Analysis of the HPV Challenge: Industry Violations and EPA Negligence
by Nicole Cardello, M.H.S. Executive
Summary
The Environmental Protection Agency’s
(EPA’s) High Production Volume (HPV) Challenge encourages chemical
companies to volunteer to conduct screening-level animal toxicity
tests on 2,800 industrial chemicals. Pursuant to this program, companies
have committed to provide available toxicity information and present
proposals to test chemicals. Researchers at the Physicians Committee
for Responsible Medicine (PCRM) have reviewed all 24 HPV Challenge
test plans submitted to the EPA as of August 1, 2001, evaluating
each test plan for compliance with the original HPV Challenge framework
as well as animal welfare guidelines issued by the EPA. This review
found major and recurring flaws in the program’s execution
as well as its fundamental design. Of the 24 test plans submitted,
18 (75 percent) violate fundamental terms of the program. Many participating
companies have failed to conduct comprehensive analyses of available
data and have instead proposed unnecessary and inhumane tests. Furthermore,
the program’s exclusion of human health and exposure data is
leading to irrelevant experiments that will not protect human health
or the environment.
Introduction
In April 1998, former Vice President
Al Gore announced the "Chemical Right-to-Know Program,"
a national initiative aimed at making hazard information about chemicals
available to the public. In October of that year, the Environmental
Protection Agency (EPA) announced, as part of this initiative, the
High Production Volume (HPV) Challenge. This plan encourages chemical
manufacturers, importers, and distributors to volunteer to conduct
screening-level animal toxicity tests on 2,800 industrial chemicals
produced or imported in the United States in amounts exceeding 1
million pounds per year.
The impetus for the HPV Challenge came
from a July 1997 report entitled Toxic Ignorance, issued
by the environmental group Environmental Defense and the April 1998
EPA report Chemical Hazard Data Availability Study: What Do We
Really Know About the Safety of High Production Volume Chemicals?
These papers suggested that basic information about many HPV chemicals
is not available. In December 1998, PCRM conducted its own analysis,
finding data in publicly available databases overlooked by Environmental
Defense and the EPA that were sufficient for basic hazard assessments
of the majority of the chemicals. This information is documented
in PCRM’s report entitled Availability of HPV Chemical Data.
During its design, the HPV Challenge
was never subjected to scientific and public comment. Only later
did a notice appear in the Federal Register, but no comments
were solicited. PCRM president Dr. Neal Barnard testified before
the House Subcommittee on Energy and the Environment in June 1999
and demonstrated that much of the data the EPA thought were lacking
were, in fact, readily available. Dr. Barnard also pointed out that
the list of chemicals slated for testing included chemicals with
well-known toxicity, such as carbon tetrachloride, turpentine, tetraethyl
lead, and cyclonite rat poison.
HPV Program Implementation
The HPV Challenge is a voluntary program,
wherein chemical manufacturers, producers, and importers volunteer
to submit testing proposals to the EPA for certain chemicals. The
agency makes these test plans publicly available and allows comments
from interested parties for 120 days thereafter. Under the program,
participating companies are to provide all available information
that would be relevant to the hazard characterization to avoid unnecessary
or duplicative testing. More than 400 companies are not participating
in the program, and in-house data from these non-participating companies
is not being used in the program. The HPV Challenge ignores all
human toxicity and exposure data.
The series of tests the EPA has requested
under the HPV program is based on the Organization of Economic Cooperation
and Development’s (OECD’s) Screening Information Data
Set (SIDS), which calls for experiments in six areas: acute toxicity,
repeat dose toxicity, developmental and reproductive toxicity, mutagenicity,
ecotoxicity, and environmental fate. This test battery includes
as many as seven animal tests and two in vitro tests. (See
Appendix A.)
As a result of concerns raised by PCRM
and other organizations, a compromise was reached among chemical
manufacturers, the EPA, and environmental, health, and animal protection
groups in October 1999. This October Agreement took the form of
a guidance letter from the EPA to all HPV participants dated October
14, 1999, which delineated principles for minimizing irrelevant
and repetitive tests and considering animal welfare provisions.
The ten points of the agreement include, in part:
1. In analyzing the adequacy of existing
data, participants shall conduct a thoughtful, qualitative analysis
rather than use a rote checklist approach. Participants may conclude
that there are sufficient data, given the totality of what is
known about a chemical, including human experience, that certain
endpoints need not be tested.
