United States
Environmental Protection
Agency
Wfl
Toxicology
II
-------
PESTICIDE REREGISTRATION
REJECTION RATE ANALYSIS
TOXICOLOGY
-------
TABLE OF CONTENTS
CHAPTER PAGE
I. INTRODUCTION
H. SCOPE OF ANALYSIS 2
m. TOXICOLOGY CHAPTER 4
IV. DESCRIPTION OF DISCIPLINE 5
V. CURRENT REJECTION RATE 22
VI. REJECTION FACTORS 51
SUMMARY TABLE OF
REJECTION FACTORS 167
CONCLUSIONS 172
IX. RECOMMENDATIONS 173
X. APPENDIX A - List of EPA
Guidance Documents 176
-------
LIST OF FIGURES
Figure 1 - Rejection and Supplementary Rates For All Toxicology Guideline
Requirements
Figure 2 - Current (Post-1988) Rejection Rates By Guidelines - Acute Toxicity
Figure 3 - Current (Post-1988) Rejection Rates By Guideline - Subchronic Toxicity
Figure 4 - Current (Post-1988) Rejection Rates By Guideline - CORT Studies
Figure 5 - Current (Post-1988) Rejection Rates By Guideline - Other Toxicology Studies
Figure 6 - Toxicology Guidelines with Lower Rejection Rates Over Time
Figure 7 - Toxicology Guidelines with Increasing Rejection Rates Over Time
Figure 8 - Guideline 81-1 Acute Oral Toxicity (Rat)
/•
Figure 9 - Guideline 81-2 Acute Dermal Toxicity (Rabbit/Rat)
Figure 10 - 81-3 Acute Inhalation Toxicity (Rat)
82-4 Subchronic 90-day Inhalation (Rat)
Figure 11 - Guideline 81-4 Acute Eye Irritation (Rabbit)
Figure 12 - Guideline 81-5 Acute Primary Dermal Irritation
Figure 13 - Guideline 81-6 Dermal Sensitization
Figure 14 - Guideline 82-1 (a) Subchronic 90-day Feeding Study (Rat)
Figure 15 - Guideline 82-l(b) Subchronic Feeding Study (Dog)
Figure 16 - Guideline 82-2 Subchronic 21-day Dermal (Rat/Rabbit)
Figure 17 - Guideline 82-3 Subchronic 90-day Dermal (Rodent)
Figure 18 - Guideline 83-l(b) Chronic Toxicity (Dog)
Figure 19 - Guideline 83-3(a) Developmental Toxicity (Rat)
Figure 20 - Guideline 83-3(b) Developmental Toxicity (Rabbit)
Figure 21 - Guideline 83-4 Two-Generation Reproduction (Rat)
-------
Figure 22 - Guideline 83-5 [83-l(a) and 83-2(b)] Combined Chronic/Onco (Mouse)
Figure 23 - Guideline 83-5 [83-l(a) and 83-2(a)] Combined Chronic/Onco (Rat)
Figure 24 - Guideline 84-2(a), 84-2(b), 84-4 Mutagenicity
Figure 25 - Guideline 85-1 General Metabolism
Figure 26 - Guideline 85-2 Dermal Absorption
-------
REJECTION RATE ANALYSIS
L INTRODUCTION
This rejection rate analysis has been undertaken by the Special Review and
Reregistration Division (SRRD), the Health Effects Division (HED) and the Environmental
Fate and Effects Division (EFED) in the Office of Pesticide Programs (OPP) of the
Environmental Protection Agency (EPA). The purpose of this guideline-by-guideline analysis
is to identify those factors that most frequently cause toxicology guideline studies required
for reregistration to be rejected. This information will enable OPP to (a) provide registrants
with information on rejection factors to minimize their reoccurrence in future studies, (b)
reassess the adequacy of its guidance, (c) determine the appropriate regulatory response to a
future rejected study, and (d) make any internal changes in process, procedures or criteria
deemed appropriate.
The decision to analyze these factors was made after a FIFRA Reregistration
recosting analysis, conducted in the Spring of 1991, indicated that rejected studies posed the
most significant potential for delays in the production of Reregistration Eligibility Documents
(REDs). Reregistration eligibility decisions require that reasonable risk assessments be
performed for all relevant human health and ecological end points for each chemical.
Performing such risk assessments requires a "substantially complete" data base. A
"substantially complete" data base requires that registrants submit toxicology studies of
acceptable quality. Also, a significant reduction in rejection rates for most disciplines is
required for OPP to be able to meet its production schedule for REDs.
-------
BL SCOPE OF ANALYSIS
Initially, the scope of this analysis was limited to studies which were characterized by
the Agency's reviewers as supplementary, invalid, or unacceptable. The supplementary
classification included studies that were upgradable or fatally flawed (i.e., not upgradable and
rejected). Additionally, supplementary studies that were later upgraded to core minimum,
core guideline, or acceptable were included.
The analysis included rejected studies because such studies require additional time and
resources of the Agency for review of replacement studies. A review with a finding of
invalid, unacceptable, or supplementary - fatal classification was considered a rejected study
for the purposes of this analysis.
The potentially upgradable studies (10% [845/8250] of the reviewed studies) were
also analyzed because the evaluation of new data to upgrade a study can take as much as or
more time than the original evaluation.
Finally, this analysis included all studies on reregistration lists A through D as well as
new chemicals (January 1985 to June 1992). Excluded from this analysis were range-finders,
studies that were submitted for other than guideline purposes, and interim reports. Since the
scope of analysis for toxicology is different from that used for the other study disciplines
which relied on an analysis of List A rejected studies only), comparison of toxicology's
rejection rates with other discipline rejection rates may not be appropriate.
Process
First the Agency reviewed the data evaluation records (study reviews) on a guideline-
by-guideline basis in order to:
(1) identify those factors that most frequently caused each guideline study to be
rejected;
(2) determine the rejection rates and trends (where the sample size was adequate)
for each guideline requirement;
(3) assess the adequacy of EPA's guidance documents with respect to each
rejection factor; and
(4) determine if each rejection factor is "avoidable."
Secondly, a draft was provided to an industry workgroup of toxicologists for review
and comment in order to (1) obtain from a user's perspective the adequacy of EPA's
guidance documents corresponding to each rejection factor, and (2) better understand why the
rejection factors occur. The industry workgroup included: Ray Cardona (Uniroyal), Rick
-------
Stanton (Valent), Gail Arce (Elf Atochem), Jim Bus (DowElanco), Linda Dulak (Hoechst-
Roussel), Susan Hurt (Rohm & Haas), Joel Kronenberg (Monsanto), Jim Killeen (Ricerca),
Larry Kier (Monsanto), Val Roloff (Monsanto), Jane McCarty (FMC), Mitch Sauerhoff
(CIBA-GEIGY), Has Shah (CMA - Biocides Panel), and Abe Tobia (BASF). Industry and
EPA scientists met on February 17 - 18, 1993 to discuss the problem areas in order to
develop a better understanding of them.
The revised toxicology chapter explicitly includes industry comments on each
rejection factor and EPA's response to them.
-------
HI. TOXICOLOGY CHAPTER
This chapter examines the results of the toxicology rejection rate analysis. The
following information is discussed: (1) a description of the discipline of toxicology, (2) the
current rate of rejection of toxicology studies, (3) a list of the most common factors that have
led to the rejection of these studies, (4) industry comments on each rejection factor, (5) EPA
response to comments, and (6) conclusions and recommendations.
-------
IV. DESCRIPTION OF DISCIPLINE
Toxicology data are used by the Agency to assess hazards to humans and domestic
animals. The data are derived from a variety of acute, subchronic, and chronic toxicity tests,
developmental/reproductive tests, and tests to assess mutagenicity and pesticide metabolism.
The following is a description of the toxicology studies that are required by EPA for
registration and reregistration purposes:
ACUTE TESTING
Acute Oral Toxicity - Rat (guideline 81-1). The purpose of an acute oral toxicity
study is to obtain insight into the lexicological symptoms associated with a single oral
ingestion of a pesticide, and to determine the median lethal dose (i.e., LD50). The
incidences of mortalities at each dose level are then used to determine an oral LD50 value
(frequently presented with 95% confidence limits). There are usually 5 males and 5 females
per dose group, although the Agency has stated that it will accept studies with fewer animals,
as well as those using only one sex (if that sex is the more sensitive one to the effects of the
test substance). The Agency also accepts studies demonstrating that no mortality occurs
following the administration of that material to 5 males and 5 females at the limit dose of 5
g/kg-
Acute Dermal Toxicity - Rabbit (guideline 81-2). The purpose of the acute dermal
toxicity study is to determine the dermal LD50, its statistical limits, and slope using a single
exposure covering 10% of the total body surface area over a 24-hour period with a 14-day
post-exposure observation period. This study must be performed on all test substances unless
the substance is corrosive, or is a gas or a highly volatile substance that can not be
administered dermally. It is also an initial step in establishing a dosage regimen in
subchronic and other studies and may provide information on dermal absorption and the
mode of toxic action of a substance by the dermal route.
Acute Inhalation Toxicity - Rat (guideline 81-3). The purpose of an acute inhalation
toxicity study is to provide an assessment and evaluation of the toxic characteristics of an
inhalable material, such as a gas, volatile substance, or aerosol/particulate. It also provides
information on possible health hazards via the inhalation route. An acute inhalation toxicity
study determines the median lethal concentration (LC50), its statistical limits and slope using
a single exposure, usually of 4 hours, and a 14-day post-exposure observation period. The
LC50 is expressed as weight or volume of test substance per volume of air (e.g., mg/L or
ppm). It is also an initial step in establishing a dosage regimen in subchronic and other
studies and may provide additional information on the mode of toxic action of a substance.
As an alternative to whole body, oro-nasal (head only exposures) may be used if the
animals exposed in the chambers are excessively coated with test substance.
-------
Primary Eye Irritation - Rabbit (guideline 81-4). The purpose of an acute eye
irritation assay is to determine the effects of ocular exposure to a pesticide. The Subdivision
F Guidelines specify that at least six animals (usually albino rabbits) should be used. An
aliquot (0.1 ml or 100 mg) of the test substance is usually instilled into one eye of each
rabbit; the eyes are examined 1, 24, 48, and 72 hours later, and scored according to the
method of Draize. If there is no evidence of irritation at 72 hours, the study may be ended.
Otherwise, the eyes should be observed and scored until clear or until day 21. If there is
information indicating that the test substance is corrosive (pH <_ 2 or >_ 11.5, or corrosion
or if severe irritation was observed in a dermal application study) then eye testing is not
necessary, as it is assumed the substance would produce a severe effect in the eye.
Primary Dermal Irritation - Rabbit (guideline 81-5). Primary dermal irritation
studies are performed to determine if the test substance is an irritant and/or has any corrosive
effects on the skin of mammals. Information derived from this test serves to indicate the
existence of possible hazards likely to arise from exposure of the skin to test substance. A
small area of skin (approximately 6 cm2) is exposed for a period of 4 hours. Then the test
substance is removed. The degree of irritation is read and scored at specified intervals
(within 30-60 minutes, then at 24, 48, and 72 hours after patch removal) and is further
described to provide a complete evaluation of the effects. The duration of the study should
be sufficient to permit a full evaluation of the reversibility or irreversibility of the effects
observed, but should not exceed 21 days.
Dermal Sensitization - Guinea pig (guideline 81-6). The purpose of a dermal
sensitization study is to assess and evaluate the toxic characteristics of a pesticide which
result from repeated human skin contact under normal conditions of use. Dermal
sensitization is distinct from dermal irritation in that irritation is a direct result of the
chemical characteristics of the chemical at the local site of application. Sensitization is a
more complex immunologically mediated cutaneous reaction to a substance. In general, initial
exposures to a sensitizing substance may result in some local signs of irritation but
subsequent exposures to lower concentrations of the same substance may occur in more
intense dermal responses (and in some cases of cross sensitization to related substances).
There are currently seven types of dermal sensitization studies that may be used to
fulfill this requirement. At present these same methods are also approved by the
Organization for Economic Cooperation and Development (OECD) 1981 guidelines. [In the
1991 draft guideline #406, the OECD indicated that the Guinea Pig Maximization and
Buehler tests are the preferred tests.] There are, however, real differences in opinion among
toxicologists within EPA and OECD regarding the best method to use for each chemical.
The methods currently regarded as acceptable by Health Effects Division are:
- Optimization Test
- Freund's Complete Adjuvant Test
-------
- Guinea Pig Maximization Test
- Split Adjuvant Technique
- Buehler Test
- Open Epicutaneous Test
- Footpad technique
Each of these tests utilizes the guinea pig as the experimental animal. This species
has wide acceptability as the species of choice. Each test has an induction phase in which
the animal is treated repeatedly to activate the immunological sensitization mechanisms. The
techniques of induction are the important differences among these tests. The techniques
range from applying the test substance to the surface of the skin to making intradermal
injections with an adjuvant containing dead bacterial cells known to cause sensitization. The
induction phase is followed by a challenge phase in which the test animal is dosed with an
equal or smaller dose of the test substance. The animal is then evaluated for reactions
occurring within 48 hours (usually). The more intricate details of each of these tests are
described in the Standard Evaluation Procedure for Evaluation of Dermal Sensitization.
It should be noted that if the study is positive, the study is in most circumstances
acceptable and the test substance is considered a sensitizer in that study.
[Note: Data from acute studies serve as a basis for labeling and packaging requirements.]
SUBCHRONIC TESTING
90-Day Feeding Study - Rodent (guideline 82-1 (a)). The purpose of a subchronic
feeding study is to establish a no-observed effect level (NOEL) and to further identify and
characterize the specific organ(s) affected by the test substance after repeated administration
over a period of 90 days. In many cases, this study is conducted as a "preliminary" to the 2-
year chronic rodent toxicity study. Under these conditions, the findings in the subchronic
study are frequently used to justify dietary exposure levels in the chronic study. If a rodent
chronic toxicity study on an active ingredient is acceptable, but the subchronic study is not,
then usually the Agency does not require that the subchronic study be repeated. However, in
the absence of an acceptable chronic study for an active ingredient, a subchronic study is
required under conditions relating to potential human exposure (refer to 40 CFR 158.340).
90-Day Feeding Study - Non-Rodent (guideline 82-l(b)). The subchronic 90-day
feeding study was designed to determine the no-observed-effect level (NOEL) and the toxic
effects associated with continuous or repeated exposure to a substance for a 90-day period.
The study is also useful for determining the dose levels for the chronic non-rodent study. The
non-rodent species in most studies is usually the dog and, in particular, the beagle breed.
-------
21-Day Dermal Toxicity (guideline 82-2). The purpose of the 21-day dermal study is
to provide information on possible health hazards likely to arise from repeated exposures by
the dermal route over a limited period of time. It serves to indicate the need for further
longer term studies. It can also provide information on the selection of dose levels for
longer term studies. A systemic NOEL should be determined from this type of study and is
expressed in terms of the weight of a substance given daily/unit weight of test animal
(mg/kg). The preferred species is the albino rabbit. The test substance is applied to an area
of skin (no less than 10 cm2 surface area) for at least 6 hours a day for a period of 21 days.
90-Day Dermal Toxicity (guideline 82-3). The 90-day dermal study has been
designed to determine the toxic effects associated with continuous or repeated dermal
exposure to a test substance for a 90-day period; however, it is not capable of determining
those effects that have a long latency period. In subchronic toxicity tests, it is desirable to
have a dose-response relationship as well as a systemic NOEL. Test animals are treated with
test substance for at least 6 hours/day on a 7-days-per week basis (5-days-per week is also
acceptable), for 90 days.
90-Day Inhalation Toxicity - Rat (guideline 82-4). The subchronic inhalation study
is designed to determine the NOEL and toxic effects associated with continuous or repeated
exposure to a test substance for a period of 90 days. The test is not capable of determining
those effects that have a long latency period for development (e.g., carcinogenicity and life
shortening). It provides information on health hazards likely to arise from repeated exposure
by the inhalation route over a limited period of time. It should provide information on target
organs, the possibilities of accumulation, and can be used to select dose levels for chronic
studies and to establish safety criteria for human exposure. Hazards of inhaled substances
are influenced by the inherent toxicity and by physical factors such as volatility, particle size,
and exposure.
[Note: Guideline §81-3 and §82-4 were considered together due to the relatively small
number of §82-4 supplementary studies (six). See the section §81-3 for the rejection factors
associated with this guideline.]
CHRONIC TESTING
Chronic Feeding Studies (guidelines 83-1 (a,b) and 83-5). The objective of a
chronic toxicity study is to determine the effects of a substance in a mammalian species
following prolonged and repeated exposure. Under the conditions of this test, effects which
require a long latent period or are cumulative should become apparent. The application of
these standards should generate data on which to identify the majority of chronic effects and
to determine dose-response relationships. Ideally, the design and conduct should allow for
the detection of general toxicity including neurological, physiological, biochemical, and
hematological effects and exposure-related morphological (pathology) effects. Two chronic
studies are required for most food-use pesticides. One should be in non-rodents, preferably
in the dog, and be at least 12 months in duration. The second study should be in the rodent,
8
-------
preferably in the rat and be of at least 24 months duration. The study in rats can be and is
usually combined with one of the oncogenicity studies. Guidance for this study is found
under guideline 83-5.
Oncogenicity Studies (guideline 83-2). The objective of a long-term oncogenicity
study is to observe test animals for a major portion of their life span for the development of
neoplastic lesions during .or after exposure to various doses of a test substance by an
appropriate route of administration. Two rodent studies are required for most food-use
pesticides. Preferred animals are the rat and mouse. The rat oncogenicity study is usually
combined with the chronic study (guideline 83-5) and the mouse oncogenicity study is usually
run alone.
DEVELOPMENTAL AND REPRODUCTION TOXICITY
Developmental Toxicity - Rodents and Non-Rodents (guideline 83-3 a & b). The
objective of a developmental study is to determine the effects of a pesticide when significant
exposure of acute duration occurs to human females. If the product is intended for food use,
two developmental studies are required: one in rodents (rats) and another in non-rodents
(rabbits). The developmental studies are designed to stu4y the adverse effects on the
developing organism which may result from exposure from one or both parents during
prenatal development. The test method is as follows: the test substance is administered in
graduated doses, for at least part of the pregnancy covering the period of organogenesis
(organ development), to several groups of pregnant animals. Shortly before the expected
date of delivery, the pregnant females are sacrificed, the uteri removed, and the contents
examined for embryonic or fetal deaths, and live fetuses. At least 20 pregnant rats or 12
pregnant rabbits are recommended per dose group to ensure that a sufficient number of litters
(pups) are produced to permit a meaningful evaluation. Information gathered from the
developmental studies may be the only information that is available regarding the effects of
in utero exposure to the pesticide or it may be supplemented by the results of the
reproduction study.
A newly revised SEP for developmental toxicity studies has been drafted, and public
comments will be sought within the next 3-6 months.
Reproduction (guideline 83-4). The objective of a two-generation reproduction study
is to determine the effects that are likely to result from human exposure to a pesticide over a
portion of the human lifespan, which is significant in terms of the frequency, magnitude,
and/or the duration. The reproduction study is designed to provide general information
concerning the effects of a test substance on gonadal function, estrus cycles, mating
behavior, conception, parturition, lactation, weaning, and the growth and development of the
offspring. The study may also provide information about the effects of the test substance on
neonatal morbidity, mortality, and preliminary data on teratogenesis. The rat is the preferred
species for this study. The test method is as follows: the test substance is administered to
the parental (P) animals (males and females) for at least 8 weeks prior to mating, through
-------
mating, during the resulting pregnancies, and through the weaning of their first generation
(Fj) offspring. The test substance is then administered to selected F, offspring during their
growth into adulthood, mating, and production of a second (F2) generation, up until the F2
generation is 21 days old (weaned). At least 20 male rats and a sufficient number of females
to yield at least 20 pregnant females at or near term are recommended per dose group to
ensure that sufficient number of litters (pups) are produced to permit a meaningful
evaluation.
A SEP for Reproductive toxicity studies has been drafted and the Agency hopes to
release the document for public comment within the next 12 months.
MUTAGENICITY TESTING
Mutagenicity (guideline 84-2 & 84-4). On March 1, 1991 new mutagenicity
guidelines were proposed by the Agency. While the new Subdivision F Mutagenicity
Guidelines are now finalized, the revision to 40 CFR Part 158 (data requirements for
registration) that provides new Mutagenicity Guidelines with legal standing are not finalized
at this time. This means that the old guideline has legal basis for registration purposes. If a
registration is currently underway (e.g., registration package is in the Agency), then the old
guideline applies, even if 40 CFR Part 158 is changed while the mutagenicity test are being
reviewed. If a registrant is currently planning to perform mutagenicity testing, the new
guidelines should be used.
The purpose of the mutagenicity testing is to assess the potential to affect the
qualitative or quantitative integrity of the mammalian cell's genetic components. The assays
are selected: 1) to detect, with sensitive assay methods, the capacity of a chemical to alter
genetic material in cells; 2) to determine the relevance of these mutagenic changes to
mammals, and when mutagenic potential is demonstrated; 3) to incorporate these findings in
the assessment of heritable effects, carcinogenicity and possibly, other health endpoints.
Chemicals are often non-mutagenic unless converted to an active mutagen by
metabolic processing, but the reverse can also occur. Therefore, a metabolic activation
system must be incorporated into any test system other than intact mammals and insects.
The test substance must be tested both in the presence and the absence of mammalian
tissue extracts (with appropriate cofactors) which have been demonstrated to convert a wide
range of chemical "promutagens" (substances which are mutagenically-inactive in the absence
of the tissue extracts) to mutagenically-active substances. Rat liver extracts are preferred.
The tissue must be pre-induced for the relevant enzymatic activities when appropriate. The
inducer must be effective for the class of compounds under test. Other tissue extracts must
be used in addition to liver extracts when the principal site of metabolism of the test
substance is known not to be in the liver; or when other tissues, including plant tissue, are
known to give positive results in chemicals which are structurally-related.
10
-------
The test substance may also be exposed to metabolic processing in intact mammals by
a host-mediated system in which the target cells are inserted into host tissues or body
cavities. Hepatocytes may also be used to provide metabolic processing, either as a co-
culture with a target cell, or as the primary assay system.
All assays must be run with concurrent positive and negative controls with the
possible exception of the mouse specific locus test. Positive control compounds must be
selected to demonstrate both the sensitivity of the indicator organism and the functioning of
the metabolic activation system. Positive controls must also be selected to demonstrate the
sensitivity of the indicator cells or organisms to a compound with chemical characteristics
similar to those of the test substance. For instance, an alkylating agent must be used as a
control for an expected alkylator, and an intercalating agent for a suspected intercalator.
When applicable, the positive control must be administered by the same route as the test
substance.
Both a solvent and where applicable, a non-solvent negative control, should be
included.
Current guidelines require one test in each of three categories (Subdivision F, pages
147-151). The selection of tests to satisfy each category was essentially left up to the
registrant unless the OPP had requested a specific test for a category. All tests were listed in
Subdivision F under Series §84-2. There were two mutagenicity assays that were
inappropriately categorized in the Subdivision F Guidelines. These are noted in the
following list:
A) Gene mutations
Microorganisms:
a) Salmonella typhimurium reverse mutation assay at the
histidine locus (Ames assay)
b) Escherichia coli. WP2 and WP2 uvrA
c) Bacillus subtilisTKJ 5211, TKJ 6321
Submammalian organisms:
a) Drosophila melanogaster
Mammalian cells forward or reverse mutations (positive findings usually involve point
mutations, but may be due to chromosomal deletions and/or chromosomal
rearrangements).
11
-------
b) Chinese hamster lung (V79 cell line) (hgprt locus)
c) Chinese hamster ovary (CHO cell line) (hgprt or xprt
locus)
d) Mouse Lymphoma (L5178Y cell line) (tk locus)
Specific locus:
a) Mouse
B) Structural chromosome aberrations
Eucaryotic microorganisms:
a) Aspergillus nidulans
b) Neurospora crassa
Submammalian organisms:
a) Drosophila melanogaster
Mammalian cells in culture:
a) Sister chromatid exchange (Note: this assay is more
appropriately classified under Other genotoxic effects.)
b) Cytogenetic analysis
Mammals:
a) Micronucleus test
b) Sister chromatid exchange (SCE)
c) Cytogenetic analysis
d) Dominant lethal; mouse or rat
e) Heritable translocation; mouse
12
-------
C) Other genotoxic effects (§84-4 a typographical error in the data requirements;
it should be listed under §84-2. Categories "A" & "B" do not exist either.)
DNA damage and repair:
a) Differential toxicity in bacteria:
Escherichia coli pol A+/pol A-
Bacillus subtilis H17/M45
b) Mitotic recombination in eucaryotic organisms
Saccharomyces cerevisiae
Aspergillus nidulans
c) Unscheduled DNA synthesis:
Rat hepatocytes in culture
d) DNA alkaline elution
e) Sister chromatid exchange
Numerical chromosomal aberrations
a) Mitotic segregation in yeast
b) Micronucleus test (this assay is more appropriately classified as a
structural chromosomal aberration test).
Mammalian cell transformation
a) Cells in culture
Target organ/cell analysis
a) Sperm morphology
b) DNA synthesis inhibition
c) DNA alkylation
13
-------
The proposed 1993 guidelines require three assays, restricted to the following, which
the Agency believes would provide the greatest amount of useful information to give an
initial assessment of mutagenic potential:
1) Salmonella typhimurium reverse mutation assay at the histidine locus (Ames
assay).
2) Mammalian cells in culture forward gene mutation assay (positive responses
would usually involve point mutations, but could be due to chromosomal
deletions and/or chromosomal rearrangements).
a) Mouse lymphoma L5178Y cells, thymidine kinase gene locus (the preferred
assay); or,
b) Chinese hamster ovary (CHO) or Chinese hamster lung fibroblast (V79)
cells, hypoxanthine-guanine phosphoribosyl transferase (HGPRT) gene locus;
or,
c) Chinese hamster ovary (CHO) cells strain AS52, xanthine-guanine
phosphoribosyl transferase (XPRT) gene locus.
3) In vivo cytogenetics (usually rat bone marrow):
a) Metaphase analysis (aberrations); or,
b) Micronucleus assay.
GENERAL METABOLISM
General Metabolism (guideline 85-1). Data on biotransformation of pesticides are
required to support the registration of each manufacturing use product and each end use
product when the following conditions apply as outlined in 40 CFR §158.135: 1) Chronic
toxicity or carcinogenicity data are required for registration of the pesticide, or 2) positive
evidence of mutagenicity is obtained from in vivo or in vitro studies with the pesticide.
Testing of the disposition of a pesticide is designed to obtain adequate information on
the absorption, distribution, biotransformation, and excretion as a function of dose. Basic
pharmacokinetic parameters determined from these studies will also provide information on
the potential for accumulation of the pesticide in tissues and/or organs and the potential for
induction of biotransformation as a result of exposure to the pesticide. These data can be
used in conjunction with chronic toxicity and/or oncogenicity studies of the pesticide to aid in
the determination of the mechanism(s) of toxicity and/or carcinogenicity if such effects are
observed.
14
-------
Data from biotransformation studies are also used 1) to assess the potential toxicity of
plant and/or animal metabolites to mammalian organisms, and 2) if toxicity is found, in the
setting of appropriate tolerance levels for those metabolites in raw agricultural commodities.
SPECIAL TESTING
Domestic Animal Safety (guideline 86-1). The domestic animal safety tests are
generally conducted only when cats, dogs, cattle, pigs, sheep, or other domestic animals will
be exposed to a given pesticide through direct application for pest control and/or
consumption of treated feed. These studies are required on a case-by-case basis.
Dermal Penetration (guideline 85-2). Dermal absorption studies are required on an
individual basis for compounds having a serious toxic effect, identified by oral or inhalation
studies, for which a significant route of human exposure is dermal and for which the
assumption of 100% absorption does not produce an adequate margin of safety.
The determination of dermal absorption in relation to dose and time of exposure allow
the calculation of margin of safety or risk for systemic toxic effects, which have not been or
cannot be practically tested by the dermal route.
The dermal absorption study is not a toxicology study because it tells one nothing
about the potential for harm inherent to a particular pesticide. Instead it is a type of kinetic
study indicating the amount of pesticide passing through the skin from a specified exposure.
It is useful only as part of a risk assessment procedure.
Delayed Neurotoxicity (acute and 28-day). In the assessment of organophosphorus
substances, studies of delayed neurotoxicity using the adult hen as the test animal and
including behavioral observation of gait, histopathological assessment of the brain, peripheral
nerve, and spinal cord, and neurochemical assessment of inhibition of acetylcholinesterase
(AchE) and neurotoxic esterase (NTE) are needed to identify and characterize these potential
effects.
Flow Chart 1 - Food Uses
The purpose of this section is to explain the sequencing and interdependence behind
the toxicology studies and regulatory outcomes. The guidelines can be broken down into
eight testing groups: (1) acute, (2) subchronic, (3) chronic, (4) developmental/reproduction,
(5) mutagenicity, (6) metabolism, (7) special testing, and (8) neurotoxicity. The data
requirements for food uses will be described here (see Chart 1). Similar logic can be applied
to the non-food use patterns. Chart 2 describes the data requirements for the antimicrobials.
The initial step in the evaluation of the toxic characteristics of a pesticide is to
determine its acute toxicity. As depicted in Chart 1, these studies are as follows: acute oral
(81-1), acute dermal (81-2), acute inhalation (81-3), primary eye irritation (81-4), primary
15
-------
dermal irritation (81-5), and dermal sensitization (81-6). All of the acute studies are required
but individual studies within the acute battery can be waived based on certain physical and
chemical properties of the test substance. These studies provide information on health
hazards likely to arise soon after, and as a result of, a single exposure. Data from the acute
studies (i.e., LD50, LC50, eye effects, and skin effects) are used to classify the pesticides in
Toxicity Categories I - IV (see Chart 1 - Hazard assessment); and to provide a basis for
precautionary labeling (i.e., reentry intervals for farmworkers, protective clothing for
applicators). These data also provide information for establishing appropriate dose levels in
the subchronic and other studies, provide initial information on the mode of toxic action(s) of
a substance, and determine the need for child resistant packaging.
The next sequence of testing, subchronic, provides information on health hazards that
may arise from repeated exposure to a chemical over a limited period of time. These studies
are : 90-day feeding (82-1), 21-day dermal (82-2), 90-day dermal (82-3), and 90-day
inhalation (82-4). These tests are required only under the conditions shown in Chart 1. The
data from the subchronic studies provide information on target organs and accumulation
potential. The data are also useful in selecting the maximum tolerated dose levels for
chronic studies and for establishing safety criteria for human exposure.
The chronic toxicity studies are intended to determine the effects of a substance in a
mammalian species following prolonged and repeated exposure. Each of the following
chronic tests are required: chronic feeding-rodents (83-1 (a)) and chronic feeding - non-
rodents (83-l(b)). Effects which have a long latency period or are cumulative should be
detected from these studies. The no-observed-effect-level (NOEL) and lowest-effect-level
(LEL) are determined from this data. The oncogenicity studies in rats (83-2(a)) and in mice
(83-2(b)) allow the test animals to be observed over most of their life span for the
development of neoplastic lesions during or after exposure to various dosages of a test
substance.
The developmental testing sequence is designed to determine the potential of the test
substance to induce structural and/or other abnormalities to the fetus as a result of exposure
to the mother during pregnancy. The required testing includes developmental - rat (83-3(a))
and developmental - rabbit (83-3(b)). The two-generation reproduction test (83-4) is
designed to provide information concerning the general effects of a test substance on gonadal
function, estrus cycles, mating behavior, conception, parturition, lactation, weaning, and the
growth and development of the offspring. The data generated from the developmental/
reproduction studies is the NOEL, LEL, developmental toxicity potential, and the margins of
exposure for dietary and non-dietary exposure.
Gene mutation (§84-2(a)), structural chromosomal aberration (§84-2(b)), and other
genotoxic effects (formerly §84-4) are required to assess a substance's potential to affect
mammalian cells genetic components. The data generated includes: (1) gene mutation, (2)
chromosome changes, (3) an assessment of the mechanism of carcinogenicity, (4) possible
16
-------
heritable effects, and (5) additional information for the weight-of-evidence when combined
with carcinogenicity.
The group of studies listed under Special Testing includes: General Metabolism (§85-
1), Domestic Animal Safety (§86-1), and Dermal Penetration (§85-2). The purpose of a
metabolism study is to produce data that identifies potentially toxic metabolites, to show the
tissue distribution of metabolites, and to allow for a comparison of plant vs. animal
metabolites. Dermal penetration studies are required conditionally and give an indication of
what percentage of the chemical is dermally absorbed. Domestic animal safety is only
required on a case-by-case basis.
On March 15, 1991, the Agency provided new guidance for neurotoxicity testing.
Therefore, neurotoxicity was not included in the rejection rate analysis. However, an outline
of the individual studies has been provided in Chart 1 for reference. The acute neurotoxicity
study (81-7) provides a basic assessment of neurotoxicity functions. Guideline (81-8),
neurotoxicity screening battery, generates an assessment of neurobehavioral functions. The
28-day neurotoxicity study (82-6) provides an assessment of neurobehavioral functions and an
estimate of the reference dose. Data generated from the 90-day neurotoxicity study (82-7)
are used to assess neurotoxicological functions. Data from the post-natal development
toxicity study (83-6) are used to assess neurobehavioral functions in developing organisms.
Finally, 83-l(c), neurotoxicity screening battery provides additional data on the neurotoxicity
associated with chronic exposures.
Each of the studies in the subchronic, developmental, reproduction, chronic,
carcinogenicity, and neurotoxicity categories are ultimately reviewed by the RfD Peer
Review Committee which consists of a panel of Agency scientists. The RfD Committee
determines the adequacy of the studies, the data evaluation records (DERs), and the cancer
assessment or developmental/reproductive toxicity assessment. The RfD Committee also
establishes the reference dose and can classify a chemical as a D or E carcinogen, if
appropriate.
After review by the RfD Committee, the analysis may follow one of two paths
depending on the trigger (i.e., cancer or developmental/reproductive toxicity). If there is
evidence of cancer, the chemical is sent to the HED Cancer Peer Review Committee where it
can be classified as an A, B, C, or D carcinogen. If the chemical is classified as an A, B, or
C quantifiable carcinogen, then the Committee will determine the cancer potency or Ql*. At
this point, the chemical goes through two more analyses. A Dietary Risk Evaluation System
(DRES) analysis is run to determine the risk for certain populations from each commodity.
A worker exposure analysis is performed also which will determine the need for label
restrictions and to estimate cancer and worker risk such that individual risk is less than 1 x
to 1 x 10-6.
If there is evidence that the chemical has the potential to cause developmental/
17
-------
reproductive toxicity, a worker exposure analysis is done to calculate the worker margin-of-
exposure (MOE). A DRES analysis is also conducted when there is evidence of
developmental/reproductive toxicity. Here, the acute dietary MOE is determined. If the
acute dietary MOE is less than 100, then the Agency may cancel certain uses of the pesticide
or implement use restrictions such that the MOE is greater than 100.
In the case where a chemical shows no evidence of cancer or
developmental/reproductive toxicity, a DRES analysis is performed. A tolerance is
established so that the risk of total dietary exposure does not exceed 100% of the RfD.
Flow Chart 2 - Antimicrobials
The antimicrobial testing depicted in Chart 2 is a tiered data requirement system that
gives the registrant of an antimicrobial pesticide the option of generating a complete
toxicology data base or developing a basic first tier of acute, mutagenicity, and
developmental studies (one species), subchronic dermal studies, and also exposure
information. Based on the results of the toxicity tests and the exposure analysis, further
studies, up to and including long-term testing, may be required.
