sciencemag.org SCIENCEINSIGHTS | POLICY FORUM
cals. The agency is still working on this as-
signment, with about 2000 chemicals left
to review (see SM). ClPFPECAs may well
be in this queue, but confirming whether
they are is not easy; it likely requires a for-
mal request under the federal Freedom of
Information Act. We can therefore extract a
second lesson about the regulatory system:
Some chemicals may fall through the cracks
in the public tracking system in the United
States, not because they are adequately as-
sessed for toxicity but for other reasons.
Yet just because ClPFPECAs may not
be in public view does not mean that the
chemicals are out of range of U.S. regula-
tors. Under TSCA, it is possible that even if
the product is protected as a trade secret,
EPA is actively overseeing the ClPFPECAs.
Under this “extensive regulation” scenario,
EPA could ask for more testing or other
negotiated restrictions as a condition to
approving ClPFPECAs under TSCA. Or
the agency may be satisfied—after con-
ducting its own analysis—that the risks
of ClPFPECAs are not unreasonable. EPA
might even impose new restrictions on the
manufacture of ClPFPECAs over time if it
learns of new “adverse effects,” which man-
ufacturers are required by law to report.
Conversely, even if ClPFPECAs were sub-
ject to premanufacture regulatory review,
it is also possible that EPA decided to pass
ClPFPECAs into commerce without much,
or any, testing or analysis. Only about 15% of
premanufacture notifications for new chemi-
cals submitted to EPA include any health or
safety test data at all [( 15 ), p. 11], and EPA’s
own statistics show that only 10% of the new
chemicals entering commerce between 1979
and 2016 involved restrictions or testing or-
ders (see SM). In the case of the ClPFPECAs,
then, EPA may have been concerned that
there was insufficient scientific evidence
available at the time to support an order de-
manding more testing. Under the pre-2016
law, EPA was required, within a brief 90-day
period, to produce some evidence of poten-
tial risk as a predicate to ordering additional
toxicity tests from the manufacturers. This
catch-22 resulted in a paucity of testing or-
ders ( 1 , 2 ). [Congress removed this legal im-
pediment in the 2016 TSCA amendments, but
that came too late for the ClPFPECAs ( 2 ).]
We thus don’t know much, if any-
thing, about the regulatory oversight of
ClPFPECAs in the United States. This mys-
tery provides yet a third lesson about chem-
ical regulation: For the 40,000-plus chemi-
cals in commerce, the burden of chemical
assessment rests almost entirely on a small
group of EPA regulators ( 1 ). As a result,
some, perhaps many, chemicals likely fall
through the cracks. Indeed, despite the
amendments to TSCA in 2016, chemical
manufacturers are still not required to an-
ticipatorily test or assess their chemicals
as a condition to marketing in the United
States ( 1 , 2 ). Instead, it is the regulators
who bear responsibility for identifying the
most hazardous chemicals, identifying the
relevant scientific literature bearing on tox-
icity, ordering new tests, and synthesizing
and analyzing the available information
bearing on the safety of individual chemi-
cals (not to mention determining proper
policy responses) ( 1 , 2 ). Manufacturers’ pri-
mary role is to serve as respondents. In this
role, they not only provide the information
that EPA requests, but also have the right
to lodge extensive and critical comments on
the agency’s work. Manufacturers can also
sue the agency in court, arguing that some
aspect of the agency’s analysis might be ar-
bitrary (see SM). And the new powers EPA
gained from the 2016 amendments—such as
the ability to require additional testing and
prioritize chemicals of concern—came with
substantial procedural impediments ( 2 ).
What about Europe? ClPFPECAs are of-
ficially registered in two EU regulatory pro-
grams. As mentioned above, EFSA approved
ClPFPECAs in the manufacture of nonstick
coating products ( 5 ). And ClPFPECAs are
listed—along with five hazard classifica-
tions—in the EU’s notification (CLP) data-
base (see SM). In neither case, however, is
the supporting research behind these regula-
tory findings readily available to the public.
Additionally, in 2005, the European Union
implemented REACH, a more aggressive
chemical regulatory program than the U.S.
TSCA program. Unlike TSCA, REACH re-
quires manufacturers to conduct a compre-
hensive literature search on the toxicity of
each of their chemicals sold above threshold
quantities in the European Union. If the ex-
isting scientific information proves incom-
plete under REACH standards, manufactur-
ers are also required to conduct additional
toxicity testing ( 1 ). Yet a search by CAS
number returns no results for ClPFPECAs
on the European Chemical Agency’s web-
site that lists registered REACH chemicals.
