On Selecting a Minimal Set of In Vitro Assays to Reliably Determine Estrogen Agonist Activity
Citation:
Judson, R., K. Houck, Eric Watt, AND R. Thomas. On Selecting a Minimal Set of In Vitro Assays to Reliably Determine Estrogen Agonist Activity. REGULATORY TOXICOLOGY AND PHARMACOLOGY. Elsevier Science Ltd, New York, NY, 91:39-49, (2017).
Impact/Purpose:
• Agency Problem: Congress directed the EPA to establish the Endocrine Disruptor Screening Program (EDSP) for characterizing chemicals’ potential to disrupt hormones via estrogen, androgen and thyroid signaling pathways. The EPA uses the EDSP Tier 1 battery of 11 in vitro and in vivo assays to screen chemicals for endocrine disrupting activity. With an estimated 10,000 chemicals to screen, it would take more than 100 years to collect relevant chemical safety data. EPA needs alternative approaches to increase efficiency and throughput to ensure chemical safety and protect public health. • Approach: Using alternative approaches to prioritize chemicals for Tier 1 tests, and ultimately replace Tier 1 with high-throughput alternatives, is the objective of the EDSP21 program. Under EDSP21, ORD and OCSPP collaborated to develop a predictive model of ER activity, using a suite of 16 in vitro assays. This ER model was the first example of using in vitro high-throughput assays to replace a guideline in vivo test for regulatory decisions. One limitation of the full ER model is the large number of assays used. This work examined if a subset of the 16 in vitro assays provided equally accurate predictions of ER activity when compared to the full model. • Results: This work developed an approach to select a minimum battery of ER assays for classifying estrogenic agonist activity of a chemical. Starting with the full model based on 16 assays, a simple linear model identified multiple subset models to accurately predict the estrogenic potential of a chemical. There are subsets that include as few as 4 of the original 16 assays. These subset models correctly predicted estrogenicity potential of reference chemicals with a balanced accuracy of >90%, with subsets using 7 in vitro assays achieving 96% balanced accuracy. • Impact to the Agency: This work describes a method for detecting ER activity using minimal sets of in vitro assays. If this method is suitable for the EDSP21 regulatory framework, the EPA could make chemical safety decisions more efficiently and more cost-effectively.
Description:
The US EPA is charged with screening chemicals for their ability to be endocrine disruptors through interaction with the estrogen, androgen and thyroid axes. The agency is starting to explore the use of high-throughput in vitro assays to use in the Endocrine Disruptor Screening Program (EDSP), potentially as replacements for lower-throughput in vitro and in vivo tests. The first replacement is an integrated computational and experimental model for estrogen receptor (ER) pathway activity, to be used as an alternative to the Tier 1 in vitro ER binding and transactivation assays and the in vivo uterotrophic bioassay. The experimental component of the ER agonist model uses a set of 16 in vitro assays that incorporate a variety of technologies and cell lines and probe multiple points in the ER pathway. Here, we demonstrate that it is possible to achieve equivalent levels of performance against both in vitro and in vivo reference chemical sets as the full ER agonist model using various subsets of assays. The simplest “subset” model that achieves maximum accuracy against multiple metrics uses only 4 assays. There are multiple accurate subsets, allowing flexibility in the construction of a multiplexed assay battery. We also discuss the issue of challenging chemicals, i.e. those that tend to give false positive results in certain assays, and could hence be more problematic when only a few assays are being used.