2. Participants shall maximize the
use of existing data and scientifically adequate data to minimize
further testing.
3. Participants shall maximize the
use of scientifically appropriate categories of related chemicals
and structure activity relationships.
4. Consistent with the Screening Information
Data Set (SIDS) program of the Organization for Economic Cooperation
and Development (OECD), participants shall not conduct any terrestrial
toxicity testing.
5. Participants are encouraged to use
in vitro genetic toxicity testing to generate any needed
genetic toxicity screening data, unless known chemical properties
preclude its use.
6. Consistent with the OECD/SIDS program,
participants generally should not develop any new dermal toxicity
data.
7. Participants shall not develop sub-chronic
or reproductive toxicity data for the HPV chemicals that are solely
closed system intermediates, as defined by the OECD/SIDS guidelines.
8. In analyzing the adequacy of screening
data for chemicals that are substances Generally Recognized as
Safe (GRAS) for a particular use by the Food and Drug Administration
(FDA), participants should consider all relevant and available
information supporting the FDA’s conclusions. Participants
reviewing the adequacy of existing data for these chemicals should
specifically consider whether the information available makes
it unnecessary to proceed with further testing involving animals.
As with all chemicals, before generating new information, participants
should further consider whether any additional information obtained
would be useful or relevant.
9. Because validated non-animal tests
for some SIDS endpoints may be available soon, participants shall
make the following revisions to the sequence of testing:
(a) Testing of closed system intermediates,
which present less risk of exposure, shall be deferred until
2003;
(b) Individual chemicals (i.e. those
HPV chemicals not proposed for testing in a category) that require
further testing on animals shall be deferred until November
2001.
10. Companies shall allow 120 days
between the posting of tests plans and the implementation of any
testing plans.
The EPA’s Susan Wayland, Charles
Auer, and Oscar Hernandez reiterated these principles in October
2000 in letters to all HPV participants and trade associations.
The EPA further emphasized this guidance in the Federal Register
Notice 65 Fed. Reg. 81686 (December 26, 2000) entitled Data Collection
and Development on High Production Volume (HPV) Chemicals.
Research Methods
PCRM has evaluated each test plan that
has been submitted to the EPA and subjected to public comment under
the HPV Challenge program as of August 1, 2001, to evaluate compliance
with the original HPV framework and the October Agreement. Specifically,
PCRM has examined each test plan along the following parameters:
adequate presentation of existing data, appropriate use of chemical
categories, testing within the scope of the HPV program, and consideration
of whether or not proposed experiments would advance the understanding
of the test substances’ potential health risks.
PCRM evaluations entailed searches of
publicly available databases for relevant toxicity studies, analyses
of physical and chemical properties, environmental fate and transport,
human exposure assessments, and epidemiological studies. The databases
searched include PubMed, Chemfinder, Toxline, Hazard Substances
Database, the EPA’s Integrated Risk Information System, and
the National Toxicology Program study results database. Additionally,
PCRM ascertained whether the sponsored chemicals were already regulated
by other federal agencies, such as the Food and Drug Administration
(FDA) or the Occupational Safety and Health Administration (OSHA).
PCRM identified exposure limits recommended by these agencies as
well as the National Institute for Occupational Safety and Health
(NIOSH). PCRM also performed basic Internet searches and consulted
toxicology textbooks.
After reviewing available information
on these chemicals, PCRM researchers assessed whether or not the
tests proposed by chemical companies would expand relevant knowledge
of the risk posed by the chemicals to the environment or public
health.
Results
Of the 24 test plans reviewed by the
EPA as of July 26, 2001, 18 (75 percent) contained substantive violations.
These violations are presented in Table 1 and Table 2. These problems
recur with such sufficient frequency that they cast serious doubt
on the participants’ ability to carry the program through and
the EPA’s commitment to properly administering the program.
Most notably:
1. Chemical manufacturers have repeatedly
failed to report existing hazard information.
2. Opportunities to group structurally
or toxicologically similar chemicals are not being utilized.
3. Many companies are proposing animal
tests beyond the scope of the HPV program.
4. The EPA is not enforcing the original
HPV framework or the principles outlined in the October Agreement.
