Tiered Testing Under TSCA Section 4 Has Literally ‘Gone to the Birds’

In January 2021 and March 2022, the Environmental Protection Agency (EPA) issued test orders under Section 4 of the Toxic Substances Control Act (TSCA) that included sediment testing and even testing in vertebrate animals as avian ecotoxicity testing (both acute and chronic/reproductive). These test orders were issued by EPA for high priority chemicals prioritized for risk evaluation under TSCA Section 6. The test orders have ruffled feathers in the regulatory community and raised some questions, specifically: How does EPA decide what orders to issue under TSCA Section 4? What about statutory language to use tiered approaches and reduce vertebrate animal testing? 

In answer to the first question, there is no formal EPA policy regarding how it will issue a Section 4 test order. In the absence of this, we can look only at what the statute says, other EPA guidance, and (retrospectively) at what EPA has done. We do all of these below.

In answer to the second question, Congress provided clear direction in Section 4 that a tiered approach to testing is required unless more advanced testing is justified:

When requiring the development of new information under this subsection, the Administrator shall employ a tiered screening and testing process, under which the results of screening-level tests or assessments of available information inform the decision as to whether 1 or more additional tests are necessary, unless information available to the Administrator justifies more advanced testing of potential health or environmental effects or potential exposure without first conducting screening-level testing [15 U.S.C. § 2603(a)(4)].

Section 4 also makes clear that other information, including toxicity and computational toxicology information, must be considered before issuing an order for vertebrate animal testing.

We at ACC are not aware of any information provided by EPA indicating that TSCA Section 4 avian toxicity testing (or any other testing) is justified. Indeed, we are not aware of any tiered approaches used by the Agency in deciding what studies to order under TSCA Section 4 for high priority chemicals. As such, we posit that the first, necessary step is to consider what a tiered screening and testing approach might look like and how EPA might make decisions about whether to require new testing, particularly in vertebrate animals.  

Best Available Science

There are numerous examples in the scientific literature of tiered toxicity testing approaches such as that described by Plunkett et al (2010) who note:¹

Tiered toxicity testing is ordered testing where less complex tests are initially used to obtain data, with the results then evaluated to inform the study design of more complex tests. In contrast, sequential testing is where each and every test specified in a battery of tests must be performed.

In a true tiered toxicity testing framework, decision criteria, or “toxicity triggers” are used to determine whether to proceed with further testing, as well as to determine which specific type of test should be conducted.

The advantages enumerated are consistent with TSCA Section 4(h)’s mandate regarding reduction of testing on vertebrates:

• more efficient use of resources, animals, and time to identify chemicals that are of highest concern for public and environmental health;
• more flexibility to allow tailoring of testing for specific toxicities, populations, or other risk assessment and risk management needs;
• readily interpretable testing results because tests are pre-sorted into toxicity endpoints or effect pathways;
• timely evaluation and identification of chemicals that may pose certain unacceptable hazards/risks; and
• rapidly available results for chemicals that present lower toxicity potential.

Nearly a decade ago, a tiered testing approach by which such new data sources could be used for risk assessment was proposed,² and application of such an approach is being implemented by regulators in Canada and elsewhere.³ An Integrated Approach to Testing and Assessment (IATA) is a flexible framework for chemical hazard characterization that relies on an analysis of existing information coupled with the generation of new information, often using tiered or iterative testing strategies. The Organization of Economic Cooperation and Development (OECD) has sponsored a project highlighting successful IATA case studies since 2014.⁴

In addition, there are examples of existing EPA guidance and precedents that could be applied to the TSCA risk evaluation process. 

Existing EPA Guidance and Precedents

There are several guidance documents and policy precedents in the EPA Office of Chemical Safety and Pollution Prevention (OCSPP) where early screening may result in waiving of certain testing given particular circumstances.

