NTP Database Summarization And Evaluation

Environmental Health Sciences

Investigators

Linked publications & trials

Abstract

The goal of this project is to summarize, evaluate, and interpret data contained in NTP carcinogenicity databases. We contributed to NTP Technical Reports on benzophenone, transplacental exposure to AZT, bromodichloromethane, benzophenone, and the Toxic Equivalency Factor studies of PCB 153, the binary mixture of PCB 126 and PCB 153, the binary mixture of PCB 126 and PCB 118. These reports were reviewed and approved by the NTP Technical Reports Subcommittee in December, 2004. We are also contributing to NTP Technical Reports on 4-methylimidazole, divinylbenzene-HP, diisopropylcarbodiimide, dibromoacetic acid, and methyl isobutyl ketone, along with NTP Genetically Modified Mouse Reports on diisopropylcarbodiimide and the water disinfection compounds of sodium bromate, dichloroacetic acid, bromodichloromethane. These reports will be reviewed by the NTP Technical Reports Subcommittee in September, 2005. To better represent genetic diversity, the NTP is considering using multiple rodent strains in carcinogenicity and toxicity testing. In June 2005, NTP held a workshop on rodent stocks and strains, with consideration of multiple strain designs as a major issue. In preparation for this workshop, we compared the statistical power of multi-strain and single-strain designs to detect a carcinogenic effect over a wide range of plausible options and we presented the findings at the workshop. In brief, we found that an analysis that pools the results of multiple strains has a similar power to a single strain design having the same overall number of animals unless one or more strains are particularly sensitive to the chemical. In this case, the multi-strain design is more powerful. Not surprisingly, separate analyses of each strain in a multi-strain design provide less power to detect a carcinogenic effect than does a single strain design. NTP is considering these power calculations, along with the logistics of conducting multi-strain studies, in their deliberations about possible use of multi-strain designs. NTP is interested in reducing, replacing and refining the use of rodents in laboratory studies. We provided advice to the NTP on experimental design and statistical methods for evaluating alternative methods for toxicity and carcinogenicity testing. These include: ? advising NTP Interagency Center for the Evaluation of Alternative Toxicological Methods (NICEATM) on statistical methods for validating the use of in vitro cell cultures to predict toxicity in laboratory animals, ? designing a study to estimate false positive and false negative rates of a dog model for testing whether chemicals prolong the QT wave in the heart beat which may ultimately lead to lethal arrhythmias, ? continuing our evaluation of the usefulness of growth, reproduction, feeding, and movement of C. elegans for assessing toxicity, ? developing statistical methods for analyzing Salmonella mutagenicity assays, and ? reconsidering existing statistical methods for analyzing micronucleated erythrocyte assays in light of the underdispersion observed in the data. The effects of certain drugs and herbal remedies on the heart and cardiovascular system have received much publicity recently. We reviewed our database to identify and characterize the chemicals that caused thrombosis in mice and rats. We provided a reconciliation of apparently conflicting results from two sodium fluoride studies. We contributed to the fine-tuning of current methods for toxicity and carcinogenicity testing through: ? a reassessment of organ/tissue pooling in the Toxicology Data Management System (TDMS), and ? a comparison of micronucleated erythrocyte counts as measured by flow cytometry vs. on microscope slides.