In addition, implementation of the program
to date shows these major flaws in the program’s fundamental
design:
5. The HPV program excludes exposure
and use information that would make hazard information meaningful.
6. The HPV program is focused exclusively
on nonhuman data, even though human toxicity, exposure, and epidemiological
studies have been performed on many of the sponsored chemicals.
7. Available in vitro tests
using aquatic microorganisms are being neglected.
These problems of the HPV Challenge program
are discussed below:
1. Chemical manufacturers have repeatedly
failed to report existing hazard information.
Under the HPV program, chemical companies
have committed to conducting comprehensive literature reviews of
existing data to determine if any further animal tests should be
conducted. This idea was reiterated in the October Agreement, which
states that "participants shall conduct a thoughtful, qualitative
analysis rather than use a checklist approach.…Participants
shall maximize the use of existing and scientifically adequate data
to minimize further testing."
Of the 24 test plans that underwent the
120-day public review as of August 1, 2001, 15 called for new testing.
Of these, PCRM found additional available hazard information that
had been overlooked or otherwise omitted for 8 of the 15 (53 percent)
chemicals. (See Table 1, Table 2.) Because companies have, in these
instances, not been thorough in their review of existing data, many
studies will be unnecessarily duplicated. In many cases, the available
information on the toxicity and metabolism of chemicals far surpassed
the basic information that can be ascertained by the SIDS battery.
For example, the American Petroleum Institute
(API) proposed a series of animal tests on petroleum coke, a solid,
nearly pure carbon product resulting from petroleum processing and
refining. In its test plan, the API failed to provide a complete
review of data on the known health and environmental effects of
these substances. Numerous toxicological studies have been conducted
on animals. Additionally, epidemiological studies of petroleum workers
exposed to petroleum coke and other related substances from petroleum
coker facilities, carbon black operations, and coal coke dust document
known human health effects.
The API also failed to present available
information in its test plan for petroleum gases. Petroleum gases
are products from natural gas processing and petroleum refining
operations, and include chemicals such as methane, ethane, propane,
and butane. At high doses, they displace oxygen. However, extensive
human and nonhuman animal data already indicate that these compounds
are relatively non-toxic. Moreover, people are generally exposed
to very low doses. Illustrative of the generally benign chemical
nature of these compounds is the fact that propane, n-butane, and
isobutane are all classified as Generally Recognized as Safe (GRAS)
for appropriate uses in food products by the Food and Drug Administration.
The American Gas Association specifically requested that these compounds
be exempt from the HPV program in a letter to the EPA, as these
naturally occurring substances are already sufficiently well characterized
and present a very low health risk. In defense of the HPV program,
Dr. William Sanders, director of the EPA Office of Pollution Prevention
and Toxics, testified before the U.S. House Science Subcommittee
on Energy and the Environment in June 1999 that the EPA was "not
requiring testing on butane." Nevertheless, the API has proposed
an unnecessary test series on ethane, propane, butane, and isobutane
that will not contribute to the understanding of the sponsored substances’
toxicological characteristics, and the EPA failed to reject this
excessive and inappropriate test plan.
In yet another example of inappropriate
testing, the American Chemistry Council (ACC) proposed re-testing
mixtures containing butadiene and isoprene, both of which are potential
human carcinogens. Butadiene is a synthetic chemical used primarily
in the production of rubber and plastics, and is also present in
gasoline. The effects of butadiene chemicals have already been evaluated
in humans in numerous occupational studies. These chemicals are
known to be sufficiently dangerous. As such, screening-level animal
tests are clearly unnecessary.
Exposure to the synthetic form of isoprene
occurs in the manufacturing of rubber. Isoprene also occurs naturally
in the environment as emissions from vegetation. The crude, screening-level
tests proposed in this test plan will provide no insight that would
facilitate the regulation of isoprene in the workplace. Extensive
toxicological work is already being conducted on the kinetics of
isoprene, and human metabolism, toxicological mechanisms, human
biomarkers, and human physiologically based pharmacokinetic (PBPK)
models should be further investigated. There is no appropriate role
for the HPV test series in further evaluation of this chemical.
General Electric (GE) Plastics submitted
three test plans for individual chemicals without providing even
minimal information about the structure and use of the compounds.