1. EPA’s Office of Pesticide Programs (OPP), the sister office to the Office of Pollution Prevention and Toxics (OPPT, where the TSCA programs reside) within OCSPP, issued its Guiding Principles for Data Requirements in 2016 to
   
   help guide the identification of data needs, promote, and optimize full use of existing knowledge, provide consistency in the data request process across all scientific disciplines and all OPP divisions, and focus on the data needed to allow for a scientifically sound and credible characterization of a specific pesticide’s risk profile for the exposure scenarios of interest (p. 1).
   
   These principals support OPP’s strategy of using Integrated Approaches to Testing and Assessment, which promote a hypothesis based, systematic, integrative use of exposure and hazard information.
    
2. With respect to avian ecotoxicity testing, EPA OPP concluded, based on the retrospective analysis of 119 pesticides, that waivers may be granted for avian sub-acute dietary testing unless one of several specific conditions occurs, particularly regarding a unique mode of action or high potential for bioaccumulation.^5,6
    
3. EPA OCSPP’s background document regarding the Ecological Effects Test Guidelines for terrestrial wildlife⁷ recommends that
   
   for industrial chemicals, the maximum amount of toxicity test information should be obtained from the initial or lower tier test, Avian Acute Oral Toxicity Test (the OCSPP 850.2100 guideline)… For non-toxic or low toxicity chemicals (based on the results of the OCSPP 850.2100 test) it is likely that no further higher tier testing… would be supported or recommended (p. 3).

EPA Sediment Toxicity Tiered Screening Approach

There are other examples in EPA OCSPP of more quantitative tiered screening approaches for decision-making on requests for additional data.  EPA OPP has Toxicity Testing and Ecological Risk Assessment Guidance for Benthic Invertebrates that provides direction to its ecological risk assessors on when to require whole sediment toxicity tests.⁸

Whole sediment toxicity tests are conditionally required for a pesticide as specified in 40 CFR Part 158 Subpart G based on a series of toxicity testing ‘triggers’ that pertain to:

1. The likelihood of chemical exposure in aquatic ecosystems;
2. The extent the chemical partitions to sediment particles;
3. The persistence of the pesticide in the aquatic environment;
4. The toxicological relevance of predicted or monitored exposure concentrations; and
5. The likelihood of chemical exposure in estuarine/marine ecosystems.

In this respect, there are more quantitative screening assessments of exposure, fate, and toxicity with off-ramps for waiving of sediment testing where the information dictates it is appropriate.

These same principles and toxicity testing triggers could be applied for other aquatic organisms or terrestrial wildlife (e.g., avian sub-acute and reproductive toxicity testing). 

Higher Tier Screening

There are ample data available from which higher tier screens may be developed to inform decisions regarding the development of new toxicity data. One example is EPA’s ECOTOX Knowledgebase which currently boasts having 1,113,714 results for 12,540 chemicals from 53,257 references.^9,10 Moreover, there are readily available statistical distribution approaches to derive environmental threshold concentrations.¹¹ For example, a chemical toxicity distribution (CTD) is a probabilistic method where the toxicity of a group of chemicals is assessed in one species with a common endpoint.¹² The ECOTOX Knowledgebase contains hundreds of results for acute toxicity of Northern bobwhite quail that could serve as the basis for a CTD to inform the need for new acute dietary toxicity testing of avian species for high priority chemicals.

Next Steps

There are numerous opportunities for EPA to apply tiered screening approaches in its decision-making for orders for new studies. Many of those approaches have already been developed and validated by EPA OPP and could be translatable to chemicals other than pesticides.  As such, they are “on the shelf” and ready for use by EPA OPPT. 

However, EPA appears to be moving, inexplicably, in the other direction.  Recently, EPA announced that it is rescinding a long-standing policy to impose exposure release thresholds for new chemicals in favor of expending precious Agency resources to quantify risks that it knows are small and not likely to be unreasonable. 

It is ironic that, in the interest of protecting wildlife, EPA will sacrifice birds in avian toxicity testing studies, even when it is far from clear that this is necessary.  EPA’s issuance of test orders under TSCA Section 4 needs to be transparent and consistent with tiered approaches and protections for vertebrate animals specified by TSCA.  Right now, tiered testing under TSCA has literally ‘gone to the birds,’ but best available scientific approaches on tiered toxicity testing, existing Agency guidance and precedents, and use of higher tier analysis on existing data provide an opportunity for EPA to make effective decisions regarding the need for test orders that are protective and improve the efficiency of the risk evaluation process.

