EXPERIMENTAL DESIGN AND DATA ANALYSIS METHODOLOGY

点击翻译按钮获取中文摘要

This project is concerned with statistical methodology issues in the design, analysis, and interpretation of laboratory animal experiments, especially the long term rodent carcinogenicity studies carried out by the National Toxicology Program (NTP). Making use of the NTP rodent carcinogenicity database, we (1) developed mathematical models to predict control tumor incidence in rats and mice based on age, body weight, and other factors; (2) evaluated the frequency of and factors responsible for chemically-related tumor decreases in NTP studies; and (3) carried out an evaluation of the frequency of nasal cavity carcinogens in the NTP database. The knowledge gained through this research will increase our understanding of the value and limitations of long term rodent studies. Another investigation in this area evaluated hepatocarcinogenicity among mice exposed to chlordane. We found that both the proportion of treated mice with liver carcinomas, and the average number of liver tumors per treated mouse, decreased significantly after chlordane exposure was stopped. In addition, the proportion of treated mice experiencing toxic changes closely paralleled the temporal development of hepatocellular neoplasia, with the severity of toxicity increasing initially and then decreasing after the point of maximum tumor burden. The possibility that exposure-induced tumors might regress after treatment is stopped could have a large impact on the design, analysis, and interpretation of NTP studies.