For example, in its test plan for tris(nonylphenol)phosphite (TNPP),
GE failed to provide such basic information as the structure and
use of the chemical, its physical properties, or the test method
for each health endpoint. Additionally, the test plan disregarded
the environmental fate and transport of the compound. Existing data
on food products indicate that foods in contact with TNPP additives
may contain levels of free nonylphenol, a product of a hydrolysis
reaction of TNPP. Nonylphenol is a potential endocrine disruptor
for which an abundance of data exists on its toxicological and physicochemical
properties. Since nonylphenol may well be the environmentally relevant
moiety, a critical review of its properties and behavior is essential
for a thorough analysis of the potential health effects of TNPP.
However, application of the crude SIDS battery will not advance
our understanding of the health risks in any meaningful way.
2. Opportunities to group structurally or
toxicologically similar chemicals are not being utilized.
The
October Agreement states, "Individual chemicals (i.e. those
chemicals not proposed for testing in a category) that require testing
on animals shall be deferred until November 2001." However,
5 of the 15 (33 percent) test plans calling for animal tests were
for individual chemicals, in clear violation of the October Agreement.
(See Table 1, Table 2.) The failure to form chemical categories
and apply structure activity relationships will result in repetitive
and inefficient testing. For example, the individual chemicals submitted
by GE could be grouped with other compounds. TNPP could be assessed
in the context of a larger group of phenyl-phosphorous antioxidant
stabilizers. p-Cumylphenol could be included in a larger substituted
or alkylphenol category. In addition to reducing the costs and numbers
of animals used in screening evaluations, formation of chemical
categories provides a contextual basis to evaluate toxicity, providing
a deepr understanding about compounds’ toxicity.
In another example, the API failed to
coordinate with other chemical companies to include all related
compounds in its analysis of petroleum coke. The composition of
petroleum coke is very similar to other HPV chemicals, such as "coal,
anthracite, calcined CAS #68187597" and coke, (coal) CAS #65996772,
which were not included in this group. Using information about these
other compounds would support a more in-depth analysis of the potential
hazard of petroleum coke.
The API’s test plan for petroleum
gases represents a missed opportunity to apply structure activity
relationships to evaluate chemical toxicity. Petroleum gases include
substances such as methane (natural gas), ethane, butane, isobutane,
and propane. These gases are simple organic compounds with a simple
progression of structures and identical functional groups throughout,
meeting all criteria for application of structure activity relationships
as outlined in EPA guidance. Yet, the EPA refused to support the
category construction and the opportunity to use existing data that
would both reduce animal use and overall cost.
In contrast, the test plans submitted
by the Flavor and Fragrance High Production Volume Consortia represent
thoughtful analyses of well-studied chemicals that have been used
in structure activity relationships for many years. The Consortia
demonstrated a good understanding about toxicokinetics of these
compounds, many of which are classified as GRAS food ingredients
by the FDA. Given that these compounds have been consumed by millions
of people over many years, the resultant effects, if any, are available
from human observation without the need to extrapolate from rodents.
Regrettably, in the face of all the evidence presented, the EPA
still is not convinced the category is reasonable.
3. Many companies are proposing animal tests
beyond the scope of the HPV program.
The October Agreement states, "Participants
are encouraged to use in vitro genetic toxicity testing to
generate any needed genetic toxicity screening data, unless known
chemical properties preclude its use." However, 6 of the 24
test plans (25 percent) propose to use less sensitive genetic toxicity
tests on animals without justification. (See Table 1, Table 2.)
The October Agreement also states, "Consistent
with the OECD/SIDS program, participants generally should not develop
any new dermal toxicity data." Yet, four test plans (17 percent)
called for these lethal dermal dose tests. (See Table 1, Table 2.)
Although terrestrial toxicity testing
is specifically proscribed, the API’s test plan for petroleum
coke called for this test. (See Table 1, Table 2.)
Some test plans have offered no rationale
for testing at all. GE Plastics submitted test plans for three individual
chemicals that were mere checklists, with no rationale for conducting
experiments and little information about its sponsored chemicals.
GE proposed in vivo genotoxicity studies and dermal toxicity
tests, in violation of the agreement.
The FMC Corporation submitted test plans
for two individual chemicals, methallyl chloride (MAC) and 2,3-dihydro-2,2-dimehtyl-7-benzofuranol
(7OH). The test plan for 7OH called for acute dermal and in vivo
genetic toxicity tests. The FMC Corporation volunteered these chemicals
on March 10, 1999, yet did not submit test plans until February
13, 2001, after the majority of the tests had been completed.
