The U.S. Environmental Protection Agency (EPA) announced on January 4, 2023, that it issued the next Toxic Substances Control Act (TSCA) test order requiring testing on per- and polyfluoroalkyl substances (PFAS) under its National PFAS Testing Strategy. The test order requires companies to conduct and submit testing on trifluoro(trifluoromethyl)oxirane (HFPO), a perfluoroalkyl substance used in making plastics. EPA states that this is “the second test order under the strategy and the latest action taken under EPA’s PFAS Strategic Roadmap to confront contamination from forever chemicals nationwide.” According to the press release, EPA will use the information received under the order to improve its understanding of human health effects of HFPO, as well as “the effects of dozens of PFAS that are structurally similar to HFPO and in the same Testing Strategy category of PFAS, improving the agency’s overall data on PFAS.”
According to EPA, HFPO is used in making plastics and in organic chemical manufacturing. More than 1,000,000 pounds of HFPO are manufactured each year, according to TSCA Chemical Data Reporting (CDR) rule reports.
EPA states that “[a]fter thoroughly examining existing hazard and exposure data,” it concluded that HFPO may present an unreasonable risk of injury to health or the environment. According to EPA, the potential hazards from exposure to HFPO could include neurotoxicity, reproductive effects, and cancer. EPA notes that it also found there is insufficient information to determine the effects on human health from inhalation of HFPO (which is a gas at room temperature). EPA’s test order will address this data need.
The Chemours Company FC LLC, DuPont De Nemours Inc., E. I. du Pont de Nemours and Company, and 3M Company are the recipients of the test order. According to EPA, prior to issuing the test order, EPA engaged in discussions with the recipient companies “to encourage them to voluntarily submit existing data on HFPO to EPA.” EPA states that based on information in existing studies provided by cooperating companies, EPA determined that certain data were no longer needed and thus excluded the relevant testing requirements from the issued order. EPA made the data voluntarily provided by these companies publicly available in docket EPA-HQ-OPPT-2021-0910.
EPA states that it will continue to communicate with manufacturers and processors before issuing a test order to encourage voluntary data submissions that could refine its categorization of PFAS included in the Testing Strategy, accelerate the availability of data to EPA and the public, or eliminate the need for testing.
Companies subject to the test order may either conduct the tests as described, including testing of physical-chemical properties and health effects following inhalation, or provide EPA with existing information that they believe EPA did not identify in its search. EPA states that it “encourages companies to jointly conduct testing to avoid unnecessary duplication of tests and will also consider possible combinations of tests that cover all required endpoints to diminish the amount of time, animal subjects and costs required.”
According to EPA, the order employs a tiered testing process, as required by TSCA. The results of all the first-tier testing are required to be submitted to EPA within 415 days of the effective date of the order and will inform EPA’s decision as to which additional tests are necessary. EPA will make the order and any data submitted in response publicly available on its website and in the online docket subject to confidentiality considerations under TSCA Section 14.
Bergeson & Campbell, P.C. (B&C®) is pleased to see that EPA properly designated the second PFAS TSCA Section 4(a)(1) test order on HFPO as such, rather than incorrectly calling it a TSCA Section 4(a)(2) test order as EPA did on the test order for 6:2 Fluorotelomer sulfonamide betaine (6:2 FTSB). The test order on HFPO, however, reflects some of the issues of concern we flagged in our review of the first PFAS test order on 6:2 FTSB (see our memorandum dated June 8, 2022), specifically with regard to EPA’s inadequate bases for the ordered testing and inconsistent justifications. We highlight several examples below.
As part of the Tier 1 testing, the test order on HFPO required “Hydrolysis as a Function of pH,” according to EPA and Organization for Economic Cooperation and Development (OECD) test guidelines. The test order further included the requirement for “testing at pH 1.2 for physiological purposes and reporting products of hydrolysis.” EPA’s stated need for these tests was that “Hydrolysis as a function of pH influences stability in the respiratory tract and environmental media, as well as other parameters important for determining inhalation dosimetry and toxicity.” EPA did not, however, provide an adequate basis for ordering these tests.
For example, the EPA and OECD test guidelines for hydrolysis both state that the test guidelines are “…applicable to slightly volatile and non-volatile compounds of sufficient solubility in water.” The test guidelines further state that “The test should not be applied to chemicals that are highly volatile from water…” We note that EPA concluded in the test order that “HFPO has been determined to be an insoluble and/or highly volatile gas [emphasis added].” EPA does not elaborate why it views this test guideline as appropriate for a substance that EPA views as insoluble and highly volatile.
More generally, the test order begs the question of why EPA ordered these studies. HFPO has physicochemical properties that outside the boundaries of substances intended for testing according to OECD TG 111 and HFPO’s residence time in water are expected to be minimal in the non-hydrolyzed state. Therefore, the potential hazard concerns to the stomach from ingesting non-hydrolyzed HFPO from water are insignificant based on existing data.