Perhaps ClPFPECAs are produced in low
enough quantities (less than 1 tonne/year)
to be exempted from REACH, or perhaps
ClPFPECAs satisfy other REACH exemp-
tions, such as those governing impurities or
polymers as defined under REACH. Publicly
available information again does not re-
solve this question (see SM).A LONG WAY TO GO
Public health and environmental concerns
led to the phase-out of PFOA and its clos-
est chemical relatives. It seems obvious
that in such a scenario, society should
want to be reasonably certain that the re-placement chemical “cure” is not worse
than the phased-out chemical “disease.” Yet
the researchers from EPA and New Jersey
DEP found substitute polyfluorinated com-
pounds, ClPFPECAs, in their environmental
samples ( 3 ). At present, there is little in the
public scientific record to indicate whether
the environmental dissemination of these
substitute PFASs is benign or harmful, and if
harmful, how harmful. Our examination of
the U.S. and European regulatory programs
raises more questions than answers about
the extent to which ClPFPECAs are being
tracked, studied, and regulated. Certainly
the European Union has made more prog-
ress than has the United States in this re-
gard ( 5 ). Still, the toxicological mysteries of
ClPFPECAs—and thousands of other poten-
tially toxic chemicals that are regulated (or
perhaps not regulated) in ways that remain
effectively inscrutable—suggest that we
have a long way to go in designing effective
and accountable chemical regulation, par-
ticularly in the United States. jREFERENCES AND NOTES- J. G. Applegate, Ecol. Law Q. 35 , 721 (2008).
- V. J. Watnick, Harv. Envtl. L. Rev 43 , 373 (2019).
- J. Washington et al., Science 368 , 1103 (2020).
- C. Tonelli, A. Di Meo, R. Picozzi, J. Fluor. Chem. 128 , 46
(2007). - EFSA Panel on Food Contact Materials Enzymes,
Flavourings and Processing Aids, EFSA J. 8 , 1519 (2010). - M. Mazzoni et al., Norman Bull. No. 4, p. 5 (2015).
- EPA docket on PFOA voluntary stewardship program,
docket number EPA-HQ-OPPT-2006-0621;
https://www.regulations.gov/document?D=EPA-HQ-
OPPT-2006-0621-0005 - Y. Wang et al., Ecotoxicol. Environ. Saf. 182 , 109402
(2019). - New Jersey Department of Environmental Protection,
“Comments on USEPA Draft Human Health Toxicity
Assessments for GenX and PFBS” (2019); https://down-
loads.regulations.gov/EPA-HQ-OW-2018-0614-0020/
attachment_1.pdf - California Department of Toxic Substances Control,
Product-Chemical Profile for Carpets and Rugs
Containing Perfluoroalkyl or Polyfluoroalkyl Substances,
Final Version (2019); https://dtsc.ca.gov/wp-content/
uploads/sites/31/2020/02/FinalProduct-Chemical
Profile_Carpets_Rugs_PFASs_a.pdf - M. Goot, North Carol. Law Rev. 98 , 629 (2020).
- P. Grandjean, Environ. Health 17 , 62 (2018).
- M. M. Jacobs, T. F. Malloy, J. A. Tickner, S. Edwards,
Environ. Health Perspect. 124 , 265 (2016). - Environmental Working Group, Off the Books II:
More Secret Chemicals (2016); http://www.ewg.org/
research/off-the-books-ii-more-secret-chemicals/
introduction#_edn1 - U.S. Government Accountability Office, Chemical
Regulation: Options Exist to Improve EPA’s Ability to
Assess Health Risks and Manage Its Chemical Review
Program (2005); http://www.gao.gov/new.items/d05458.pdf - https://comptox.epa.gov/dashboard/dsstoxdb/
results?search=329238-24-6#exec_sum
ACKNOWLEDGMENTS
We thank anonymous reviewers and A. Gore, J. Scott, and V.
Watnick for valuable feedback and comments. S.C.G. and
W.E.W. declare no competing interests.SUPPLEMENTARY MATERIALS
science.sciencemag.org/content/368/6495/1066/suppl/DC110.1126/science.abc12501068 5 JUNE 2020 • VOL 368 ISSUE 6495
Published by AAAS