During normal conditions of use, humans may be exposed to these antimicrobial
pesticides orally, dermally, through the eye, and by inhalation. Based on the use patterns
and chemical/physical properties of the chemical, the exposure generally falls into one of the
following categories based on the duration, frequency and the magnitude of the exposure:
Category 1 High exposure - high and frequent (weekly or more)
Category 2 Medium exposure - high but infrequent (weekly or less) or low
but frequent
Category 3 Low exposure - low and infrequent
The reasoning behind this tiered approach is that it is less expensive and the
development of exposure information uses fewer laboratory animals than long-term
toxicology testing.
The following is a discussion of the tiered approach for antimicrobial testing
requirements. The acute studies (guidelines 81-1 through 81-6) are required for all
antimicrobial use pesticides. These tests follow the exact same format as the acute studies
for food use pesticides (see previous discussion). Each registrant must submit the first-tier
toxicity studies on each active ingredient (a.i.). The registrant must also agree to submit data
from second-tier and/or third-tier toxicity studies to the extent that the results from the first-
tier studies indicate a need for data from higher-tier studies.
18
-------
The first-tier studies and the conditions under which they are required are as
follows:
(1) Mutagenicity (84-2. 84-4). These studies are required of all active ingredients.
If the overall results of the mutagenicity tests suggest that the a.i. may pose
carcinogenic risk, then the oncogenicity studies (83-2) are required. (See tier
3).
(2) 90-Day Dermal (82-3). This study is required for all active ingredients except
for certain dermally corrosive antimicrobials.
(3) 90-Dav Inhalation (82-4). This study is required only if the a.i. is a gas at
room temperature or if the use of the product results in respirable droplets (15
microns of less in diameter) and the use results in repeated inhalation
exposure.
(4) Developmental - first species: rat or rabbit (83-3). This test is required of all
active ingredients. High exposure may trigger the need for the second study.
The second-tier studies and the conditions under which they are required are as
follows:
(1) 90-Day Feeding (82-1). This study is required if the NOEL from the 90-day
dermal toxicity study (82-3) or the dermal exposure equivalent level calculated
from the 90-day oral study is less than 1000 times higher than the human
dermal exposure to the active ingredient. This test is also required if the
NOEL from the 90-day inhalation toxicity study (82-4) is less than 1000 times
higher than the human inhalation exposure to the active ingredient. The
findings from the subchronic toxicity study will be used to establish the dose
levels for the chronic feeding and oncogenicity studies.
(2) Developmental - second species (83-3). This study is required if the data from
the first-species developmental study suggest embryo or developmental
toxicity.
(3) Dermal Absorption (85-2). If the results of the exposure study and/or other
data show an unacceptable developmental, reproductive, or carcinogenic risk,
then a dermal absorption study is required. For developmental and
reproductive risks, the evaluation is based on the effects seen in the first
developmental study. For carcinogenic risk, the exposure data are used to
compare the a.i. with ethylene dibromide (EDB). EDB is used for comparison
as a worst-case example.
19
-------
The third-tier studies and the conditions under which they are required are:
(1) Oncogenicity - rat and mouse (83-2). These studies are required if the overall
results of the mutagenicity test battery suggest that the a.i. may pose an
carcinogenic risk and/or, if based on dermal exposure (as shown by lower tier
studies or other data and not negated by the actual dermal absorption study)
that the calculated risk exceeds that of ethylene dibromide (EDB).
(2) Chronic Feeding - rat and dog (83-IV These studies are required if the MOE
is less than 1000 based on subchronic data and if exposure cannot be reduced.
(3) Metabolism (85-IV This study is required only if additional information on
metabolism of the chemical is necessary to clarify unusual effects observed in
chronic or reproduction studies or to clarify issues concerning structure
activity relationships.
(4) Reproduction (83-4). This study is required if developmental toxicity and/or
adverse effects on reproductive organs were observed in the 90-day dermal or
inhalation study.
20
-------
Summary
From a rejection rate perspective, the most critical toxicology guideline requirements
are the CORT studies (Chronic, £)ncogenicity (i.e., carcinogenicity), Reproduction, and
Teratogenic) since they can take up to four years to complete and any rejected studies would
most likely cause a delay in the RED. Secondly, the decision not to accept a study does not
receive nearly the amount of attention as do accepted studies which go through an extensive
peer review process.
21
-------
V. CURRENT RETECTION RATE
The following graphs demonstrate the current and historical rejection rates and
supplementary rates for each of the toxicology guidelines. None of the results reported in
this section have been tested for statistical significance, and therefore caution should be
exercised in their interpretation. The purpose here is not to develop an empirically
defensible rejection rate or supplementary rate value. Rather, the intent is to use rejection
rates and supplementary rates as the best indicators available of where additional
Agency/registrant attention and efforts are warranted to improve the quality of the studies.
Further, caution is warranted in using the rejection and supplementary rates given the
interpretation that is required to determine whether the study is accepted, upgradable, or
rejected. For the mutagenicity and metabolism studies, the review results in either an
acceptable or unacceptable rating, although not all unacceptable studies have to be repeated.
For the other guidelines, the review can result in one of the following ratings: (1) core
guideline, (2) core minimum, (3) core supplementary, or (4) invalid. The core guideline
rating indicates that everything is acceptable. The core minimum rating indicates that while
some things are missing, the study still fulfills the guideline requirements. The core
supplementary rating indicates a significant deficiency or that additional information is
required and that the study may be upgradable or may be fatal (must be repeated). The
invalid rating indicates that the study has guideline deficiencies such that the study is
irrevocably damaged. Judgement has been exercised in categorizing studies in this analysis.
Studies rated as core supplementary have been categorized in this analysis as either rejected
(i.e., fatal) or supplementary (i.e., upgradable).
Figure 1 illustrates the overall rejection rate and supplementary rate for toxicology.
The current fatal rejection rate (post 1988) is approximately 7%, and has remained relatively
constant across the three time periods. The time frames include studies submitted in (a)
1985, (b) 1986-1988, and (c) post 1988 (through 6/92). The current (post 1988)
supplementary (i.e., upgradable) rate is 12% and has risen in each of the three time periods.
While the rejection rates for toxicology are relatively low and constant across the three time
periods, the substantial number of studies rated supplementary and the rising supplementary
rate over time warrants concern.
Figures 2 through 5 illustrate what the Agency believes is the current (post 1988)
rejection rate for each guideline. The current rate is defined as all studies that have been
submitted, reviewed by EPA and rejected since 1988. The rejection rate is given at the top
of each guideline bar, and the number of rejected studies over the number of studies
reviewed are listed inside each bar. Some toxicology guidelines with an insufficient number
of studies were omitted. These guidelines include Acute Delayed Neurotoxicity [81-7],
90-Day Neurotoxicity [82-5(a) and 82-5(b)], and Domestic Animal Safety [86-1].
Figure 6 illustrates those guidelines whose rejection rates appear to have dropped
significantly over the three time frames. It is encouraging to note that all of the CORT
22
-------
(Qhronic, Qncogenic, Reproductive, Teratogenic) guidelines have shown a drop in their
rejection rates since most of these are four-year studies and would likely delay the RED date
(Reregistration Eligibility Determination) if they had to be repeated. Substantial decreases in
rejection rates have also occurred in the Mutagenicity studies [84-2(a), 84-2 (b), and 84-4]
and the General Metabolism study [85-1].
Figure 7 illustrates that the Metabolism study's [85-2] current rejection rate is 15%
which has increased over the previous period's rate of 13%. The Subchronic 90-day Feeding
- Rat study's [82-l(a)] current rejection rate of 23% has increased over the previous period's
rate of 15%. The Acute and Subchronic Inhalation studies [81-3 and 82-4] current rejection
rate of 19% represents a substantial increase over the previous period's rate of 7%. Finally,
the Dermal Sensitization study [81-6] has shown an increase from less than 1% to 10% (post
1988).
Figures 8 through 26 characterize each guideline's rejection rates and supplementary
rates over time. The Agency reviewed the amount of time that had elapsed in attempting to
upgrade supplementary studies and found, in some cases, that it took five, six, or even seven
years, which is significantly more time than is required to repeat the study. The long
elapsed times of four years or more occur for studies submitted between 1985 and 1988 or
before the initiation of FIFRA 88. While the long lapse times may accurately reflect past
Agency/registrant performance, there is good reason to believe that the results do not
accurately reflect current performance. First, FIFRA 88 desk top computers have been
provided to all chemical review managers (CRMs) and product managers (PMs) and data
tracking systems have been developed and implemented. Prior to FIFRA 88, only a limited
access main-frame tracking system was available. Thus, the Agency's capability to track the
timeliness of responses and reviews has been greatly enhanced. Second, prior practice did
not require imposing time limits on registrant response for supplementary data. In
reregistration, all supplementary data requests have a time limit imposed. Third, in
reregistration all supplementary-data-request correspondence is sent as certified mail to
ensure that the registrant receives it.
In order to determine whether a systematic problem might be causing these delays,
the Agency conducted a qualitative analysis of the CORT studies as a subset of the
toxicology studies. Studies that had been outstanding for 3 or more years were identified and
individually investigated to define their current status. Approximately one-third of these
studies had actually been upgraded and accepted by the Agency but were not recorded in the
HED's internal tracking system (not to be confused with the reregistration tracking systems).
Another one-third were studies for chemicals that had been cancelled, withdrawn, or
unsupported but their status had not been recorded. In some cases, the food uses were
cancelled, and therefore the study was no longer required. Approximately 35% of these
studies are pending Agency review. Several reasons were revealed as to why these reviews
are still pending. In some cases the registrants response was reviewed, but all of the
deficiencies were not adequately addressed. This required another Agency letter and
registrant response. This may go back and forth several times thus creating the "the ping-
23
-------
pong syndrome." In some cases, the registrants were submitting data that were not required
to satisfy a guideline requirement. The Agency considers these nonessential data and does
not give it a high priority for review. In other cases, a delay in review of a study resulted
from internal Agency reorganization and shifting of priorities.
Summary
Key implications that might be drawn from these graphs include:
(1) Overall rejection rates in toxicology are low and have remained relatively
constant.
(2) A substantial number of toxicology studies have been rated supplementary
(i.e., upgradable), and the supplementary rate appears to be rising.
(3) The time elapsed in upgrading studies has frequently exceeded the amount of
time that would have been required to repeat the studies; however, the Agency
has good reason to believe that these substantial time lapses more accurately
reflect past Agency/registrant performance than current performance.
(4) The CORT studies have shown substantial declines in their rejection rates.
(5) The Mutagenicity studies and the General Metabolism study have also shown
decreasing rejection rates.
(6) Dermal Penetration, Subchronic 90-day Feeding -rat, Acute and Subchronic
Inhalation and Acute Dermal Sensitization studies have all shown substantial
increases in their rejection rates over time.
(7) None of the implications discussed above are based on statistically significant
results, and therefore caution should be exercised in interpreting them.
24
-------
20 -i
• ___
CD
215
Q)
a.
coio H
DC
O 5
0
"5T
a:
0
Figure 1
Rejection and Supplementary Rates
For All Toxicology Guideline Requirements
6%
Rejected Studies
Supplementary Studies
12%
11%
1985
1986-1988 Post 1988
Year Study Submitted
* # rejected studies/# studies reviewed
Note: Studies reviewed include studies from List A,B,C, & D as well as from chemicals registered after 1984.
-------
c
-------
^ 10° ~
4^
CD
O 80 -
CD
CD
•M
CO
oc
o
CD
"5T
DC
60 -
40 -
20 -
0
Figure 3
Current (Post-1988) Rejection Rates
By Guideline - Subchronic Toxicity
100%
82-1 (a) Subchronic 90-day Feeding - Rat
82-1{b) Subchronic 90-day Feeding - Dog
82-2 Subchronic 21-day Dermal - Rat/Rabbit
82-3 Subchronic 90-day Dermal - Rodent
82-4 Subchronic 90-day Inhalation - Rat
82-5(a) Subchronic 90-day Neurotoxicity - Hen
82-5(b) Subchronic 90-day Neurotoxicity - Mammal
23%
4%
0/16^
82-3
82-1 (a) 82-2 82-1 (b)
Guideline Number
Rejected Studies
* # studies rejected/total # studies reviewed
Note: Insufficient data to evaluate 82-5{a) and 82-5{b); 82-4 combined with 81-3 in Figure 2.
-------
^_J
f~
-------
20 -i
0
O
0
Q-
0
15 -
CO
cc
c
O
10 -
O
0
"cF
DC
5 -
0
Figure 5
Current (Post-1988) Rejection Rates
By Guideline - Other Toxicology Studies
15%
85-2
12%
84-2(a) Gene Mutation - Ames
84-2(b) Structured Chromosome Aberration
84-4 Other Genetic Effects
85-1 General Metabolism
85-2 Dermal Penetration
86-1 Domestic Animal Safety
2%
/*&'•;'"M
JT.{^» i.*A
2/107*
84-2(a) 85-1
Guideline Number
7 X
Rejected Studies
* # studies rejected/total # studies reviewed
Note: Insufficient data to evaluate 86-1
-------
Figure 6
Toxicology Guidelines with Lower
Rejection Rates over Time
50 -i
0 40
O
-------
30 -,
C 25
0)
O
20 -
0
03 15
cc
C
.2 10
o
0
'oT
DC
5 -
0
Figure 7
Toxicology Guidelines with Increasing
Rejection Rates Over Time
8/60
85-2
8/35 <
45/234'
1985
1986-1988
/\ Post-1988
23/2361
27/372
2/405*
2/154*
82-1 (a) 81-3
82-4
Guideline Numoer
81-6
# rejected studies/# studies reviewed
-------
10 -H
Figure 8
Guideline 81-1
Acute Oral Toxicity
8/277* 121/277*
18/428* II22/428*
1985 1986-1988 Post-1988
Year Study Submitted
Rejected Studies 0j Supplementary Studies
* # rejected studies/^ studies reviewed
-------
0)
o
CD
CL
0)
03
o:
c
10 H
8
6 -
4 -
o
0
'5T
cc
0
Figure 9
Guideline 81-2
Acute Dermal Toxicity
Rabbit & Rat
1985
1986-1988 Post-1988
Year Study Submitted
7/j: Rejected Studies
* # rejected studies/* studies reviewed
Supplementary Studies
-------
20 -i
-------
0
O
CD
Q.
CD
+-•
CO
o:
o
0)
"5T
DC
10 -
Q _
6 -
4 -
Figure 11
Guideline 81-4
Acute Eye Irritation
Rabbit
1985 1986-1988 Post-1988
Year Study Submitted
K/j Rejected Studies
* # rejected studies/^ studies reviewed
Supplementary Studies
-------
0
o
CD
Q.
CD
CO
DC
c
o
10 -i
8 -
6 -
4 -
o
CD 2
'5T
o:
Figure 1Z
Guideline 81-5
Acute Primary Dermal Irritation
3/184 *
3/370'
1985 1986-1988 Post-1988
Year Study Submitted
Rejected Studies H Supplementary Studies
* # rejected studies/I studies reviewed
-------
14 -
0) 12 H
(2 10 -
i ^
GC
c 6H
o 4H
0
DC
2 -
0
2/154*
Figure 13
Guideline 81-6
Dermal Sensitization
2/154*
4%
2/405
14%
1985 1986-1988 Post-1988
Year Study Submitted
K/j Rejected Studies
# rejected studies/* studies reviewed
Supplementary Studies
-------
25 -i
Figure 14
Guideline 82-1 (a)
Subchronic 90-day Feeding Study
1985 1986-1988 Post-1988
Year Study Submitted
H Rejected Studies B Supplementary Studies
* # rejected studies/^ studies reviewed
-------
40
c
-------
20 -i
0
CL
0
to 10 H
cc
o
0
"5T
CC
5 -
0
Figure 16
Guideline 82-2
Subchronic 21-day Dermal
Rat/Rabbit
13%
0/30 *
1985
15%
1986-1988 Post-1988
Year Study Submitted
Rejected Studies
# rejected studies/,? studies reviewed
Supplementary Studies
-------
0)
o
CD
0-
0
•M
CO
DC
o
CD
'5?
DC
100 -,
80 -
60
40 -
20 -
0
Figure 17
Guideline 82-3
Subchronic 90-day Dermal
Rodent
1985
75%
100%
1986-1988
Year Study Submitted
Post-1988
K/j Rejected Studies
* # rejected studies/# studies reviewed
Supplementary Studies
-------
30 -i
§ 25-
&_
0)
0- 20 -
0
to 15 -
CC
10 H
o
CD 5 _j
••~^ ^
CD
DC
0
Figure 18
Guideline 83-1 (b)
Chronic Toxicity
Dog
23%
27%
£/j Rejected Studies
# rejected studies/^ studies reviewed
1986-1988 Post-1988
Year Study Submitted
Supplementary Studies
-------
30 n
Figure 19
Guideline 83-3(a)
Developmental Toxicity
1985 1986-1988 Post-1988
Year Study Submitted
Rejected Studies B Supplementary Studies
* # rejected studies/^ studies reviewed
-------
30 -i
-------
25 -i
O 20
u.
0
Q.
CD
CD
o:
O
0
'5T
DC
15 -
10 -
5 -
0
Figure 21
Guideline 83-4
Two-Generation Reproduction
21%
1985
Rat
14%
12%
'.KVSt •.,• /
2%
7/59^
1986-1988 Post-1988
Year Study Submitted
¥/\ Rejected Studies
# rejected studies/# studies reviewed
Supplementary Studies
-------
30 -i
0 25
O
0
CL 20
0
CD 15
oc
o 10
O
0 5 -|
•— ' •* ^
0
a:
o
Figure 22
Guideline 83-5
[83-1 (a) and 83-2(b)j
Combined Chronic/Onco
Mouse
26%
22%
5%
//
777 X
22%
J&'ff r/''" ^'.
1985 1986-1988 Post-1988
Year Study Submitted
K/j Rejected Studies
* # rejected studies/^ studies reviewed
Supplementary Studies
-------
Figure 23
25 -i
Guideline 83-5
[83-1(a) and 83-2(a)]
Combined Chronic/Onco
Rat
1985 1986-1988 Post-1988
Year Study Submitted
K/j Rejected Studies || Supplementary Studies
# rejected studies/^ studies reviewed
-------
20 -i
C
0)
2 15
CD
0.
CO 10
or
c
o
S 5
0
0
Figure 24
Guidelines 84-2(a), 84-2(b), 84-4
Mutagenicity
1985
16%
13%
1986-1988 Post-1988
Year Study Submitted
Rejected Studies jjj Supplementary Studies
# rejected studies/* studies reviewed
-------
0)
O
30 -i
25 -
20 -
0
to 15
a:
c
1°
o
0)
"55*
DC
0
Figure 25
Guideline 85-1 |
General Metabolism I
1985
24%
1986-1988 Post-1988
Year Study Submitted
Y/\ Rejected Studies
\..sf....'. *
* # rejected studies/# studies reviewed
Supplementary Studies
-------
25 -i
§ 20
CD
a.
-------
VL REJECTION FACTORS
The following toxicology guidelines were analyzed to determine the most common
reasons that studies submitted to meet these guidelines were rejected (i.e., factors that caused
studies to be rated invalid as well as supplementary). EPA scientists listed these rejection
factors below in rank order according to the frequency of their occurrence. The number in
parenthesis represents the number of times that factor was observed.
After each rejection factor, specific references to EPA guidance addressing that factor
are given. The EPA guidance is analyzed and is referred to in this report, to determine if
the guidance documents available to registrants adequately cover the areas where problems
have occurred. (A list of all guidance documents available for toxicology studies is provided
in Appendix A, at the end of this document).
After each rejection factor and the corresponding references to EPA guidance, an
Industry Comment section has been provided with industry scientists' (a) assessment of the
adequacy of the EPA guidance, (b) explanation of technical difficulties, if any, associated
with the rejection factor, and (c) recommendations. Following each Industry Comment
section is EPA's response to that comment. In the future, registrants could face regulatory
action should studies be rejected because of "avoidable" factors.
INDUSTRY OVERVIEW AND RECOMMENDATIONS
In preparing their comments on the analysis, the Industry identified four basic factors
which they believe contribute to rejection and feel the Agency should address, these are: (1)
lack of clarity or shifting guidelines, (2) the need for a no-observed-effect level (NOEL) in
certain studies, (3) reviewer inconsistency, and (4) the Agency use of the maximum tolerated
dose (MTD).
Guideline Clarity
Studies such as dermal penetration, acute inhalation, dermal sensitization studies and
mutagenicity, have either shown substantial increases in their rejection rates or have high
rejection rates. We believe this is due to a combination of frequent changes in guidance,
lack of clarity or use of standards which are difficult to meet. This is a primary source of
frustration which is reflected in our comments under the specific rejection factors.
NOEL
The rejection rate for 90-day feeding studies in rodents has also increased. The main
rejection factor cited is that a NOEL was not established. This was also a factor in several
other types of studies. Latitude should be granted by the Agency if the biological response
found at the lowest dose tested in the study is either not adverse (No-Observed-Adverse-
51
-------
Effect Level - NOAEL) or if NOELs can be set using longer tern studies. In cases where
this is not possible, instead of rejecting studies outright, a NOEL may be calculated via the
use of linear regression analysis or application of larger safety factors. Agency guidelines
should clearly reflect this latitude.
Reviewer Inconsistency
Another concern which surfaced during our discussions of the rejection factors is
inconsistency between reviewers. What appears to be adequate for one reviewer is
inadequate for another. Comprehensive internal communication of policy, training, better
communication of where there is flexibility in the guidelines, acceptance of OECD's
protocols and peer review are activities which can improve this situation. Enhanced informal
communication between the reviewer and the registrant in the early stage of Agency review
would also help prevent formal study rejection and avoid unnecessary delays. Minor
deviations or unanticipated concerns which can be handled during the initial review process
can save time and energy for both the Agency and the registrant and ultimately reduce the
rejection rates. We have noted some big improvements in these areas over the last few
years; however, paradoxically the peer review process seems to generate more rejections and
apparent inconsistencies in some cases. There should be more consistency between primary
review and peer review.
Maximum Tolerated Dose
Although the rejection rates for CORT studies remain low, there are serious concerns
over the Agency's criteria for the selection of the highest dose (maximum tolerated dose or
MTD). The Agency has determined that a major rejection factor in chronic
feeding/oncogenicity studies is failure to meet current maximum tolerated dose criteria.
Controversy over use of the MTD in carcinogenicity bioassays and its relevance to human
risk assessment has been high in recent years. The MTD issue has been a major focus of the
National Academy of Science (NAS) Committee on Risk Assessment Methodology (CRAM),
as well as the National Toxicology Program's (NTP) Board of Scientific Counselors who
have suggested that the NTP reexamine its criteria for setting the top dose. In this regard,
the NACA Toxicology Roundtable prepared a paper entitled "Comments on the Use of
Maximum Tolerated Dose for Carcinogenicity Bioassay" and submitted in to the NAS
CRAM Committee on May 14, 1991 for its consideration. A copy of this paper which
summarizes registrants views on MTD testing is attached.
The controversy over use of the MTD revolves around trying to maximize the
sensitivity of the rodent bioassay on one hand while attempting to identify relevant human
carcinogens on the other hand. Some of this controversy has been generated by inconsistent
definitions or interpretations of MTD. Some scientists and regulators (OECD) feel the
highest dose tested in a carcinogenicity bioassasy should be a "Minimally Toxic Dose" rather
than a "Maximum Tolerated Dose." This view is taken because dose levels which produce
physiologically or metabolically compromised animals are much less relevant to hazard
52
-------
identification. Doses that exceed pharmacokinetic saturation levels are often not relevant for
dose-response extrapolation to the much lower dose levels to which humans are exposed.
Therefore, the criteria for selecting the highest dose in a bioassay should focus on identifying
the maximum dose level relevant to assessment of risk at human exposure levels. In practice
this can mean identifying a maximum dose at which the test animals health is not
compromised by exposure to the test substance or one which is in a linear pharmacokinetic
range or one that represents some multiple of exposure levels.
Changing the MTD concept would help reduce false positive findings and
identification of hazards that are not relevant to human exposure. A review of National
Toxicology Program (NTP) studies published in 1985 revealed that 2/3 of chemicals judged
positive were considered carcinogenic based on effects observed solely at the highest dose
(MTD) level. Thus about 2/3 of the chemicals considered positive would have been
considered negative if the maximum dose tested would have been 1/2 MTD. Follow up
publications disclosed that this number was somewhat smaller but still very large. These
reviews also ignore to some extent the fact that in some tests, animals were badly stressed at
the lower doses too (i.e., the mid-dose may also be an MTD). Some implicit assumptions
used by regulators in interpreting results from studies conducted at the MTD include the
following: (a) pharmacokinetics are not dose dependent, (b) dose-response is linear, (c) DNA
repair is not dependent on dose, (d) response is not age dependent, and (e) a test need not
bear a relationship to human exposure. In addition, the interpretation of these studies
assumes that a on-threshold mechanism of action exists. These assumptions are often not
valid./ We believe many results at the MTD are "false positives" and/or not relevant to
human hazard. Our view, which is shared by many others in the scientific community, is
that it makes sense to alter the carcinogenicity bioassay by testing at doses lower than an
MTD (as currently defined by the EPA) when justified on the basis of human exposure
considerations. A "minimally toxic dose" as clarified by OECD may be most appropriate as
it is widely used and has the support of many scientists.
Using lower top doses would not compromise safety or the ability to detect human
health hazards. Apostolou (1990) has presented data which suggests that most of the known
human carcinogens can be detected in animals at dosage levels well below the MTD. There
is therefore some reason to believe that a reduction in the highest dose tested would be
sufficient to detect important human health hazards and at the same time not detect
carcinogenic hazards which are insignificant. It would be easy to accept "false positives" as
conservative regulatory values if the Agency has the discretion to use them in an appropriate
manner. However, in many cases (i.e., Delaney) the identifications are misused at
considerable cost to the American consumer, farmers, and industry.
Since the MTD is chosen empirically, controversy over its definition and
interpretation is difficult to resolve. When a carcinogenic response occurs at toxic doses, the
finding should be considered irrelevant to human hazard identification and risk assessment,
especially if target organs appear to be compromised during the study. EPA should seriously
consider revising its guidelines to use a minimally toxic dose or even use some other
53
-------
mechanism to set doses in oncogenicity studies. This would be consistent with existing
international schemes and recent proposals at NTP. Registrants would welcome the
opportunity to further discuss the MTD issue, including the relevance of pharmacokinetics
and metabolism data in the dose selection process, with the Agency in a workshop or open
forum.
Finally, the Agency requested the registrant's view of why some studies listed as
supplementary have taken in some cases four to seven years to upgrade. We essentially
agree with the Agency's analysis of this problem. At times the Agency is at fault due to
substantial delays in review or failure to communicate the need for further information or
data to the registrant after the review is completed. On the other hand, the registrant's
response may be delayed. Regardless of who is at fault we agree with the Agency's belief
that these substantial time lapses more accurately reflect past Agency/registrant performance
than current performance. Hopefully, this process will significantly improve the time from
submission of studies to review and acceptance.
Summary
The following overall points concerning toxicology test guidelines and protocols
should be considered by the Agency to help deal with the problems discussed in this
overview. First, international harmonization (acceptance of international protocols) is
required in order to avoid needless repetition of studies and overuse of animals by
registrants. Considerable flexibility in meeting the objectives of the study should be built in.
Second, consistent testing guidance should be easily accessible to all registrants in a' single
document, explicit as possible while providing sufficient latitude to permit use of sound
scientific principles by reviewers. Internal policy codifications should be quickly and widely
distributed to registrants. If there is enough confusion within the Agency to require a policy
note, there is probably confusion in industry. Third, the Agency must adequately, and in a
timely fashion, notify registrants when it proposes to change testing guidelines. Reviewers
often begin to use draft guidance documents before they are publicly available. Studies
which are in progress at the time when guidelines are changed should not be rejected based
on the new requirements. It would help prevent confusion if future changes in the guidelines
were issued in a less piecemeal fashion. It would be best if clarifications or updates were
incorporated directly into the original document and a revised document issued. If this is not
possible, formal addenda should be issued and widely disseminated to registrants. The
addenda should indicate how they are different from old guidance.
In addition, many of the rejection factors have been labeled as "avoidable" by the
Agency. Industry believes that many of these are in fact unavoidable and should be
considered on a case-by-case basis.
54
-------
GUIDELINE 81-1 ACUTE ORAL TOXICITY
1. Rejection Factor; Lack of characterization of the test substance (i.e.. purity
and/or composition and/or stability) (32)
EPA Guidance on this Factor: Guidance appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 pages 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
manufacturer's lot number of the test substance with relevant properties of the substance
tested, such as physical state, pH, stability and purity, and the identification and composition
of any vehicles used in administering the test substance are required by the Agency. For
each test, control, or reference substance that is mixed with a carrier, tests shall be
conducted by appropriate analytical methods to determine uniformity of the mixture as well
as concentration and stability in the carrier.
Industry Comment: This is the most commonly occurring criteria for rejection of
acute studies. We feel that the guidance to registrants is sufficient in the case of technical
materials but it is lacking for the end use products. The Agency needs to provide guidance
for the characterization of formulated products. Obviously such areas as the identity of
impurities must be handled differently in the case of the technical versus end use pesticide.
For formulation studies reporting the batch number and the percent active ingredient should
be adequate. Complete and accurate records of the preparation of the dosing solutions/
suspensions should be adequate to demonstrate that the animals were dosed with approxi-
mately the intended amount of test substance. Slight deviations from the intended dose
administered will be lost amongst the much larger biological variability from animal-to-
animal and study-to-study and would rarely affect the labeling classification for a given test
substance. It should be acceptable to reference the product chemistry or the Confidential
Statement of Formula in the report.
A similar problem exists with the stability of the technical grade pesticide versus the
end use product. Determining only the active ingredient's stability should be sufficient for
studies with formulated product.
Industry Recommendation: The Agency needs to provide guidance for the charac-
terization of formulated products. It should be acceptable to reference the product chemistry
or Confidential Statement of Formula in the report. Determining only the active ingredient's
stability should be sufficient for studies with the formulated product.
EPA Response: It is noteworthy that the registrants consider Agency guidance for
characterization of the technical materials as sufficient; but for formulated (end use) products
as inadequate. The Technical Support Section of the Registration Division performs the
55
-------
majority of the acute toxicity reviews for the formulated products; and the Toxicology
Branches of the Health Effects Division (HED) perform the reviews for the acute toxicity on
the technical products. The rejection rate analysis was performed primarily on the HED
reviews, and thus on the technical active ingredients. The issue of guidance for the
formulated products is then somewhat beyond the manageable scope of this document.
However, the list below would pertain to all studies submitted to the Agency for review
(technicals and hopefully end use products).
The Agency is requesting each registrant to include a cover sheet in each study which
includes the following information (if applicable):
Code name, trade name, chemical name, and CAS number (only for technical
product - TGAI),
Percentage active ingredient,
Structure,
Lot and/or batch number,
Physical state,
Stability, or stability and homogeneity in the diet (when the test substance is
administered as part of the diet),
If isomers are present, give percentages,
Major impurities (if of toxicological significance),
A reference to the product chemistry volume,
(For formulations, a reference to the Confidential Statement of Formula or
batch number).
This should greatly reduce studies from being rejected due to this factor in the future.
The toxicology reviewer generally receives only toxicology studies in their packages from
Registration Division. Often there is a statement within the conducting laboratory's report to
the effect that: "the supplier of the test substance is responsible for its characterization and
composition," with no further indication as to lot and/or batch number, stability, and
percentage active ingredient (most laboratories report the physical state and appearance of the
test substance, and they will usually report any number(s) on the package, as received).
56
-------
2. Rejection Factor; Inadequate (or insufficiently spaced) dose levels to calculate a
LD50 value (211
EPA Guidance on this Factor: Guidance on this topic appears in the Subdivision F
Guidelines Hazard Evaluation: Human and Domestic Animals (§80-4 p. 31, §81-1 pp.s 34,
36, and 38). As noted in the Subdivision F Guidelines, the LD50 value shall always be
considered in conjunction with the observed toxic effects and any necropsy findings. The
LD50 value is a relatively coarse measurement useful only as a reference value for
classification, labeling purposes, and packaging, for expressing the possible lethal potential of
the test substance by the ingestion route. Several oral LD50 studies were rejected because of
complete (or nearly complete) mortality at the one dose level (usually 5000 mg/kg) at which
the test substance was administered. It is stated on page 35 of the Subdivision F Guidelines
that "if a test at one dose level of at least 5000 mg/kg body weight.. .produces no compound-
related mortality, then a full study using a minimum of three dose levels might not be
necessary." Also, it is stated on page 36 that "at least three dose levels should be used and
spaced appropriately to produce test groups with a range of toxic effects and mortality rates.
The data should be sufficient to produce a dose response curve and, where possible, permit
an acceptable determination of the LD50." Note that, to some extent, this stipulation has
been modified by the Agency's position that other methods of dose level utilization
(particularly those using fewer animals) may be acceptable.
Industry Comment: The guideline needs to be revised. A great deal of confusion
exists with respect to the number of animals and doses required. The guidelines need
language that defines what is acceptable with respect to the use of a single sex, the limit
dose, fixed doses, etc. Some reviewers have rejected studies because a definitive LD50
value was not determined in both sexes. This does not seem consistent with Agency policy
on minimizing of animal use.
Some labs have revised standard LD50 protocols in order to comply with the EPA
document entitled Revised Policy for Acute Toxicity Testing date September 22, 1988. This
document clearly states that EPA has revised its guidelines for acute toxicity testing and that
animals should not be used merely to obtain an LD50 value.
This problem occurs in part because there is conflict between the Guidelines, the
Phase III Guidance Document and the Agency's Revised Policy for Acute Toxicity testing.
According to the Guidelines, this study should produce test groups with a range of toxic
effects and mortality rates. The data should be sufficient to produce a dose response curve
and, where possible, permit an acceptable determination of the LD50. The Phase III
Guidance Document indicates that the doses tested should be sufficient to determine the
Toxicity Category or use a Limit dose. The Revised Policy on Acute Toxicity Testing
appears to allow a variety of methods to estimate the acute toxicity and goes on to say that
"We plan to accept only newly generated industry data that conform with our revised
guidance unless an adequate rationale accompanies the submission."
57
-------
The Agency's public statements encouraging the use of methods that use fewer
animals may not be compatible with the concept of producing a representative dose response
curve. In using the minimal number of animals to "estimate" the LD50 value some
compromise must be accepted by the Agency. It is suggested that the Agency consolidate
and clarify its guidance on reducing the numbers of animals versus what are the minimal
objectives of this test.
While the guideline indicates that a limit test at 5000 mg/kg is acceptable only if no
compound related mortality occurs, if only 1/5 animals/sex die at this dose, the LD50 would
still exceed 5000 mg/kg body weight. What additional information would be gained by
dosing animals at greater that 5000 mg/kg body weight? Additionally, could limit tests be
used to classify materials into labelling categories rather than definitive LD50s? Perhaps
animals could be dosed at label cutoffs such as 2000 mg/kg, 500 mg/kg and 50 mg/kg body
weight.
