Testing for endocrine disruption: Are we there yet?

Cal Baier-Anderson, Ph.D., is a Health Scientist.

After long delays, the EPA’s Office of Pesticide Programs recently issued endocrine disruptor screening test orders for dozens of high-priority pesticide ingredients.  Endocrine disruptors are chemicals capable of interfering with the action of hormones that regulate biological processes such as development, growth, reproduction and metabolism.  The test orders require pesticide manufacturers to evaluate their chemicals using a specific battery of tests.

Identifying which chemicals are endocrine disruptors can help protect people and the environment from harmful exposures.  So, with test orders now in the hands of pesticide manufacturers, will we finally get the data we need?


In the Food Quality Protection Act of 1996, Congress mandated the development of the Endocrine Disruptor Screening Program (EDSP), with the goal of developing test methods to identify those chemicals with the potential to interact with the estrogen, androgen, or thyroid hormone systems.  Now, more than 12 years later, the first tier of the comprehensive test battery has been finalized.  And EPA has just issued test orders for dozens of chemicals (mostly pesticide active ingredients) it has identified as priorities for screening using the Tier 1 test battery.

Throughout this process, many chemical manufacturers have complained that the new testing is superfluous: conventional test methods required by EPA, they say, are already sufficient to capture any “meaningful” endocrine disruptor effects (for example, see 5th paragraph here and 5th paragraph here).

Most independent scientists disagree.  They argue that perturbations of endocrine systems are themselves adverse effects, and are also associated with a broad spectrum of complex diseases and disorders, including infertility, cancer, obesity, diabetes mellitus, and cardiovascular disease (for examples, see here, here and here).  Therefore, more specialized tests are required to detect endocrine-disrupting substances.

Meanwhile, in another corner of EPA, new rapid test methods are being developed for use in chemical prioritization that may be able to identify chemicals that require the more in-depth testing.  The emergence of such high-throughput test methods could greatly accelerate our ability to screen tens of thousands of chemicals for hazardous properties.  But it remains to be seen if they are adequate to identify endocrine-disrupting substances.  More on that in a minute.

The EDSP tests are needed to protect human health

Data derived using the EDSP Tier 1 test battery are designed to be evaluated as a whole: the collective results can help to identify impacts on a range of important hormone-related processes. Using both in vitro (“test tube”) and in vivo (whole animal) studies, the tests evaluate a variety of mechanisms by which a chemical can interfere with hormone function.  The principal focus is on the functions of estrogen, androgen and to a lesser extent, thyroid hormones, although many more hormones exist.

The Tier 1 tests are used to determine if additional testing is needed, and if so, the chemical will be subjected to more detailed Tier 2 testing.  According to EPA, the Tier 2 tests take longer and cost more, but are also more robust, because they are designed to encompass multiple critical life stages and processes, as well as a broad range of doses.

The conventional test methods that many in industry argue are sufficient are primarily reproductive and developmental toxicity tests conducted on laboratory animals.  In these studies, animals are exposed to a chemical, either prior to mating or during fetal development, and reproductive success and offspring development are evaluated.

Such conventional test methods can identify chemicals that cause overt reproductive and developmental effects such as infertility and birth defects.  But more sensitive test methods are needed to capture the more insidious and multi-faceted effects of endocrine disruption.  The recognition of the need for more sensitive and specific tests is what led to the development of the EDSP in the first place.

Data from the first round of EDSP Tier 1 test orders are not due until 2011.  Chemicals that “fail” the Tier 1 tests will then proceed to Tier 2.   So it will be quite a while before anyone knows which chemicals will be officially designated under the EDSP as endocrine disruptors.

And we can probably expect that some companies will challenge the need for the required testing, claiming that the conventional test methods can provide sufficient information about endocrine disrupting properties.  My view:  EPA should not accept this rationale because the conventional tests submitted to comply with pesticide regulations did not measure important endpoints included in the EDSP tests.

Alternative test methods hold promise for pre-screening

Because testing for endocrine disruption is time-consuming and costly and requires the use of laboratory animals, ideally only those chemicals that are likely to be endocrine disruptors should be so tested.  But how can we know in advance which chemicals are candidates without testing?

Scientists in EPA’s Office of Research and Development (ORD) have been plugging away at the design of a large collection of alternative test methods, referred to as ToxCast, for use in chemical prioritization.  Among them are tests that could be used to identify potential endocrine disruptors.

ToxCast is EPA’s effort to modernize chemical safety testing by developing rapid and efficient methods to evaluate the tens of thousands of chemicals in commerce.  The tests involve minimal use of laboratory animals, relying instead on tests using cells and cell components to determine whether a chemical can perturb or disrupt specific kinds of biological activity.

A PowerPoint presentation on Endocrine Profiling developed by EPA scientists for a stakeholder group describes how 309 chemicals screened using the ToxCast assays were analyzed for endocrine-disrupting properties.  It focused on tests involving the estrogen (5 tests), androgen (4 tests) and thyroid (4 tests) signaling pathways, as well as other nuclear receptors and chemical metabolizing enzymes (70 tests) that have potential relevance to endocrine signaling.

The chemicals were then ranked by their ability to perturb these pathways.  Interestingly, the notorious BPA was ranked second highest out of 309 chemicals.

Some concerns remain

This all sounds promising.  But one of my concerns is that the limited scope of current ToxCast assays may miss key aspects of endocrine disruption.  While it is relatively easy to develop high-throughput methods to test chemicals for their capacity to interact with hormone receptors, it is much harder to capture other interactions that reflect the more complex biological interactions that occur within a whole animal.  Some of the EDSP tests are designed to detect the latter.