This company side-stepped the public review process and violated
the October Agreement as well as the original framework that the
HPV participants agreed to follow, both of which called for a comment
period between the submitting of proposed test plans and their implementation.
4. The EPA is not enforcing the original
HPV framework or the principles outlined in the October Agreement.
Despite
its stated commitments to minimizing irrelevant tests, the EPA continues
to make it difficult for companies to reduce testing and has endorsed
rote checklist test plans from companies that propose testing beyond
the scope of the HPV program.
According to the October Agreement, "[I[n
analyzing the adequacy of screening data for chemicals that are
food substances Generally Recognized as Safe (GRAS) for a particular
use by the Food and Drug Administration (FDA), participants should
consider all relevant and available information" and should
specifically "consider whether the information available makes
it unnecessary to proceed with further testing involving animals."
Yet, the EPA has supported the Flavor and Fragrance High Production
Consortia’s proposals to unnecessarily dose fish with GRAS
food ingredients, including components of cinnamon oil and terpenoid
alcohols naturally occurring in fruits and vegetables.
The EPA deemed the aforementioned GE
test plans that represented violations of the October Agreement
to be "minimally acceptable" and pointed out some of the
problems. It then failed to follow up on the matter.
The EPA did point out some of the violations
in the FMC test plan in a letter to the company signed by Oscar
Hernandez dated March 7, 2001, in which the EPA requested specifically
that "FMC consider these concerns and advise the Agency within
30 days of any modifications to its submission." However, the
EPA did not follow up with the company directly, but simply reiterated
concerns about violations in comments on its Web site on June 14,
2001. Only after PCRM intervened did the company agree to drop some
tests.
The American Forest and Paper Association
(AF&PA) HPV Work Group submitted a test plan for spent pulping
liquor, a highly alkaline, corrosive mixture that causes tissue
damage immediately upon contact. Human or ecological exposures to
these substances are highly unlikely, as the material remains on
the manufacturing site and has limited commercial use. The AF&PA
did not propose any mammalian tests with caustic substances, citing
OECD documents clearly stating that any testing on animals that
would cause pain and suffering need not be carried out. However,
the AF&PA did propose tests on fish with a neutralized mixture,
despite the fact that the mixture’s fundamental toxic property
is lost after neutralization. Any tests with neutralized spent pulping
liquor will produce meaningless results. A comment from the AF&PA’s
test plan essentially acknowledges these problems and that testing
is being conducted simply to be a "team player":
"It will be necessary to neutralize
the test material in order to bring it to a pH that is compatible
with survival of the test organisms in order to perform the testing.
This will affect the composition of the material and the results
therefore may not represent the original substance. However, AF&PA
will undertake the testing in the spirit of the HPV program."
The EPA did not reject this test plan.
The EPA also accepted the API’s test plan on petroleum gases,
which included a series of tests on ethane, propane, butane, and
isobutane, despite the fact that, as noted above, Dr. William Sanders
testified before the U.S. House Science Subcommittee on Energy and
the Environment in June 1999 that the EPA was "not requiring
testing on butane."
The above points show the HPV Challenge
participants are openly and frequently violating the program guidelines
and the October Agreement. The EPA has ignored their violations,
allowing excessive and inappropriate testing to proceed.
A review of the 24 test plans shows fundamental
flaws in the HPV Challenge design. These problems are discussed
below:
5. The HPV program excludes exposure and
use information that would make hazard information meaningful.
The underlying
assumption of the HPV program is that substances produced in large
quantities automatically lead to high exposures and therefore threaten
public health. The EPA states in the Federal Register Notice
65 Fed. Reg. 81686 (December 26, 2000) that "it is generally
accepted that chemicals having a high level of production have an
increased potential for exposure," but does not provide any
support for this assumption. Typically, it has not proven true.
The aminosilanes, sponsored by the Silicones
Environmental Health and Safety Council, break down immediately
when released into the environment and therefore do not present
a hazard. Yet, repeat dose and reproductive toxicity tests are being
run for physically irrelevant endpoints. The petroleum gases are
found in ambient and occupational environments at levels that are
orders of magnitude below exposure limits.