Industry Recommendation: The guideline needs to be revised. The Agency should
establish a policy that does not require retesting if the Agency can put the material into a
label category (I, II, III, or IV). It is suggested that the Agency consolidate and clarify its
guidance on reducing the numbers of animals versus what are the minimal objectives of this
test. Perhaps animals could be dosed at label cutoffs such as 2000 mg/kg, 500 mg/kg and 50
mg/kg body weight.
EPA Response: The Agency has prepared further guidance concerning the LD50 and
associated issues. Please refer to Attachment 1, a memorandum from P. Hurley, dated
12/14/92. It should be noted that the Agency has not rejected any data on the basis of its
failing to conform with the Revised Policy on Acute Toxicity Testing. If the Agency can
adequately determine a Toxicity Category from the data submitted, then the study will not be
rejected on the basis of too few animals or deaths at 5000 mg/kg.
58
-------
GUIDELINE 81-2 ACUTE DERMAL TOXICITY
1. Rejection Factor: Lack of characterization of the test substance (i.e.. purity
and/or composition and/or stability) (21)
EPA Guidance on this Factor: Guidance appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 p. 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
manufacturer's lot number of the test substance with its relevant properties of the substance
tested, such as physical state, pH, stability and purity, and the identification and composition
of any vehicles used in administering the test substance are required by the Agency. For
each test, control, or reference substance that is mixed with a carrier, tests shall be
conducted by appropriate analytical methods to determine uniformity of the mixture as well
as concentration and stability in the carrier.
Adequate guidance was provided for this factor, therefore this rejection could be
avoided.
Industry Comment: See comments under Guideline 81-1, Rej. Factor 1.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
2. Rejection Factor; Inadequate percentage of body surface area exposed in each
dose group (7)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-2, p. 44), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 86-87. Subdivision F states that "the test
substance should be applied uniformly over an area which is approximately 10% of the total
body surface area." If a substance is known to be highly toxic, the surface area covered may
be less.
Guidance on this factor is clear and rejection could be avoided.
Industry Comment: The Agency's guidance is clear on this criteria. Additional
questions were raised in the Rejection Rate meeting in February of 1993 concerning the
surface area of the rat (for different strains) and the necessity for further guidance on
defining the animals by age or weight.
59
-------
EPA Response: The Agency suggested that test animals (Sprague Dawley, Wistar,
and Fischer 344) should be in the 6-10 week age range when received from supplier and no
older than 8-12 weeks of age at initiation of dosing for acute studies (see Attachment 2).
The Agency suggests that the area starting at the scapulae (shoulders) to the wing of the
ileum (hipbone) and half way down the flank on each side of the animal should be shaved in
order to dose approximately 10% of the body surface.
3. Rejection Factor; No Quality Assurance Statement (6)
EPA Guidance on this Factor: Guidance on this appears in the 40 CFR 160.35, and
in the Phase 3 Technical Guidance document dated December 24, 1989, 5-6. A signed and
dated Quality Assurance Statement should be included specifying dates of inspections.
Adequate guidance is provided in the above mentioned document. Rejection could be
avoided.
Industry Comment: The guidance is adequate on this point.
EPA Response: The Agency and Industry agree on this point.
4. Rejection Factor; Improper number of animals tested per dose group (5)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-2, p. 43), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 86-87. Subdivision F indicates there
should be at least 10 (5 male and 5 female) animals used per dose group.
Adequate guidance is provided in the above mentioned document. Rejection could be
avoided.
Industry Comment: We do not feel that guidance on this issue is clear. The
Revised Policy on Acute Testing encourages the use of alternate methods for acute toxicity
testing. Included in the methods described in this policy statement is the up-and-down
method which does not use a dose group of 5/sex/group. Is this policy statement still
effective or are studies being evaluated only by the published guidelines?
See also comments under Guideline 81-1, Rej. Factor 2.
EPA Response: The Agency has prepared further guidance concerning the number of
animals which should be tested per group. Please refer to Attachment 1, a memorandum
from P. Hurley, dated 12/14/92, page 3.
60
-------
5. Rejection Factor; Only one sex tested (3)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-1, p. 37a), and in the Phase 3 Technical
Guidance document dated December 24, 1989, D-4. Subdivision F states "equal numbers of
animals of each sex with healthy intact skin are recommended for each dose level."
However, the 'Revised Policy for Acute Toxicity Testing, September 22, 1988' indicates that
consideration will be given to limiting studies to the more sensitive sex, using the previous
history on the chemical class to make the determinations, and testing of a few animals of the
other sex for confirmation.
Adequate guidance is provided in the above mentioned document. Rejection could be
avoided.
Industry Comment: Guidance on this factor is not completely clear. It is probable
that at least some of these rejections (Factors 4 and 5) arose from differences between
international guidelines. The OECD guidelines require: "At least 5 animals are used at each
dose level" and it further states "Where information is available demonstrating that a sex is
markedly more sensitive, animals of this sex should be dosed." It is clear that the OECD
and the EPA guidelines are not compatible. If toxicity studies are conducted outside the
USA, registrants are often restricted by regulations in some countries that prohibit conducting
studies that exceed the OECD guidelines. The specific guidelines referenced above are
considered adequate in most of the world except in the USA. If a registrant wanted to meet
the "letter" of the EPA guideline, studies conducted with OECD guidelines would have to be
discarded and a new study using ADDITIONAL animals would have to be conducted for the
EPA. It is doubtful that these repeat studies could be justified in terms of results or animal
usage. EPA in its own discussion of this rejection factor acknowledges that its own policies
allow for acceptance of only one sex. The Agency should consider whether slight
differences in the sensitivity of different sexes to the test substance would be of sufficient
magnitude to change the labelling of the material. Acute data are usually accompanied by
data from longer term studies in a single submission. These data could also be used to
determine if there was indeed a significant difference in the sensitivity of the sexes to the test
substance before requiring additional testing.
Industry Recommendation: Unless there is substantial scientific evidence to support
utilizing both sexes in this type of study, EPA should as a matter of policy and considering
the need to harmonize guidelines accept studies conducted according to OECD guidelines.
The Agency should consider whether slight differences in the sensitivity of different sexes to
the test substance would be of sufficient magnitude to change the labelling of the material.
EPA Response: The Agency agrees that studies should not be rejected on the basis
of only one sex being tested. Therefore this rejection factor was not necessarily avoidable.
This issue will be visited by the OECD harmonization process.
61
-------
6. Rejection Factor: Omitted source, age, weight or strain of test animal (3)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-2, pp. 29 & 43), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 86-87. Subdivision F states that adult rats
(200-300 g), rabbits (2-3 kg), or guinea pigs (350-450 g) may be used. The albino rabbit is
preferred because of its wide data base, size, strain, and ease of handling. Justification for
the use of other than these three should be provided. The species, strain, and source of the
test animal is also required.
Adequate guidance is provided in the above mentioned document. Rejection could be
avoided.
Industry Comment: If the rejection is as stated, there is agreement on the adequacy
of the guidance. It should be pointed out that although the rabbit may be the preferred
species based partially on size and ease of handling, many laboratories around the world
prefer the rat.
Industry Recommendation: Studies conducted in the rat should be acceptable to the
Agency if adequately conducted. In some cases there is a rationale for the use of the rat
instead of the rabbit.
EPA Response: The use of the rat is acceptable without justification and will not be
the basis for rejecting a study. Also, the Agency will be more flexible on the weight
restriction for acute studies. If the animals meet the age requirements, then a weight
category will not apply for acute studies.
62
-------
GUIDELINES 81-3 AND 82-4 ACUTE AND 90-DAY INHALATION TOXICITY
Industry General Comment; The draft policy document" Interim Policy for Particle
Size and Limit Concentration Issues in Inhalation Toxicity Studies" addresses many of the
issues discussed below. We generally agree with the recommendations in that report with
respect to lower limit dose and an MMAD (mass median aerosol diameter) of 1 to 4 fim for
acute studies and 1 to 3 /xm for longer term studies.
1. Rejection Factor: Less than 25 percent of particles were < 1 pm (72)
EPA Guidance on this Factor: Guidance for this appears in the Hazard Evaluation
Division, Standard Evaluation Procedure (SEP) for Inhalation Toxicity Testing, August 1988
(p. 15). The SEP indicates "it would seem appropriate that at least 25 percent of the particle
distribution used in these studies be in the submicron range for acute and repeat exposures.
When studies are carried out using large particle distributions (median diameters greater than
3.0 ftm), judgement is necessary in determining whether the study should be repeated using a
smaller particle size range...If the test results show minimal toxicity via other routes, then
the acute inhalation testing should be repeated using smaller particle sizes."
A 1992 Society of Toxicology (SOT) position paper was presented to the Agency, and
subsequently published in Fundamental and Applied Toxicology, recommending "that the
limit test concentration should be the highest concentration (up to 5 mg/L) that can be
achieved while still maintaining a particle size distribution having an MMAD between 1-4
urn." The paper based this recommendation on the following:
1. Aerosol particle generation and behavior limit the size of particles in an
exposure atmosphere at a concentration of 5 mg/L.
2. Particle size influences deposition site in the respiratory tract.
3. Damage to any area of the respiratory tract can produce lethality.
4. It is not possible to predict the most responsive region of the tract or the most
harmful particle size of an untested agent (i.e., smaller particles do not always
cause the most damage).
5. "In practice, it has not been possible to achieve the theoretical limit within a
factor of two (i.e., 25% of particle mass { 1 /urn, as recommended by the
U.S.E.P.A.) at concentration of 5 mg/L."
6. "It is often very difficult to reduce the size of starting materials below 2-3
because of the kinetic energy needed for deagglomeration."
63
-------
The SOT paper recommends using particles having aerodynamic diameters that could
deposit throughout the rodent respiratory tract (between 1-4 /xm). However, some EPA
scientists are skeptical that this range of particles are indeed deposited throughout the rodent
respiratory tract. In a memorandum by Stanley Gross dated April 18, 1989, it is stated: "If
the laboratory is having difficulty in achieving the required diameters, the study needs to
indicate what they did and why they were unable to provide the small particles." If the
explanation is adequate the study would be considered acceptable.
In conclusion, the SEP requirement that 25% of particles must be less that 1 micron
appears to be frequently unattainable, and the requirement should be changed to more closely
reflect the SOT position paper. EPA is of the opinion that this requested change applies to
the acute inhalation (§81-3) and the subchronic inhalation (§82-4) studies. As lower doses
are used in a subchronic study it should generally be possible to achieve smaller particles,
and all efforts should be taken to generate submicron particles. However, any changes in
this requirement await Industry comment.
Industry Comment: The Society of Toxicology position paper, "Recommendations
for the Conduct of Acute Inhalation Limit Tests", was prepared by leading experts in the
field of inhalation toxicology and takes an extremely realistic look at inhalation toxicity
testing. The document clearly states that the EPA SEP particle size requirements are usually
unattainable. The document also demonstrates that test atmospheres with mass median
aerodynamic diameters in the range of 1 to 4 /*m are deposited throughout the entire
respiratory tract of the rat. This document also explains some of the problems which can be
encountered when attempting to reduce test atmosphere particle size while maintaining the
high chamber concentrations required by the EPA.
Industry Recommendation: It is our belief that inhalation studies should be
considered acceptable when the particle size is reduced to every extent possible and when the
report documents scientifically valid justification. In addition, we believe that a change in
the particle size requirements should be made to the Inhalation Toxicology SEPs as
recommended in the draft EPA policy (see General Comments).
EPA Response: The Agency has prepared a draft memorandum (see Attachment 3,
Section 1) entitled "Interim Policy for Particle Size and Limit Concentration Issues in
Inhalation Toxicity Studies" from J. Whalan and J. Redden, dated December 12, 1992. The
document has been reviewed by a variety of Inhalation Toxicity experts from academia,
contract laboratories, and registrants. The document recommends 1 - 4 /*m MMAD particle
size for acute studies and 1 - 3 /*m MMAD for longer term studies. The range suggested for
longer term studies is based on limited data and may be revised if additional data warrants a
change. In addition, the limit dose has been reduced to 2 mg/L. The new particle size
requirements for acute inhalation toxicity studies are largely based on the Society of
Toxicology position paper "Recommendations for the Acute Inhalation Limit Tests." This
interim policy is currently in effect.
64
-------
In Attachment 3, Section 2 is a memorandum entitled "Comments on Standard
Evaluation Procedure, Inhalation Toxicology Testing (SEP/Inhalation)," April 18, 1989 (by
Dr. S. Gross). This document addresses Industry's concern "that inhalation studies should
be considered acceptable when the particle size is reduced to every extent possible and when
the report documents scientifically valid justification." This memorandum of April 18, 1989
states "if the laboratory is having difficulty in achieving the required diameters, the study
authors should indicate what they did and why they were unable to provide the small
particles." If the explanation is deemed adequate the study would be considered acceptable.
In Attachment 3, Section 3 is a memorandum entitled "Policy on Acute Inhalation
Toxicity Data Waivers", by P. Fenner-Crisp dated December 8, 1991." This document
addresses specific reasons for granting inhalation toxicity waivers. It also contains guidance
for reviewers to determine if a registrant's explanation about problems generating a test
substance are justifiable.
2. Rejection Factor; Three exposure concentrations were not used; LC50 could
not be calculated, or was not tested to limit concentration;
highest concentration achieved did not produce toxicity and
it was not apparent from report if a higher concentration
could be achieved (38)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§81-3 p. 52). Subdivision F states three
exposure concentrations should be used and spaced appropriately to produce test groups with
a range of toxic effects and mortality rates. The data should be sufficient to produce a dose-
response curve and permit an acceptable estimation of the median lethal concentration.
Range finding studies may help to estimate the positioning of the test groups so that no more
than three doses will be necessary.
Also, Subdivision F on p. 49 states "EPA recommends the following means to reduce
the number of animals used to evaluate acute effects of chemicals exposure while preserving
its ability to make reasonable judgements about safety:
• Attempt to use existing data on structurally related chemicals.
• If data for calculating an LC50 are needed, perform an acute toxicity study whereby
the value of the data derived from the investment of animals lives is enhanced. EPA
does not encourage the use of animals solely for the calculation of an LC50.
• Use methods that minimize the number of animals in the test."
65
-------
The SOT position paper also questions the value of a limit concentration of 5 mg/L
because:
1. "lack of lexicological or real-world exposure relevance, and the possibility of
airway obstruction and suffocation." It has not been possible to achieve 25%
of particle mass < 1 /*m at a concentration of 5 mg/L.
In at least one review a study was judged as supplementary because the limit dose was
not achieved (4.9 mg/L analytical concentration). Clearly, grading this study supplementary,
on this factor alone, was an error on the part the Agency.
The limit concentration, in light of the SOT comments, may have to be lowered after
the industry workshop comments.
2. No explanation was given in the study that the dose selected was the maximum
achievable concentration.
Guidance for this appears in the Gross memorandum dated April 18,1989. "In order
to declare the concentrations as the maximum attainable, the registrant needs to indicate what
efforts were made to reach the 5 mg/L concentrations, what problems were encountered and,
if possible, try to explain why higher concentrations were not achievable."
EPA guidance for the limit dose is adequate and therefore was avoidable on' the part
of the registrant.
3. No LC50 was established; female LC50 could not be determined, or LC50 for
each sex not reported; with only two doses it is impossible to do probit
analysis.
Clearly, some of the LC50 rejection reasons seem to be in conflict with this policy.
Only if one sex is more sensitive to a chemical, as in some organophosphates, should the
registrant be required to do more testing.
EPA guidance here is inadequate and rejection may have been unavoidable.
Industry Comment: Most laboratories employ all available methods in an attempt to
produce the highest respirable concentration for limit test exposures as mandated by the EPA
Standard Evaluation Procedures.
Studies should not be rejected because the limit dose was not achieved. In many
cases this may be a reflection of the difficulty of achieving these doses at the specified
particle size distribution. We agree with the recommendations in the draft EPA policy; the
limit dose should be reduced to 2 mg/L for aerosols and particle size requirements should be
66
-------
changed to 1 to 4 pm MM AD (see General Comments). If the Agency can categorize the
product with respect to a toxicity category, no additional testing should be required.
Some laboratories have had acute inhalation toxicity studies rejected by the EPA
because they failed to determine a definitive LC50 value in both sexes. This appears to
contradict the EPA's document entitled Revised Policy for Acute Toxicity Testing dated
September 22, 1988. This document clearly states that EPA has revised its guidelines for
acute toxicity testing and that animals should not be used merely to obtain an LC50. The
Agency should clarify guidance on use of animals.
The revised policy document was purported to have been disseminated to industry,
governmental bodies, scientific societies and animal welfare groups, however, in at least one
case, a laboratory was informed by an EPA reviewer responsible for reviewing acute toxicity
studies that he had no knowledge of the document. They were also informed by two EPA
reviewers that this document was not universally accepted at EPA and that studies conducted
under these guidelines may or may not be acceptable to individual reviewers. As registrants,
we are only able to satisfy either the standard guidelines or the newer modifications in any
given study design. We would like written EPA clarification on which approach will yield
satisfactory EPA reviews. The current disparity among EPA reviewers is resulting in
delayed registrations and is forcing the use of increased rather than decreased animal usage
because studies that are rejected have to be repeated.
Industry Recommendation: Studies should not be rejected because the limit dose
was not achieved. If the Agency can categorize the product with respect to a label category,
no additional testing should be required. We would like written EPA clarification on which
approach will yield satisfactory EPA reviews.
EPA Response: The Agency's draft memorandum (see Attachment 3, Section 1)
entitled "Interim Policy for Particle Size and Limit Concentration Issues in Inhalation
Toxicity Studies" proposes the reduction of the limit dose to 2 mg/L,
The Agency is requiring that both sexes be tested unless justification can be given to
test one sex over another (i.e., if a long-term inhalation study had been performed). Also, if
the Agency can adequately determine a Toxicity Category from the data submitted, then the
study will not be rejected. A question arose as to whether a study performed for a
formulated product would be acceptable if there was no technical product ever manufactured.
The Agency responded that it would depend on the percentage of active ingredient in the
formulated product.
67
-------
3. Rejection Factor: Inadequate reporting of exposure methodology [Examples;
inadequate sampling & generation, lack of information on
distribution of test substance in chamber, missing calibration
data, missing quality assurance statement, missing daily
observations, body weights, necropsy results, and the strain
of rat testedl (20)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-3 p. 54, 550 and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 88-89 and 106-108. Subdivision F
addresses the description of the exposure apparatus and the required animal data that should
be reported which includes body weights, necropsy findings, and clinical observations. PR
Notice 86-5, July 29, 1986 requires a Certification of Good Laboratory Practice (p. 6)
Adequate guidance is provided for this factor. Rejection based on this factor was
avoidable.
Industry Comment: All of the information on exposure is obviously necessary to
meet good scientific input for the study. The EPA guidance for the most part is adequate,
however, some problems have been encountered in nose-only studies concerning sampling
procedures and frequency of sampling.
Industry Recommendation: The Agency may want to look at nose-only studies to
determine if this will be a rejection problem, since nose-only is recommended for aerosols
and particles in the draft inhalation policy.
EPA Response: The Agency has recently been in communication with the SOT
Inhalation Toxicology work-group (December 10, 1992). The work-group had concerns with
a certain commercial nose-only apparatus called "Flow-Pass", which is not well suited for
high aerosol concentrations. Therefore, at the suggestion of the work-group, the language on
page 13 of the interim policy (see Attachment 3, Section 1) has been modified:
"3. It is recommended, but not required, that nose-only or head-only exposure be
used for aerosol studies in order to minimize oral exposure resulting from
animals licking test substance off their fur. Individual housing must be used
during whole-body exposure to prevent filtering by the fur which occurs when
animals huddle together."
Nose-only exposure will be addressed in detail in the upcoming revision of the SEP
for Inhalation Toxicology.
68
-------
4. Rejection factor; Protocol errors (15)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-3 p. 51). Example: animals underweight or
under age. Subdivision F indicates that young adult animals should be used and weight
variation of animals or groups should not exceed ± 20% of the mean weight for each sex.
Example: excessive relative humidity or temperature, or temperature & relative
humidity was not reported. Guidance on this appears in Subdivision F - Hazard evaluation:
human and domestic animals (§81-3 p. 52). Subdivision F states that ideally the temperature
in the chamber should be maintained at 22 C° ± 2° for the rat. Ideally, relative humidity
should be maintained between 40% and 60%. The method of generating and using water as
a vehicle may preclude this condition.
Example: chambers were not at equilibrium at start of dosing or only 1 hour of
exposure at equilibrium. Guidance on this appears in Subdivision F - Hazard evaluation:
human and domestic animals (§81-3 p. 52). Subdivision F states the duration of the
exposure should be at least four hours allowing appropriate additional time for exposure
equilibrium.
Example: missing data for food consumption in the subchronic study. Guidance on
this appears in Subdivision F - Hazard evaluation: human and domestic animals (§82-3 p.
96). Food consumption is required weekly if body weight is affected.
Sufficient guidance is provided on this factor and rejection was avoidable on the part
of the registrant.
Industry Comment: EPA guidance on this factor is generally adequate, however, it
should be noted that differences in exposure time (1 hour vs. 4 hours) reflect past guideline
changes. We note that studies were rejected because animals were underweight or under
age. The only guidance that EPA gives for age is the term "young adult." This term is too
subjective in that it can mean many things to different people.
Industry Recommendation: If EPA considers that age is critical to acute toxicity
studies, it should give more detailed advice on what it considers "young adult" specifying age
limits in days or weeks.
EPA Response: The Agency suggested that test animals (Sprague Dawley, Wistar,
and Fishcer 344) should be in the 6 - 10 week age range when received from the supplier
and no older than 8-12 weeks of age at initiation of dosing for acute studies. The Agency
suggested that in order to dose approximately 10% of the body surface, the area starting at
the scapulae (shoulders) to the wing of the ileum (hipbone) and half way down the flank on
each side of the animal should be shaved.
69
-------
5. Rejection Factor; Lack of characterization of the test substance (purity and/or
composition and/or stability) (16)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 p. 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities and contaminants) should be supplied to the Agency. Also, the Agency
requires the manufacturer's lot number of the test substance with its relevant properties of the
substance tested, such as physical state, pH, stability and purity, and the identification and
composition of any vehicles used in administering the test substance. For each test, control,
or reference substance that is mixed with a carrier, tests shall be conducted by appropriate
analytical methods to determine uniformity of the mixture as well as concentration and
stability in the carrier.
EPA guidance on this factor is adequate and rejection was avoidable on the part of the
registrant.
Industry Comment: See comments under Guideline 81-1, Rej. Factor 1.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
6. Rejection factor; Questions regarding test substance preparation (12)
EPA Guidance on this Factor: Example: should the test substance be milled?
Guidance on this appears in Subdivision F - Hazard evaluation: human and domestic animals
(§81-3 p. .51). Subdivision F states it may be necessary to reduce the size of the aerosol
particles to inhalable size for the animal tested.
Example: which material was tested - the technical or end-use product; and what
vehicle was used and was it the most appropriate vehicle. Guidance on this appears in
Subdivision F - Hazard evaluation: human and domestic animals (§81-3 p. 51). Subdivision
F indicates that the manufacturing-use product and, if it is different, the technical grade of
each active ingredient, shall be tested to support the registration of a manufacturing-use
product and the end-use product shall be tested to support the registration of an end-use
product. If possible, the chemical composition and physical state of the test substance should
be the same as that which is encountered in the product.
Example: was a solvent necessary? Guidance on this appears in Subdivision F -
Hazard evaluation: human and domestic animals (§81-3 p. 52). Subdivision F indicates if a
vehicle is required it should not alter the chemical or lexicological properties of the test
substance. Otherwise, a vehicle control group should be used.
70
-------
The guidance found in the cited document is adequate; therefore, rejection was
avoidable.
Industry Comment: EPA guidance on this is fairly adequate. However, the need to
use a test substance which represents the form being presented in the end-user situation is
most important and germane to the acute inhalation test. Preparation of the material needs to
be consistent with the physical and chemical properties of the product.
Industry Recommendation: Grinding of granular pesticides makes little sense. For
materials that are too large to be respirable, waivers should be granted.
EPA Response: In Attachment 3, Section 3 is a memorandum entitled "Policy on
Acute Inhalation Toxicity Data Waivers." This document addresses reasons for granting
inhalation toxicity waivers. It also contains guidance for reviewers to determine if a
registrant's explanation about problems generating a test substance are justifiable. This
policy will be incorporated into the upcoming SEP for Inhalation Toxicology. EPA will
clarify the conditions for "fines" of granular formulations in the revised SEP for inhalation.
7. Rejection factor; Actual chamber concentration not measured or only nominal
concentration measured (8)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-3 p. 55f) and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 88-89 and 106-108. Subdivision F
recommends that exposure data be tabulated and presented with mean values and a
measurement of variability (e.g., standard deviation) be included. This includes information
on airflow rates, temperature, nominal concentrations, total concentrations, and particle size
distributions.
EPA guidance is adequate for this factor. Rejection was avoidable on the part of the
registrant.
Industry Comment: EPA guidance on this factor is adequate, however, please note
comments on nose-only exposure under this guideline, Rej. Factor 3.
EPA Response: Refer to response under Guideline 81-3, Rej. Factor 3.
71
-------
GUIDELINE 81-4 PRIMARY EYE IRRITATION - RABBIT
1. Rejection Factor: Lack of characterization of the test substance (purity and/or
composition and/or stability) (26)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 pp. 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities and contaminants) should be supplied to the Agency. Also, the Agency
requires the manufacturer's lot number of the test substance with its relevant properties of the
substance tested, such as physical state, pH, stability and purity, and the identification and
composition of any vehicles used in administering the test substance. For each test, control,
or reference substance that is mixed with a carrier, tests shall be conducted by appropriate
analytical methods to determine uniformity of the mixture as well as concentration and
stability in the carrier.
Industry Comment: See comments under Guideline 81-1, Rej. Factor 1.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
72
-------
GUIDELINE 81-5 PRIMARY DERMAL IRRITATION
1. Rejection Factor; Lack of characterization of the test substance (purity and/or
composition and/or stability) (17)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 p. 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
Agency requires the manufacturer's lot number of the test substance with its relevant
properties of the substance tested, such as physical state, pH, stability and purity, and the
identification and composition of any vehicles used in administering the test substance. For
each test, control, or reference substance that is mixed with a carrier, tests shall be
conducted by appropriate analytical methods to determine uniformity of the mixture as well
as concentration and stability in the carrier.
Adequate guidance for this guideline was provided, therefore, rejection was
avoidable.
Industry Comment: See comments under Guideline 81-1, Rej. Factor 1.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
2. Rejection Factor: No Quality Assurance Statement and/or No Good Laboratory
Practice (GLP) Statement (11)
EPA Guidance on this Factor: Guidance on this appears in the 40 CFR 160.35, and
Phase 3 Technical Guidance dated December 24, 1989, 5-6. A signed and dated Quality
Assurance Statement should be included specifying the dates of inspections.
Adequate guidance for this guideline was provided, therefore, rejection was
avoidable.
Industry Comment: It is agreed there is adequate guidance on this point.
EPA Response: The Agency and Industry agree on this point.
73
-------
3. Rejection Factor; Improper test substance application/preparation (10)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§80-4 p. 30, §81-5, p. 57), and in the Phase 3
Technical Guidance document dated December 24, 1989, C 92-93. Subdivision F states that
when testing solids, the test substance should be moistened with water or, where necessary, a
suitable vehicle to ensure good contact with the skin. When vehicles are used, the influence
of the vehicle on irritation of the skin by the test substance should be taken into account.
Liquid test substances are generally used undiluted. The test substance should be applied to
a small area (approx. 6 cm2) of skin and covered with a gauze patch for a maximum of 4
hours.
Adequate guidance for this guideline was provided, therefore, rejection was
avoidable.
Industry Comment: EPA's response to this rejection criteria quotes Subdivision F:
"when vehicles are used, the influence of the vehicle on irritation of the skin by the test
substance should be taken into account." Unfortunately, the Agency is not clear how this
should be done. In many reviews of studies which utilize vehicles other than water, adverse
comments are made on the choice of vehicles.
Industry Recommendation: The Agency needs to give some guidance on selection
of non-aqueous vehicles which are often needed for agrichemicals. Either a suggested list of
vehicles or specific guidance on how to determine the influence of vehicles should be
supplied.
EPA Response: Some of the common non-irritating vehicles known to EPA are:
distilled water, saline, ethanol + water, gum arabic, carboxymethylcellulose, glycerol, and
mineral oil. If the vehicles are unknown to EPA, it would save time by providing an
explanation in the study report as to why that a particular vehicle was chosen. Also,
justification should be given when an irritating vehicle is used. The following points should
be considered when selecting a vehicle: solubility and suspendability, high volatility,
toxicity, and permeability.
4. Rejection Factor; Omitted source, age, weight, or strain of test animal (4)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§80-4 p. 30, §81-5 p. 57). Subdivision F states
under "Reporting of Data" that the source of the animals shall be given and that the albino
rabbit is the preferred species. Justification should be given if another mammalian species is
selected.
74
-------
Adequate guidance for this guideline was provided, therefore, rejection was
avoidable.
Industry Comment: In the EPA's response, it is emphasized that the rabbit is the
preferred species. Although this is the most frequently used species in the United States,
many Europeans and others have used the rat. It is unclear what type of justification would
be necessary if a registrant wanted to submit data in the rat. This is an issue of international
harmonization.
Industry Recommendation: Rat should be acceptable.
EPA Response: The use of the rat is acceptable without justification and will not be
the basis for rejecting a study. Also refer to Guideline 81-2, Rejection Factor #2.
5. Rejection Factor: Missing individual/summary animal data (4)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§80-4 p. 28-33, §81-5 p. 57-58), and in the Phase
3 Technical Guidance document dated December 24, 1989, C 92-93. Subdivision F states
"The test report shall include all information necessary to provide a complete and accurate
description and evaluation of the test procedures and results. A test report should contain at
least three parts: a summary and evaluation of the test results; a description of the test
procedures; and the data and information required by each applicable section of this
subdivision." Summary data shall be submitted in tabular form, showing for each individual
animal irritation scores for erythema and edema at 30 to 60 minutes, 24, 48 and 72 hours
after patch removal.
Adequate guidance for this guideline was provided, therefore rejection was avoidable.
Industry Comment: It is agreed that adequate guidance has been provided.
EPA Response: The Agency and Industry agree on this point.
75
-------
GUIDELINE 81-6 DERMAL SENSITIZATION
1. Rejection Factor; No positive control, naive control, or other control problem
(30)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-6 p. 61-62) and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 94-95. Subdivision F recommends the
periodic use of a positive control substance with an acceptable level of reliability for the
selected test system. In some studies the positive control was included but did not elicit a
positive response or the study itself did not employ a positive control. Some study reports
did not describe the results of a positive control experiment conducted by the same method in
the same laboratory with the strain of test animal used. Other studies did not include a naive
control. Some of the seven study protocols do not require the inclusion of naive controls
since the part of the guinea pig not treated during the induction phase often serves as the
control. The requirement for a naive control, thus, depends on the study type and reactions
to the positive control in the test animals used.
This rejection factor may have been unavoidable. The guidance document needs to
state more clearly when naive controls should be included for each of the several dermal
sensitization study types.
Industry Comment: We agree that this factor is not avoidable. There is inconsis-
tency among reviewers on the acceptability of positive controls which were not run
concurrently with the study. Some reviewers will accept a periodic positive control to
establish the reliability of the test system. However, other studies are rejected without a
concurrent positive control even though periodic positive controls are run every 3-6 months
and submitted with the study. One reviewer indicated that the periodic positive control was
acceptable only if the test substance produced sensitization in the test. If the test substance
did not produce sensitization in the test, a concurrent positive control was necessary. The
study was classified at supplementary and a new study with a concurrent positive control was
required. This seems like an unnecessary use of animals. The EPA should clarify when and
if concurrent positive controls are needed.
EPA's discussion of the use of naive controls is confusing. It is unclear as to what
purpose naive controls would serve in the induction phase. Naive controls are very useful,
but not essential to the challenge part of these studies.
Industry Recommendation: We concur with the conclusions in the Agency analysis:
the Agency needs to better define when "naive" controls are required and clarify the use of
data from this group. The EPA should clarify when and if concurrent positive controls are
needed.
76
-------
EPA Response: The Agency has recommended that the guidelines be revised to state
that one of two options are acceptable: 1) a concurrent positive control group should be used
with each dermal sensitization study or 2) documentation of the laboratories training program
for their technicians should be provided to the Agency. The inclusion of a positive control is
considered an essential part of quality control for a study of this type. Historical positive
controls will be accepted by the Agency providing they were performed within 6 months of
the definitive study and used the same strain and supplier as the definitive study,
Under circumstances where a dermal sensitization study is "clearly negative", the
Agency will not reject the study when it has been submitted without naive controls
(challenge).
2. Rejection Factor; Dosing level problems (i.e.. too much or too little irritation)
(28)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-6 p. 60). Subdivision F indicates that if the
test substance concentration chosen causes marked irritation, it should be diluted with
physiological saline until a concentration is found which produces only slight irritation.
Acceptability of the study depends upon the dose levels of the test substance used for
induction and challenge. In some studies the test substance produced too much dermal
irritation such that a reaction due to the challenge could not be readily distinguished from the
irritant properties of the test substance. A pretest should be performed for test substances
that are irritating substances in order to determine the minimally irritating dose levels. On
the other hand, dose levels which are far below the minimally irritating dose level should not
be used. Highly concentrated nonirritating materials do not require a pretest if high
concentrations are used for induction and challenge.
Guidance regarding appropriate dosing levels is clearly stated in the guidance
document. Therefore, rejection was avoidable.
Industry Comment: This is a major rejection factor; although guidance is clear, this
factor is not avoidable in the sense that irritation in the narrow range specified by the Agency
can be difficult to achieve. The Agency needs to be more flexible in accepting studies
showing irritation outside the specified range. Moderate irritation in virgin controls would
admittedly invalidate a study if it occurred in the challenge phase but it is difficult to
understand why the Agency often considers it to invalidate a study if it occurs in the
induction phase.
Another biological fact that reviewers often disregard is animal variability. This is
again one of the difficulties experienced when extrapolating data from a few pre-test animals
to the larger numbers used in the main study. Occasionally, there are animals within a group
77
-------
which show a lower threshold to irritation than would be expected from the general popula-
tion. With some chemicals, there is a wider irritation threshold than with other chemicals.
The Agency needs to consider these factors when reviewing a study. It is often not
possible to produce textbook data when conducting studies in the laboratory, and unless a
study is flawed in a manner that makes the data invalid, every attempt should be made to
accept a study. This is consistent with good science and the policy to reduce the unnecessary
use of animals in repeat studies.
Industry Recommendation: The Agency needs to be more flexible in accepting
studies showing irritation outside the specified range. The Agency needs to consider these
factors when reviewing a study.
EPA Response: The Agency recommends a dose level showing minimal irritation to
be used for the induction phase; however, the Agency realizes this is often difficult to attain
and stay within a narrow range of acceptable irritation. The registrants should submit to the
Agency the range-finding data, if the doses which were selected in the main study proved to
be irritating.
When a test substance produces too much irritation at challenge (or irritation at
induction that can persist through to challenge), the study is compromised and a repeat study
using diluted test substance is required because the reviewers cannot distinguish the
difference between a primary irritation reaction and a sensitization reaction.
HED and its individual reviewers will be more flexible in accepting studies in which
there is no perceptible irritation especially when range-finding studies present the dose levels
which result in minimal irritation. One important factor is that animal variability between
the group used for the range-finding study and the definitive study may differ slightly in the
initial dermal irritation response.