If ToxCast tests fail to identify important endocrine-disrupting processes, then they could lead to the mis-classification of endocrine disrupting chemicals as safe.  Such “false negatives” would be unacceptable, particularly in a chemical screening program.

One solution:  Compare the results of ToxCast testing to the results of the EDSP Tier 1 test battery.  If this comparison demonstrates that the number of false negatives is sufficiently low, that would bolster the case for using ToxCast as a predictor of chemicals needing to be subjected to the more detailed EDSP testing.

Another limitation is that, unlike the EDSP tests, ToxCast does not incorporate tests that are relevant for non-human organisms.  Given that many endocrine-disrupting substances can be found in the environment, where they have been shown to impact fish and frogs, EDSP’s more integrated approach to chemical testing is a distinct advantage.

Finally, the ToxCast tests have not yet been subjected to the same level of validation as have the EDSP tests to ensure their sensitivity and accuracy.  The comparison to results of EDSP testing could help in this regard as well.


The EDSP Tier 1 test battery, which has been under development for over a decade, is designed to characterize the effects of chemicals on critical hormonal pathways – information that cannot be obtained through conventional toxicity testing.

The ToxCast testing program promises efficient pre-screening of chemicals, but whether it is adequate for use in identifying candidate chemicals for EDSP testing remains to be seen.  EPA must take steps to properly validate the ToxCast assays and compare the results to the full EDSP test battery to make this determination.

ToxCast data are now available for a large number of chemicals.  With the issuance of the test orders for the EDSP, I hope that EPA will soon be in position to compare data from these two programs and make better-informed decisions about where to go from here.

This entry was posted in Emerging testing methods, Health science and tagged , , , , , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.


  1. Posted January 5, 2010 at 12:11 pm | Permalink

    Cal Baier-Anderson provides an excellent overview of where EPA is today in detecting chemicals that are endocrine disruptors, which is the very first step toward regulation. Today, EPA is in the process of issuing test orders to pesticide manufacturers to use 11 screens (Tier 1) designed to detect only a small component of the possible actions a chemical might display if it interfered with the endocrine system. The biochemical changes detected during screening will be used for identification but will not provide the information needed to regulate a chemical. Similarly, the results of EPA’s literature-derived TOXCAST program designed to prioritize chemicals for testing will not be sufficient for regulation. And under the current plan, chemicals that get caught in the EDSTP Tier 1 safety nets will then go on for further testing using Tier II protocols that were designed over 12 years ago and still not standardized or validated.

    While all this has been going on, a program at the NIEHS Division of Extramural Research and Training (DERT) has been funding research on the developmental and reproductive health effects of endocrine disrupting chemicals. This research, based on the principles of endocrinology, using a human disease-based approach, dosing animals at relevant ambient levels, accounting for realistic inverted U dose response curves, taking a serious look at critical timing of exposure, determining internal dose in animals, monitoring the internal dose in humans, and measuring function from the molecular to the cell and whole animal is on the verge of fulfilling the need for full blown hazard identifications for chemicals. This is the kind information needed for regulatory purposes.

    BPA may be second to the top on the TOXCAST “top ten” list of bad chemicals, but when the rubber hits the road, what is needed to get the chemical behind bars is a full blown hazard ID about the chemical. Congress needs to take advantage of what has been going on at the NIEHS, especially in light of the national health crises of endocrine related disorders. Society cannot wait. The timing could not be better to take advantage of the stage of readiness at the EPA and that of NIEHS to address these serious health problems. Cooperation and collaboration among the programs can turn this health crisis around. Separating the research on the health effects of chemicals from any form of conflict of interest would be a giant step forward. This will truly be a 21st century approach, and for once prevention will get the attention it deserves.

    To lend your voice to a growing list of supporters of a recently introduced federal bill to allow the research at NIEHS to inform the regulation of endocrine disrupting chemicals, go to http://www.endocrinedisruption.com/endocrine.edlaw.php

    Theo Colborn

  2. Cal Baier-Anderson
    Posted January 6, 2010 at 11:39 am | Permalink

    Thank you, Theo, for weighing in on this important issue.

    I agree wholeheartedly that the current test battery included in the Endocrine Disruptor Screening Program (EDSP) is inadequate in scope. For example: Its main focus is on estrogen- and androgen-regulated functions and barely touches on thyroidal effects. It does not address hormones that can influence metabolism, or those that interact with neurological and immunological systems.

    Still, positive results in the limited number of EDSP assays would provide important information, even if negative results cannot and should not be used to clear a chemical of having endocrine-disrupting effects.

    I also agree that the most appropriate use for ToxCast is for screening, which is less than what is needed for regulatory purposes. It is worth emphasizing, however, that the value of ToxCast is that by using high-throughput assays, tens of thousands of chemicals can be screened and evaluated for biological activity associated with numerous hazard endpoints (including, but not limited to, developmental, reproductive, endocrine, cardio-respiratory, carcinogenic, immunological, neurological and metabolic effects).

    The technology allows that the array of specific assays can be enhanced as our understanding of the biology changes. There is no reason that ToxCast could not accommodate an examination of dozens of additional endocrine-relevant endpoints.

    One note of caution about the assays being developed under DERT is that if they are to be suitable for use for regulatory purposes, there will be a need for standardization and validation, which will delay their availability for use. This is obviously problematic, because although it certainly shouldn’t have to take the ten years EDSP took to validate its assays, it will still take some amount of time and money.

    In the meantime, given that important data are being generated under DERT, it would be wonderful to have these data entered into a curated, searchable database to facilitate ready access and use. And to have regulators incorporate them into their assessments, alongside any and all other credible and relevant data.