The Flavor and Fragrance High Production
Volume Consortia’s test plans for terpenoid tertiary and primary
alcohols underscore this fundamental flaw of the HPV program. The
ingestion of the natural form of the terpenoid tertiary and primary
alcohols in fruits and vegetables is more than an order of magnitude
greater than the ingestion of the manufactured form of these substances
used as food flavoring additives. These HPV chemicals have been
studied and classified as GRAS food ingredients by the FDA, showing
that chemicals produced in large quantities are not necessarily
harmful.
The EPA is asking for tests with the
cinnamyl derivatives and the terpenoid teritary and primary alcohols.
These chemical categories, proposed by the Flavor and Fragrance
High Production Volume Consortia, include common household substances
found in fragrance and food flavoring products. These tests are
difficult to justify.
The EPA approved the American Forest
and Paper Association HPV Work Group’s test plan for spent
pulping liquor, a highly alkaline, corrosive mixture that causes
tissue damage immediately upon contact. As noted above, human or
ecological exposures to these substances are highly unlikely, as
the material remains on the manufacturing site and has limited commercial
use. The hazards associated with this mixture are already well-known
and any information resulting from the SIDS battery would not change
the understanding or handling of these corrosive chemicals.
Exposure information could be used to
prioritize the 2,800 HPV chemicals so that companies are not wasting
resources on studying naturally occurring substances and food additives,
or chemicals that are present at undetectable levels in the environment.
6. The HPV program is focused exclusively
on nonhuman data, even though human toxicity, exposure, and epidemiological
studies have been performed on many of the sponsored chemicals.
The
HPV Challenge program neglects human data that are already available
and are most relevant for predicting human toxicity. In fact, many
of the chemicals, such as butadiene, butane, spent pulping liquor,
and others are already monitored or controlled in the occupational
environment. Existing occupational studies of miners and refinery
workers would provide relevant information for the API’s petroleum
coke test plan. The ACC’s characterization of 1,3-butadiene
is incomplete without summaries of the available human exposure,
biomarker, and epidemiological studies. The API’s petroleum
gas plan calls for testing on animals with common substances, such
as propane, butane, methane, and ethane, which come from natural
gas processing and petroleum refining operations. By the API’s
own admission, extensive human and nonhuman animal data already
indicate that these compounds are relatively non-toxic.
The neglect of human data presents major
problems for interpretation. Results from animal experiments must
be extrapolated from high to low doses, and then from rats and mice
to humans. The uncertainties involved in these assumptions reduce
the validity and accuracy of risk assessments. For example, significant
intra- and interspecies differences in effects of butadiene and
isoprene have hindered the understanding of the behavior of these
potential carcinogens in humans. This failure to incorporate existing
human health data is currently impeding reassessment and regulatory
action by the EPA.
7. Available in vitro tests using
aquatic microorganisms are being neglected.
Protozoan members of Ciliophora, such
as the Tetrahymena, are frequently used as a measure of aquatic
toxicity in ecological risk assessments (Larsen et al, 1997). The
biochemistry and physiology of Tetrahymena have been thoroughly
investigated since the 1950s, and Tetrahymena, especially T.
pyriformis, have been used for aquatic toxicity testing since
the 1970s. Moreover, the genomics of the organism are currently
being elucidated. The T. pyriformis population growth test
is quick, easy, and inexpensive, and has considerable breadth (Schultz,
1997). It allows the examination of a large number of independent
organisms that possess features of both single eukaryotic cells
and multicellular organisms. Studies at varying concentration levels
can easily be repeated, and many chemicals can be examined in a
short period of time. Range-finding tests allow accurate approximation
of both the highest concentration with no observed effect on population
growth and the lowest concentration with total inhibition of cell
replication. Fish toxicity tests are less economical, less humane,
slower, and more labor intensive.
The EPA has a massive database on the
acute toxicity of more than 600 organic chemicals to fish called
"Acute Toxicities of Organic Pollutants to Fathead Minnows
(Pimephales promelas)." Comparisons of toxicity test results
from the in vitro TETRATOX assay and the EPA’s fish
acute toxicity data have yielded good correlation between the two
methods (Sinks et al, 2001). Similarly, good correlation was observed
between ciliate and guppy fish toxicity (Seward et al, 2001). Moreover,
where there is not good agreement, there is a logical rationale
for this departure (Bearden and Schultz, 1998). Evaluation of in
vitro and in vivo aquatic toxicity data has allowed researchers
to develop models to predict toxicity based on quantitative structure
activity relationships (QSARs) (Schultz, 1999; Schultz and Cronin,
1999; Niculescu et al, 2000). Both the in vitro TETRATOX
assay as well as QSARs provide more humane, efficient methods to
predict aquatic toxicity at the screening level.