3. Rejection Factor; Lack of characterization of the test substance (purity and/or
composition and/or stability) (23)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 p. 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
Agency requires the manufacturer's lot number of the test substance with its relevant
properties of the substance tested, such as physical state, pH, stability and purity, and the
identification and composition of any vehicles used in administering the test substance. For
each test, control, or reference substance that is mixed with a carrier, tests shall be
78
-------
conducted by appropriate analytical methods to determine uniformity of the mixture as well
as concentration and stability in the carrier.
Adequate guidance is provided in the cited documents. Therefore, this rejection
factor was avoidable.
Industry Comment: See comments under Guideline 81-1, Rej. Factor 1.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
4. Rejection Factor: Unacceptable protocol or other protocol probJem (21)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-6 p. 60-61) and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 94-95. Subdivision F lists 7 acceptable
testing methods. These studies were rejected because a different test method than the 7 listed
in Subdivision F were submitted to the Agency. These studies had significant variations
from the established protocols such that interpretation of the data were compromised. In the
Phase 3 Technical Guidance document there is a category titled "other test accepted by
OECD." If the registrant chooses one of the OECD guideline study types, and if proper
justification is indicated, then the Agency should handle these on a case-by-case basis.
The guidance is clearly stated and the rejection factor was avoidable on the part of the
registrant.
Industry Comment: The Agency in its comments says that if a registrant chooses
one of the OECD study types, its use must be properly justified. There are a number of
valid methods for conducting dermal sensitization.
Industry Recommendation: When the Agency lists which methods are valid it
should not then require additional justification for the use of a method. Study designs which
have been accepted by OECD should be acceptable.
EPA Response: In essence, the Agency agrees with the Industry comments. In the
future, individual reviewers will not be permitted to ask for additional studies without
suitable justification; particularly if the registrant submits an acceptable study which used an
acceptable protocol. The Agency recognizes OECD protocols and will accept any of the 7
studies recommended.
79
-------
5. Rejection Factor; Individual animal scores or data missing (12)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, p. 28-33). Subdivision F states that "the
test report shall include all information necessary to provide a complete and accurate
description and evaluation of the test procedures and results. A test report should contain at
least three parts: a summary and evaluation of the test results; a description of the test
procedures; and the data and information required by each applicable section of this
subdivision." The requirement for individual data is not specifically stated in the guideline;
however, this rejection factor is based on the adherence to Good Laboratory Practices (CFR
§160). The lack of such data adds difficulty to the review process, for without it, variations
in response of each animal to treatment as well as verification of the mean data presented
cannot be determined.
Guidance on this factor is clearly stated and rejection could be avoided.
Industry Comment: Based on the EPA discussion it does not appear that guidance
on this is clear. Individual animal scores are not required in the guidelines for the induction
phase of the study. If the Agency feels these scores are required for some cases, this should
be specified in the guidelines. We agree that individual animal scores should be supplied at
challenge.
Industry Recommendation: If the Agency feels these scores are required for some
cases, this should be specified in the guidelines.
EPA Response: The requirement for ajl individual animal data is not specifically
stated in the current guidelines. The Agency will recommend that the revised harmonized
guidelines specify the inclusion of all data sets including individual animal scores from both
the induction and challenge phases.
6. Rejection Factor; Scoring method or other scoring problem (11)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§81-6 pp.s 61-62) and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 94-95. Subdivision F indicates that a brief
description of the grading system be submitted. In some studies no basis was given for the
scoring method used.
The description of scoring methods is unclear in the guidance documents; therefore, a
rejection based on this factor may have been unavoidable.
80
-------
Industry Comment: The requirement for a description of scoring methods may be
unclear in the guidance document, but this information should reasonably be expected to be
included in the methods section of any scientifically valid report.
Industry Recommendation: The Agency should clarify its guidance on this point.
EPA Response: Both the Industry and the Agency agree on this point. The current
guidelines are unclear on this issue. The registrant should submit a validated system with the
description of the scoring method, a proper reference, and the criteria for considering a
chemical positive. This issue will be addressed as part of the harmonization project with
OECD.
7. Rejection Factor; Reporting deficiencies or no Quality Assurance Statement
(10)
EPA Guidance on this Factor: Guidance on this appears in the 40 CFR §160.35,
and in the Phase 3 Technical Guidance document dated December 24, 1989, 5-6. A signed
and dated Quality Assurance Statement should be included specifying the dates of
inspections.
Adequate guidance is provided for this factor and rejection was avoidable on the part
of the registrant; however, the requirement for a Quality Assurance Statement was initiated
later and is excusable for earlier studies dated 1985 through 1987.
Industry Comment: Guidance is clear.
EPA Response: Industry and EPA agree on this point.
81
-------
GUIDELINE 82-1 (a) 90-DAY FEEDING STUDIES - RODENT
1. Rejection Factor; A NOEL was not established (13)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§82-l(a) p. 68), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 98-100. Subdivision F states that "the
lowest dose level should not produce any evidence of toxicity."
Studies rejected on this factor may have been unavoidable.
Industry Comment: "The lowest dose level should not produce any evidence of
toxicity." The term toxicity is most relevant in this rejection factor. There may be evidence
of a "biological response" by the rodent in a 90-day study at the lowest dose level. The
question then becomes whether the biological response is adverse. The determination of the
adversity of a biological response frequently resides with the Agency scientist reviewing the
report. There are responses by a rodent that may give rise to confusion. For example, liver
weight increases in the absence of confirming clinical chemistry alterations or histopathology,
or alterations in clinical chemistry (e.g., lower serum sodium levels). The Agency should
strive to eliminate different interpretations by individual scientists for this rejection factor.
There may be instances when a subchronic study (rodent) legitimately does not have a
no-effect level. This may be dealt with several ways. The registrant may choose not to
repeat the study, rather proceeding to the rodent chronic and use the NOEL in the chronic
study to determine the RfD.
Industry Recommendation: As an alternative, the Agency may consider estimation
of a NOEL using linear regression or use of an extra safety factor. The study should be
judged acceptable if it provides sufficient information for regulatory purposes. The Agency
should strive to eliminate different interpretations by individual scientists for this rejection
factor.
EPA Response: In the following paragraphs is the Agency's guidance on determining
when a NOEL is and is not necessary for toxicology studies (subchronics, chronics,
developmental, reproduction). It is to the registrants benefit to have NOEL's established in
their studies solely for the purpose of the establishing a reference dose (RfD). While the
Agency prefers that NOEL's be established, the lack of NOEL will not be an automatic
rejection criterion in the future. The Agency will attempt to estimate a NOEL based upon
the available data; however, this might not always be possible and rejection might occur.
82
-------
In theory a NOEL would not be necessary if:
1) sufficient additional information is available to make a determination for an
RfD,
2) the study is used only as a range-finder (e.g., a subchronic study for a
chronic/oncogenicity study),
3) it is possible to determine a NOEL using a dose-response analysis (i.e.,
mathematical extrapolation of the toxicological findings) where only a LOEL is
determined,
4) the RfD can be determined with the addition of an extra uncertainty factor
from an LOEL (lowest-observed-effect-level).
With regard to issue number 3, it must be clearly understood that the effect selected
for the dose-response analysis has indeed a dose-response curve which will give a
conservative/reasonable basis for establishing the benchmark dose and subsequent RfD.
Situations in which a NOEL should be determined, after a careful examination of all
available data, include:
1) when chronic (long-term) studies are not required due to use patterns (e.g.,
non-food or antimicrobials) and the subchronic study is the basis for
establishing an acceptable level of exposure,
2) the study is one among several studies available and has the most sensitive
toxicological response noted (e.g., comparison of a subchronic study against
reproductive and developmental studies indicates that they have higher effect
levels),
3) the pesticide is an ADI exceeder, and/or,
4) the test substance is very toxic or has unusual toxic properties which make
extrapolation of a NOEL or the use of additional safety factors a questionable
approach to regulating exposure to the chemical (shape or steepness of dose-
response curve(s)).
Additionally, the Agency will identify methods of estimating NOELs using a
mathematical extrapolation.
83
-------
2. Rejection Factor; Lack of characterization of the test material or incorrectly
reported (purity and/or diet composition and/or stability)
(10)
EPA Guidance on this Factor: Guidance on this topic appears in the Subdivision F
Guidelines (Hazard Evaluation: Human and Domestic Animals) on pages 22, 23, 29, and 30,
and 40 CFR Part §160.105, 160.107, and 160.113 Good Laboratory Practices. The chemical
name, molecular structure, and a quantitative and qualitative determination of its chemical
composition (including impurities and contaminants) should be supplied to the Agency.
Also, the manufacturer's lot number of the test substance with its relevant properties of the
substance tested, such as physical state, pH, stability and purity. In particular, the
Subdivision F Guidelines state (p. 22) that: "If the test or control substance is to be
incorporated into feed or another vehicle, the period during which the test substance is stable
in such a mixture shall be determined prior to the start of the study. No mixture of test or
control substance with the feed or vehicle shall be maintained or used during a period
exceeding the known stability of the test or control substance in the mixture. Alternatively,
determinations of the stability of the test or control substance in statistically randomized
samples of the diet or vehicle mixture shall be made periodically during the study to ensure
that proper mixing, formulation, and storage procedures are being followed and that the
appropriate concentration of the test or control substance is contained in the mixture..."
Also: "If the test substance is incorporated into feed or another vehicle, its
homogeneity and concentration shall be determined prior to the start of the study and
periodically during the study... Statistically randomized samples of the mixture shall be
analyzed to ensure that the proper mixing, formulation, and storage procedures are being
followed, and that the appropriate concentration of the test or control substance is contained
in the mixture."
Industry Comment: The lack of characterization is separate from the factor that the
test substance was not stable or homogeneous in the diet. Characterization data are often not
included in 90-day rodent feeding studies. Frequently, the characterization information is
elsewhere in the submission (e.g., Product Chemistry section), and all that should be
required is to have knowledge that the material has been characterized. It is clear, however,
that characterization data are required. All that should be required in the report is lot
number, batch number, physical state, stability, and percent active ingredient.
We would be interested in a breakout of the number of studies that have been rejected
because of lack of dietary stability/homogeneity data. Individual registrants may approach
the question of stability/homogeneity differently. Currently there is significant latitude in
how the analysis of test substance in diet can be approached.
Industry Recommendation: The Agency may wish to provide more guidance which
provides sufficient latitude to permit use of scientific judgment if this remains a problem
84
-------
area. All that should be required in the report is lot number, batch number, physical state,
stability, and percent active ingredient.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
3. Rejection Factor; Lack of (or insufficient) clinical chemistry and/or lack of (or
insufficient) histopathology (4)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§82-l(a) pp.s 70-71), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 98-100. Subdivision F indicates a number
of clinical biochemistry determinations which should be carried out for rodents at the end of
the test period. The histopathology requirements are specified on pages 74-75.
Industry Comment: Without details, it is difficult to comment on this rejection
factor. However, several of the clinical chemistry and histopathology "suggestions" or
requirements are, in our opinion, unnecessary and should not be cause for rejection (e.g.,
histopathology requirements for liver, lungs, kidneys, and all gross lesions from low and
mid-dose animals even if no lesions were noted in high-dose animals). Please refer to the
NACA Comments on the Subdivision F Guidelines (1/17/91) for additional details.
EPA Response: Only four studies were rejected for these deficiencies. The Agency
agrees that several of the clinical chemistry requirements (particularly serum sodium,
potassium, and chloride) are probably irrelevant. There is no indication that studies have
been rejected for not reporting these particular parameters. The major problem appears to
have occurred with the lack of reporting of histopathology of a target organ in low and mid-
dose animals when there were findings (or possible findings) in high-dose animals.
The Agency is considering a change in the clinical chemistry guidelines in order to
make them more appropriate and useful. Clinical chemistry is considered useful because it
often precedes more major toxicological events.
The Agency has the following general comments on performing clinical chemistry
(both rodent & non-rodent):
1) Clinical chemistry values may vary with bleeding site.
2) Clinical chemistries at the end of a 2 year rodent study are of little value due
to the effects of aging. The Agency is open to suggestions for the sampling
intervals.
85
-------
3) Changes in serum electrolytes should not be used as lexicological endpoints
unless there is an association or correlation with other pathophysiological
activities.
4) Hematology should be assessed using a total picture approach, and the
significance of changes in single parameters should be correlated with other
pathophysiological findings.
5) Reference ranges are variable in rodents with respect to strain and sex.
Comparisons should be made between concurrent controls and dosed animals.
Historical control values are a good assessment of quality control.
The following suggestions are being put forward for comments:
DOGS rNON-RODENTS only):
Hepatic - SGPT
ALP
Total Protein
Albumin
BUN
Total Bilirubin
Cholesterol, triglycerides
For renal function, one should keep in mind there is no singularly sensitive parameter
in the urine that could assess renal damage. Collective tests for renal function include:
BUN, creatinine, N-acetyl beta glucosaminidase (rodents only), specific gravity (following
water deprivation), protein/albumin, and inorganic phosphorus.
The Agency will be proposing changes to the clinical chemistry guidelines for
rodents, which is included below:
RODENTS (only)
In rodents, serum chemistries should be tailored such that a minimum of two tests are
used to assess each of the following:
1) LIVER CELL INJURY: ALT, sorbitol dehydrogenase, AST, ornithine
carboxyl transferase;
2) CHOLESTASIS: 5'-nucleotidase, AP, GOT, and urine bilirubin; and
3) LIVER FUNCTION: bile acids, albumin, and other proteins, direct and
indirect bilirubin.
86
-------
GUIDELINE 82-l(b) 90-DAY FEEDING STUDIES - NON-RODENTS
1. Rejection factor: Reporting deficiencies [Examples; inconsistencies in data in
summary tables, individual animal data, no gross pathology.
results of statistical tests missing, report^errors. misuse of
terms - such as AChE "activity" and "levels"! (9)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§80-4 pages 28-33, and §82-1 pp.s 66-76) and in
the Phase 3 Technical Guidance document dated December 24, 1989, C 98-100. Subdivision
F states "the test report shall include all information necessary to provide a complete and
accurate description and evaluation of the test procedures and results. A test report should
contain at least three parts: a summary and evaluation of the test results; a description of the
test procedures; and the data and information required by each applicable section of this
subdivision." The requirement for individual animal data is not specifically stated in
Subdivision F §82-1, but it is listed in the Phase 3 Technical Guidance document dated
December 24, 1989. This factor is based more on scientific judgment and adherence to
Good Laboratory Practice. Most of these errors appear to be oversights on the part of the
registrant and could be avoided prior to submission.of final reports.
Guidance on this factor is adequate and rejection was avoidable on the part of the
registrant.
Industry Comment: EPA guidance seems adequate.
EPA Response: The Agency and Industry agree on this point.
2. Rejection factor; An investigation^ parameter missing (9)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§82-1 pp.s 66-76), and in the Phase 3 Technical
Guidance document dated December 24, 1989 pp. C 98-100. The Phase 3 Technical
document as well as Subdivision F list all the data points which should be submitted to the
Agency for a subchronic study. Several studies were classified as supplementary because
some of the clinical chemistry, hematology, or histopathology of certain tissues were not
assessed properly.
Sufficient instruction is provided in the guidance document. Rejection was avoidable
on the part of the registrant.
Industry Comment: Without details, it is difficult to comment on this rejection
factor. However, this appears to be similar to Rejection Factor 3 for Guideline 82-l(a).
87
-------
EPA Response: Refer response under Guideline 82-1 (a), Rej. Factor 3.
3. Rejection factor; Lack of characterization of the test substance (i.e.. purity
and/or composition and/or stability) (4)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (pages 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
manufacturer's lot number of the test substance with its relevant properties of the substance
tested, such as physical state, pH, stability and purity, and the identification and composition
of any vehicles used in administering the test substance are required by the Agency. For
each test, control, or reference substance that is mixed with a carrier, tests by appropriate
analytical methods shall be conducted to determine uniformity of the mixture as well as
concentration and stability in the carrier.
Guidance on this factor is adequate and rejection was avoidable on the part of the
registrant.
Industry Comment: See comments under Guideline 82-1 (a), Rej. Factor 2.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
4. Rejection factor; A NOEL was not established (3)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§82-1 pages 66-76) and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 98-100. The guideline states that it is
desirable to have a dose response relationship as well as a NOEL and that the lowest dose
level should not produce any evidence of toxicity.
The guidance document is unclear and the discretion to accept or reject the study on
the basis of not establishing a NOEL is up to the individual reviewer.
Industry Comment: See Comments under Guideline 82-l(a), Rej. Factor 1.
EPA Response: Refer to response under Guideline 82-1 (a), Rej. Factor 1.
-------
5. Rejection factor; Information on the pilot study and other problems
associated with dose level selection (2)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§82-1 pp.s 66-76), and in the Phase 3 Technical
Guidance document dated December 24, 1989 pp. C 98-100. The guidelines indicate that the
highest dose tested should not cause any fatalities in non-rodents, and that it is desirable to
have a dose-response relationship. The reviewers requested information on the pilot studies
to assure the basis for selection of the dose levels for the main studies.
The guidance document is unclear and the discretion to accept or reject the study on
this basis is up to the individual reviewer.
Industry Comment: We agree that guidance is unclear. Although the guideline
states that the high dose should not cause any fatalities in non-rodents, it must be realized by
the Agency that dose selection is often more of an art than a science. No matter how well a
pilot study was conducted, it is almost impossible to guarantee that the high dose level will
cause significant toxicity without any fatalities or that a good dose response relationship will
occur.
Industry Recommendation: Agency reviewers should not reject studies which do
not meet the ideal targets but rather accept studies that meet criteria for an effect and no
adverse effect. This is a good example of individual reviewers exchanging the word
"should" for "must."
EPA Response: Information on pilot studies should be included or appended to the
final report of the definitive study. [If the pilot study was not conducted as a GLP study, the
registrant should indicate either on the pilot study itself or in the front of the study report,
that it is not a GLP study.] This information is useful in aiding the reviewer in deciding if
appropriate dose levels have been selected. If a study shows effects at the lowest test dose,
the pilot study may provide information on a lower dose not showing the effect. On the
other hand, if no effects are noted in a study, the pilot study should provide information
regarding which doses will show an effect.
89
-------
GUIDELINE 82-2 21-DAY DERMAL TOXICITY
1. Rejection Factor; Lack of characterization of the test substance (purity and/or
composition and/or stability) (16)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 pp.s 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
manufacturer's lot number of the test substance with its relevant properties of the substance
tested, such as physical state, pH, stability and purity, and the identification and composition
of any vehicles used in administering the test substance are required by the Agency. For
each test, control, or reference substance that is mixed with a carrier, tests by appropriate
analytical methods shall be conducted to determine uniformity of the mixture as well as
concentration and stability in the carrier.
The guidance in the cited documents is adequate. Rejection was avoidable.
Industry Comment: See comments under Guideline 81-1, Rej. Factor 1.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
2. Rejection Factor; Raw data (See EPA Response) analyses incomplete or
missing (10)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§82-2 pp. 82-83) and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 101-102. As the guideline indicated
summary data shall be submitted in tabular form, showing for each test group the number of
animals at the start of the test, the number of animals showing lesions, the type of lesions
and the percentage of animals displaying each type of lesion.
The guidance in the cited documents is adequate. Rejection was avoidable.
Industry Comment: The criteria (summary data) and the rejection factors (missing
raw data) do not appear to match. Some clarification is required.
EPA Response: The Agency agrees that clarification is needed on this rejection
factor. Agency requirements are that individual animal data and summary tables are to be
included in the study reports. Raw data, however, are not required to be included in the
study, but should be maintained by the sponsor in compliance with the Good Laboratory
90
-------
Practices. [Note: the term "raw" data was used incorrectly in this factor, the correct term
should have been "individual data."]
3. Rejection Factor; A systemic NOEL was not established (6)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§82-2 p. 78), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 101-102. The guideline recommends that
the lowest dose tested should not produce any evidence of toxicity.
Industry Comment: See Comments under Guideline 82-1 (a), Rej. Factor 1.
We agree that the guidance document is clear, however, some flexibility is warranted
in accepting or rejecting this study for lack of a NOEL. For example, no NOEL should be
required for an effect seen in the oral studies, so long as the oral route is more or equally
sensitive (for the effect in question as established by the dose response curve) and a NOEL
was established in the oral study. Also, Agency policy on the need for dermal NOEL in this
study needs clarification. Although the analysis does not indicate a problem several
registrants have reported some problems.
Industry Recommendation: Some flexibility is warranted in accepting or rejecting
this study for lack of a NOEL. Agency policy on the need for dermal NOEL in this study
needs clarification.
EPA Response: Refer to response under Guideline 82-1 (a), Rej. Factor 1. For
further guidance, a systemic NOEL is preferred, but a dermal NOEL is not required and will
not be the basis for future rejections.
4. Rejection Factor; Inadequate percentage of body surface area exposed in each
dose group (5)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§82-2 p. 79), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 101-102. The Subdivision F indicates that
the test substance should be applied evenly over an area which is approximately 10% of the
total body surface area. If a substance is known to be highly toxic, the surface area covered
may be less.
The guidance in the cited documents is adequate. Rejection was avoidable.
Industry Comment: We agree that guidance is adequate, however, the long duration
of the study sometimes presents operational problems because the dose is repeatedly applied
91
-------
to the same area of skin. Use of rabbits may allow dosing areas to be alternated to prevent
unusual levels of skin damage.
In addition, there may be uncertainty or disagreement as to how this area is to be
determined. Therefore, we suggest that the Agency indicate the approximate areas (cm2) to
be treated for both species (rat and rabbit) that are typically used for these studies. Approxi-
mate areas corresponding to 10% of the body surface area of the rabbit and rat are listed
below:
Approx. Area Equiv.
to 10% Body Surface
Species Approx. Body Wt. Range (cm2)
Rabbit 2 to 3 kg 159 to 231
Rat 200 to 300 g 32 to 42
The recommended body surface area numbers have been derived from the surface law
equation, s=kMb°-67 and the values of the Meeh coefficient, k, for the species of test animals
(Knut, et ah, 1984, Scaling: Why is Animal Size so Important?. Cambridge University
Press, Cambridge, England, pp. 75-89).
Industry Recommendation: We suggest that the Agency indicate the approximate
areas (cm2) to be treated for both species (rat and rabbit) that are typically used for these
studies.
EPA Response: The Agency is suggesting that test animals (Sprague Dawley,
Wistar, and Fischer 344) should be in the 6-10 week age range when received from
supplier and no older than 8-12 weeks of age at initiation of dosing for acute studies. Also,
the Agency is suggesting that in order to dose approximately 10% of the body surface, the
area starting at the scapulae (shoulders) to the wing of the ileum (hipbone) and half way
down the flank on each side of the animal should be shaved.
5. Rejection Factor; Insufficient number of dose levels tested (3)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§82-2 p. 78), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 101-102. The guidelines recommend at
least 3 dose levels, with a control and, where appropriate, a vehicle control group to be used
per study.
The guidance in the cited documents is adequate. Rejection was avoidable.
92
-------
Industry Comment: We agree guidance is adequate, however, in the interest of
minimizing use of animals, the Agency may want to amend the guidelines to allow the
reviewer some latitude with respect to accepting studies conducted with less than 3 doses or
without a solvent control.
Industry Recommendation: See comments.
EPA Response: The Agency requires three treated dose levels plus a vehicle (or
solvent) control in order to obtain a dose-response, unless the limit dose of 1 g/kg/d is used
for the study.
93
-------
GUIDELINE 82-3 90-DAY DERMAL TOXICITY
1. Rejection Factor; Lack of characterization (purity and/or composition and/or
stability) of the test substance (5)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 pp.s 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
manufacturer's lot number of the test substance with its relevant properties of the substance
tested, such as physical state, pH, stability and purity, and the identification and composition
of any vehicles used in administering the test substance are required by the Agency. For
each test, control, or reference substance that is mixed with a carrier, tests by appropriate
analytical methods shall be conducted to determine uniformity of the mixture as well as
concentration and stability in the carrier.
Adequate guidance was provided for this guideline, therefore this was an avoidable
rejection factor.
Industry Comment: See comments under Guideline 81-1, Rej. Factor 1.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
2. Rejection Factor; A systemic NOEL was not established (5)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§82-3 p. 85), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 103-105. The guideline recommends that
the lowest dose tested should not produce any evidence of toxicity. Also the guideline states
it is desirable to have a dose response relationship as well as a NOEL.
Industry Comment: See comments under Guideline 82-2, Rej. Factor 3.
EPA Response: Refer to response under Guideline 82-l(a), Rej. Factor 1.
3. Rejection Factor; Incomplete/missing raw animal data analyses (4)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§82-2, p. 91), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 103-105. The guideline states that
94
-------
summary data shall be submitted in tabular form, showing for each test group the number of
animals at the start of the test, the number of animals showing lesions, the type of lesions
and the percentage of animals displaying each type of lesion.
Adequate guidance was provided for this guideline, therefore this was an avoidable
rejection factor.
Industry Comment: See comments under Guideline 82-2, Rej. Factor 2.
EPA Response: Refer to response under Guideline 82-2, Rej. Factor 2.
4. Rejection Factor; Insufficient number of dose levels tested (2)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§82-2 p. 85), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 103-105. The guideline states at least 3
dose levels, with a control and, where appropriate, a vehicle control group, should be used
in each study.
Adequate guidance was provided for this guideline, therefore this was an avoidable
rejection factor.
Industry Comment: See comments under Guideline 82-2, Rej. Factor 5.
EPA Response: Refer to response under Guideline 82-2, Rej. Factor 5.
5. Rejection Factor: Poorly controlled test environment (2)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 p. 25) and the 40 CFR Part §160.41.
Each testing facility shall be of suitable size and construction to facilitate proper conduct of
studies. Structural requirements and environmental control of rooms housing animals should
be well regulated.
Adequate guidance was provided for this guideline, therefore this was an avoidable
rejection factor.
Industry Comment: We agree, guidance is adequate.
EPA Response: The Agency and Industry agree on this point.
95
-------
GUIDELINES 83-l(a) AND 83-2(a) CHRONIC FEEDING/
ONCOGENICITY - RATS
1. Rejection factor; Missing histopathologv information (29)
EPA Guidance on this Factor: Guidance on this topic appears in the 1984
Subdivision F guidelines, as well as the Standard Evaluation Procedure (Subchronic and
Chronic Exposure) pp.s 30-31, (EPA-540/9-85-020) June, 1985, and the Standard Evaluation
Procedure (Oncogenic Potential) p. 15, October, 1985, and the Phase 3 Guidance Document
dated December 24, 1989, pp.s. D-6 and D-7, and the Acceptance Criteria for Subdivision
F, page 39 of a memo dated November 7, 1989. A signed pathology report, which includes
both gross and microscopic histopathology is required for a chronic and/or oncogenicity
study in rats. Summary tables for gross and microscopic pathology must be presented, as
well as the individual data preferably in the form of an individual pathology sheet for each
animal. All tissues listed on pp.s. 114 and 115 of the Subdivision F guidelines should be
examined histopathologically, or an explanation given as to why a tissue should be excluded,
should accompany the pathology report. An "adequate" number of tissues should be
examined histopathologically. An "adequate" number is not defined precisely in the
Subdivision F guidelines, except to say "all" tissues. The FDA 1982 "Red Book" p. 57
defines the criteria for an acceptable negative lifetime study as no more than 10% of any
group (animals or tissues) being lost to autolysis, cannibalism, or management problems.
Subdivision F guidelines (p. Ill) states there should be no more than 10% of the animals in
any test group lost from the test due to cannibalism, autolysis of tissues, misplacement, and
similar management problems. The actual number of tissues examined for each organ/dose
for histopathology should be given. Histopathology, according to the Subdivision F
guidelines, p. 115, should be performed on all tissues for all rats in the control and high dose
groups and target organs in all animals. Supplementary classifications of studies has
occurred when the reviewers consider a response seen at the high dose an effect, and request
histopathology on the low and mid dose tissues of that organ.
Guidance in the cited documents appears adequate. However, when there is a
difference of opinion on what is considered an effect, the factor is difficult to call avoidable.
Industry Comment: Agency guidance seems adequate, however, the cited guidance
dated November 7, 1989 does not appear to be available to registrants.
EPA Response: This memorandum is the acceptance criteria document for
Subdivision F guidelines, dated November 7, 1989, which was appended as part of the
references for the Rejection Rate document.
96
-------
2. Rejection Factor; Missing information in study reports (22)
EPA Guidance on this Factor: Guidance on this topic appears in the 1982
Subdivision F Guidelines, as well as the Standard Evaluation Procedure (Subchronic and
Chronic Exposures) pp.s. 18-30, the Acceptance Criteria for Subdivision F, p. 39, of Nov.
7, 1989 memo, and in the Phase 3 Technical Guidance Document dated December 24, 1989,
p. D-8. Many studies were called "supplementary" because clinical chemistry information
was missing, or not tabulated in summary form, or not accompanied by individual data.
Guidance as to which clinical chemistry parameters should be investigated are clearly
tabulated on p. 112 of the Subdivision F guidelines. Food consumption should be
determined weekly during the first 13 weeks of a study, and then approximately monthly
thereafter (Subdivision F guidelines, p. 111). Clinical observations, survival and mortality,
body weights along with body weight gains should accompany the submitted study as
outlined on pp.s 110-111 of Subdivision F guidelines. Urinalysis and ophthalmologic exams
should also accompany the study report (Subdivision F p. 113). Individual data as well as
statistical analyses are very briefly outlined on p. 117 of the Subdivision F guidelines, and
should also accompany the study report.
Guidance in the above cited documents appears clear. These rejections could be
avoided.
Industry Comment: Without more details it is difficult to comment. However, as
presented in NACA's Comments on the Subdivision F Guidelines (1/17/91), there are several
clinical pathology and histopathology requirements that, in our opinion, are not necessary and
should not be required. For example, the lack of urinalysis is cited as a reason for rejection.
It is difficult to understand how a study can be rejected because of the lack of a minor
determination that is not even required in the 90-day study. The scientific utility of urinaly-
sis data is very questionable. Studies should only be rejected if they have a significant flaw
that would jeopardize the scientific merit of the study. The omission of urinalysis data from
a chronic rodent study would not meet this criteria. We recommend that the Agency
harmonize its guidance in one document and provide guidance to reviewers on which
measurements are crucial and which may be discretionary.
Industry Recommendation: There are several clinical pathology and histopathology
requirements that, in our opinion, are not necessary and should not be required. Studies
should only be rejected if they have a significant flaw that would jeopardize the scientific
merit of the study. We recommend that the Agency harmonize its guidance in one document
and provide guidance to reviewers on which measurements are crucial and which may be
discretionary.
EPA Response: The Agency agrees that clarification is necessary concerning missing
information. The Agency is considering a change in the clinical chemistry guidelines (please
refer to 82-l(b) Rejection Factor #2). This will also be addressed in the OECD
harmonization.
97
-------
3. Rejection Factor; MTD was not achieved (20)
EPA Guidance on this Factor: The Toxicology Branches have encountered many
carcinogenicity studies which were considered tested at too low a dose to evaluate the
carcinogenic potential of a chemical. The branches' 1987 MTD document attempted to
define the concept of a "maximum tolerated dose" or MTD in order to bring greater
consistency of review, decreasing the subjectivity of the term, and let the regulated industry
know what was acceptable concerning dose selection for carcinogenicity studies. Many
studies were started, unfortunately, before the policy was made publicly available, were
submitted to the Agency after the policy was being used, and were classified as
"supplementary" because they (and their subchronic range-finding studies) had not achieved
an MTD. In addition, confusion concerning interpretation of some of the endpoints on the
regulated industry's part led to other studies being considered inadequate. The number of
studies in each year that were called "supplementary" for this rejection factor reflects this
early confusion. For this rejection factor 3/21 studies were considered "supplementary" in
1985, 15/21 in 1986-1988, and 3/21 in 1989-1991. Very few studies in the last several years
have been considered tested at too low a dose to assess carcinogenicity.
Guidance for this factor was not good until 1987. Thereafter, studies rejected for this
factor increased as mentioned above, but are now very low. This factor is avoidable,
especially if the registrant submits subchronic data which will help to choose the appropriate
doses for the long-term studies.
Industry Comment: Although the rejection rates for CORT studies remain low,
there are serious concerns over the Agency's criteria for the selection of the highest dose
(maximum tolerated dose or MTD). The Agency has determined that a major rejection
factor in chronic feeding/carcinogenicity studies is failure to meet current maximum tolerated
dose criteria. Controversy over use of the MTD in carcinogenicity bioassays and its
relevance to human risk assessment has been high in recent years. The MTD issue has been
a major focus of the National Academy of Science (NAS) Committee on Risk Assessment
Methodology (CRAM), as well as the National Toxicology Program's (NTP) Board of
Scientific Counselors who have suggested that the NTP reexamine its criteria for setting the
top dose. In this regard, the NACA Toxicology Roundtable prepared a paper entitled
"Comments on the Use of Maximum Tolerated Dose for Carcinogenicity Bioassay" and
submitted to the NAS CRAM Committee on May 14, 1991 for its consideration. A copy of
this paper, which summarizes registrants views on MTD testing, is attached (see Attachment
4).
The controversy over use of the MTD revolves around trying to maximize the
sensitivity of the rodent bioassay on one hand while attempting to identify relevant human
carcinogens on the other hand. Some of this controversy has been generated by inconsistent
definitions or interpretations of MTD. Some scientists and regulators (OECD) feel the
highest dose tested in a carcinogenicity bioassay should be a "Minimally Toxic Dose" rather
than a "Maximum Tolerated Dose." This view is taken because dose levels which produce
98
-------
physiologically or metabolically compromised animals are much less relevant to hazard
identification. Doses that exceed pharmacokinetic saturation level are often not relevant for
dose-response extrapolation to the much lower dose levels which humans are exposed.
Therefore, the criteria for selecting the highest dose in a bioassay should focus on identifying
the maximum dose level relevant to assessment of risk at human exposure levels. In practice
this can mean identifying a maximum dose at which the test animals' health is not compro-
mised by exposure to the test substance or one which is in a linear pharmacokinetic range or
one that represents some multiple of exposure levels.
Changing the MTD concept would help reduce false positive findings and identifi-
cation of hazards that are not relevant to human exposure. A review of National Toxicology
Program (NTP) studies published in 1985 revealed that 2/3 of chemicals judged positive were
considered carcinogenic based on effects observed solely at the highest dose (MTD) level.
Thus about 2/3 of the chemicals considered positive would have been considered negative if
the maximum dose tested had been one-half MTD. Follow up publications disclosed that this
number was somewhat smaller but still very large. These reviews also ignore to some extent
the fact that in some tests, animals were severely stressed at the lower doses too (i.e., the
mid-dose may also be an MTD). Some implicit assumptions used by regulators in interpret-
ing results from studies conducted at the MTD include the following: (a) pharmacokinetics
are not dose dependent, (b) dose-response is linear, (c) DNA repair is not dependent on
dose, (d) response is not age dependent, and (e) a test need not bear a relationship to human
exposure. In addition, the interpretation of these studies assumes that a non-threshold
mechanism of action exists. These assumptions are often not valid. We believe many results
at the MTD are "false positives" and/or not relevant to human hazard. Our view, which is
shared by many others in the scientific community, is that it makes sense to alter the
carcinogenicity bioassay by testing at doses lower than an MTD (as currently defined by the
EPA) when justified on the basis of human exposure considerations. A "minimally toxic
dose" as clarified by OECD may be most appropriate as it is widely used and has the support
of many scientists.
