ze, evaluate, and interpret data contained in the NTP rodent carcinogenicity database and other large databases. The goal of this project is to summarize, evaluate, and interpret data contained in the NTP rodent carcinogenicity database and other large databases. One recent database evaluation investigated the role of transgenic mouse models in carcinogen identification. Data from 99 experiments involving the three most extensively used transgenic mouse models - Trp53+/-, Tg/AC and RasH2 - were evaluated. The ability of these models (used singly or in combination) to predict human carcinogenicity (as determined by the NTP Report on Carcinogens and/or the IARC Monographs) was investigated. Although the transgenic models had good overall concordance (ranging from 74-81%) with known/probable human carcinogenicity, the errors tended to be in the direction of missing known or suspected human carcinogens (i.e., """"""""false negatives""""""""). This performance was improved somewhat (approximately 85% correct determinations with no """"""""false negatives"""""""") when a """"""""mixed strategy"""""""" of a transgenic model in conjunction with the rat bioassay was used. Overall, the transgenic models performed well, but important issues of validation and standardization need further attention to permit their regulatory acceptance and use in human risk assessment. Dr. Shyamal Peddada and I recently completed our participation in the second phase of an Organization for Economic Co-operation and Development (OECD) validation program for the uterotrophic bioassay. This bioassay is intended to identify the in vivo activity of compounds that are suspected agonists or antagonists of estrogen, and to assist the prioritization of positive compounds for further testing. Two model systems, the immature female rat and the adult ovariectomized rat, were tested and compared. Data from nineteen participating laboratories using seven different chemicals, and both a blinded and unblinded protocol were evaluated (each leading to papers that will appear in Environmental Health Perspectives). The overall conclusions were that both model systems appear robust, reproducible and transferable across laboratories and able to detect weak estrogen agonists.
Showing the most recent 10 out of 38 publications