Unfortunately, there is no mechanism
in the HPV program for eliminating testing on fish based on the
physical properties of the chemical. Therefore, substances with
low observed toxicity or compounds with low water solubility, such
as the cinnamon oil components, have been proposed for unnecessary
testing on fish. Highly volatile compounds, such as the petroleum
gases, are unlikely to pose a hazard to fish, yet the API proposed
in vivo aquatic toxicity tests.
Conclusion
As of August 1, 2001, 24 test plans have
been submitted to the EPA and undergone the full public comment
period under the HPV Challenge program. PCRM has submitted comments
on nearly all test plans, yet the EPA has failed to respond to even
one set of comments. Despite clear-cut guidelines for the conduct
of this program, many participating companies have provided test
plans that fail to bring forth relevant data and have instead proposed
unnecessary, expensive, and inhumane tests. Furthermore, the program’s
exclusion of existing human health and exposure data is leading
to duplicative and irrelevant tests that will not protect public
health or the environment.
The program is estimated to cost the
EPA $16 million each year to administer. EPA officials have expressed
concern about their ability to handle the deluge of incoming data
and they have been unable to appropriately follow up on the 24 test
plans to date. They have provided mixed messages to companies by
repeatedly expressing a commitment to reducing irrelevant tests,
but then requesting unnecessary animal tests for substances with
low toxicity, such as cinnamon oil and ethane. EPA officials have
requested additional tests on potential carcinogens instead of recommending
regulatory action. They have asked for tests on neutralized solutions
of corrosive substances, when the hazards of the toxic compounds
are well understood and tightly controlled. In other words, the
EPA has insisted—contrary to the October Agreement and good
science—that companies "check the box" rather than
perform a thoughtful, useful analyses. The EPA’s presumption
is that more animal testing is usually required and that obstacles
continue to be raised for those companies that attempt to use categories,
structure activity relationships, or existing data.
Experience to date with the HPV Challenge
indicates it is poorly administered and haphazardly conducted, and
will do nothing to protect human health or the environment.
References
Bearden AP, Schultz TW. Comparison
of Tetrahymena and Pimephales toxicity based on mechanism
of action. SAR QSAR in Environmental Research 1998;9:127-53.
Dimitrov SD, Mekenyan OG, Schultz TW. Interspecies modeling of narcotics
toxicity to aquatic animals. Bulletin of Environmental Contamination
and Toxicology 2000;65:399-406.
Niculescu SP, Kaiser LE, Schultz TW. Modeling the toxicity of chemicals
to Tetrahymena pyriformis using molecular fragment descriptors
and probabilistic neural networks. Archives of Environmental Contamination
and Toxicology 2000;39:289-98.
Schultz TW. Tetratox: Tetrahymena pyriformis population growth
impairment endpoint—a surrogate for fish lethality. Toxicological
Methods 1997;7:289-309.
Schultz TW. Structure-toxicity relationships for benzenes evaluated
with Tetrahymena pyriformis. Chemical Research in Toxicology
1999;12:1262-7.
Schultz TW, Cronin MTD. Response-surface analyses for toxicity to
Tetrahymena pyriformis: reactive carbonyl-containing
aliphatic chemicals. Journal of Chemical Information and Computer
Sciences 1999;39:304-9.
Seward JR, Hamblen EL, Schultz TW. Regression comparison of Tetrahymena
pyriformis and Poecilia reticulata toxicity. Chemosphere
2001;(in press).
Sinks GD, Schultz TW. Correlations of Tetrahymena and
Pimephales toxicity: evaluation of 100 additional compounds.
Environmental Toxicology and Chemistry 2001;20:917-21.
Media
Center | Health | Research
| About PCRM | Catalog
| Join Us | Search
| Site Index | Home
The site does
not provide medical or legal advice. This Web site is for information purposes
only.
Full Disclaimer | Privacy Policy
|
Animal
Experimentation Issues 