Using lower top doses would not compromise safety or the ability to detect human
health hazards. Apostolou (1990) has presented data which suggests that most of the known
human carcinogens can be detected in animals at dosage levels well below the MTD. There
is therefore some reason to believe that a reduction in the highest dose tested would be
sufficient to detect important human health hazards and at the same time not detect carcino-
genic hazards which are insignificant. It would be easy to accept "false positives" as
conservative regulatory values if the Agency has the discretion to use them in an appropriate
manner. However, in many cases (i.e., Delaney) the identifications are misused at consider-
able cost to the American consumer, farmers and industry.
Since the MTD is chosen empirically, controversy over its definition and interpreta-
tion is difficult to resolve. When a carcinogenic response occurs at toxic doses, the finding
should be considered irrelevant to human hazard identification and risk assessment, especially
if target organs appear to be compromised during the study. EPA should seriously consider
99
-------
revising its guidelines to use a minimally toxic dose or even use some other mechanism to set
doses in oncogenicity studies. This would be consistent with existing international schemes
and recent proposals at NTP. Registrants would welcome the opportunity to further discuss
the MTD issue with the Agency, including the relevance of pharmacokinetics and metabolism
data in the dose selection process, in a workshop or open forum.
Finally, the Agency requested the registrant's view of why some studies listed as
supplementary have taken in some cases four to seven years to upgrade. We essentially
agree with the Agency's analysis. At times the Agency is at fault due to substantial delays in
review or failure to communicate the need for further information or data to the registrant
after the review is completed. On the other hand, the registrant's response may be delayed.
Regardless of who is at fault we agree with the Agency's belief that these substantial time
lapses more accurately reflect past Agency/registrant performance than current performance.
Hopefully, this process will significantly improve the time from submission of studies to
review and acceptance.
EPA Response: Health Effects Division (HED) has committed to holding a
workshop with ILSI to discuss alternative approaches to setting the upper dose for
carcinogenicity studies in rodents. In addition, the NAS CRAM committee has issued their
recommendations to the Agency on the MTD issue. In conjunction with this endeavor, the
Agency is in the process of revising the metabolism guidelines which will contribute to these
decisions. The Agency encourages registrants to discuss this issue with them on a case-by-
case basis.
At the meeting in February (between EPA and Industry) there was discussion
regarding Industry's perception that different EPA reviewers conclude differently on studies
or that the outcome of the study changes after the Cancer Peer Review. EPA responded that
the Cancer Peer Review/RfD committee should help to assure consistency among the
reviewers, since the Cancer Peer Review/RfD committee meetings take place after the Data
Evaluation Records for the studies are completed. HED has resolved that the reviews for
high priority items and the RfD meetings will coincide; and the reviews with their
accompanying RfD will go out to the Registration Division (RD) together.
Until the MTD issue is resolved, registrants are advised by EPA to meet with EPA
to discuss dose levels based on the 90-day subchronic studies.
4. Rejection Factor; Missing historical control data (14)
EPA Guidance on this Factor: Guidance on this topic appears partially in the
Subdivision F guidelines, pages 9, 24, and 124, and more completely in the Phase 3
Technical Guidance document dated December 24, 1989, p. D-7. As stated in the Phase 3
Technical Guidance document, historical control data must be provided where possible for
neoplastic lesions which are significantly increased in treated animals. Also, particularly
100
-------
unusual non-neoplastic lesions identified in treated animals must be accompanied by historical
control data. This historical control data should be presented by individual studies, tumor
types, laboratories and sex (i.e., data presented as a range are not sufficient). It is felt by
the review staff that if historical control data are needed for a particular study, then the study
should be considered "supplementary" until the historical data are submitted and reviewed.
Otherwise there is little incentive for the regulated industry to generate and submit the data
requested.
Guidance for this factor is clear; however, the need for historical control data can be
a matter of interpretation. This rejection factor may be difficult to avoid.
Industry Comment: Historical control data are extremely valuable. However, the
current EPA guidance is not sufficient to determine when the historical control incidence data
will be required and when they will not be necessary. Historical control data are often
requested by reviewers even when there are no specific findings which are increased in
treated animals. If a strain other than the Sprague-Dawley rat is used in these studies,
reviewers often request historical control data when control and treated data are similar but
show deviations from the typical Sprague-Dawley incidence. The guidelines indicate that
historical data should be submitted when pertinent. It is difficult for a laboratory to
determine when these data are pertinent. This is especially the case, if they are requested,
even when there are no increased incidences above the controls in treated animals, and only
if the incidence deviates from some EPA standard incidence.
Currently, if registrants submit historical control data from the laboratory animal
vendor or from the literature (other labs), the EPA has not accepted this data. They consider
this type of data to be of limited value. We disagree, because this information from mainly
the vendor is the largest database that exists for neoplastic and non-neoplastic tumors and
should be used in data evaluation.
Industry Recommendation: Since as the Agency states, "...the need for historical
control data can be a matter of interpretation," we would suggest that the types of historical
data that are useful and the types which should be routinely submitted should be discussed
with registrants.
EPA Response: Historical control data are extremely valuable to the reviewers in
helping to determine whether an increase is actually a treatment-related effect. Often a study
with only concurrent controls can not resolve some of these issues. Sometimes historical
control data are requested for studies where animals are not familiar to the reviewer, because
a low or slight response is difficult to interpret. Familiarity with all strains of animals, both
U.S. and European, is difficult to achieve. The Agency would like to see historical data on
positive and borderline responses and will write a requirement for this into the revised
guidelines as part of the OECD harmonization process. However, in the interim, there are
several approaches that can be taken either individually or concurrently which could keep
101
-------
studies from being delayed because of this factor. Additionally the Agency has agreed to
entertain comments on clinical chemistry and histopathology.
1) The Agency could receive historical control data from the testing laboratories on a
regularly updated basis, and refer to this data bank in-house when a question arose on
a particular parameter from that laboratory. This would eliminate the need to ask the
registrant for the data, except under rare circumstances.
2) The reviewer, along with the section head, could contact the product manager and
contact the registrant directly for the necessary information, giving a grace period of
(for instance) one month before declaring the study supplementary until historical
control data came in. These data could be faxed to the reviewer directly as well as
submitted through regular channels to the product manager for the record.
In the meantime, the Agency has started to analyze the in-house carcinogenicity data
to compile historical controls from studies received by registrants on pesticide compounds.
For analysis of data, the most important information is still the concurrent control
data, followed by the laboratory's own historical controls, followed by the vender's historical
controls. This is also in agreement with NTP's use of these data.
Listed below is what is generally requested for historical control tumor incidence
data:
1) data listed by study (with study date listed),
2) historical control data for the same testing facility,
3) data using the same strain of animals, (substrain identified if different),
4) data for a time period (when possible) for 2 years prior to and subsequent to
the study of concern,
5) the N of the historical control studies,
6) data should be separated by sex,
7) tumor data on benign, malignant and benign/malignant combined,
8) data using the same route of administration,
9) data from studies of the same duration,
10) data from studies using the same vehicle where possible or at least vehicle
noted, and,
11) individual studies should be reported as well as the ranges (mean and median)
of responses for summary data.
102
-------
5. Rejection Factor: Lack of characterization (purity and/or composition and/or
stability) of the test substance (11)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (pages 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
manufacturer's lot number of the test substance with its relevant properties of the substance
tested, such as physical state, pH, stability and purity, and the identification and composition
of any vehicles used in administering the test substance are required by the Agency. For
each test, control, or reference substance that is mixed with a carrier, tests by appropriate
analytical methods shall be conducted to determine uniformity of the mixture as well as
concentration and stability in the carrier.
Guidance on this factor is clear and a rejection could be avoided.
Industry Comment: See comments under Guideline 82-l(a), Rej. Factor 2.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
6. Rejection Factor; Deficiencies in reporting the study data (8)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, pages 28-33) and in the Phase 3
Technical Guidance document dated December 24, 1989, D-10. Reporting of the study data
is essential in the evaluation or interpretation of the results of a study. Summary tables may
be missing, or difficult to interpret. Calculation and technical errors may exist and need
clarification. Several studies were rejected because the summary tables did not reflect the
information in the individual animal data, and the data had to be recalculated by the
registrant in order to make the tables interpretable.
Guidance on this point is not entirely clear. Avoidance of a rejection on this factor is
difficult because it is subject to interpretation.
Industry Comment: We agree that guidance and industry practice are not always
clear.
EPA Response: The Agency and Industry agree on this point.
103
-------
GUIDELINE 83-l(b) CHRONIC FEEDING - DOG
1. Rejection factor: Reporting deficiencies or missing data [Examples; some of
the clinical chemistry or hematologv data are missing, lack
of a signed Quality Assurance Statement or Good Laboratory
Practice Statement, statistical procedures not reported or
referenced, individual animal data missing, inadequate
description of an assay procedure (i.e.. ChE/AChE).
questions on elapsed time between sacrifice and assays, no
histopathologv on low and/or mid dose animals (when
indicated)! (15)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 pages 28-33 and §83-1 pages 107-117),
and in the Phase 3 Technical Guidance document dated December 24, 1989, C 111-113.
Subdivision F states "the test report shall include all information necessary to provide a
complete and accurate description and evaluation of the test procedures and results. A test
report should contain at least three parts: a summary and evaluation of the test results; a
description of the test procedures; and the data and information required by each applicable
section of this subdivision."
The guidance on this factor is clearly stated and rejection was avoidable on the part of
the registrant.
Industry Comment: EPA guidance seems adequate.
EPA Response: The Agency and Industry agree on this point.
2. Rejection Factor: A NOEL was not established (13)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§83-1 pp.s 108-109), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 111-113. The guideline indicates it is
desirable to have a dose-response relationship as well as a NOEL and that the lowest dose
tested should not produce any evidence of toxicity.
The guidance document is unclear; therefore, this rejection factor was unavoidable.
Industry Comment: We agree that guidance is not clear on the need for a NOEL in
this study. Under some circumstances, a NOEL should not be required for this study (for
example: dog is less sensitive than a species that shows a NOEL for the same effect as
evidenced by the dose-response curves). See also comments under Guideline 82-l(a), Rej.
Factor 2 (p. 17).
104
-------
Industry Recommendation: The Agency should provide guidance to reviewers con-
cerning those instances in which a NOEL is not established and the study is acceptable.
EPA Response: Refer to response under Guideline 82-1 (a), Rej. Factor 1.
3. Rejection factor: Lack of characterization of the test substance (i.e. purity
and/or composition and/or stability) (11)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 pp.s 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
manufacturer's lot number of the test substance with its relevant properties of the substance
tested, such as physical state, pH, stability and purity, and the identification and composition
of any vehicles used in administering the test substance are required by the Agency. For
each test, control, or reference substance that is mixed with a carrier, tests by appropriate
analytical methods shall be conducted to determine uniformity of the mixture as well as
concentration and stability in the carrier.
Guidance is adequate for this factor and rejection was avoidable on the part of the
registrant.
Industry Comment: See comments under Guideline 82-l(a), Rej. Factor 2.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
4. Rejection factor; Historical control or similar data sought (9)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 pp.s 9, & 24), and in the Phase 3
Technical Guidance document dated December 24, 1989 p. D-7. Since dog studies have a
limited number of animals per sex per dose group, when the presence of 1 or 2 lesions are
found in the treated groups more than in the control group it is often difficult to interpret
without a knowledge of the spontaneous occurrence of that lesion. The historical controls,
while not run concurrently with the test substance under study, yield valuable comparisons on
whether the test animals are typical for their species and strain for various incidences and
statistical measures for each laboratory. Historical control data are not required to be
submitted with the study, but such data should be made available upon the Agency's request.
The guidance document is unclear; therefore, this rejection factor was unavoidable.
105
-------
Industry Comment: We agree with the Agency that guidance on this factor is
unclear and that discrepancies between reviewers occur. Historical control data can be very
useful in studies where there are large numbers of animals in order to establish the true
spontaneous incidence of a response.
See also comments under Guidelines 83-1 and 83-2(a), Rej. Factor 4.
EPA Response: Refer to the response under guideline 83-1 (a) & 83-2(a), Rej.
Factor 4.
5. Rejection factor; AnLJnyestigational parameter missing (9)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§83-1 pp.s 107-117), and in the Phase 3
Technical Guidance document dated December 24, 1989 pp. C 111-113. The Phase 3
Technical document as well as Subdivision F list all the data points which should be
submitted to the Agency for a chronic study.
Guidance on this factor is clearly stated in the cited documents and rejection was
avoidable on the part of the registrant.
Industry Comment: Guidance is adequate.
EPA Response: The Agency and Industry agree on this point.
6. Rejection Factor: Dose selection, pilot study problems, or test animal problems
(5)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§83-1 pp.s 107-117), and in the Phase 3
Technical Guidance document dated December 24, 1989 pp. C 111-113. These factors
included questionable health of the dogs, source of dogs was not provided and/or no systemic
effects were noted at high dose levels and basis for its selection questioned. The guidelines
indicate that the highest dose tested should not cause any fatalities in non-rodents, and that it
is desirable to have a dose-response relationship.
The guidance document is unclear; therefore, this rejection factor was unavoidable.
Industry Comment: We agree that guidance on this factor is not clear. The Agency
should issue guidance that allows reviewers some discretion to accept studies if adequate
information for regulatory purposes is available, even when no effects were noted at high
dose levels etc.
106
-------
See also comments under for Guideline 82-l(b), Rej. Factor 5.
EPA Response: Refer to response under Guideline 82-l(b), Rej. Factor 5.
107
-------
GUIDELINE 83-2(b) ONCOGENICITY - MICE
1. Rejection Factor; Histopathologv information missing (22)
EPA Guidance on this Factor: Please refer to rejection factor #1 for rat chronic/
oncogenicity studies [guideline 83-1 (a)].
Industry Comment: Several factors are cited in the review which are required for
chronic studies but are not required for the mouse oncogenicity study. In general guidance is
adequate in this area.
EPA Response: Refer to response in Guidelines 83-l(a) & 83-2(a), Rej. Factor 1.
2. Rejection Factor: MTD was not achieved (18)
EPA Guidance on this Factor: Please refer to rejection factor #3 for rat
chronic/oncogenicity studies [guideline 83-1 (a)]. The confusion seen in the chronic/onco rat
studies was also reflected in the numbers of mouse studies rejected in the years 1985-1991.
For this factor, 2/18 studies were considered "supplementary" in 1985, 14/18 in 1986-1988,
and 2/18 in 1989-1991.
Industry Comment: See comments under Guidelines 83-1 and 83-2(a), Rej.
Factor 3.
EPA Response: Refer to response in Guidelines 83-l(a) & 83-2(a), Rej. Factor 3.
3. Rejection Factor; Lack of historical control data (14)
EPA Guidance on this Factor: Please refer to rejection factor #4 for rat chronic/
oncogenicity studies [guideline 83-1 (a)].
Industry Comment: See Guidelines 83-1 and 83-2(a), Rej. Factor 4.
EPA Response: Refer to response in Guidelines 83-l(a) & 83-2(a), Rej. Factor 4.
4. Rejection Factor; Information missing in study reports (9)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§83-1, pp.s 121-125), the Standard Evaluation
Procedure (subchronic and chronic exposure) pp.s 18-30, June 1985, and the Phase 3
108
-------
Guidance Document dated December 24, 1989, p.s. D-7. For carcinogenicity as well as
chronic studies food and water consumption should be determined weekly during the first 13
weeks of the study and then approximately monthly thereafter. Organ weights should be
taken at necropsy and are often missing in the final reports of mouse carcinogenicity studies.
NTP studies have been submitted to the Agency, but are not considered acceptable for
regulatory purposes, since so much information is missing from them. For example, they
lack organ weights, food and water consumption and clinical chemistry.
Guidance on this point is clear. These rejection factors could be avoided.
Industry Comment: Some factors are cited which are required for chronic studies
but are not required for the mouse oncogenicity study. For example organ weights are only
required for liver, kidney, brain and testes at term in 10 animals/sex in the oncogenicity
study. There is also no requirement for clinical pathology information other than a blood
smear and differential.
In the EPA comments, it is stated that food and water should be measured during the
first 13 weeks. The guidelines clearly state food or water should be measured. It has
always been assumed that this requirement was associated with the vehicle used for the test
substance. It is difficult, if not impossible, to get adequate water consumption data in mice.
In view of the limited importance of this parameter, it is difficult to understand why EPA
would reject a carcinogenicity study for not having water consumption data.
See also comments under Guideline 83-1, Rej. Factor 2.
EPA Response: Refer to response in Guideline 82-l(a), Rej. Factor 3 and Rej.
Factor 2 in Guideline 82-l(b). However, if water consumption was requested for this type of
study, it was because it was a drinking-water study, and the actual dose given could not be
calculated. This would also apply to food consumption, if it was requested.
5. Rejection Factor; Lack of characterization (purity and/or composition and/or
stability) of the test substance (8)
EPA Guidance on this Factor: Please refer to rejection factor #5 for rat chronic/
oncogenicity studies [guideline 83-l(a)].
Industry Comment: See comments under Guideline 82-l(a), Rej. Factor 2.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
109
-------
6. Rejection Factor; Deficiencies in reporting of study data (6)
EPA Guidance on this Factor: Please refer to rejection factor #6 for rat
chronic/oncogenicity studies [guideline 83-1 (a)]. In addition to those comments made on the
rat studies, tremendous data variation on particular endpoints have caused concern to
reviewers who have asked for clarification when they see deficiencies. Also, if data points
such as tumors or organ weights are excluded from final counts, those reasons for exclusion
should be detailed in the study.
Industry Comment: See Guideline 83-1 and 83-2(a), Rej. Factor 6.
EPA Response: The Agency and Industry agree on this point. Also, some further
guidance is given in Guideline 83-l(a) and 83-2(a), Rej. Factors 1 and 4.
110
-------
GUIDELINE 83-3(a) DEVELOPMENTAL TOXICITY - RODENTS
1. Rejection Factor; Historical controls are needed from the performing
laboratories to aid in the assessment of potential
developmental effects (20)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 & §83-3, pages 9, 24, & 129), and in the
Phase 3 Technical Guidance document dated December 24, 1989, D-10. Historical control
data are not required to be submitted with the study, but such data should be made available
upon the Agency's request. The historical controls, while not run concurrently with the test
substance under study, yield valuable comparisons on whether the test animals are typical for
their species and strain for various incidences and statistical measures for each laboratory.
Sometimes the best rationale for interpretation of skeletal or visceral variations or
malformations or any other apparent effects is with a comparison with the historical control
data collected from various studies with the same species, strain of animals, vehicle, and
conducted in a similar time frame. Other examples in which the Agency needs historical
control data are when a decision is needed on the justification of a non-treatment-related
status of a significant increase of a particular effect, or backing up a judgement based on
scientific interpretation, or justifying a false positive that can occur because the concurrent
controls are unusual for a particular finding (low incidence).
Currently, the guidance does not specifically require historical controls to be
submitted with the study. But if the registrant is trying to explain a specific fetal fmding(s)
and they use the historical controls for justification, then historical control data should be
appended to the study report. This rejection factor was not always "avoidable" on the part of
the registrants.
Industry Comment: We do not feel that this factor is always avoidable. See
comments under Guideline 83-1 and 83-2(a), Rej. Factor 4 (p. 23).
We agree with the Agency that historical controls are important in aiding in the
interpretation of findings.
It has not been conventional or required for registrants to submit historical control
data in reproduction studies except when the study director has relied on it for interpretation
of a particular parameter. Comprehensive submission of raw data on all measured
parameters would be too bulky for routine submission. Are there specific parameters such as
fertility data, which it would be helpful to routinely provide?
The EPA Guidance on this factor notes that historical control data may be used in
"justifying a false positive." We believe that the data may be used to help determine if an
111
-------
apparent change is likely to be treatment related or to avoid a false conclusion (negative or
positive).
EPA Response: The subject of historical control data will be addressed more fully in
the Developmental Standard Evaluation Procedure presently under draft. The uses of such
data is subject to professional judgement but is often required for several situations (e.g.,
positive data which is not necessarily dose-related, determination of the background rate for
common skeletal or visceral variations or findings, background rate for uncommon variations
or malformations, determination of genetic drift or changes in necropsy/examination
procedures for identifying developmental effects).
The Agency encourages all testing laboratories to develop historical control data for
their own benefit as well as for use in interpretation of specific studies. Inevitably, historical
control data will be requested from a testing lab for certain test substances. (See related
comments on historical control data for chronic/carcinogenicity studies, guideline 83-1 (a),
Rej. Factor 4.)
The following parameters should be included for developmental toxicity data (taken in
part from Table 9, Feusnner et al. (1992), Teratology 46:349-365):
Number of animals tested
Pregnancy rate
No. does/dams copulated or inseminated
No. does/dams pregnant
Pregnancy rate
Implantation frequency
No. does/dams
No. implantations
Implantations per doe/dam (Mean ± S.D.)
Percentage values for abortions
Percentage of abortions (and premature deliveries) expressed in terms of
implantations (excludes those that died)
Percentage values for resorptions
Does/dams with early resorptions
Does/dams with late resorptions
Does/dams with early and/or late resorptions
Early resorptions expressed as percentage of implantations
Late resorptions expressed as percentage of implantations
Early and late resorptions expressed as percentage of implantations
112
-------
Number of Caesarean-delivered fetuses
No. does/dams with live fetuses
No. does with dead fetuses
No. live fetuses
Live fetuses per dam (Mean ± S.D.)
Live fetuses expressed as percentage of implantations
Dead fetuses expressed as percentage of implantations
Average weights of fetuses/dam or doe (Mean ± S.D.) (combined and separate sexes)
Sex ratios of fetuses
Litter and fetal incidence of all gross, visceral and skeletal findings
Other
Dates study initiated and completed
Source of animals
Source of nutrition and any other variable (e.g., environmental, use of
different types of caging, etc.) which might potentially impact study
Vehicle(s) used
Method of administration
Individual studies should be reported as well as the ranges (mean and median) of
responses for summary data. When historical control data is submitted with a specific study
it is preferred that the data be within ± 2 years of initiation of that study.
The Agency has compiled a list of historical control parameters most often requested
(see Attachment 5).
2. Rejection Factor; Lack of characterization (purity and/or composition and/or
stability) of the test substance (18)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (pages 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
manufacturer's lot number of the test substance with its relevant properties of the substance
tested, such as physical state, pH, stability and purity, and the identification and composition
of any vehicles used in administering the test substance are required by the Agency. For
each test, control, or reference substance that is mixed with a carrier, tests by appropriate
analytical methods shall be conducted to determine uniformity of the mixture as well as
concentration and stability in the carrier.
113
-------
Guidance is very clear on this rejection factor. Therefore, this rejection factor was
"avoidable" on the part of the registrant.
Industry Comment: See comments under Guideline 82-l(a), Rej. Factor 2.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
3. Rejection Factor; Information missing or requiring clarification of the
laboratories methods (examples; staining techniques.
randomization & mating methods, or if the uteri were
stained with ammonium sulfite in the non-pregnant rats)
(17)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, pages 28-33) and in the Phase 3 Technical
Guidance document dated December 24, 1989, D-10. Subdivision F states "the test report
shall include all information necessary to provide a complete and accurate description and
evaluation of the test procedures and results. A test report should contain at least three
parts: a summary and evaluation of the test results; a description of the test procedures; and
the data and information required by each applicable section of this subdivision." Some of
these factors are essential in the evaluation or interpretation of the results from a study.
Also, at any time if the reviewer feels the study report lacks clarity as presented, the Agency
can request further information from the registrants.
This rejection factor is "unavoidable" on the part of the registrant to some degree.
However, registrants should try to have the study reports as complete as possible so that
delays are minimized.
Industry Comment: Historically, the Agency has accepted referenced methodology.
Currently, the Agency appears to require the detailed description of the method within the
report, vs. citation of the published and scientifically acceptable method.
Industry Recommendation: We would like clarification on this from the Agency.
EPA Response: The Agency does accept referenced methodology, but the state-of-
the-science is always changing and techniques are usually modified. Also, there are times
when short cryptic methods are written in the reports which confuse instead of simplify.
In most instances, the majority of the missing information could be obtained by
contacting the company agent by telephone so that the information could be faxed to the
reviewer as well as sent through the normal channels. If the study authors do not want to
include the specific methods, then a copy of the Standard Operating Procedure (SOP's) from
the performing laboratory could be appended to the study report. If the technique is new or
114
-------
has never been submitted to the Agency, then the registrant should append the SOP with the
study report.
4. Rejection Factor; Information missing or requiring clarification in the
laboratories results (examples: appendices, preparation
techniques for the test substance, source of animals, dosing
dates, and individual animal data) (17)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, pages 28-33), the 40 CFR Part §160
(GLP) and in the Phase 3 Technical Guidance document dated December 24, 1989, C 117-
118. Subdivision F states "the test report shall include all information necessary to provide a
complete and accurate description and evaluation of the test procedures and results. A test
report should contain at least three parts: a summary and evaluation of the test results; a
description of the test procedures; and the data and information required by each applicable
section of this subdivision." These rejection factors are often just an oversight by the
registrant and lack of some types of information on a study can seriously hinder the Agency
reviewer in analysis, interpretation or confirmation of study results. The requirement for
individual data is not specifically stated in the guideline; however this rejection factor is
based on the adherence to Good Laboratory Practices (CFR §160). The lack of such data
adds difficulty to the review process, for without it, variations in response of each animal to
treatment as well as verification of the mean data presented cannot be determined.
Guidance on these data requirements is clearly stated in the Agency's guidance
documents. Therefore, this rejection factor is "avoidable" on the part of the registrant.
Industry Comment: The information on preparation techniques, source of animals,
dates and individual data should be in every report. As noted in the EPA comments on this
Factor, several separate guidance documents exist concerning these data requirements.
Supplying the documents together, on a more formal basis may aid in making all aware that
these are basic data needed in any report of a developmental toxicity study.
EPA Response: The Agency is preparing a new SEP for Developmental Toxicity
and anticipates this document will resolve many of these factors.
5. Rejection Factor; Information missing from the laboratories results (examples:
summary tables, fetal or litter incidences, and gravid uterine
weights) (16)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, pages 28-33) and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 117-118. Subdivision F states "the test
115
-------
report shall include all information necessary to provide a complete and accurate description
and evaluation of the test procedures and results. A test report should contain at least three
parts: a summary and evaluation of the test results; a description of the test procedures; and
the data and information required by each applicable section of this subdivision."
Guidance is very clear on the summary table and fetal incidences, but it is generally
considered a laboratory standard to take gravid uterine weights. Therefore, this rejection
factor is "avoidable" on the part of the registrants.
Industry Comment: We agree with the Agency conclusion that this rejection factor
is avoidable, but feel that missing one parameter should not always be the criteria for rejec-
tions. For example, gravid uterine weights are not necessary when individual fetal weights
are presented. The taking of gravid uterine weights is not as accurate a measurement as
individual fetal weights. The subtraction of the gravid uterine weight from the maternal
weight does not always provide a good measure of maternal toxicity because of the period
between the end of dosage and necropsy allows for the recovery of the maternal animal from
the effects of the test substance.
Industry Recommendation: Missing one parameter should not always be the criteria
for rejections.
EPA Response: The Agency agrees that missing one parameter should not be the
only criteria for rejecting a study and will try to be more flexible in the future. However,
the Industry must realize that certain types of data are more critical to the study's
acceptability than others. The Agency will address these issues in the new SEP for
Developmental Toxicity.
6. Rejection Factor; Doses selected in the studies did not produce toxicity such
that maternal and/or developmental NOELs could be
established (10)
EPA Guidance on this Factor: Example: developmental NOEL could not be
established, due to effects at the lowest dose tested. Guidance on this appears in Subdivision
F - Hazard Evaluation: Human and Domestic Animals (§83-3, pages 126 & 127), the
Teratology SEP, p. 21, and in the Phase 3 Technical Guidance document dated December
24, 1989, D 9-10. The guideline recommends that the lowest dose tested should not produce
any evidence of toxicity. However, in many studies developmental effects were seen at the
lowest dose tested and a NOEL for developmental toxicity could not be established. It seems
that the study authors discounted the effects seen at the lowest dose, and the EPA reviewers
established this dose as the LOEL. Studies rejected on the basis of this flaw are very rarely
upgradable.
116
-------
Example: no maternal effects were noted in the highest dose tested. The guideline
recommends that the highest dose tested should induce some overt maternal toxicity such as
slight weight loss to ensure that the test was performed at a lexicologically responsive
exposure level. Only the lowest dose tested should not produce any evidence of toxicity.
Lack of both maternal and developmental effects at the high dose level would indicate the
need for an additional study at higher doses. However, if the test substance was dosed at the
limit dose of 1000 mg/kg and no evidence of maternal or embryo toxicity was observed, then
other studies at other dose levels will not be necessary. In addition, if a NOEL for maternal
toxicity was not determined, but a NOEL for developmental toxicity was identified in a
study, an additional study at a higher dose level is not necessary.
Example: no dose-related effects were noted in any dose group and the study was not
performed at the limit dose (1000 mg/kg). The guideline recommends that the highest dose
tested should induce some overt maternal toxicity such as slight weight loss. This ensures
that the test was performed at a lexicologically responsive exposure level. In some instances
the range-finding studies were requested from the registrant to aid in the justification of the
dose levels chosen for the main study.
This rejection factor was "unavoidable" on the part of the registrant mostly due to the
fact that there will probably always be a "scientific" disagreement on the establishment of
maternal or developmental NOELs.
Industry Comment: We agree with the Agency that this is probably an unavoidable
rejection factor. It is apparent from EPA's comments that the desire for a dose response in
these studies is considered an easily attained goal. It should be recognized that especially for
materials with steep dose-response curves, it may be difficult to establish a level that
produces maternal toxicity without mortality. Recently, EPA has rejected studies for
insufficient numbers of litters at the high dose. It therefore appears that the maternal toxicity
at the high dose must not be so severe as to produce abortion. This seems to create a Catch-
22 situation. Unfortunately, it is usually not possible to choose perfect doses before a study
starts. It is often difficult after it is completed. The Agency should show more latitude in
accepting or rejecting studies for these factors.
Several registrants noted cases where reviewers failed to consider other data which
had been submitted to the Agency. For example, studies have been conducted at doses
which have failed to produce a developmental NOEL. Additional studies have been
conducted at lower doses which have then failed to demonstrate maternal or developmental
toxicity. Both studies may be rejected when, considered together, sufficient data have been
developed in both studies to meet the guidelines. The discussion of the combined data
submitted by the registrant apparently does not always reach the reviewer along with the
newly submitted study. This creates a great deal of confusion before the studies are finally
accepted.
117
-------
Lack of a maternal NOEL should not always be cause for study rejection. It should
be kept in mind that a developmental study is conducted primarily to determine a
developmental NOEL. If a NOEL is available from other studies, then the lack of a
maternal NOEL in a developmental study should not result in rejection unless the pregnant
animal is much more sensitive than the non-pregnant animal.
Industry Recommendation: The Agency should show more latitude in accepting or
rejecting studies for these factors. Lack of a maternal NOEL should not always be cause for
study rejection.
We would also like to have EPA confirmation that teratology studies conducted at the
limit dose of 1000 mg/kg/day are acceptable, if no maternal toxicity or fetotoxicity is
observed. This is clearly stated in the guideline, however, registrants have informed us that
EPA has rejected studies which did not show maternal toxicity at the limit dose.
EPA Response: The Agency prefers that both a maternal and developmental NOEL
be established in the most sensitive species, but the lack of maternal NOEL will not be an
automatic rejection criterion. However, in the following circumstances a NOEL would be
needed for the risk assessment purposes: 1) when the effect noted in the pregnant animals is
substantially different from the non-pregnant animals, 2) if maternal toxicity severely
compromises the outcome of the study; and 3) if the developmental toxicity study was the
most sensitive study used to calculate a RfD. Refer to the response under Guideline 82-l(a),
Rej. Factor 1 for additional guidance on NOEL's.
The Agency agrees with Industry that confusion does sometimes exist when two or
more studies are being considered together to establish a developmental NOEL. The Agency
would like to request that registrants provide the Agency with the MRID numbers for the
additional studies and/or submit a copy of the other study(s) with their written request.
The Agency will accept 1000 mg/kg as the limit dose for Developmental toxicity
studies when no maternal toxicity or fetotoxicity has been observed. The Agency would like
to suggest that (for studies not performed at the limit dose) the range-finding study be
appended to the study report as additional justification for the dose levels.
7. Rejection Factor; Questions on statistics need to be answered, or further
statistics need to be performed, and/or the litter was not the
experimental unit (8)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, page 31), and in the Phase 3 Technical
Guidance document dated December 24, 1989, D-10. Subdivision F states "appropriate
statistical methods shall be used to summarize experimental data, to express trends, and to
evaluate the significance of differences in data from individual test groups. The methods
118
-------
used shall reflect the current state of the art." Additionally, the Phase 3 Technical Guidance
document specifically states the "litter is considered the most relevant unit for statistical
assessment." Sometimes it was not clear to the reviewers exactly what animals were
excluded (included) from tests or why.
In the FR - Developmental Risk Assessment (1991), p. 63808 (f), the litter is
considered the experimental unit in most developmental studies. The statistical analyses are
generally designed to analyze the relevant data based on incidence per Utter or on the number
of litters with a particular end point. If the litter is not used as the experimental unit, the
reviewer cannot properly evaluate the study.
Guidance is clear on the usage of the litter as the experimental unit for statistics.
This rejection factor was "avoidable" on the part of the registrants.
Industry Comment: We agree that current scientific thinking supports the use of the
litter as the relevant unit for statistical analyses. It is interesting that the Agency requires
fetal incidence and litter incidence to be reported for gross, soft and skeletal alterations. It
would be helpful if the Agency clarified their position on statistical analysis of fetal incidence
data.
The inclusion of the number of animals in a particular analysis should be specified
clearly in the report. It is almost always appropriate to exclude non-pregnant animals from
body weight and feed consumption analyses but some laboratories include analyses of all
animals in a statistical analysis of clinical signs. Animals that abort should be excluded from
analysis of body weight.
Industry Recommendation: It would be helpful if the Agency clarified their position
on statistical analysis of fetal incidence data.
EPA Response: The current Agency policy is that the relevant unit for statistical
analysis is the litter. On occasion , the Agency would like to have the fetal indices for
gross, soft, and skeletal alterations with accompanying statistical analysis in the report
because it can be important to aid the reviewer on elucidating a finding. The new SEP for
Developmental toxicity will help address this rejection factor.
8. Rejection Factor; Did not use conventional assessments for skeletal or visceral
examinations (4)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§83-3, page 129). "For rats, mice and hamsters,
one-third to one-half of each litter should be prepared and examined for skeletal anomalies,
and the remaining part of each litter should be prepared and examined for soft tissue
anomalies using appropriate methods." In two studies the fetuses were examined by x-ray
119
-------
techniques and this is not recognized by the Agency as an acceptable method for
examination. The SEP (pages 18 & 19) lists two methods of soft tissue (visceral)
examinations (Staples and Wilson) and the Dawson technique of Alizarin red S staining for
the skeletal examination.
Guidance on this rejection factor is clearly stated. This rejection factor was
"avoidable" on the part of the registrant.
Industry Comment: There are several acceptable methods for preparation and
evaluations of fetuses for skeletal or visceral examination. Although X-ray is not routinely
used in this country it has been used in Europe and an experienced laboratory can evaluate
fetal skeletons by this method.
Industry Recommendation: Some consideration should be given to accepting this
technique in certain cases.