EPA also ordered in vitro respiratory toxicity testing as part of the Tier 1 testing. Specifically, the test order states: “Portal-of-entry effects in the lung have been demonstrated in animal models but their relevance for use in human health assessment is uncertain. EPA is requiring an in vitro toxicity study using cells of human origin to examine portal-of-entry effects in human tissue [emphasis added].” We find this statement uncompelling given that EPA’s Integrated Risk Information System (IRIS) has developed numerous inhalation reference concentrations (RfC) as part of its human health assessments using animal studies where portal-of-entry effects in the respiratory tract were used for deriving the inhalation RfCs. See, e.g., acrolein IRIS Chemical Assessment Summary. EPA does not explain why animal studies are appropriate for evaluating portal-of-entry effects in humans for IRIS reviews, but are not appropriate for evaluation under TSCA.
We also note that in EPA’s first PFAS TSCA Section 4(a)(1) test order on 6:2 FTSB, EPA only ordered in vivo inhalation toxicity testing because it concluded that “No scientifically valid non-vertebrate test method of equivalent or better scientific quality and relevance currently exists to determine/measure inhalation exposure dosimetry and toxicity for this test substance [which EPA concluded ‘is expected to be an insoluble solid substance and therefore may present concern for portal-of-entry effects for inhalation exposures.’].”
EPA further stated in the test order on 6:2 FTSB that “[it] considered in vitro respiratory toxicity models and found that currently available in vitro respiratory tract cell culture models are only relevant to water-soluble and gaseous substances [emphasis added].” In comparison, EPA stated the following in the TSCA Section 4(a)(1) test order on HFPO as the supporting basis for ordering in vitro respiratory toxicity testing on HFPO: “This test uses air-liquid interface cultures that have been successfully used with gaseous test articles (Bowers et al., 2021; Upadhyay and Palmberg, 2018) and will examine respiratory tract remodeling effects.” EPA appears to argue both that the cell model method is inappropriate (for insoluble 6:2 FTSB) because the in vitro model is only relevant for water-soluble and gaseous substances and yet it is appropriate for HFPO even though HFPO is also insoluble.
It is difficult to reconcile EPA’s mutually exclusive scientific rationales; it appears that EPA is simply seeking to order testing for the sake of ordering testing rather than to generate data to fill a legitimate data need.
As for the Tier 2 testing on HFPO, EPA reviewed the available toxicity data on HFPO, which “…reported hazards for acute toxicity, reproductive and developmental toxicity, specific target organ toxicity, and neurotoxicity.” EPA further noted that the “…available data from an OECD 422 Combined Repeated Dose Toxicity Study with the Reproduction/Developmental Toxicity Screening Test in rats showed severe neurotoxicity, including vacuolization and/or necrosis of brain neuronal cells. Further, acute toxicity and respiratory tract irritation, or portal of entry effects, were evident.” EPA considered these data and the regulatory testing requirements for pesticides under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) and ordered the following in vivo inhalation toxicity studies: a 2-generation reproduction toxicity study, a developmental neurotoxicity study, a subchronic neurotoxicity study, and combined chronic toxicity/carcinogenicity studies. While these high-tier tests may refine EPA’s understanding, it is not clear that EPA needs this additional information in advance of performing a risk evaluation using uncertainty factors, especially since this testing will take about six years, thereby delaying significantly EPA taking any regulatory action. EPA also provides no insight into how these data will inform its PFAS Testing Strategy because HFPO (as an epoxide) is so very different from other PFAS.
Second, EPA’s use of the testing requirements under FIFRA as a supporting basis for ordering testing on HFPO is clearly in conflict with Congressional intent in TSCA. If Congress had intended for EPA to evaluate every possible toxicity endpoint on TSCA regulated chemical substances, it would have established a required list, yet it did not.
Finally, we do not understand EPA’s logic in issuing the PFAS test orders thus far. The PFAS Testing Strategy was designed to advance EPA’s understanding expeditiously of the potential properties, including toxicity of these substances. Therefore, a more pragmatic approach, in our view, would be for EPA to evaluate the existing data, much of which it clearly overlooked, to determine common toxicity endpoints across the various PFAS, and to order testing to evaluate potential data needs on those endpoints for those substances that present the greatest concern. In contrast, EPA appears to be taking an approach of researching these substances one at a time, an approach that the PFAS Testing Strategy was intended to avoid.
B&C is concerned about EPA’s track record with issuing test orders under TSCA Section 4(a)(1) and TSCA Section 4(a)(2). As discussed here and in our June 8, 2022, memorandum, EPA’s approach for identifying existing data on PFAS overlooked a considerable amount of information for its TSCA Section 4(a)(1) test orders. Further, as we noted in our 2022 Forecast, EPA’s TSCA Section 4(a)(2) test orders suffered comparable shortcomings, including EPA ordering testing on substances when data were already available in the peer-reviewed literature and even in EPA’s possession from previous regulatory actions.
We appreciate that EPA is actively taking steps to understand better the potential concerns for PFAS and other existing chemical substances. If EPA believes, however, that PFAS, and HFPO in particular, pose the public health concerns as EPA has publicly stated, then EPA should be taking action expeditiously to regulate these substances, rather than researching all possible hazards while the substances remain actively used in commerce. After all, EPA expressed concerns for these endpoints for HFPO 40 years ago.