applicable to both large-scale and smaller experimental designs.