EPA Response: In the Agency's experience, the use of X-rays for skeletal
examinations is not a validated technique and will not be accepted by the Agency at this
time.
120
-------
GUIDELINE 83-3 (b) DEVELOPMENTAL TOXICITY - NON-RODENTS
1. Rejection Factor; Clarification of laboratories procedures or interpretation of
the data (examples: method of intracranial exams.sacrifice
order, questions on artificial insemination, mating records.
or need range-finding study) (25)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, pages 28-33), and in the Phase 3
Technical Guidance document dated December 24, 1989, C 117-118 & D-10. Subdivision F
states "the test report shall include all information necessary to provide a complete and
accurate description and evaluation of the test procedures and results. A test report should
contain at least three parts: a summary and evaluation of the test results; a description of the
test procedures; and the data and information required by each applicable section of this
subdivision." Some of these factors are essential in the evaluation or interpretation of the
results from a study. For example, whether the heads of the fetuses properly evaluated for
malformations or variations depends greatly on how the heads were examined (what
technique was used). Range-finding studies are requested to ascertain whether the proper
dose level had been selected for the main study and for information on maternal wastage.
Also, the developmental effects noted in the range-finding study can aid the reviewer in
making a more critical evaluation of the main study. If, at any time, the reviewer feels the
study reports lack clarity as presented, the Agency has the right to request further
information from the registrants.
This rejection factor was "unavoidable" on the part of the registrant to some degree,
although registrants should try to have the study reports as complete as possible so that
delays are minimized.
Industry Comment: See comments under Guideline 83-3(a), Rej. Factor 1.
EPA Response: Refer to responses under Guideline 83-3(a), Rej. Factors 3 and 4.
2. Rejection Factor; Either maternal or fetal individual data are required for
proper evaluation of results (18)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, pages 28-33), and in the Phase 3
Technical Guidance document dated December 24, 1989, C 117-118. Subdivision F states
"the test report shall include all information necessary to provide a complete and accurate
description and evaluation of the test procedures and results. A test report should contain at
least three parts: a summary and evaluation of the test results; a description of the test
procedures; and the data and information required by each applicable section of this
121
-------
subdivision." The requirement for individual data is not specifically stated in the guideline;
however, this rejection factor is based on the adherence to Good Laboratory Practices (CFR
§160). But the lack of such data adds difficulty to the review process, for without it,
variations in response of each animal to treatment as well as verification of the mean data
presented cannot be determined. This is especially important when sex or dose-related
differences in development are present.
Guidance on this rejection factor is not entirely clearly stated. Therefore, this
rejection factor was not always "avoidable" on the part of the registrant.
Industry Comment: See comments under Guideline 83-3(a), Rej. Factor 5.
EPA Response: Refer to response under Guideline 83-3(a), Rej. Factors 5 and 7.
3. Rejection Factor; Historical controls are needed from the performing
laboratories to aid in the assessment of potential
developmental effects (17)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 & §83-3, pages 9, 24, & 129), and in the
Phase 3 Technical Guidance document dated December 24, 1989, D-10. Historical control
data are not required to be submitted with the study, but such data should be made available
upon the Agency's request. The historical controls, while not run concurrently with the test
substance under study, yield valuable comparisons on whether the test animals are typical for
their species and strain for various incidences and statistical measures for each laboratory.
Sometimes the best rationale for interpretation of skeletal or visceral variations or
malformations or any other apparent effect is with a comparison with the historical control
data collected from various studies with the same species, strain of animals, vehicle, and
conducted in a similar time frame. Other examples of when the Agency needs historical
control data are on the justification of a non-treatment-related status of a significant increase
of a particular effect, or backing up a judgement based on scientific interpretation, or
justifying a false positive that can occur because the concurrent controls are unusual for a
particular finding (low incidence).
Currently, the guidance does not specifically require historical controls to be
submitted with the study. If the registrant is trying to explain specific fetal finding(s) and
they use historical controls for support, then historical control data should be appended to the
study report. This rejection factor was always "avoidable" on the part of the registrants.
Industry Comment: See comments under Guideline 83-3(a), Rej. Factor 1.
EPA Response: Refer to response under Guideline 83-3(a), Rej. Factor 1.
122
-------
4. Rejection Factor: Lack of characterization (purity and/or composition and/or
stability) of the test substance (17)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (pages 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
manufacturer's lot number of the test substance with its relevant properties of the substance
tested, such as physical state, pH, stability and purity, and the identification and composition
of any vehicles used in administering the test substance are required by the Agency. For
each test, control, or reference substance that is mixed with a carrier, tests by appropriate
analytical methods shall be conducted to determine uniformity of the mixture as well as
concentration and stability in the carrier.
Guidance on this rejection factor is clearly stated. Therefore, this rejection factor was
"avoidable" on the part of the registrant.
Industry Comment: See comments under Guideline 82-1 (a), Rej. Factor 2.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
5. Rejection Factor; Excessive maternal toxicity (7)
EPA Guidance on this Factor: Guidance on the appropriate number of litters that
should be produced per dose level are found in Subdivision F - Hazard Evaluation: Human
and Domestic Animals (§83-3, page 127) - "at least 12 pregnant rabbits are recommended at
each dose level and control group." When excessive maternal toxicity occurs in a study, not
enough litters are produced to ensure sufficient pups for meaningful evaluation of the
teratogenic potential of the test substance. This can be prevented, to some extent, by
performing a range-finding study to identify proper doses. In certain instances, when
toxicity occurs, the Agency will address the issues on a case-by-case basis.
Industry Comment: We agree that the Agency should evaluate this on a case by
case basis. In the EPA guidelines the only criteria for the numbers of animals is the one that
states that "at least 12 pregnant rabbits." The guidelines do not give a minimum number of
litters. The Agency appears to assume that all pregnant animals will have litters. Under
even the most optimistic criteria this is not always possible. Especially in rabbits, maternal
toxicity often results in abortion. It is curious that the guideline was written specifying the
numbers of pregnant rabbits rather than the numbers of litters.
Industry Recommendation: We feel that loss of one or 2 litters should not be a
criteria for rejection. As long as there is a dose level without maternal toxicity and dose
123
-------
levels are appropriately spaced, excessive maternal toxicity at the high dose should not be
cause for rejection. The registrant should be given the option to justify the statistical power
of a test with fewer litters before having the study rejected.
EPA Response: The Agency requires at least 12 litters per dose level, but the
Agency also realizes this is often difficult to obtain in rabbits. Those studies with less than
12 litters per dose level will be handled on a case-by-case basis. The Agency agrees with
Industry that excessive toxicity in the high dose and this will not be grounds for future
rejections. The Agency will clarify under what conditions maternal mortality could be used
as a rejection factor in the Developmental SEP.
6. Rejection Factor; Doses selected in the studies did not produce toxicity such
that maternal and/or developmental NOELs could be
established (7)
EPA Guidance on this Factor: Example: developmental NOEL could not be
established, due to effects at the lowest dose tested. Guidance on this appears in Subdivision
F - Hazard Evaluation: Human and Domestic Animals (§83-3, pages 126 & 127), the
Teratology SEP, p. 21, and in the Phase 3 Technical Guidance document dated December
24, 1989, D 9-10. The guideline recommends that the lowest dose tested should not produce
any evidence of toxicity. However, in many studies developmental effects were seen at the
lowest dose tested and a NOEL for developmental toxicity could not be established. It seems
that the study authors discounted the effects seen at the lowest dose, and the EPA reviewers
established this dose as the LOEL. Studies rejected on the basis of this flaw are very rarely
upgradable.
Example: no maternal effects were noted in the highest dose tested. The guidelines
recommend that the highest dose tested should induce some overt maternal toxicity such as
slight weight loss to ensure that the test was performed at a sensitive level. Only the lowest
dose tested should not produce any evidence of toxicity. Lack of both maternal and
developmental effects at the high dose level would indicate the need for an additional study at
higher doses. However, if the test substance was dosed at the limit dose of 1000 mg/kg and
no evidence of maternal or embryo toxicity was observed, then other studies at other dose
levels will not be necessary. In addition, if a NOEL for maternal toxicity was not
determined, but a NOEL for developmental toxicity was identified on a study, an additional
study at a higher dose level is not necessary.
This rejection factor was "unavoidable" on the part of the registrant mostly due to the
fact that there will probably always be a "scientific" disagreement on the establishment of
maternal or developmental NOELs.
Industry Comment: See comments under Guideline 83-3(a), Rej. Factor 6.
124
-------
EPA Response: Refer to response under Guideline 83-3(a), Rej. Factor 6.
7. Rejection Factor; Questions on statistics need to be answered, or further
statistics need to be performed, and/or the litter was not the
experimental unit (4)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, page 31), and in the Phase 3 Technical
Guidance document dated December 24, 1989, D-10. Subdivision F states "appropriate
statistical methods shall be used to summarize experimental data, to express trends, and to
evaluate the significance of differences in data from individual test groups. The methods
used shall reflect the current state of the art." Sometimes it was not clear to the reviewers
exactly what animals were excluded (included) from tests or why.
In the FR - Developmental risk assessment (1991), p. 63808 (f), the litter is
considered the experimental unit in most developmental studies. The statistical analyses are
generally designed to analyze the relevant data based on incidence per litter or on the number
of litters with a particular end point. If the litter is not used as the experimental unit, the
reviewer cannot properly evaluate the study.
Guidance is clear on the usage of the litter as the experimental unit for statistics.
This rejection factor was "avoidable" on the part of the registrant.
Industry Comment: See comments under Guideline 83-3(a), Rej. Factor 7.
EPA Response: Refer to response under Guideline 83-3(a), Rej. Factor 7.
125
-------
GUIDELINE 83-4 REPRODUCTION
1. Rejection Factor: Information missing from the laboratories results (examples;
appendices and individual animal data) (9)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, pages 28-33), and in the Phase 3
Technical Guidance document dated December 24, 1989, C 119-120. Subdivision F states
"the test report shall include all information necessary to provide a complete and accurate
description and evaluation of the test procedures and results. A test report should contain at
least three parts: a summary and evaluation of the test results; a description of the test
procedures; and the data and information required by each applicable section of this
subdivision."
Guidance on this rejection factor was clearly stated. Therefore, this rejection factor
'avoidable" on the part of the registrant.
was "
Industry Comment: It is clearly stated that all individual data should be provided,
and we agree with the Agency that this is an avoidable rejection factor. Studies conducted
prior to the introduction of automated data collection programs or in laboratories that "hand-
record" data for multigeneration studies have a difficult time preparing all individual data in
the current EPA format for reports.
Industry Recommendation: Some consideration should be given to allowing
individual data to be submitted in forms other than the prescribed format. In no case should
a study that is scientifically acceptable be repeated because the individual data was not in an
acceptable format for the submitter to include in the report.
EPA Response: The Agency agrees with Industry that a study which is scientifically
acceptable should not be rejected. However, the study should be submitted in the prescribed
format or a format which closely resembles the EPA format so that an independent analysis
of the data can be performed.
2. Rejection Factor: Lack of characterization (pnritv and/or composition and/or
stability) of the test substance (8)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (pages 22, 23, & 29) and 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices. The chemical name, molecular
structure, and a quantitative and qualitative determination of its chemical composition
(including impurities, and contaminants) should be supplied to the Agency. Also, the
manufacturer's lot number of the test substance with its relevant properties of the substance
126
-------
tested, such as physical state, pH, stability and purity, and the identification and composition
of any vehicles used in administering the test substance are required by the Agency. For
each test, control, or reference substance that is mixed with a carrier, tests by appropriate
analytical methods shall be conducted to determine uniformity of the mixture as well as
concentration and stability in the carrier.
Guidance on this rejection factor was clearly stated. Therefore, this rejection factor
was "avoidable" on the part of the registrant.
Industry Comment: See comments under Guideline 82-l(a), Rej. Factor 2.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
3. Rejection Factor; Information missing or requiring clarification of laboratory's
methods or results (examples: range-finding studies were
requested, how many rats were mated, reporting errors, and
questions on missing pups and culling procedures) (8)
EPA Guidance on this Factor: Some of these factors are essential in the evaluation
or interpretation of the results from the study. Range-finding studies are requested to
ascertain whether the proper dose levels had been selected for the main study and for any
information on maternal wastage. Also, any reproductive effects noted in the range-finding
study can aid the reviewer in making a more critical evaluation of the main study. In some
studies, the reviewers had difficulty in tracking the mating procedures, numbers of pups born
in each litter, or how many pups were culled when the F2 generation was initiated on study.
In some reports there were numerical errors in the summary tables when compared to the
individual data. Additionally, if the reviewer feels the study report lacks clarity as
presented, the Agency has the right to request further information from the registrants.
Guidance on this rejection factor is clear and rejection could have been avoided on the
part of the registrant.
Industry Comment: See comments under Guideline 83-3(a), Rej. Factor 3.
EPA Response: Refer to responses under Guideline 83-3(a), Rej. Factor 3. The
Agency is preparing a new SEP for Reproduction Toxicity and hopes to have a draft copy for
industry to review and comment on soon. This document should resolve many of these
factors.
127
-------
4. Rejection Factor; Historical controls are needed from the performing
laboratories to aid in the assessment of potential
reproductive effects (4)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4, pages 9 & 24). Even though historical
control data are not required to be submitted with the study, such data should be made
available upon the Agency's request. The historical controls, while not run concurrently
with the test substance under study, yield valuable comparisons on whether the test animals
are typical for their species and strain for various incidences for each laboratory. Sometimes
the best rationale for interpretation of fertility effects is with a comparison with the historical
control data collected from various studies with the same species, strain of animals, vehicle,
and conducted in the similar time frame.
Currently, the guidance does not specifically require historical controls to be
submitted with the study. Therefore, this rejection factor was "unavoidable" on the part of
the registrant.
Industry Comment See comments under Guideline 83-3(a), Rej. Factor 3.
EPA Response: The subject of historical control data will be treated in the
Reproductive Standard Evaluation Procedure presently under draft. In general, historical
control data should not be used as a substitute for concurrent control results unless clearly
justified and should not be routinely requested. However, testing laboratories are encouraged
to routinely collect historical control data.
The reasons for requesting historical control data are similar to those for
developmental toxicity studies:
1) Historical control data can provide a guide for determining the biological
significance of statistically significant differences observed in a reproductive study.
Such data may indicate whether a concurrent control group incidence (e.g., pregnancy
rate, dams with 100% resorptions, etc.) are unusually low for the test species or if
nominally increased incidences for a treatment group are within the normal range seen
for that strain/species. While the range of reported values is important, when mean
historical control data are supplied, it is presumed that findings outside two standard
deviations of the mean are outliers and therefore may not be acceptable for
comparison against treatment groups. Collection of historical developmental anomaly
data in offspring in reproduction studies may also be of value to the Agency.
2) Historical control data also may indicate trends in the overall vigor, fertility, or
litter size of a particular strain/species, which relate to genetic drift and can aid in the
interpretation of apparent unusual findings.
128
-------
Some specific examples of when historical control data would be necessary to clarify
the findings in a reproduction study include:
1) Unusual concurrent control findings, leading to difficulty in interpretation of the
validity of significant findings in treated groups.
a) A clear dose response for decreases in fertility, increased abortions or
premature births, smaller live litters, increased number of dead pup/litter, etc.,
is not evident but is elevated above the concurrent control such that statistical
or apparent biological significance is achieved in one or more treatment
groups.
b) Fertility appears to be unusually low in the concurrent control and/or
treatment groups and is suspected of impacting the overall validity of the
study.
2) Variations between generations
a) Differences in various reproductive parameters are noted between control
generations (Fl, F2) (e.g., litter sizes, mean pup weights).
b) If developmental landmark time frames are recorded (e.g., pinna unfolding,
anogenital distance) they may vary between generations and this variation may
appear biologically or statistically significant.
3) Unusual findings
a) Developmental anomalies or variations are noted in a treatment group, but
not the concurrent control, and the findings are not dose-related. Background
incidence data from either reproduction or developmental toxicity studies may
assist in determining biological significance.
b) An apparently high incidence of an observation such as pup mortality is
observed at all dose levels, including control, although no significance is
demonstrated in treated groups.
Examples of the types of historical control data necessary to effectively evaluate a
reproduction study are given below. It is understood that individual study results as well as
summary statistics (i.e., means, median values, standard deviations, ranges, etc.) must be
submitted. Individual studies should be identified by the dates of initiation and completion,
vehicles utilized (if any), method of administration, changes in feed or animal suppliers,
environmental conditions, or other significant factors which change over time. Since there
are many different methods of calculating reproductive indices, what the Agency considers
129
-------
the most critical information for a complete interpretation of reproductive toxicity studies has
been presented. Therefore, this is not a comprehensive listing of reproductive indices.
Reproductive performance varies widely within the same species due to genetic and
environmental factors. Although concurrent control data are normally the most appropriate
for comparison, on occasion, historical control data for the same strain/species provide the
reviewer with valuable information regarding the background rate for various reproductive
parameters as well as normal variations and trends. For both developmental and
reproductive toxicity studies, it is preferable that historical control data be supplied within at
least a ± 2 year range (Draft Developmental Toxicity SEP, 1992). However, it is generally
acceptable for reproduction studies to have a longer time period (± 5 years) since the types
of reproduction data requested (see table below) require longer times to generate and are not
as sensitive to subtle changes in examination procedures or diagnostic criteria. Furthermore,
a longer time frame allows for potential genetic drifts to be noted. Studies that are not
representative or typical, either in study design and conduct, or in the condition (health) of
the animals, should not be included in the historical control data set.
Historical data sets derived from multi-generation reproduction and/or developmental
toxicity studies conducted in multiple laboratories and published in the scientific literature
(e.g., Clemens et al., 1992, MARTA poster at Teratology meeting) may be considered by
the reviewer in the interpretation of study results. Generally, however, these results must be
regarded with a certain degree of caution, since there may be unknown variability in study
conditions and conduct between various laboratories, as well as inapparent differences in
technical procedures, interpretation of effects, and calculation of indices.
SAMPLE: Table of Historical Control Parameters for Reproduction Studies*
General Information
Number of males and females mated
Number of dams/does pregnant
Number of confirmed pregnancies/number of confirmed matings
Number of males and females surviving to scheduled termination
Number of females with abortions
Number of females with premature deliveries
Number pregnant up to parturition
Adult Reproduction Indices^
Copulatory Index
Mating Index
Fecundity Index
Male Fertility Index
Female Fertility Index
130
-------
Parturition Index
Gestation Index
Pup Reproductive/Indices and Parameters#
Live Birth Index
Sex Ratio
Live Litter Size Index
Sex Index
4-Day Survival Index (viability index)
21-Day Survival Index (weaning index [days 0-21]/lactation index [days 4-21])
Preweaning Index
Pup birth Weight (day 0)
Pup birth weight (day 4/pre-standardization)
Pup birth weight (day 4/post-standardization)
Pup birth weight (day 7)
Pup birth weight (day 14)
Pup birth weight (day 21)
Incidences of developmental anomalies
* Historical control data for each generation of a given study should be submitted.
# See discussion in SEP on reproductive indices for definitions.
5. Rejection Factor; No reproduction NOEL was established due to effects at the
lowest dose tested (3)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§83-4, page 131), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C-120. The guideline recommends that the
lowest dose tested should not produce any evidence of toxicity. However, in some studies
reproductive effects were seen at the lowest dose tested and a NOEL for reproductive
toxicity could not be established. It seems that the study authors discounted the effects seen
at the lowest dose, and the EPA reviewers established this dose as the LOEL. Studies
rejected on the basis of this flaw are very rarely upgradable.
This rejection factor was "unavoidable" on the part of the registrant mostly due to the
fact that there will probably always be a "scientific" disagreement on the establishment of a
reproductive NOEL.
Industry Comment: We agree this is often unavoidable due to the usual scientific
disagreement on what is a NOEL or NOAEL.
131
-------
EPA Response: Refer to response under Guideline 82-l(a), Rej. Factor 1.
6. Rejection Factor: Low fertility was experienced in dose groups and/or an
inadequate number of animals were used per dose level (3)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§83-4, page 131). The guidelines recommend
that "each test and control group should contain at least 20 males and a sufficient number of
females to yield at least 20 pregnant females at or near term." If low fertility occurs in the
second (Fj) generation, the Agency will review this issue on a case-by-case basis, and if
justification can be given to the Agency on the low fertility, the study may be upgraded.
However, if low fertility occurs in the first generation (F^ then a meaningful evaluation of
reproductive effects cannot be performed.
Industry Comment: The Agency appears to want 20 litters per group but has guide-
lines that require 20 pregnant females. The Agency should understand that in most studies it
is not possible to get 20 litters from 20 pregnant females. In other studies occasional poor
mating will not always yield 20 pregnant females. It is difficult to understand a fixed criteria
that places 20 pregnant females in the acceptable column but says 19 is not acceptable, since
the statistical power of the test is not significantly impacted if one has fewer than twenty
litters. Some flexibility must be allowed for natural variation as well as test substance
effects. Studies should be accepted unless there is a serious scientific flaw.
There should be a sufficient number of litters to allow evaluation of possible test
substance effects. Ideally at least 20 litters should be evaluated. However, studies having
17, 18 or 19 litters in a particular dosage group should not lead to rejection especially when
the reason for the slightly lowered number of litters is adequately addressed in the report.
Although the guidance on this factor states that "case by case basis" exceptions will be made,
it is generally felt by registrants that EPA considers 20 a magic number. If there are less
than twenty Utters, additional offspring can be selected from randomly selected litters to
provide enough adequate numbers for the second generation.
Generally, lower fertility rates are a continuing problem with some strains of rats.
Some recognition of these problems by the Agency would be helpful.
EPA Response: The Agency and Industry agree on this point.
7. Rejection Factor: No systemic NOEL was established in the absence of
reproductive effects (3)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§83-4, page 131). The guideline recommends
132
-------
that the lowest dose tested should not produce any evidence of toxicity. However, in many
studies reproductive effects were seen at the lowest dose tested and a NOEL for reproductive
toxicity could not be established. It seems that the study authors discounted the effects seen
at the lowest dose, and the EPA reviewers established this dose as the LOEL. If no
reproductive toxicity is indicated, and if the animals are tested at the limit dose of 1000
mg/kg (memorandum dated 7/27/88: Selection of an Upper Limit Dose in Multigeneration
Reproduction Studies, Theodore Farber, Ph.D, Toxicology Branch, Health Effects Division),
then the study is acceptable. But just because the study does not have a systemic NOEL will
not cause a study to be rejected.
Guidance on this point is not entirely clear; avoidance of a rejection on this factor is
difficult because it is subject to interpretation.
Industry Comment: We agree with the Agency that avoidance of a rejection on this
factor is difficult because it is subject to interpretation. We do not feel that a systemic
NOEL should be required.
Industry Recommendation: A reproduction study which has a NOEL for reproduc-
tive effects should not be rejected due to the lack of a NOEL for adult systemic effects,
provided that NOELs were established for the effect in other studies or other methodologies
can be used to estimate an RfD for the reproduction study.
EPA Response: Refer to response under Guideline 82-1 (a), Rej. Factor 1.
133
-------
GUIDELINE 84-2 MUTAGENICITY TESTING
Industry General Comments and Recommendations: One general area of concern
that is evident from reading the mutagenicity section is that the guidance for conducting
studies is located in a hodgepodge of testing guidelines, Standard Operating Procedures,
Technical Guidance Documents and even internal memoranda. In some cases guidance is not
consistent between the different documents and some of the documents are probably not
readily available (e.g., internal memoranda). This highlights a clear need for testing
guidance that is explicit, consistent, readily available and contained in as few separate
sources as possible. This guidance should be harmonized to the fullest extent possible within
the Agency and with other national and international guidance. The guidance should be
evaluated by scientific experts and should be periodically reviewed and updated to reflect
new scientific data and concepts.
In several places, OTS Health Effects Testing Guidelines (50 FR 39439) are listed as
providing guidance for the conduct of mutagenicity assays. It should be noted that these
guidelines were originally proposed for TSCA purposes and, as far as we are aware, it was
not until 1989 that reliance on these guidelines for FIFRA purposes was proposed by OPP.
Prior to this, the Pesticide Assessment Guidelines Subdivision F Hazard Evaluation: Human
and Domestic Animals (Revised edition, November 1984) did not provide specific guidelines
for mutagenicity assays, but referenced EPA Gene-Tox Program papers and an EPA/SRI
International Project "In vitro Mutagenicity Studies of Environmental Chemicals", 1982.
The referenced Phase 3 Technical Guidance document (which has very limited guidance
information) was also not issued until 1989. Therefore, it is probably inaccurate to assume
that the Health Effects guidelines were generally recognized as providing guidance for
mutagenicity studies submitted for FIFRA purposes or that study rejection necessarily reflect
failure to follow mutagenicity guidelines established for FIFRA studies.
Formulation of explicit guidelines for mutagenicity tests is, of course, an essential
component for minimizing study rejection rate, and EPA's recent efforts to harmonize
guidelines within the Agency and with international guidelines such as the OECD guidelines
are commendable in this context. We compliment and endorse continued efforts by EPA and
OPP to subject new guidelines and guideline revisions to comprehensive comment and review
by knowledgeable scientists and to perform periodic review and revision of guidelines. This
places the guidelines on a solid scientific basis and serves Agency, industry and the public
well. EPA should work to issue comprehensive guidance that brings all of the various
current guidance documents together.
To the extent that this document proposes new regulations (regulations that are
different than what is found in the current guidelines), Industry would like to be able to
formally comment on these changes and meet with the Agency to discuss these issues.
EPA Response to General Comments and Recommendations: Industry suggests
that the guidance for conducting studies is randomly located in various guises (i.e., different
134
-------
documents, internal memoranda). This may be the case for much guidance over the past
years. Recent efforts to provide guidance have provided a much clearer picture and have
been placed in more convenient formats. The requirements to satisfy mutagenicity testing
have been placed in the Pesticide Assessment Guidelines, Subdivision F, Hazard Evaluation:
Human and Domestic Animals, Series 84, Mutagenicity, Addendum 9 (EPA-540/09-91-122;
published 1991, NTIS Publication No. PB91-158394). The availability of this document was
described in the Federal Register (56 FR 21366 - 21367, May 8, 1991). In §84-2 are the
specific tests required for initial testing and clear decision tracts for confirmatory and/or
further testing when mutagenicity concerns are identified. This revised mutagenicity
guidance in Subdivision F, Addendum 9 has undergone rigorous peer review, public
comment and examination by the Scientific Advisory Panel (SAP). The background and
responses to the public comments and SAP are found in the support document which has
been formally published in a peer reviewed journal (Dearfield et al., Mutation Research 258:
259-283, 1991). These requirement changes in Addendum 9 will be codified in Part §158 of
the CFR Volume 40 as a result of the upcoming revisions to Part § 158.
Within Section §84-2 of the Subdivision F, Addendum 9 guideline is the OPP
statement on where to obtain actual guidance to perform the various mutagenicity tests. The
OPP uses the Health Effects Testing Guidelines (40 CFR Part §798, Subpart F, Genetic
Toxicity) for protocol guidance. It is agreed with industry that this has not been apparent
until recently. The protocols are evaluated by scientific experts and are periodically updated
as the state of the science progresses. Furthermore, they undergo public comment as well
before finalization. These protocols are now harmonized within the Agency and are
considered guidance for both OPP and OPPT (i.e., all of OPPTS). Current efforts are
concentrating on harmonizing these protocols with OECD guidelines.
135
-------
A) Gene mutations (§84-2): [Tests include - Mammalian cell culture, forward or reverse
mutations at specific loci: (a) Chinese hamster ovary (CHO); (b) Chinese hamster lung (V79);
(c) mouse lymphoma (15178Y); (d) diploid human lymphoblast (WI-12 OR WQ-38).
1. Rejection Factor; Purity, batch numbers, stability, or analytical concentration
information missing (70)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 pp. 22, 23, & 29), 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices, and the Standard Operation
Procedure #2000, Toxicology Branch/HED/OPP, Revised 9/30/85 with corrections from
Federal Register 52: 19080-19081, May 20, 1987. The chemical name, molecular structure,
and a quantitative and qualitative determination of its chemical composition (including
impurities, and contaminants) should be supplied to the Agency. Also, the manufacturer's
lot number of the test substance with its relevant properties of the substance tested, such as
physical state, pH, stability and purity, and the identification and composition of any vehicles
used in administering the test substance are required by the Agency. For each test, control,
or reference substance that is mixed with a carrier, tests by appropriate analytical methods
shall be conducted to determine uniformity of the mixture as well as concentration and
stability in the carrier.
In an internal Agency memorandum dated July 5, 1991, (see Attachment 6) it was
noted that, for in vitro mutagenicity testing, dosing mixtures or solutions are usually
prepared only once, then are applied to the test organism. Concentrations are usually
prepared via dilutions from the top concentration or from a stock solution. The
recommendation is that analysis be performed on the top concentration solution. If the test
substance is soluble in aqueous solution, then analysis would be performed with a carrier that
is the same as the aqueous medium the test organisms are grown in. However, test
substances are frequently dissolved in carriers such as DMSO or acetone, and are then
applied to the aqueous environment containing the test organisms. In this case, analysis
should be done on the top dosing mixture found in the test situation.
In the absence of analysis, a positive mutagenic response or (lacking a positive
response) occurrence of cytotoxicity might be used as an indication that the test substance
was present at a sufficiently high concentration. However, if neither a positive response nor
cytotoxicity was observed at the highest dose (including situations when the test substance
was administered up to a limit dose, such as 5000 /xg/plate in the Ames assay) the study
could be judged unacceptable. Also, in order to comply with GLP's, analytical chemistry
should be performed for all chemicals.
In the case of in vivo mutagenicity testing, a positive mutagenic response or (lacking
a positive response) symptoms of toxicity in the test animal could be used as an indication
136
-------
that the test substance was present at a sufficiently high concentration. However, in the
absence of these findings, analytical data could be necessary.
It is emphasized that the material in the July 5, 1991 memorandum represents
suggestions for dealing with the issue of analytical determinations in mutagenicity testing,
rather than final Agency policy on this subject.
Industry Comments: The referenced internal Agency memorandum dated July 5,
1991 contains very significant guidance on analysis of dosing solutions for in vitro
mutagenicity studies that is not present in the other guidance documents. It is not clear
whether this guidance is publicly available.
As noted by the Agency, in vitro mutagenicity studies are frequently conducted using
solutions of test substance that are prepared and administered once to the test system.
Furthermore, there is a common practice to use dose solutions that are prepared on the day
of use in both in vitro and in vivo studies. Under these circumstances, we agree with the
Agency's suggestion that analytical measurements are not necessary when the upper dose
level exceeds solubility or generates toxic effects. For mutagenicity studies, these endpoints
represent clear evidence that an appropriate exposure to the test substance has occurred and
no analytical measurements should be required.
We also agree with the suggestion that demonstration of toxicity in an in vivo study
clearly demonstrates administration of an appropriate upper dose level of test substance and,
in such cases, analytical measurements are not necessary.
Industry Recommendation: It will be extremely important for the Agency to
communicate any policy decisions relating to analytical determinations in mutagenicity testing
as quickly as possible to avoid unnecessary rejection of otherwise acceptable studies. Any
analytical work that has been proposed that would meet GLP requirements should follow the
recommendation found in the GLP standard. Additional work should be unnecessary.
EPA Response: It is emphasized that the memorandum (see memorandum from K.
Dearfield, dated July 5, 1991 in Attachment 6) was written in response to GLP requirements
for characterization of the test substance, and as such, it represents an attempt at encouraging
HED reviewers to show some flexibility. The memorandum provides rationales for
acceptance of some studies that might have been rejected otherwise.
There should be a reasonable description of the analytical method used. If the
analytical work is done in a laboratory different from the one in which the mutagenicity
study was conducted, there should be adequate supporting documentation (including a
statement that the analysis was done under GLP's), as well as information that confirms that
the composition of the sample analyzed represents what was actually tested.
137
-------
In evaluating results from in vitro structural chromosomal aberration assays,
particularly when findings are negative, the Agency may take into consideration any
additional data that might be available on the test substance. This might include metabolism
data (information as to whether the test substance reaches the target organ/tissue),
gastrointestinal absorption data (whether the test substance and/or its metabolites are readily
absorbed), and the anticipated human exposure route (dermal and/or inhalation and/or oral)
as a result of pesticide end-use. Additionally, the Agency may also consider the lack of
indications of a carcinogenic response from long-term animal feeding studies and/or human
epidemiological data in not asking for further mutagenicity studies.
Also, for further guidance refer to response under Guideline 81-1, Rej. Factor 1.
2. Rejection Factor: MTD issue, no range-finding study; inadequate high dose; no
evidence of toxicitv at any dose; insufficient (or no)
cytotoxicitv and limit-dose level (5000 ^g/plate) not reached
and/or test substance not tested up to solubility limits (64)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39439-39458, September 27, 1985, Federal
Register 52: 19078-19079, in the Phase 3 Technical Guidance document dated December 24,
1989, C 124-126, as well as 40 CFR Part §798. The guideline states that the highest
concentration shall produce a low level of survival and the survival in the lowest
concentration shall approximate the negative control. The highest concentration is limited by
toxicity (10-20% relative survival), solubility or 5000 /tg/mL. Range-finding studies are not
specifically addressed in the guidelines. However, range-finding studies should be submitted
to the Agency in order for the reviewer to determine if the maximum tolerated dose has been
attained.
For Ames studies, it is stated in the Federal Register that toxicity may be evidenced
by a reduction in the number of spontaneous revertants, a clearing of the background lawn,
or by the degree of survival of treated cultures. Relatively insoluble test substances should
be tested up to the limits of solubility. For freely soluble nontoxic chemicals, the upper test
chemical concentration should be determined on a case by case basis.
With respect to cytotoxicity in the Ames assay, most laboratories do a preliminary
assay only with strain TA 100, and then use the findings to set (the same) exposure levels for
all four strains. There does not seem to be any readily available EPA guidance as to whether
or not this is permissible, or whether it is more appropriate to test all four strains
individually for cytotoxicity before initiating mutagenicity testing. However, there is no
indication that the Agency has rejected a study because only strain TA 100 was used in the
preliminary cytotoxicity assay.
138
-------
Industry Comment: The first paragraph states that a limit dose of 5000 jig/ml can
be used for freely soluble, relatively non-toxic chemicals for in vitro gene mutation assays.
This guidance is contained in the Phase 3 Technical Guidance Document, but does not appear
in the Health Effects guidelines. We support the concept of a limit dose for all mutagenicity
assays and recommend inclusion of limit doses in the EPA guidelines.
The rejection factor discussion indicates that "range-finding studies should be
submitted to the Agency in order for the reviewer to determine if the maximum tolerated
dose has been attained." This statement is not consistent with the guideline requirements and
is an unnecessary requirement in the case of gene mutation assays in cultured mammalian
cells. The Health Effect guideline for this type of assay does not indicate that the highest
concentration tested should be a "maximum tolerated dose" as stated in the rejection factor
discussion. Instead, it states that the highest concentration tested should produce a low level
of survival, be at the limit of solubility, or be handled on a case-by-case basis for freely
soluble relatively non-toxic chemicals. In the latter case, the Phase 3 Technical Guidance
document indicates that a limit dose of 5000 /xg/ml can be used. Given these criteria,
determination of whether or not an appropriate upper dose has been tested can be made
without reference to range-finder data. While it may be desirable or preferable to include
range-finding data in the report, omission of these data should not be sufficient reason to
reject a study if the highest dose level tested meets the criteria of the guideline.
