The Center for Alternatives to Animal Testing is an academic center affiliated with the Division of Toxicological Sciences in the Department of Environmental Health Sciences of the Johns Hopkins University Bloomberg School of Public Health.

 

Johns Hopkins School of Public Health

Abstract for TestSmart -- Endocrine Disruptors

Current Status of Efforts to Validate the EDSTAC Tier 1 Screening Tests

Errol Zieger
Chapel Hill, NC

The proposed EPA-EDSTAC Tier 1 screening tests (T1S) includes 5 in vitro tests (estrogen receptor binding and reporter gene assays; androgen receptor binding and reporter gene assays; steroidogenesis assay with minced testes), and 5 in vivo tests (rodent 3-day uterotrophic assay; rodent 20-day pubertal female with thyroid, rodent 5-7-day Hershberger assay, frog metamorphosis assay; fish gonadal recrudescence assay). In addition, the report recommends one possible alternative in vitro assay (placental aromatase assay) and 3 in vivo assays (modified rodent 3-day uterotrophic assay with i.p. dosing; rodent 14-day intact male assay with thyroid; and, rodent 20-day thyroid/pubertal male assay). These assays, protocol outlines, and the rationales for their selection, are described to a limited extent in the EDSTAC report. No specific data or references to validation studies with these chemicals were included in the report, although there are general statements made regarding reproducibility and anticipated false positives and negatives for some of the tests.

There is a general U.S. Government (and European Union) policy that new or revised tests, or traditional tests that are proposed for new uses be shown to be validated for their intended use before they are recommended or accepted to support regulatory decisions. In this sense of the use, 'validation' has been defined as the determination of the reliability and relevance of a test for a specific purpose. The reliability of a test is a measure of its intra-and inter-laboratory reproducibility; the relevance of a test is judged by its ability to measure what it is designed to measure. Validation studies typically involve more than one laboratory (too examine reproducibility) and the testing of coded chemicals. Chemicals are selected for testing according to the endpoint the test is designed to measure. In the case of the T1S tests, the reference endpoint would be the effect in the proposed T2S tests, or known endocrine effects in mammals and/or humans. The chemicals selected should have various levels of potency, including chemicals that have no effect.

Although a test may be highly reproducible, it may not be highly predictive for the event it is designed to measure. Therefore, it may not be suitable for its intended purpose. Although the reliability and relevance of a test can be fairly accurately measured, there are no agreed-upon values for reproducibility and predictivity that define a test as acceptable for regulatory use. The acceptable values are highly subjective, and will be based on the type of test being evaluated, the uses to which the test would be put and the decisions it will support, and the corresponding values for the test, if any, it is designed to replace or supplement. Essentially, a new test is accepted for regulatory use when it provides the regulator with an acceptable level of comfort that the test can be used to support regulatory decisions of hazard or risk.

The only formal, publicly available, evaluations of T1S tests are those currently being performed by the Organisation for Economic Co-operation and Development (OECD) on the rat uterotrophic and Hershberger assays. There is also an EPA effort to examine chemicals using an in vitro androgen receptor binding assay.

The OECD was directed by its member countries (including the U.S.) to evaluate the in vivo rat uterotrophic and Hershberger assays, toward the eventual goal of developing Test Guidelines for the assays. As part of this effort, the test protocols were to be standardized, and the assays' reproducibility and relevance determined. A Validation Management Group (VMG) was formed and, in February 1999, recommended test protocols for the two procedures.

Uterotrophic assay. The EDSTAC document separated the assay into a recommended oral-exposure route and an alternative i.p.-route procedure. One question that arose in the VMG was whether to use sexually immature females or ovariectomized females. The initial study design used rats, one strong known positive estrogen (ethinyl estradiol; EE) and a comparison of i.p. vs. oral routes, and immature vs. ovariectomized animals. Nineteen laboratories worldwide participated on a voluntary basis. The results showed good reproducibility among all labs, and that, for this EE, the i.p. route led to a slightly higher response, as did the use of ovariectomized animals. A second study was then designed, using a refined protocol, and 7 coded potent and anticipated weak substances; 20 laboratories participated. This testing has been completed and the results are being evaluated.

Hershberger assay. Similar questions to those addressed in the uterotrophic assay; what route of administration should be used, and whether immature or castrated animals be used. A study with 1 strong positive control chemical was designed to examine these variables and the reproducibility of the procedure, and determine a dose response curve; 17 laboratories worldwide volunteered to participate. This testing is underway at this time.

In vitro assays. The EPA is having approximately 500 chemicals tested for in vitro estrogen binding activity using a cell free test system. In a separate study, the EPA is also examining 500 chemicals for in vitro cell-free androgen binding activity. These studies will start early this year. Although not designed as validation studies, the results from these tests will be available for use in any formal in vitro validation studies. At the present time, there does not appear to be any validation studies using reporter gene assays.

In September 2000, the EPA issued a contract proposal for endocrine disrupter screening and testing standardization and validation support for tests recommended by EDSTAC. All recommended tests have been included. The contract is expected to be awarded around the beginning of 2001.

Summary. With the exception of the OECD's efforts with rodent assays, there have been no systematic validation efforts for endocrine disrupter assays. A joint effort between the U.S. EPA and the U.S. National Toxicology Program Interagency Center for the Evaluation of Alternative Toxicological Methods (NICEATM) is currently underway to identify in vitro endocrine disrupter tests that are sufficiently well developed so that they can be put into validation studies. At the same time, the EPA, through its new contract, will be develop (where necessary), standardize, and validate methods identified by EDSTAC for Tier 1 and Tier 2 screening.