You are here:
SCREENING CHEMICALS FOR ESTROGEN RECEPTOR BIOACTIVITY USING A COMPUTATIONAL MODEL
Citation:
Browne, P., R. Judson, W. Casey, N. Kleinstreuer, AND R. Thomas. SCREENING CHEMICALS FOR ESTROGEN RECEPTOR BIOACTIVITY USING A COMPUTATIONAL MODEL. ENVIRONMENTAL SCIENCE & TECHNOLOGY. American Chemical Society, Washington, DC, 49(14):8804-8814, (2015).
Impact/Purpose:
EPA is accepting ToxCast ER model data for 1812 chemicals as alternatives for the EDSP Tier 1 ER binding, ER transactivation, and uterotrophic assays. The use of high-throughput and computational methods dramatically increases EPA’s ability to rapidly screen chemicals for endocrine bioactivity, and provide an alternative to animal-based EDSP Tier 1 ER binding and uterotrophic assays. The application of these innovative tools for screening chemicals for endocrine bioactivity represents the first step in a paradigm shift for chemical safety testing, and the first systematic application of ToxCast data in an EPA regulatory program.
Description:
The U.S. Environmental Protection Agency (EPA) is considering the use high-throughput and computational methods for regulatory applications in the Endocrine Disruptor Screening Program (EDSP). To use these new tools for regulatory decision making, computational methods must be appropriately validated. Traditional validations of toxicity tests are time intensive, evaluate a relatively small number of chemicals, and are not well-suited to high-throughput methods. Here we describe a multi-step, performance-based validation establishing scientific confidence in new computational methods and demonstrating these tools are sufficiently robust to be used in a regulatory context. Results from 18 estrogen receptor (ER) ToxCast high-throughput screening assays, measuring different points along the signaling pathway with different assay technologies, were integrated into a computational model. The resulting ToxCast ER model scores range from 0 (no activity) to 1 (bioactivity of the native ligand, 17β-estradiol) and can discriminate ER bioactivity from assay-specific interference and cytotoxicity. ToxCast ER model performance was evaluated for 40 in vitro and 43 in vivo reference chemicals. ToxCast ER model results were also compared to EDSP Tier 1 screening assays in current regulatory practice for a diverse set of more than 100 chemicals. ToxCast ER model accuracy was 95% when compared to the large set of in vitro and in vivo reference chemicals. In addition, the ToxCast ER model predicted the outcomes of EDSP Tier 1 guideline and other uterotrophic studies with > 90% accuracy. The performance of the high-throughput assays and ToxCast ER model predictions of agonist bioactivity demonstrates these methods are sensitive, specific, quantitative, and efficient; and thus protective of human health and the environment. EPA is accepting ToxCast ER model data for 1812 chemicals as alternatives for the EDSP Tier 1 ER binding, ER transactivation, and uterotrophic assays. The use of high-throughput and computational methods dramatically increases EPA’s ability to rapidly screen chemicals for endocrine bioactivity, and provide an alternative to animal-based EDSP Tier 1 ER binding and uterotrophic assays. The application of these innovative tools for screening chemicals for endocrine bioactivity represents the first step in a paradigm shift for chemical safety testing, and the first systematic application of ToxCast data in an EPA regulatory program.