The paragraph on Ames studies maximum dose incorrectly states that for freely
soluble non-toxic chemicals the upper test chemical concentration should be determined on a
case-by-case basis. In fact, 5 mg/plate is indicated as an appropriate upper limit for this
assay. The case-by-case statement appears in the Health Effects guideline for gene mutation
assays in cultured mammalian cells.
As noted in the document, it is a very common practice to conduct toxicity range-
finder studies for the Ames assays with only one strain (usually TA100). As there is no
requirement for a range-finder assay and because evaluation of the appropriateness of highest
concentrations tested will usually not require examination of range-finder data, there is no
need to provide further guidance on strain selection for range-finder testing or institute any
requirements in this area. As noted for the in vitro gene mutation studies, there is no reason
to automatically reject studies if range-finder data are not provided.
Industry Recommendation: Omission of range-finding data should not be sufficient
reason to reject a study if the highest dose level tested meets the criteria of the guideline.
EPA Response: The term "MTD" is inappropriate for this type of study. The
Agency agrees with industry that a study in which there is no evidence of cytotoxicity at a
limit dose of 5000 /tg/plate (or 5000 ng/mL in a non-Ames study) of a freely soluble
material should be acceptable (if there were no other problems), although analytical data
might be necessary in the absence of both cytotoxicity and a positive mutagenic response (see
Attachment 7). However, when this limit dose has not been reached, and there is no
139
-------
indication of a sufficiently high level of cytotoxicity at the highest dose, then there may be
problems in accepting the study. Some studies have been rejected because there is little or
no evidence of cytotoxicity at the highest dose level; although a preliminary range-finding
study indicated complete or nearly complete cytotoxicity at this same concentration.
Generally, range-finding studies are useful in suggesting an appropriate concentration range
for exposure.
The Agency agrees that if the highest concentration tested produces an appropriately
low level of survival at the limit of solubility or is 5000 /*g/mL 0-ig/plate for Ames studies),
then the study should not be rejected on the basis of inadequately high dosing. The issue of
the limit dose is addressed in Dearfield et al., Mutation Research 258: 277, 1991, which is
appended in Attachment 6.
3. Rejection Factor; Insufficient (or inappropriate) tester strains used in Ames
assays (21)
EPA Guidance on this Factor: Guidance on this topic appears in Federal Register
50: 39439-39441 and Federal Register 52: 19078-19079, which states, in part, that the four
strains TA 1535, TA 1537, TA 98, and TA 100 should be used, and that the use of other
strains in addition to these four is left to the discretion of the investigator. It is noted that
while this was a major factor in the rejection of a number of studies in the period from 1985-
1987, there is no record that any Ames studies have been rejected on the basis of this
criterion since 1987. However, it is also possible that since that date any studies which were
deficient in this respect have not passed a screening review, and so have never been sent for
a complete review. Examination of the Subdivision F screening review.acceptance criteria
(p. C-124) indicates a potential problem, as TA 1536 (instead of TA 1537) is specified as
one of the four tester strains. Presumably the specification of TA 1536 is a typographical
error.
Industry Comment: The use of four strains is clearly specified in the Health Effects
guidelines. However, as noted earlier, the use of Health Effects guidelines for FIFRA
purposes was not proposed until 1989.
EPA Response: The Agency and industry agree on this point.
4. Rejection Factor; Tester strains not verified in Ames assays (17)
EPA Guidance on this Factor: According to Federal Register 52: 19078-19079, as
well as 40 CFR Part §798: "The requirement of histidine for growth should be demonstrated
for each strain. Other phenotypic characteristics should be checked using such methods as
crystal violet sensitivity and resistance to ampicillin. Spontaneous reversion frequency
140
-------
should be in the range expected either as reported in the literature or as established in the
laboratory by historical control values."
Industry Comment: It is possible that tests verifying test strain properties may have
been performed and not reported. One possible area of confusion might be the use of the
term "for each strain" in the Health Effect guideline for this assay. It is arguable whether
this term necessitates testing for each culture used in an assay or periodically. Many
laboratories use frozen culture aliquots prepared from a master culture to inoculate testing
cultures. If the master culture phenotype has been verified, then this should be sufficient.
Industry Recommendation: Clarification of this point would be useful.
EPA Response: The verification of tester strains for Salmonella assays is useful
when the responses of tester strains do not conform to expected standards or are unusual. If
the strains provide expected spontaneous background revertants, proper induction to positive
controls, and there are no other reasons to doubt the validity of the responses, then
verification should not be the sole reason for rejection of the study. However, if the strains
do not appear to be responding normally, then some verification of the tester strains (e.g.,
crystal violet sensitivity and resistance to ampicillin) should be performed before the
acceptance of that study. If, as industry implies, verification has been performed but not
reported, it should be reported if there appears to be any problem with the activity of the
tester strains.
5. Rejection Factor; For mammalian cells in culture, harvest time was not
determined bv cell-cvcle analysis (15)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "exposure shall be for a suitable period of time, in most cases 1 to 5 hours is effective;
exposure time may be extended over one or more cell cycles. (2) At the end of the exposure
period, cells shall be washed and cultured to determine viability and to allow for expression
of the mutant phenotype. (3) At the end of the expression period, which shall be sufficient
to allow near optimal phenotypic expression of induced mutants, cells should be grown in
medium with and without selective agent(s) for determination of numbers of mutants and
cloning efficiency."
Industry Comment: There is no apparent reason why gene mutation assays in
cultured mammalian cells would be expected to routinely have harvest times determined by
cell cycle analysis or why studies would be rejected based on this consideration. Harvest
times related to cell-cycle analysis are considerations for in vitro cytogenetic assays or SCE
assays and not for gene mutation assays using cultured mammalian cells. The only provision
141
-------
related to cell-cycle in the EPA guidance section for mammalian cell gene mutation assays is
an alternate provision that is an exposure consideration and not a harvest time consideration.
Industry Recommendation: Clarification of this point is requested, particularly as a
fairly large number of studies were apparently rejected for this reason.
EPA Response: The Agency agrees with industry that harvest times related to cell
cycle intervals are considerations generally for in vitro cytogenetic studies or SCE assays.
For mammalian gene mutation assays, cells are exposed for a limited period of time. Once
exposure is over, the next important time frame is the expression time where cells need to be
cultured for a sufficient period to allow newly induced mutations to be fixed and expressed.
In the future, mammalian gene mutation assays should not be rejected on this basis.
6. Rejection Factor; Missing protocol; missing raw data (10)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "in addition to the reporting recommendations as specified under 40 CFR Part § 792,
Subpart J the following specific information shall be reported: (i) Cell type, number of cell
cultures, methods for maintenance of cell cultures, (ii) Rationale for selection of
concentrations and number of cultures, (iii) Test conditions: composition of media, CO2
concentration, concentration of test substance, vehicle, incubation temperature, incubation
time, duration of treatment, cell density during treatment, type of metabolic activation
system, positive and negative controls, length of expression period (including number of cells
seeded and subculture and feeding schedules, if appropriate), selective agent(s). (iv)
Methods used to enumerate numbers of viable and mutant cells, (v) Dose-response
relationship, where possible."
Industry Comment: A large number of experimental conditions are mentioned in the
listed reporting requirements. It would be very helpful to know which of these are common-
ly not reported and are cause for rejection.
EPA Response: There were a total of 10 gene mutation assays which were rejected
with deficiencies in reporting. The specific factors involved in rejection (number of studies
in parenthesis) were the following: missing S9 preparation methods (4), missing procedures
for determining cell density (2), missing sufficiently detailed methods section and/or standard
protocol (2), revertant frequencies for positive controls not reported (2), missing individual
plate counts (1). The total adds up to 11 because more than one factor was involved in one
study.
142
-------
7. Rejection Factor; The results were 'Equivocal' (5)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. Usually an
equivocal study has only one dose. The guideline suggests that several concentrations
(usually four) of the test substance shall be used, and a concentration-related toxic effect
should occur. The highest dose should produce toxicity (or cytotoxicity) or should be a limit
dose (such as 5,000 jug/plate in the Ames assay).
Industry Comment: It is not clear what is meant by the term "equivocal" as a
rejection reason. The guidance section for this rejection factor states that the usual reason is
that only one dose level was tested, but testing an inadequate number of dose levels is also
specified as Rejection Factor 8 with the same number of rejected studies. Are the same
studies included in Rejection Factor 8 and Rejection Factor 7 or are there separate
"equivocal" reasons for rejection of those studies referenced in Rejection Factor 7?
EPA Response: Separate studies were used in Rejection Factor 8 and Rejection
Factor 7. Industry was unclear about the term "equivocal" as a rejection factor. Industry
stated that one reason for the confusion is that only one dose level tested was part of the
rejection consideration. Equivocal results from an initial assay should indicate that an
independent assay be subsequently performed to see if the response could be clarified (i.e.,
negative or positive). The Health Effects Testing Guidelines states that if a single positive
response is observed, a repeat assay should be performed. If, after multiple assays, the
response is still "equivocal", then no amount of additional testing with the same assay will
provide clarification. Therefore, the interpretation of the assay is questionable and a
different assay might be more appropriate. Generally speaking, when performing a
mammalian gene mutation assay in culture, several concentrations are used to help define a
concentration dependent response if the chemical is positive in the assay. It is unusual to use
only one concentration for gene mutation assays.
8. Rejection Factor; Only 1 dose was administered (5)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985. The guideline
states "several concentrations (usually four) of the test substance shall be used. These should
yield a concentration-related toxic effect."
Industry Comment: See comments for Rejection Factor 7 above.
EPA Response: Refer to response in Rej. Factor 7 above.
143
-------
B) Structural chromosome aberrations (§84-2)
a) Tests include: Mouse micronucleus assay and m vivo mammalian cytogenics assay
with rodent bone marrow
1. Rejection Factor; Dose levels were too low or no explanation of why this is the
maximum attainable concentration; no or low cvtotoxicitv
indicating that an insufficient level of test substance was
transported to the target tissue (MTD issue) (38)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "for an initial assessment, one dose of the test substance may be used, the dose being
the maximum tolerated dose (to a maximum of 5,000 mg/kg) or that producing some
indication of cytotoxicity, e.g., a change in the ratio of polychromatic to non-monochromatic
erythrocytes. Additional dose levels may be used. For determination of dose response, at
least three dose levels should be used."
Industry Comment: The discussion of this factor needs to be amplified to provide a
clearer understanding of precisely why studies were rejected. This seems particularly
important given the large number of studies affected by this rejection factor (38). Two
obvious problems are the use .of the term "maximum tolerated dose" and "some indication of
toxicity" in the guidance document. Maximum tolerated dose is a very vague term that can
have dramatically different interpretations between guidelines and between different offices
and departments of EPA. It is not particularly surprising, given the subjective nature of this
term, that differences of opinion might exist as to whether a particular dose is the "maximum
tolerated dose."
Industry Recommendation: This is an area where much more explicit guidance
would be very useful and is probably necessary to improve the acceptability of studies.
Specifically, clarification is needed on overt toxicity.
The Agency indicates that evidence of expousre to the target tissue, usually bone
marrow, is required for a fully acceptable valid test. However, this criteria is currently not
in the guidelines. If the Agency is planning to require this test, this new requirement should
be published as a proposed guideline and there should be an opportunity for public comment.
EPA Response: More explicit guidance is given concerning dose limits in terms of a
highest dose (5000 mg/kg) unless limited by toxicity (see Attachment 8, page 277). This can
be evidenced by signs of overt toxicity to the animals, or cytotoxicity to the target tissue
(e.g., significant bone marrow suppression in the frequency of PCEs in the bone marrow
micronucleus assay), or some high proportion (e.g., 50 to 80%) of the LD50. However,
using the LD50 in itself is not sufficient evidence to justify that a suitable dose level will be
144
-------
utilized. Without an indication of toxicity or dose limit, then the consideration of whether
test substance reached the target tissue comes into play. This would provide evidence that a
suitable amount of test substance was available for assaying potential effect(s) at the target
and would be used for the acceptability of the study. Chemicals of limited solubility should
be tested to the maximum amount feasible (e.g., water-insoluble chemicals may be dissolved
or suspended in other appropriate vehicles). EPA looks at signs of toxicity and is in a state
of transition regarding the evaluation of these studies. For a fully acceptable, valid test,
evidence of exposure to the target tissue, usually bone marrow (e.g., cytotoxicity, decreased
mitotic index, evidence of 14C distribution to the target site) is required. If no evidence of
exposure of the target tissue is apparent, this test may not be appropriate for analysis of the
test substance and another test protocol should be selected.
Guidance on the difference of the scenarios of acceptable tests.
It appears that some distinction is being made about degrees, or different methods, or
determining the acceptability of a mutagenicity test. It needs to be made clear that there are
no levels of acceptability. The test is either acceptable or unacceptable according to the
OECD and/or OPP protocol guideline. If the test has been performed appropriately and
there are no reasons for rejection, then it is an acceptable test.
The confusion seems to emanate from the composite review of all the mutagenicity
data and other information that may impact on what the mutagenicity data may indicate
(i.e., the weight of evidence assessment for genotoxic potential). This weight of evidence
approach for the available information (includes submitted results from the registrants and
published information) and what it means for possible additional testing is discussed in the
published Pesticide Assessment Guidelines, Subdivision F, Mutagenicity, Addendum 9 (dated
March, 1991; EPA publication no. 540/09-91-122, available from NTIS order no. PB91-
158394). In particular, section §84-2 part (c) indicates that additional testing may be
necessary to confirm or extend results from the initial mutagenicity battery. Also, additional
testing for heritable effects may be required as described in section §84-2 part (d).
A specific example was mentioned during the meeting and could help illuminate the
distinction between acceptability of individual tests and the interpretation of the results in the
overall weight of evidence assessment. For example, an in vivo micronucleus assay with
mouse bone marrow was performed. By all criteria in the protocol guideline, it is found to
be an acceptably performed and valid test (e.g., appropriate high dose, proper sampling
times, proper procedures according to GLPs). The test would be classified as acceptable.
Now, during the weight of evidence consideration using all test results from other submitted
tests and available information, it was determined that the bone marrow may not be an
appropriate target for assaying the toxicity of the test chemical (may not distribute there in
high enough concentration, or at all). Another target may be found more appropriate.
Therefore, additional testing may be required to extend the results from the available testing
as per section §84-2 part (c). In other words, the micronucleus test may be acceptably
performed and an acceptable test for that assay system, but it does not provide appropriate
145
-------
information in the overall weight of evidence. It is not rejected, but additional testing is
required to provide a better "picture" of the possible genotoxicity of the test chemical.
Guidance on criteria for overt toxicity
In many instances, the adequacy of the top dose will be examined on a case by case
basis as there may be unique circumstances surrounding particular test chemicals. The limit
dose is 5 g/kg. This is written in many of the test protocol guidelines and reaffirmed by the
OPP's Scientific Advisory Panel (SAP). The Agency realizes the OECD revision effort is
looking closely at the dose limit and that the Agency will comply with the final decision.
For now, the limit dose remains at 5 g/kg.
Another criterion for appropriate top dosing is signs of overt toxicity due to the test
chemical. Signs of adequate top dosing could include indices of cytotoxicity at the target
tissue/cells (e.g., cytotoxicity; shift in PCErNCE ratio for the micronucleus assay). Also,
clinical signs attributable to the test substance (not due to the administration of material only)
would be acceptable. Some mortality would also be acceptable as a sign of adequate top
dosing (not enough to reduce number of test animals below appropriate minimums, or too
much that would indicate dosing too high). For longer term tests, such as a dominant lethal
assay, slightly reduced body weight (e.g., > 10% body weight gain decrement over the 10
week mating period after dosing) would be adequate.
Dosing at a high fraction of a reported LD50 may not be sufficient by itself in the
absence of supporting toxic signs. The lack of any toxic signs at a high fraction of the LD50
can suggest several things, one of which is that the dosing solutions may not be correct. To
ensure that the animal system has been challenged by the test chemical at sufficiently high
doses, some sign(s) needs to be observed (e.g., cytotoxicity; overt toxicity; some mortality;
other appropriate criteria). If there are no signs and the dose is below the dose limit, this
could be a basis for rejection of the study. One should note that in most instances range-
finding experiments are critical and that the information from these experiments can provide
important insight for the dose selection.
2. Rejection Factor; Purity of test substance missing; analysis of concentration in
solution; or analysis of stability missing (12)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§80-4 pp. 22, 23, & 29), 40 CFR Parts
§160.105, 160.107, & 160.113 - Good Laboratory Practices, and the Standard Operation
Procedure #2000, Toxicology Branch/HED/OPP, Revised 9/30/85 with corrections from
Federal Register 52: 19080-19081, May 20, 1987. The chemical name, molecular structure,
and a quantitative and qualitative determination of its chemical composition (including
impurities, and contaminants) should be supplied to the Agency. Also, the manufacturer's
lot number of the test substance with its relevant properties of the substance tested, such as
146
-------
physical state, pH, stability and purity, and the identification and composition of any vehicles
used in administering the test substance are required by the Agency. For each test, control,
or reference substance that is mixed with a carrier, tests by appropriate analytical methods
shall be conducted to determine uniformity of the mixture as well as concentration and
stability in the carrier. (See Gene Mutation Tests - Rejection Factor #1 - comments
regarding the internal memorandum of July 5, 1991).
Industry Comment: See comments for Guideline 84-2(a), Rej. Factor 1.
EPA Response: Refer to response under Guideline 84-2(a), Rej. Factor 1.
3. Rejection Factor; Less than 3 dose levels were performed (10)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "for determination of dose response, at least three dose levels should be used."
Industry Comment: The guidelines do not clearly require three dose levels for
testing as implied in this rejection factor. The guidelines for in vivo bone marrow
chromosome aberration and micronucleus assays state that:
"For an initial assessment, one dose of the test substance may be used, the dose being
the maximum tolerated dose or that producing some indication of cytotoxic-
ity...Additional dose levels may be used. For determination of dose-response, at least
three dose levels should be used."
It is arguable whether a clear negative result observed at a single appropriate
maximum tolerated dose fulfills the requirement because there is no purpose in evaluating
dose-response for negative test substances.
Industry Recommendation: This point requires further discussion and more explicit
guidance on how many dose levels are required to define a clear negative response.
EPA Response: For a dose-response evaluation it is necessary to have at least 3 dose
levels. The top dose needs to be an appropriate top dose or limit dose and the other doses
selected to cover reasonable intervals on a dose response curve. For example, the second
dose may be approximately 50% of the high dose and the third dose no less than 25% of the
high dose. It is recognized that some investigators may wish to use a single dose for
screening purposes. If a single dose is used, this dose should be an appropriate top dose or
limit dose. If a positive response is found at this single dose, follow-up studies may be
necessary to define a dose-response curve. If there is a clear negative response (the single
147
-------
dose is an appropriate high dose and there are no other mitigating factors preventing a clear
negative call), then lower doses are probably not necessary.
4. Rejection Factor; Missing; individual clinical signs, body weight data, or raw
data (e.g.. route of administration, slide code information.
strain or source of animals) (8)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "in addition to the reporting recommendations as specified under 40 CFR Part §792,
Subpart J the following specific information shall be reported: (i) Species, strain, age,
weight, number and sex of animals in each treatment and control group, (ii) Test chemical
vehicle, dose levels used, rationale for dose selection, (iii) Rationale for and description of
treatment and sampling schedules, toxicity data, negative and positive controls, (iv) Details
of the protocol used for slide preparation, (v) Criteria for identifying micronucleated
erythrocytes. (vi) Dose-response relationship, if applicable."
Industry Comment: The guidance section contains a very long list of required data.
It would be very useful to know what specific items on this list are common reasons for
study rejection.
EPA Response: The specific items are: 1) missing individual clinical signs; 2)
missing body weight data; and 3) missing data. Missing data includes items such as route of
test substance administration, missing slide code information, and missing strain or source of
animals. These reasons account for approximately 11 percent of the total rejected studies for
Structural Chromosome Aberrations.
5. Rejection Factor: Inappropriate sampling times, or cells were not exposed
during the entire hematopoietic cycle (5)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "sampling times should coincide with the maximum response of the assay which varies
with the test substance." Therefore, using the highest dose, bone marrow samples should be
taken at least three times, starting not earlier than 12 hours after treatment, with appropriate
intervals following the first sample but not extending beyond 72 hours. When other doses
are used sampling shall be at the maximum sensitive period, or, if that is not known,
approximately 24 hours after treatment. One time point need only to be sampled if the most
sensitive interval is known and documented.
148
-------
Industry Comment: The Health Effects guidelines provide reasonably clear guidance
on sampling times for bone marrow assays. However, more recent scientific studies have
indicated that alternative treatment and sampling regimens are just as, if not more, useful for
these assays and these should be (and are) being considered for guideline revision.
EPA Response: The Agency agrees with industry that the Health Effects Testing
Guidelines provide reasonably clear guidance on sampling times for bone marrow assays.
There are recent advances that provide some alternative regimens and OPP is aware of these.
For example, it is understood that the 72 hour sampling time for the micronucleus assay is
no longer necessary and should not be a rejection criteria now. This will be taken into
consideration during the Agency's revision of the bone marrow assay protocols and the
harmonization with OECD. Final determination of these recent advances into the protocols
await harmonization with OECD.
b) Dominant lethal- rats or mice
1. Rejection Factor; MTD issue (4)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "normally, three dose levels should be used. The highest dose should produce signs of
toxicity (e.g., slightly reduced fertility) and slightly reduced body weight. However, in an
initial assessment of dominant lethality a single dose may be sufficient. Nontoxic substances
shall be tested at 5 g/kg or, if this is not practicable, then at the highest dose attainable."
Industry Comment: Better criteria for highest dose to be tested are preferable to the
vague term "maximum tolerated dose" used in other in vivo guidelines. An MTD may be
difficult to achieve in a material of low acute toxicity, since males are dosed for a short
duration in this study. A limit dose should be specified for relatively non-toxic test
substances. The wording quoted to define the highest dose in this document is not the same
as the wording in the Health Effects document published in the Federal Register.
Clarification of EPA guidance is needed.
EPA Response: As with all in vivo mutagenicity studies, the highest dose feasible up
to a limit dose (currently 5000 mg/kg; if this is revised during OECD harmonization, then it
will be 2000 mg/kg) is to be used in the dominant lethal assay. The other considerations for
appropriate top dosing or limitations by toxicity are similar to Rej. Factor 1 for structural
chromosome aberrations. An additional limiting consideration for dominant lethal testing
includes effects on fertility.
149
-------
2. Rejection Factor: No evidence that the test substance reached the target cells
(3)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states that the highest dose should produce signs of toxicity (e.g., slightly reduced fertility)
and slightly reduced body weight. This is evidence that the test substance has reached the
target cells. The highest dose should produce toxicity or the highest dose should be 5,000
mg/kg.
Industry Comment: The Health Effects guidelines do not state that the highest dose
should produce signs of toxicity (e.g., slightly reduced fertility) and (emphasis added)
slightly reduced body weight. They state that "The highest dose should produce signs of
toxicity (e.g., slightly reduced fertility)."
There is no provision in the guidelines for this assay that there must be a demon-
stration of the test substance reaching the target cells. The use of "e.g." clearly does not
require slightly reduced fertility or specify this as the only sign of toxicity that is an
appropriate measure for the highest dose. This reason for study rejection cannot be justified
based on the guidelines for this assay.
Industry Recommendation: The Agency should clarify their guidance in this area.
EPA Response: The Agency agrees that there should be clarification for the
appropriate top dosing of short-term in vivo tests. The parameters need to be examined
including, for example, signs of toxicity and evidence that the test substance reached the
target cells.
3. Rejection Factor; Low pregnancy rates (2)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985. The guideline
states "for example, in a typical experiment, the number of males in each group shall be
sufficient to provide between 30 and 50 pregnant females per mating interval."
Industry Comment: The guidelines provide explicit examples of an acceptable range
of pregnant females or a minimum number of pregnant females but they do not require a
specific pregnancy rate. Presumably the direct reason for rejection of these studies is that
the number of pregnant females was too low rather than a low pregnancy rate.
EPA Response: Industry comments on low pregnancy rates are correct. A minimum
number of pregnant females is necessary to provide an adequate power of significance to the
150
-------
test. If the test animals have a low pregnancy rate for whatever reason (chemical related,
reproductive effect), the study would have to take this into account and probably add
additional animals to ensure an appropriate minimum number of pregnant females. The
guidelines suggest ideally a minimum of 30 pregnant females per mating period.
4. Rejection Factor; Missing positive control (2)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "positive control substances shall be used at a dose which demonstrates the test
sensitivity."
Industry Comment: It is not entirely clear from the comment whether the rejection
factor is that no positive control data are provided or, possibly, that concurrent positive
control data were not provided. The Health Effects guideline indicates that acceptable
positive control results from experiments conducted within the last 12 months at the same
laboratory can be used instead of a concurrent positive control and rejection of studies which
provide such positive control data would be inappropriate. Agency clarification on this point
would be useful.
There are limited numbers of positive controls available. Most have been established
using the IP route, which may not be an appropriate route for a given test substance.
Industry Recommendation: The Agency should provide guidance on acceptable
positive controls.
EPA Response: The Agency wants either concurrent positive control or historical
positive control data conducted within 12 months of the definitive study at the same testing
laboratory. When feasible, the positive control should be administered in the same vehicle
and by the same route as the test chemical. When this is not feasible, an IP route of
administration is justified.
151
-------
C) Other genotoxic effects (mislabeled §84-4)
Industry Comment: Comments on the following rejection factors are the same as
Guideline 84-2, especially concerning the number of guidance documents that are available
(see comments). Agency consolidation and clarification of guidance is needed.
EPA Response: Rejection factors for this guideline have been addressed in the
preceding Mutagenicity guidelines (84-2(a) and 84-2(b)).
a) Tests include: in vitro unscheduled DNA synthesis (UDS) in rat hepatocytes, the
human Hela cell line, human fibroblasts, or rat kidney cells.
1. Rejection Factor: No analytical or stability data to define the test substance
concentration, purity, or solubility in solution question (25)
EPA Guidance on this Factor: Guidance appears in the Standard Operation
Procedure #2000, Toxicology Branch/HED/OPP, Revised 9/30/85 with corrections from
Federal Register 52: 19080-19081, May 20, 1987. In all tests, the composition of the test
substance must be precisely identified. [See § 80-3(b) (2) (iv).] When possible the technical
grade of the product should be tested. The chemical composition and physical state of the
substance should be the same as that encountered in the product. Failure to provide purity,
stability or other information on the test substance prevents the reviewer from determining if
the correct test substance was tested.
2. Rejection Factor: Missing; raw data, results for metabolic activation.
background frequencies for UDS. protocol; or that
insufficient data were presented to support conclusions (16)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "in addition to the reporting recommendations as specified under 40 CFR Part § 792,
Subpart J the following specific information shall be reported: (i) Cells used, density and
passage number at time of treatment, number of cell cultures, (ii) Methods used for
maintenance of cell cultures including medium, temperature and COz concentration, (iii) Test
chemical vehicle, concentrations used in the assay, (iv) Details of both the protocol used,
preparation of the metabolic activation system, and its use in the Agency, (v) Treatment
protocol, (vi) Positive and negative controls, (vii) Protocol used for autoradiography. (viii)
Details of the method used to block entry of cells into S phase, (ix) Details of the methods
used for DNA extraction and determination of total DNA content in LSC determinations, (x)
Historical background incorporation rates of 3H-TdR in untreated cell lines, (xi) Dose-
response relationship, if applicable."
152
-------
3. Rejection Factor; MTD issue, no evidence of cyfotoxicity. missing dose
selection data (15)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "multiple concentrations of test substance, based upon cytotoxicity and over a range
adequate to define the response should be used. For cytotoxic chemicals, the first dose to
elicit a cytotoxic response, in a preliminary assay should be the highest dose tested."
4. Rejection Factor; High cytoplasmic grain count in solvent control; repeat study
with different rat hepatocyte preparation, or lower
cvtoplasmic background, or high cytoplasmic and nuclear
grain counts, or counts were not provided (8)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "cytoplasm adjacent to the nuclear areas should be counted to determine spontaneous
background. Background counts should be subtracted to give the correct nuclear grain count.
Values should be reported as net grain per nucleus. Mean, median and mode may be used to
describe the distribution of net grains per nucleus." High cytoplasmic backgrounds make it
difficult if not impossible for the reviewer to reach any valid conclusions.
5. Rejection Factor; Duplicate cultures were not performed (8)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "six independent cultures at each concentration and control should be used in LSC
UDS determinations."
b) Tests include: In vitro transformation assay: a) BALB/3T3 (mouse); BO C3H10T 1/2
(mouse).
1. Rejection Factor; Onlv one test dose used, or MTD issue (5)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
153
-------
indicates that if only one dose is used the concentration should be limited by toxicity,
solubility or 5,000 ^gfrnl.
2. Rejection Factor; Purity and stability of test substance (2)
EPA Guidance on this Factor: Guidance appears in the Standard Operation
Procedure #2000, Toxicology Branch/HED/OPP, Revised 9/30/85 with corrections from
Federal Register 52: 19080-19081, May 20, 1987. In all tests, the composition of the test
substance must be precisely identified. [See § 80-3(b) (2) (iv).] When possible the technical
grade of the product should be tested. The chemical composition and physical state of the
substance should be the same as that encountered in the product. Failure to provide purity,
stability or other information on the test substance prevents the reviewer from determining if
the correct substance was tested.
c) Tests include: In vivo sister chromatic! exchange (SCE): (a) Chinese hamster; (b) rat
bone marrow
1. Rejection Factor; MTD issue; dose selection not supported by range-finding
study; no cytotoxicity was indicated at highest dose (16)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985, and in the
Phase 3 Technical Guidance document dated December 24, 1989, C 124-126. The guideline
states "for an initial assessment, one dose of the test substance may be used, the dose being
the maximum tolerated dose or that producing some indication of toxicity as evidenced by
animal morbidity (including death) or target cell toxicity. The LD50 is a suitable guide.
Additional dose levels may be used. For determination of dose-response, at least three dose
levels should be used." Although a range-finding study is not required by the guidelines, it
would demonstrate that the maximum tolerated dose was used.
2. Rejection Factor; No analytical data to support test substance stability.
concentration, or missing test substance purity (5)
EPA Guidance on this Factor: Guidance appears in the Standard Operation
Procedure #2000, Toxicology Branch/HED/OPP, Revised 9/30/85 with corrections from
Federal Register 52: 19080-19081, May 20, 1987. In all tests, the composition of the test
substance must be precisely identified. [See § 80-3(b) (2) (iv).] When possible the technical
grade of the product should be tested. The chemical composition and physical state of the
substance should be the same as that encountered in the product. Failure to provide purity,
154
-------
stability or other information on the test substance prevents the reviewer from determining if
the correct substance was tested.
3. Rejection Factor: Missing procedural descriptions (4)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985. An example of
missing procedural description: correlation between the morphologically altered types of foci
on monolayer cells and their potential to form tumors when injected into immunosuppressed
mice was not included. The guideline states "in addition to the reporting recommendations
as specified under 40 CFR Part §792, Subpart J the following specific information shall be
reported: (i) Species, strain, age, weight, number and sex of animals in each treatment and
control group, (ii) Test chemical vehicle, dose level used, rationale for dose selection,
toxicity data, negative and positive controls. (Hi) Route and schedule of administration of
both test chemical and BrdU. (iv) Identity of spindle inhibitor, its concentration and duration
of treatment, (v) Time of sacrifice after administration of BrdU. (vi) Details of the protocol
used for slide preparation, (vii) Criteria for scoring SCE. (viii) Dose-response relationship, if
applicable."
4. Rejection Factor; Inadequate statistical analyses (3)
EPA Guidance on this Factor: Guidance for this appears in the Health Effects
Testing Guidelines, Federal Register 50: 39442-39458, September 27, 1985. The guideline
states "data should be evaluated by appropriate statistical methods."
155
-------
GUIDELINE 85-1 GENERAL METABOLISM
1. Rejection Factor; Inadequate or missing data on identification of metabolites
(37)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§85-1 pp. 152-156), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 127-128. Subdivision F states that a
metabolism study may be performed to identify, and to the extent possible, quantify
significant metabolites, while under section "Excretion" (p. 155) it is stated that urine and
feces from all dose groups should be analyzed by suitable methods in order to determine the
extent of absorption and biotransformation and to identify the metabolites. The studies were
rejected on the basis of either: 1) an inadequate description of how urinary and fecal
metabolites were identified, 2) no collection of samples, 3) poor documentation of metabolite
identification, or 4) pooling of urine and feces samples, resulting in a lack of individual
animal data.
Although a specific percentage of excreted test substance is not mentioned as a cutoff
for identification of metabolites in urine and feces, the Agency has accepted studies in which
any given metabolite representing at least 10% of the administered dose was identified. In
addition, it has not been made clear within either subsection of §85-1 that data should be
collected from individual animals, although this is the usual methodology employed in
conducting a metabolism study. Both the requirement for the extent of metabolite
identification in urine and feces as well as the requirement for collection of individual data
are being clarified in the revision of §85-1.
Based upon the rejection factors in paragraph #1 above, it is clear that if either
complete documentation on metabolite identification or adequate methodology had been used,
that rejection of studies could have been avoided. However, it is also clear that rejection of
metabolism studies based on the above criteria are also partly the result of scientific
judgment on the part of the Agency, and not the failure to satisfy the data requirement for
identification of metabolites. Registrants have, in general, tried to comply with these data
requirement. Lack of clarity within the guideline, however, has made enforcement difficult
and open to interpretation, and is an additional reason for the current effort at revising §85-1.
Industry Comment: We agree that the guidelines are not clear in this area. While it
is agreed that the percent of administered dose in urine and feces at each time point for
individual animals is necessary, it is not obvious that metabolite identification and collection
of HPLC/TLC profiles for each individual animal adds useful data. The comment that
studies were rejected because of "pooling of urine and feces samples, resulting in a lack of
individual animal data" appears to be inconsistent with the Agency's own guidance and is
thus disturbing. The Phase 3 Technical Guidance Document, p. C#127-128, specifically
states that pooled urine and feces by dose group may be used for the qualitative analysis of
156
-------
urine and feces to detect metabolism and identify metabolites. It is recommended that the
Agency not change this guidance. Metabolite identification is already a major cost of
conducting these studies. Requiring qualitative and/or quantitative analysis of urine and feces
for metabolites on an individual animal would greatly increase the cost of these studies and is
unwarranted given the current uses of these data.
It is also recommended that the Agency not require registrants to routinely identify
minor metabolites (i.e., those representing less than 10% of dose). Identification of
metabolites is a major portion of the effort required to conduct metabolism studies.
Requiring minor metabolites to be identified on a routine basis would greatly increase the
cost of these studies and is difficult to justify scientifically.
Finally, EPA guidance is not clear on what is an acceptable recovery of radioactivity
in individual animals; do all 5 animals require >90% recovery or is an average recovery of
>90% in the group acceptable? On occasion with 5 animals/sex/group dosed, 1 animal may
have a recovery of less than 90%. Is it necessary to dose another animal to have a data set
of 5 animals with radioactivity recovery >90%? No real additional information is gained in
this process. In future revisions of this guideline, this issue should be addressed.
Industry Recommendation: It is recommended that the Agency not change the
guidance appearing in the Phase 3 Technical guidance Document regarding pooling of urine
and feces samples. It is also recommended that the Agency not require registrants to
routinely identify minor metabolites. In future revisions of this guideline, acceptable
recovery of radioactivity in individual animals should be addressed.
