A Better Approach
Early in the development of the Endocrine Disruptor Screening Program (EDSP), PETA recognized the need for a faster, more accurate, and valid screening approach that would greatly reduce the number of animals killed in tests. We proposed an alternative to the Environmental Protection Agency’s (EPA) two-tier strategy for the first time at the Society of Toxicology’s (SOT) annual meeting in 2009. Rather than testing on animals in the first tier, our alternative strategy considers physical and chemical data, existing toxicological data, and the use of validated in vitro (cell-based) tests and computational methods. As further elaborated in a presentation by PETA at the 2010 SOT annual meeting, the results of this alternative preliminary tier can be used in a weight-of-evidence approach to 1) categorize chemicals and identify those most likely to cause adverse effects and 2) design an intelligent chemical-specific strategy for further screening or testing. Such a strategy would greatly reduce the use of animal testing for identification and classification of endocrine-disrupting chemicals.
PETA’s strategy is organized into five levels, similar in structure to the Organisation for Economic Co-operation and Development’s (OECD) Conceptual Framework for the Testing and Assessment of Endocrine Disrupting Chemicals (“framework”). Although the OECD framework is not a testing strategy to be followed linearly, it nevertheless suggests a more logical approach to data gathering than the EPA’s two-tier approach and forms a basis for deciding whether it is necessary to obtain further data. The levels of the OECD framework are as follows:
- Level 1 involves the collection and assessment of all existing data, such as physical and chemical properties, structure-activity relationships, available toxicological information, and results of computational toxicology models, to evaluate a chemical’s potential as an endocrine disruptor before entering any other level of the framework.
- Level 2 consists of in vitro screening assays used for hazard detection and identification of possible mechanisms of action (MOAs). These assays are in most cases deliberately over-responsive since they are designed to provide alerts for endocrine disruption.
- Level 3 consists of shorter-term (three weeks or less) in vivo (animal) tests that yield information about specific endocrine MOAs and pathways. They are deliberately highly responsive (often using castrated/immature animals), and a chemical with a negative outcome is generally regarded as inactive against the specific mechanism evaluated by the particular assay. Although Level 3 screens do not evaluate all possible mechanisms of endocrine disruption, the results can be used to make decisions about performing higher-tier in vivo tests.
- Level 4 consists of in vivo tests that evaluate multiple MOAs and measure numerous endpoints. Data from these tests may be used for risk assessment and decision-making.
- Level 5 uses large numbers of animals in longer-term (more than three weeks), sometimes multigenerational tests that often include more dose levels and numerous measurements that provide further information about risk assessment and dose response.
In contrast to the OECD’s five-tiered approach above, the EPA currently uses a two-tier system. Tier 1 consists of five in vitro screening assays that evaluate specific MOAs (see “Screens and Tests”) and six in vivo assays of the Level 3 and 4 types. A chemical showing a positive response in Tier 1 would be subject to more testing in Tier 2 using Level 5 assays. Although the OECD framework still involves animal testing, it allows for decision-making at more levels and allows for the possibility of avoiding the longer, more animal-intensive tests and reducing the numbers of animals used overall. Because of these features, it is preferable to the EPA’s system, which requires Level 5 testing in Tier 2 after only one round of assessment in Tier 1. PETA’s goal is to see all animal testing replaced with non-animal, more human-relevant methods of assessing chemicals.
The EPA provides an option to chemical companies to submit existing information, known as Other Scientifically Relevant Information (OSRI), as a possible means for waiving some of the Tier 1 assays. OSRI usually consists of data that have already been collected from animal tests conducted to support registration of pesticides, for example. These tests look at many of the same endpoints evaluated in the Tier 1 and Tier 2 EDSP tests and may form a basis for deciding on a chemical’s potential to disrupt the endocrine system. In addition to potentially securing Tier 1 test waivers, existing information can be used to better design an integrated, chemical-specific strategy for risk analysis that further streamlines testing and reduces animal use. PETA demonstrated such a strategy at the 2010 International Congress on in Vitro Toxicology, using two List 1 chemicals as examples, and recently published a paper on this subject in the prestigious journal, Toxicological Sciences.
An integrated test strategy, such as the OECD framework and the one demonstrated by PETA, could vastly reduce the exorbitant costs associated with the EPA’s EDSP––both financially and in terms of animal suffering and death. In vitro assays using human-derived tissue are generally more sensitive, reliable, and relevant to humans than animal-based methods and produce results more quickly and at a much lower cost. In vitro assays are excellent models for investigating the MOA of potential hormone-disrupting chemicals and can be performed either individually or in large numbers through automated “high-throughput” screening (HTS).
For example, the EPA’s “Toxicity Forecaster” or ToxCastTM research program employs an innovative non-animal approach to rapidly screen and prioritize chemicals for toxicological testing through the use of HTS assays and computer models. A multiyear effort launched in 2007, ToxCastTM is a cost-effective method for prioritizing thousands of chemicals for further toxicity testing, based on their potential for adverse health effects. Currently, the EPA is using ToxCastTM to evaluate 880 chemicals (including 56 of the EDSP List 1 chemicals) in approximately 50 endocrine-related HTS assays and has published the early results and more recent results of these analyses. The advantage of the structure of the ToxCastTM program’s database is that connections can rapidly be made between in vitro assay results and existing mammalian and ecotoxicity data, greatly facilitating identification and interpretation of MOA information.
After PETA scientists had urged the EPA for many years to adopt an integrated testing approach like the one described here as well as to make better use of in vitro tests to screen chemicals, the agency unveiled a new workplan in September 2011 called EDSP21, which seeks to integrate computational and non-animal methods into a chemical prioritization and assessment scheme for endocrine disruptors. The near-term goal of the plan is to use existing data, computer models, and individual or suites of the ToxCastTM in vitro HTS assays to determine the relative order in which chemicals should be screened. The intermediate-term goal is to replace current validated Tier 1 in vitro screening assays with validated ToxCastTM in vitro HTS assays, use the results to inform and target current in vivo estrogen- or androgen-specific screening assays, and, when possible, reduce the use of animals for screening purposes. The long-term goal (>5 years) is to consider full replacement of the Tier 1 in vivo screening assays with validated ToxCastTM in vitro HTS assays and eliminate the use of animals for screening purposes. The EPA is close to achieving the near-term goal in evaluating potential disruptors of the estrogen hormone system using computational toxicology tools and HTS assays as described in a paper released in early January 2013.
Conducting the current Tier 1 test battery on one chemical uses 584 animals, costs at least $385,000, and produces information of extremely questionable usefulness in a regulatory context. (See cost table on the “Screens and Tests” page.) As the EPA moves away from a default application of the full battery of Tier 1 assays, particularly with data-rich chemicals such as pesticides, and adopts the more efficient and potentially more useful approach described in EDSP21, further testing, if any, should become more targeted to only those chemicals with the greatest potential for endocrine disruption, eventually saving tens of thousands, if not millions, of animal lives.