___________________________

[1] Plunkett, L.M., A.M. Kaplan, R.A. Becker. 2010. An enhanced tiered toxicity testing framework with triggers for assessing hazards and risks of commodity chemicals, Regulatory Toxicology and Pharmacology, Vol. 58(3): 382-394. https://doi.org/10.1016/j.yrtph.2010.08.003.

[2] Thomas et al. 2013. Incorporating New Technologies Into Toxicity Testing and Risk Assessment: Moving From 21st Century Vision to a Data-Driven Framework, Toxicological Sciences, Vol. 136(1):4–18. https://doi.org/10.1093/toxsci/kft178.

[3] Gannon, A.M., M. Moreau, R. Farmahin, R.S. Thomas, T.S. Barton-Maclaren, A. Nong, I. Curran, C.L. Yauk. 2029. Hexabromocyclododecane (HBCD): A case study applying tiered testing for human health risk assessment, Food and Chemical Toxicology, Vol. 131: 110581. https://doi.org/10.1016/j.fct.2019.110581.

[4] https://www.oecd.org/chemicalsafety/risk-assessment/iata-integrated-approaches-to-testing-and-assessment.htm

[5] USEPA. 2020. Final Guidance for Waiving Sub-Acute Avian Dietary Tests for Pesticide Registration and Supporting Retrospective Analysis. U.S. Environmental Protection Agency, Washington, DC. February 2020.  Available at: https://www.epa.gov/pesticide-registration/bridging-or-waiving-data-requirements.

[6] In a retrospective analysis of 119 pesticides with publicly available ecological risk assessments that were registered into commerce between 1998 and 2017, EPA concluded that “the sub-acute dietary test did not identify risk in greater than 99% (118 out of 119) of chemicals based on results that either the acute oral RQ was higher than the sub-acute dietary RQ, or both the acute oral and the subacute dietary tests did not generate an RQ value of concern.”  (Hilton, GM, E Odenkirchen, M Panger, G Waleko, A Lowit, AJ Clippinger. 2019. Evaluation of the avian acute oral and sub-acute dietary toxicity test for pesticide registration, Regulatory Toxicology and Pharmacology, Vol. 105: 30-35. https://doi.org/10.1016/j.yrtph.2019.03.013).

[7] Ecological Effects Test Guidelines OCSPP 850.2000: Background and Special Considerations-Tests with Terrestrial Wildlife, Available at: https://www.epa.gov/test-guidelines-pesticides-and-toxic-substances/series-850-ecological-effects-test-guidelines.

[8] https://www.epa.gov/pesticide-science-and-assessing-pesticide-risks/toxicity-testing-and-ecological-risk-assessment

[9] https://cfpub.epa.gov/ecotox/

[10] Olker, J.H., Elonen, C.M., Pilli, A., Anderson, A., Kinziger, B., Erickson, S., Skopinski, M., Pomplun, A., LaLone, C.A., Russom, C.L. and Hoff, D. (2022), The ECOTOXicology Knowledgebase: A Curated Database of Ecologically Relevant Toxicity Tests to Support Environmental Research and Risk Assessment. Environ Toxicol Chem, 41: 1520-1539. https://doi.org/10.1002/etc.5324

[11] Kienzler, A., S. Bopp, M. Halder, M. Embry, A. Worth. 2019. Application of new statistical distribution approaches for environmental mixture risk assessment: A case study, Science of The Total Environment, Vol 693: 133510. https://doi.org/10.1016/j.scitotenv.2019.07.316.

[12] Williams, E.S., Berninger, J.P. and Brooks, B.W. 2011. Application of chemical toxicity distributions to ecotoxicology data requirements under REACH. Environmental Toxicology and Chemistry, 30: 1943-1954. https://doi.org/10.1002/etc.583