EPA Response: OPP agrees there is inconsistency between the current pesticide
metabolism guidelines (Subdivision F, § 85-1) and the Phase 3 Technical Guidance
Document for reporting of individual animal data on metabolites in urine and feces.
Accordingly, it is OPP's position that the use of pooled animal data for identification of
urinary and fecal metabolites is acceptable. With regard to an acceptable level of recovery,
it is recommended that >95% of the administered dose be accounted for within each dose
group. Metabolites of administered test substance which represents >5% of the
administered dose should be identified. These recommendations are open for discussion
between OPP and Industry.
2. Rejection Factor; Improper methodology or dosing regimen (26)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
Evaluation: Human and Domestic Animals (§85-1, pages 152-156), and in the Phase 3
Technical Guidance document dated December 24, 1989, C 127-128. Several aspects of
improper methodology were rejection factors. These included: 1) no individual dosing data
or animal weight data, 2) contamination of excreta samples, 3) overall poor documentation of
laboratory procedures used, 4) use of unhealthy animals, 5) lack of one or more dose groups,
157
-------
and 6) improper reporting of excretion and distribution data. No specific statements within
§85-1 address these issues, with the exception of (5) and (6) above. For (5), it is stated in
subdivision F that four groups of animals should be studied: a single intravenous dose group
tested at the low dose; a single low oral dose; a series of repeated low oral doses followed
by a single radiolabeled dose after 14 repeated unlabeled doses; and a single oral high dose.
Submitted studies rejected on the basis of #5 above were rejected due to a lack of one or
more of these dose groups, or the attempt to substitute other studies in their place (for
example, substitution of a 28 or 90-day oral dose study for a repeated oral dose group).
There is currently no lack of clarity as to the required dose groups for a metabolism study
under §85-1. There is, however, a lack of stated purpose for such dose groups. Revised
metabolism guidelines will state the required dose groups and the specific purpose for each.
Currently, the Agency favors a tiered approach to metabolism testing in accordance with the
AIHC recommendations, and this will be taken into consideration in the revision of §85-1.
Subdivision F states on p. 155, that quantity of isotope, together with recovery of
percent administered dose in feces, urine, ... should be included in the test report. Thus,
while not entirely clear, data on excretion and distribution gathered from a metabolism study
should at least be reported in terms of % administered dose. The specific reporting
requirements for metabolism studies will be clarified in the revised metabolism guidelines.
Industry Comment: We support the proposal for a tiered approach. While the
purpose of each of the dosing regimens is fairly clear to metabolism scientists, it is also cleat
that all of the dosing regimens are not necessary for most test substances and will not yield
information that is useful for regulatory purposes.
EPA Response: OPP agrees that not all dose groups should be required in
metabolism testing. The decision of which specific dose groups to use in the study remains
to be discussed.
3. Rejection Factor; Inadequate number of animals (less than 5/sex/group) were
used in the dose groups (19)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§85-1 pp. 152-156), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 127-128. As stated on page 153 that at
least ten animals (five males and five females) should be used at each dose level. The
Agency is currently revising the requirement for 5 animals/sex/dose group, as it is evident
that in many cases, such numbers of animals are not necessary. However, until publication
of these revised guidelines, registrants will be required to use 5 animals/sex/dose group in
metabolism studies, unless adequate justification is submitted with the study for the use of
less than 5 animals/dose group.
158
-------
Industry Comment: The Agency is to be complimented on the decision to reduce
the number of animals per dose group when the guideline is revised.
Industry Recommendation: Given this decision and the fact that the OECD
guideline permits fewer animals per dose group, leniency should be exercised before
rejecting studies with fewer than 5 animals/sex/dose group.
EPA Response: OFF and Industry are in agreement that the number of animals used
in dosing groups should be reduced to harmonize with OECD guidelines. OFF will be
flexible in evaluation of submitted studies using less than 5 animals per dose group.
4. Rejection Factor; No individual animal data (13)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§85-1 pp. 152-156), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 127-128. The requirement for individual
animal data is not specifically stated in §85-1, but it is listed in the Phase 3 Technical
Guidance document dated December 24, 1989, C#127-128. This factor is based more on
scientific judgment and adherence to Good Laboratory Practice. The lack of individual
animal data is a subject which will be clarified in the revised §85-1 guidelines. Currently,
the lack of such data adds difficulty to the review process, for without it, variations in the
response of each animal to treatment as well as verification of the mean data presented
cannot be determined. This becomes particularly important when dose- or sex-related
differences in biotransformation are present.
Industry Comment: The guideline for reporting metabolism studies (Subdivision F,
Series 81-4 - Data Reporting) lists individual animal data under Optional Sections. Guidance
on this factor does not appear to be consistent. The Phase 3 Technical Guidance Document
states that pooled urine and feces by dosing group may be used to obtain a qualitative
analysis of urine and feces to detect metabolism and identify metabolites. Perhaps the
rejection rate analysis should clarify what data on individual animals is required.
Industry Recommendation: It is recommended that the Agency review this apparent
new requirements to routinely report individual data. There does not appear to be any strong
basis for this requirement.
EPA Response: OPP agrees that the reporting requirements for individual animal
data are in need of clarification. This issue can be addressed in future discussions between
OPP and Industry.
159
-------
5. Rejection Factor: Improper reporting (12)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§85-1 pp. 152-156), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 127-128. The definition of this rejection
factor, includes: 1) no statement of adherence to Good Laboratory Practice Standards; 2) no
data on purity and stability of the test chemical; 3) no justification for dose levels used; 4)
discrepancies between results reported in the body of the report and those found within data
tables; and 5) submission of interim reports.
For (2) and (3) above, some references are made in §85-1 (pages 153-154), but
clarification is required in the revised guidelines. Up until the present time, the registrant
has usually supplied data on purity and stability of test chemical and justification for dose
levels used which go beyond the current guideline requirements. However, it is clear that a
guideline specifying the submission of purity and stability data for the test substance and dose
solutions, as well as adequate explanation of the reasons for selection of dose levels for
metabolism studies is required.
As for the other categories (1, 4, and 5 above), these appear to be oversights on the
part of the registrant and could be avoided prior to submission of final reports. Clarification
of submission of interim reports is required on the part of the Agency, and will be dealt with
in the revision of metabolism guidelines. It is recommended that in the interim, any
preliminary reports or special studies on the metabolism of pesticides be submitted as
non-guideline studies unless submitted specifically in support of §85-1.
Industry Comment: Agency guidance is clear.
EPA Response: The Agency and Industry agree on this point.
6. Rejection Factor; Inadequate or missing tissue residue analysis data (11)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§85-1 pp. 152-156), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 127-128. Subdivision F indicates that the
test report shall include the following data derived from tests on animals in all dose groups:
...and the following tissues and organs of animals in all groups: bone, brain, fat, testes,
heart, kidney, liver, lung, blood, muscle, spleen, tissues with displayed pathology, uterus,
and residual carcass. The registrant should be aware of the specific requirements for tissue
residue analysis under current guidelines.
Currently, there is a lack of understanding as to why certain tissues are required and
for what purpose these data are requested. It is acknowledged that in the majority of cases,
data on terminal distribution do not provide significant information to the Agency in the
160
-------
absence of such factors as long half-life or binding of radiolabel to specific tissues or organs.
Thus, the requirement for such data is undergoing review and will be clarified in the revised
metabolism guideline. It is anticipated that if tissue distribution data are required, that these
data will be required for the early time points following pesticide administration, and not at
terminal sacrifice. In the absence of revised guidelines, it is recommended that the registrant
submit terminal tissue distribution data for the tissues as listed on page 155-156.
Industry Comment: There is no question that the guideline is clear as to which
tissues need to be collected and analyzed. However, some leniency should be exercised
before rejecting studies where not all the listed tissues have been analyzed, since the Agency
acknowledges "that in the majority of cases, data on terminal distribution do not provide
significant information to the Agency" and it is unclear how or whether the Agency has ever
used these data in a hazard or safety evaluation.
Industry Recommendation: The need for these data should be considered carefully
when the guideline is revised. Some leniency should be exercised before rejecting studies
where not all the listed tissues have been analyzed.
EPA Response: OPP agrees that while the guideline is clear as to the list of required
tissues for analysis, this list needs revision and clarification. The currently required list often
does not yield significant information to OPP in the absence of such factors as extended
half-life or binding of radiolabel to specific tissues or organs. OPP agrees to be flexible in
evaluation of studies submitted with less than the required list of tissues.
7. Rejection Factor; Testing at only one dose level (6)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§85-1 pp. 152-156), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 127-128. The original intent of this section
was to provide information on the disposition of a pesticide at a low non-saturating
concentration of test chemical as well as a high expected saturating concentration of test
chemical. In reality, two doses do not provide a meaningful picture of the possible array of
enzyme kinetics occurring at different doses of the test chemical. In the revised metabolism
guideline, it is expected that, using a tiered approach, more than 2 doses of test chemical will
be requested to obtain a more accurate picture of the enzyme kinetics as a function of dose.
Industry Comment: The Agency is to be commended for revising the guidelines.
We agree that a tiered approach is best. There are many cases where a single dose study
could provide enough information for registration and risk assessment. We also agree that a
simple multiple dose study might be used to determine information on the kinetics of
metabolism and excretion.
161
-------
EPA Response: OPP acknowledges that specific situations might exist for the use of
either single or multiple dose groups. This issue can be addressed in future discussions
between OPP and industry.
8. Rejection Factor: Only one sex of animal used (5)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§85-1 pp. 152-156), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 127-128. As stated on page 153, both
sexes of test animal are currently required for metabolism testing. The need for both sexes
has been a subject of considerable debate, and revision to §85-1 will consider the need for
testing of the metabolism of a test chemical in both sexes. Until revised guidelines appear
stating otherwise, both sexes will be required for a metabolism study.
Industry Comment: The Agency should exercise discretion in rejecting studies
which test only one sex. Unless there is information in the rest of the toxicology package
indicating large differences in toxicity between sexes, this testing probably does not add to
the risk assessment.
Industry Recommendation: The Agency should revise its guidance to allow some
discretion or describe conditions where testing on one sex can be accepted.
EPA Response: Until further discussion on the use of one or both sexes of animals
for metabolism testing, OPP will be flexible in its acceptance of metabolism studies
conducted in only one sex.
9. Rejection Factor; Lack of an intravenous dose group (4)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§85-1 pp. 152-156), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 127-128. According to Subdivision F,
page 154, a low dose intravenous dose group is currently required if the test chemical is
soluble in water or physiological saline. Clearly, many of the test chemical submitted for
registration are insoluble to some degree in these media, which usually excludes the
requirement for such a dose group. While an intravenous dose group is currently a
requirement, there is no clear purpose stated within §85-1 as to why such a dose group is
required. However, it is also clear that since the promulgation of this guideline, other dosing
vehicles are now available which are biologically compatible with intravenous dosing. The
issue of dosing by the intravenous route will be considered in the revision of existing
metabolism guidelines.
162
-------
Industry Comment: The pharmacological reason for collecting these data is known
and accepted. However, in most cases, it seems unnecessary to collect these data for
regulatory purposes.
Industry Recommendation: A consensus agreement to the usefulness of these data
should be reached before continuing this requirement.
EPA Response: OPP agrees with Industry that a consensus with regard to the
requirement for an intravenous dose group should be reached. This issue can be addressed
in future discussions between OPP and Industry.
10. Rejection Factor; No collection of I4CO? (3)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: human and domestic animals (§85-1 pp. 152-156), and in the Phase 3 Technical
Guidance document dated December 24, 1989, C 127-128. In Subdivision F page 155, it is
stated that if a preliminary study shows that no volatile label materials are exhaled during the
period of 0-24 hours after dosing, such evidence may be submitted in lieu of measuring label
in the expired air for this study. Thus, it is clear that at least preliminary work be performed
to determine the extent of excretion of test chemical by this route.
Industry Comment: These data should not be required if recoveries in the studies
are >80%.
EPA Response: OPP disagrees with the industry comment that no collection of
carbon dioxide be made if recoveries in the study are >80%. The position is maintained
that pilot studies be conducted in order to determine the extent of excretion of radiolabel by
this route. If excretion by this route is found to be less than 1 percent within a 24 hour
collection period, then no further testing for excretion by this route is necessary.
163
-------
GUIDELINE 85-2 DERMAL PENETRATION
Industry General Comments and Recommendations: In our opinion, some studies
rejected over the years have been scientifically acceptable and provided all of the information
really needed by the EPA for regulatory purposes, but did not meet the unofficial protocol.
The only official guidance for this study was recently published (1991) and is still in draft
form. However, it is our understanding that the Agency has now modified its thinking and is
requesting a much less intensive study than previously desired. Clearly, the EPA needs to
determine how the results from a study will be used prior to requiring that a study be
conducted according to a specified protocol.
1. Rejection Factor; Incomplete/missing data evaluation (carcass, feces. urine.
etc.)
EPA Guidance on this Factor: Guidance on this appears in Subdivision F - Hazard
evaluation: Dermal Absorption Studies of Pesticides (Draft) (page 6). The total test
substance must be determined in the feces, urine, blood, wash from the skin, material in or
on the protective device, material in the selected organs (if collected) and residue in the
carcass. Concentration of the test substance should be determined in the blood and any
organ samples collected.
Adequate guidance was provided for this guideline, therefore this was an avoidable
rejection factor.
Industry Comment: We tend to agree that guidance is adequate on this, however, as
pointed out in the Agency's comments, guidance exists as a draft or in a recently released
guidance document (1991). Many of the rejections reflect the considerable evolution of the
guidance between 1985 and 1991.
Industry Recommendation: The Agency should work hard to minimize the time
period when guidelines are in flux and remain flexible about accepting studies conducted
using older guidance.
EPA Response: A draft has been prepared and further work towards completion of
the document which would be suitable for publication as a guideline is in abeyance due to
lack of budgetary support.
2. Rejection Factor; Improper test substance preparation/application (6)
EPA Guidance on this Factor: Subdivision F Hazard Evaluation: Human and
Domestic Animals §85-2 Dermal Absorption Studies of Pesticides (Draft) p. 1-3. The test
substance should be chemically pure and radiolabelled, usually with carbon-14 on a "core"
164
-------
part of the test substance. Labeled test substance may not be required if a sufficiently
selective and sensitive physical/chemical test for identifying the test substance is used. Doses
should be prepared such that equal quantities of radiolabel are present in each dose. Test
substance should be spread evenly over the rats skin in an area no less than 10 cm2.
The test substance should be exposed for durations of 1/2, 1, 2, 4, 10 and 24 hours.
The application site should be gently washed with soap and water. No particular type of
soap is specified, but a suggestion would be Ivory® or Dove® dishwashing detergent. The
application site should be covered during each exposure.
Adequate guidance was provided for this guideline, therefore this was an avoidable
rejection factor.
Industry Comment: See Rejection Factor 1 above.
EPA Response: Refer to response in Guideline 85-2, Rej. Factor 1. Additionally,
the Agency agrees to be flexible about accepting studies conducted under the existing
guidance.
3. Rejection factor; Raw data missing and incomplete summary tables (6)
EPA Guidance on this Factor: Subdivision F Hazard Evaluation: Human and
Domestic Animals §85-2 Dermal Absorption Studies of Pesticides (Draft) p. 7. Study results
shall be summarized and illustrated by tables and figures presenting mean values for each
experimental group of four animals. Individual animal data should also be included in
appendices.
Adequate guidance was provided for this guideline, therefore this was an avoidable
rejection factor.
Industry Comment: See Rejection Factor 1 above.
EPA Response: Refer to response under Guideline 82-2, Rej. Factor 2.
4. Rejection factor; No signed Quality Assurance Statement (4)
EPA Guidance on this Factor: Guidance on this appears in the 40 CFR §160.35,
and Phase 3 Technical Guidance dated December 24, 1989, 5-6. A signed and dated Quality
Assurance Statement should be included specifying the dates of inspections.
This signed and dated statement should be included with any study stating that: a) the
study was conducted in accordance with this part or b) there are differences between the
165
-------
practices used in the study and those required by this part; or c) the person was not a
sponsor of the study, and does not know whether the study was conducted in accordance with
this part.
Adequate guidance was provided for this guideline, therefore this was an avoidable
rejection factor.
Industry Comment: We agree that guidance is adequate on this point.
EPA Response: The Agency and Industry agree on this point.
5. Rejection Factor; Missing purity or concentration of test substance (4)
EPA Guidance on this Factor: Guidance on this appears in the Subdivision F:
Hazard Evaluation: Humans and Domestic Animals; §85-2 Dermal Absorption Studies of
Pesticides (Draft) (pages 2 and 3.) The test substance should be of known chemical purity
and radiolabeled, usually with carbon-14, in a position which is part of the "core" test
substance. Dose preparations should be prepared so that sufficient quantities of radiolabel
are present in each dose to allow sufficient sensitivity for that dose. Also, it is critical to
know the actual concentration(s) obtained so that the actual administered dose may be known.
Adequate guidance was provided for this guideline, therefore this was an avoidable
rejection factor.
Industry Comment: We agree that guidance is adequate on this point.
EPA Response: Refer to response under Guideline 81-1, Rej. Factor 1.
166
-------
SUMMARY TABLE OF REJECTION FACTORS
GUIDELINE
REJECTION FACTOR
ACUTE ORAL TOXICITY
81-1
-Lack of characterization of the test substance
-Inadequate dose levels to calculate LDSO
ACUTE DERMAL TOXICITY
81-2
ACUTE AND 90-DAY
INHALATION
81-3 and 82-4
PRIMARY EYE IRRITATION
81-4
-Lack of characterization of the test substance
-Inadequate percentage of body surface area exposed
-No quality assurance statement
-Improper number of animals tested per dose group
-Only one sex tested
-Omitted source, age, weight, or strain of test animal
-Less than 25% of particles were < 1 fim
-Three exposure concentrations were not used; LC50 could not be
calculated; highest concentration did not produce toxicity
-Inadequate reporting of exposure methodology
-Protocol errors
-Lack of characterization of the test substance
-Test substance preparation
-Chamber concentration not measured
-Lack of characterization of the test substance
PRIMARY DERMAL
IRRITATION
81-5
DERMAL SENSITIZATION
81-6
-Lack of characterization of the test substance
-No quality assurance statement and/or no Good Laboratory Practice
(GLP) statement
-Improper test substance application/preparation
-Omitted source, age, weight, or strain of test animal
-Missing individual/summary animal data
-Control problems
-Dosing level problems
-Lack of characterization of the test substance
-Unacceptable protocol or other protocol problems
-Individual animal scores or data missing
-Scoring method or other scoring problem
-Reporting deficiencies or no quality assurance statement
167
-------
90-DAY FEEDING
RODENT
82-l(a)
90-DAY FEEDING
NON-RODENT
82-l(b)
-A NOEL was not established
-Lack of characterization of the test substance or incorrectly reported
-Lack of clinical chemistry and/or lack of histopathology
-Reporting deficiencies
-Lack of characterization of the test substance
-A NOEL was not established
-An investigational parameter missing
-Information on the pilot study and other problems associated with dose
level selection
21-DAY DERMAL TOXICITY
82-2
-Lack of characterization of the test substance
-Raw data analyses incomplete or missing
-A systemic NOEL was not established
-Inadequate percentage of body surface area exposed in each dose group
-Insufficient number of dose levels tested
90-DAY DERMAL TOXICITY
82-3
CHRONIC FEEDING/
ONCOGENICITY -
RATS
83-l(a) and 83-2(a)
ONCOGENICITY -
MICE
83-2(b)
-Lack of characterization of the test substance
-A systemic NOEL was not established
-Incomplete/missing raw animal data analyses
-Insufficient number of dose levels tested
-Poorly controlled test environment
-Missing histopathology information
-Missing information in study reports
-MTD was not achieved
-Missing historical control data
-Lack of characterization of the test substance
-Deficiencies in reporting the study data
-Histopathology information missing
-MTD was not achieved
-Lack of historical control data
-Information missing in study reports
-Lack of characterization of the test substance
-Deficiencies in reporting of study data
168
-------
DEVELOPMENTAL TOXICITY -
RODENTS
83-3(a)
-Missing historical controls
-Lack of characterization of the test substance
-Information missing or requiring clarification of the laboratories
methods
-Information missing or requiring clarification of the laboratories results
-A NOEL was not established
-Statistical problems
-Did not use conventional assessments for skeletal or visceral
examinations
DEVELOPMENTAL TOXICITY -
NON-RODENTS
83-3(b)
REPRODUCTION
83-4
MUTAGENICITY TESTING
84-2
A) Gene mutations
-Clarification of laboratory procedures or interpretation of the data
-Individual maternal or fetal data missing
-Missing historical controls
-Lack of characterization of the test substance
-Excessive maternal toxicity
-A NOEL was not established
-Statistical problems
-Information missing from laboratory results
-Lack of characterization of the test substance
-Information missing or requiring clarification of laboratory methods or
results
-Missing historical controls
-A NOEL was not established due to effects at the lowest dose tested
-Low fertility and/or inadequate number of animals were used per dose
level
-A NOEL was not established in the absence of reproductive effects
-Purity, batch numbers, stability, or analytical concentration information
missing
-MTD issue, no range-finding study; inadequate high dose; no evidence
of toxicity at any dose; insufficient (or no) cytotoxicity and limit-dose
level (5000 ^g/plate) not reached and/or test substance not tested up to
solubility limits
-Insufficient (or inappropriate) tester strains used in Ames assays
-Tester strains not verified in Ames assays
-For mammalian cells in culture, harvest time was not determined by
cell-cycle analysis
-Missing protocol; missing raw data
-The results were "equivocal"
-Only 1 dose was administered
169
-------
B) Structural chromosome aberrations
1) Tests include: Mouse micronucleus assay and in vivo mammalian
cytogenics assay with rodent bone marrow
-Dose levels were too low or no explanation of why this is the
maximum attainable concentration; no or low cytotoxicity indicating that
an insufficient level of test substance was transported to the target tissue
(MTD issue)
-Purity or test substance missing; analysis of concentration in solution;
or analysis of stability missing
-Less than 3 dose levels were performed
-Missing: individual clinical signs, body weight data, or raw data (e.g.,
route of administration, slide code information, strain or source of
animals)
-Inappropriate sampling times, or cells were not exposed during the
entire hematopoietic cycle
2) Dominant lethal - rats or mice
-MTD issue
-No evidence that the test material reached the target cells
-Low pregnancy rates
-Missing positive control
C) Other genotoxic effects
1) Tests include: in vitro unscheduled DNA synthesis (UDS) in rat
hepatocytes, the human Hela cell line, human fibroblasts, or rat kidney
cells
-No analytical or stability data to define the test substance
concentration, purity, or solubility in solution question
-Missing: raw data, results for metabolic activation, background
frequencies for UDS, protocol; or that insufficient data was presented to
support conclusions
-MTD issue, no evidence of cytotoxicity, missing dose selection data
-High cytoplasmic grain count in solvent control; repeat study with
different rat hepatocyte preparation, or lower cytoplasmic background,
or high cytoplasmic and nuclear grain counts, or counts were not
provided
-Duplicate cultures were not performed
2) Tests include: In vitro transformation assay: a) BALB/3T3 (mouse);
(BO C3H10T 1/2 (mouse)
-Only one test dose used, or MTD issue
•Purity and stability of test substance
170
-------
3)
METABOLISM
85-1
Tests include: In vivo sister chromatid exchange (SCE): (a) Chinese
hamster; (b) rat bone marrow
-MTD issue; dose selection not supported by range-finding study; no
cytotoxicity was indicated at highest dose
-No analytical data to support test substance stability, concentration, or
missing test substance purity
-Missing procedural descriptions
-Inadequate statistical analyses
-Inadequate or missing data on identification of metabolites
-Improper methodology or dosing regimen
-Inadequate number of animals were used in the dose groups
-No individual animal data
-Improper reporting
-Inadequate or missing tissue residue analysis data
-Testing at only one dose level
-Only one sex of animal used
-Lack of an intravenous dose group
-No collection of '"CO2
DERMAL PENETRATION
85-2
-Incomplete/missing data evaluation
-Improper test substance preparation/application
-Raw data missing and incomplete summary tables
-No signed quality assurance statement
-Missing purity or concentration of test substance
171
-------
vm. CONCLUSIONS
Overall, the rejection rates for toxicology are low and have remained relatively
constant over time. The 1985 rejection rate was 6 percent; the 1986-1988 rejection rate was
6 percent, and the post-1988 rejection rate is 7 percent. A substantial number of toxicology
studies have been rated supplementary (i.e., upgradable), and the supplementary rate appears
to be rising. The 1985 aggregate supplementary rate was 7 percent; the 1986-1988
supplementary rate was 11 percent, and the post-1988 supplementary rate is now 12 percent.
The CORT studies, which can take up to four years to complete, have shown
substantial reduction in their rejection rates. The 1985 rejection rate for the Chronic
Toxicity and Oncogenicity - mouse studies [83-1 (a) and 83-2(b)] was 26 percent, the 1986-
1988 rate fell to 22 percent, and the current (post 1988) rejection rate is 5 percent. For the
Developmental - rat study [83-3(a)], the rejection rate has fallen from 13 percent in 1985 to
9 percent in 1986-1988 and down to 2 percent currently (post 1988). For the Developmental
- rabbit study [83-3(b)], the rejection rate has fallen from 10 percent in 1985 to 7 percent in
1986-1988 and down to 5 percent currently. For the Two-Generation Reproduction - rat
study [83-4], the rejection rate has fallen from 14 percent in 1985 to 11 percent in 1986-1988
and down to 2 percent currently (post 1988).
The Mutagenicity studies [84-2(a), 84-2(b), 84-4] have also shown declines in their
rejection rates from 17 percent in 1985 to 15 percent in 1986-1988 and 12 percent currently.
The General Metabolism study's [85-1] rejection rate has fallen from 9 percent in 1985-1988
to 2 percent currently (post 1988).
While the reduction in these rejection rates is encouraging, it is important to point out
that these studies' supplementary rates remain high (in some cases, in fact, the supplementary
rates are rising), and the time elapsed in upgrading these studies has exceeded the amount of
time that would have been required to repeat the studies. Consequently, lower rejection rates
alone may not always result in more timely satisfaction of the guideline requirements.
Several studies have not shown improvement. The rejection rate for Dermal
Penetration [85-2] has risen from 17 percent in 1985 to 27 percent currently (post 1988).
The Subchronic 90-day Feeding study's [82-l(a)] rejection rate has increased from 17 percent
in 1985 to 23 percent currently (post 1988). The Acute and Subchronic Inhalation studies'
[81-3 and 82-4] rejection rate has risen from 0 percent in 1985 to 19 percent currently (post
1988).
Industry pointed out that the overall rejection rates in toxicology are low and in fact,
rejection rates for CORT studies have shown substantial declines. Industry identified four
basic factors which they believed contribute to rejection and which they felt the Agency
should address. These are: (1) lack of clarity or shifting guidelines, (2) the necessity for a
no-observed-effect level (NOEL) in certain studies, (3) reviewer inconsistency, and (4) the
Agency use of the maximum tolerated dose (MTD).
172
-------
PL RECOMMENDATIONS
As a result of this rejection rate analysis, EPA realizes the need for additional
guidance concerning various toxicology requirements. Guideline specific
recommendations/action items are as follows and can also be found under the appropriate
guideline and rejection factors throughout the document.
1) One recurring rejection factor throughout the guidelines is the lack of characterization
of test substance (i.e., purity and/or composition and/or stability). To dispel this problem,
the Industry has committed to design a cover sheet to attach to each study which will provide
information on test substance characterization.
2) The Agency has committed to resolve the discrepancy between the Toxicity Category
criteria in effect and those in 40 CFR 156.10 in FY '94.
3) In the Acute Inhalation Toxicity Guideline 81-3 Rejection Factor #3, Industry
scientists noted that there were some problems with the guidance concerning sampling
procedures and the frequency of sampling in nose-only chambers. The Agency has
committed to revise the SEP for Inhalation Toxicology for performance of the analytical
methodology in nose-only chambers in FY '94.
Also, in Guideline 81-3 Rejection Factor #6, it states that it may be necessary to
reduce the size of the aerosol particles to inhalable size for the animal tested. The Industry
scientists believe that preparation of the material needs to be consistent with the physical and
chemical properties of the product and that grinding of granular pesticides makes little sense.
The Agency has committed to address the issue for testing "fines" of granular formulations in
the Inhalation SEP in FY '94.
5) A list of historical control parameters most often requested has been compiled by the
Agency in response to comments received for Guideline 83-3(a) Rejection Factor #1 during
the February 1993 meeting (see Attachment 5).
6) In Guideline 83-3(b) Rejection Factor #5, there was some confusion regarding the
conditions in which excessive maternal toxicity could be used as a rejection factor. The
Agency intends to clarify these conditions in the Developmental SEP in FY '94.
7) Industry scientists asked the following question in the February 1993 meeting with
regards to Mutagenicity Testing Guideline 84-2: Is the presence of a precipitate sufficient
evidence to confirm that the test system has been exposed to the test substance, and therefore
obviate the need for further analytical confirmation?
The Agency believes that there is sufficient evidence only if the precipitate cannot be
redissolved; however, the test should include at least one dose level at which the test
substance does not precipitate. The solvent must give the highest attainable concentration of
173
-------
test substance in solution (without a permanent precipitate). The solvent needs to be chosen
on the basis of the solubility of the test substance. DMSO is often an appropriate solvent,
but if the substance is more soluble in water, then water would be more appropriate. It
should be noted that there may be a situation where this does not necessarily mean that the
highest attainable concentration has been reached, even though a temporary precipitate forms
when the test substance in a solvent such as DMSO is added to the medium. For example,
a hypothetical test substance is soluble in DMSO to the extent of 250 mg of test substance
per 100 ml DMSO. One ml of a saturated solution of the test substance in DMSO is added
to 99 ml medium, and a precipitate forms, which subsequently redissolves after stirring (the
resulting solution contains 2.5 mg test substance/100 ml) and the performing laboratory states
that this is at the highest attainable concentration based on solubility in DMSO. However,
the Agency also has received information that the test substance is soluble in water (and
presumably also the cell growth medium) at 9 mg/100 ml. If there is no evidence of either
cytotoxicity and/or a mutagenic response at the highest test dose of 2.5 mg/100 ml, the study
may then be rejected because there exists the possibility that the cells could have been
exposed to higher concentrations of test substance.
8) In addition, the Agency plans to conduct the following workshops:
General Metabolism - In the meeting with Industry scientists, one issue that
was raised by them focused on the fundamental purpose of the general metabolism study. It
was decided that a workshop on general metabolism was necessary and that the workshop
should focus on the purpose of the metabolism study in the registration context. The
workshop will be held in fiscal year 1993 or early fiscal year 1994.
Maximum Tolerated Dose (MTD1 Issue - The Agency reported that the
National Academy of Sciences report on the MTD issue did not bring closure on the MTD
concept and, therefore, was not as helpful as EPA had hoped. The issue needs to move
forward and the Agency is prepared to work on this issue. A workshop on the MTD issue
will be held sometime in fiscal year 1993.
Clinical Chemistry - Industry scientists raised a question regarding the
purpose of clinical chemistry testing (i.e., is there ever an effect that is picked up only by
clinical chemistry?). The Agency responded that sometimes an effect is only noticed in a
clinical chemistry test. It was agreed that a workshop on this subject is necessary.
9) Industry scientists recommended that the following points concerning toxicology test
guidelines and protocols should be considered by the Agency. First, international
harmonization (acceptance of international protocols) is required in order to avoid needless
repetition of studies and overuse of animals by registrants. Considerable flexibility in
meeting the objectives of the study should be built into the harmonized guidance. Second.
consistent testing guidance should be easily accessible to all registrants in a single document,
explicit as possible while providing sufficient latitude to permit use of sound scientific
174
-------
principles by EPA reviewers. Internal policy codifications should be quickly and widely
distributed to registrants. Third, the Industry scientists recommend that the Agency notify
registrants when it proposes to change testing guidelines. EPA reviewers often begin to use
draft guidance documents before they are publicly available. Studies that are in progress at
the time the guidelines are changed should not be rejected based on the new requirements. It
would be best if clarifications or updates were incorporated directly into the original
document and a revised document issued. If this is not possible, formal addenda should be
issued and widely disseminated to registrants. The addenda should indicate how they are
different from old guidance.
175
-------
X,, APPENDIX A - EPA GUIDANCE DOCUMENTS
EPA distributed the following guidance documents to guide registrants on the correct
procedures for conducting toxicology studies. Specific references to these materials are made
under each of the rejection factors listed.
Subdivision F: Hazard Evaluation: Humans and Domestic Animals; 85-2 Dermal
Absorption Studies of Pesticides (Draft)
Standard Evaluation Procedure (Subchronic and Chronic Exposure) EPA 540/09-85-020,
June 1985.
Standard Evaluation Procedure (Oncogenic Potential) October, 1985.
A Position Document of the U.S. Environmental Protection Agency, Office of Pesticides
Programs: Selection of a Maximum Tolerated Dose (MTD) in Oncogenicity Studies,
Sept. 1987.
Gross S.B. & Vocci Frank J., Hazard Evaluation Division Standard Evaluation
Procedure Inhalation Toxicity Testing, EPA-540/09-88-101, August 1988.
Gross Stanley B., Memorandum Subject: Comments on Standard Evaluation
Procedure. Inhalation Toxicology Testing (SEP/Inhalation), April 18, 1989.
Salem Harry, Inhalation Toxicology Research Methods, Applications, and Evaluation,
Marcel Dekker, Inc., New York, 1987.
Technical Committee of the Inhalation Specialty Section, Society of Toxicology
Recommendations for the Conduct of Acute Inhalation Limit Tests, Fundamental and
Applied Toxicology 18, 321-327, 1992.
Federal Insecticide, Fungicide, and Rodenticide Act
Pesticide Assessment Guidelines (1984), Subdivision F. Hazard Evaluation: Human and
Domestic Animals. Office of Pesticides and Toxic Substances, Washington, DC.
FIFRA Accelerated Reregistration - Phase 3 Guidance, December 24, 1989.
Federal Register, Vol. 56, No. 234, USEPA, Guidelines for Developmental Toxicity Risk
Assessment, 12/5/1991. 63798-63826
The Risk Assessment Guidelines of 1986. Office of Health and Environmental
Assessment, Washington DC EPA/600/8-87/045
176
-------
40 CFR Parts §152-164
PR Notice 86-5: Notice to Producers, Formulators, Distributors and Registrants
(7/29/86)
Hazard Evaluation Division (1985). Standard evaluation procedure: teratology studies.
Office of Pesticide Programs, Washington DC. EPA-540/9-85-018
Hazard Evaluation Division (1992 DRAFT). Standard evaluation procedure:
Developmental Toxicity Studies
Memorandum dated 7/27/88: Selection of an Upper Limit Dose in Multigeneration
Reproduction Studies, Theodore Farber, Ph.D., Toxicology Branch, Health Effects
Division.
Dearfield Kerry L., Memorandum, Revised Mutagenicity Guideline and Requirements.
August 5, 1991.
Health Effects Testing Guidelines Federal Register 50: 39442-39458 September 27,
1985.
Standard Operation Procedure #2000, Toxicology Branch/HED/OPP, Revised 9/30/85
with corrections from Federal Register 52: 19080-19081 May 20, 1987.
FDA 1982 "Red Book"
Standard Evaluation Procedure (Dermal Sensitization) September 16, 1987.
Memorandum dated 6/28/91: Clarification of Data Requirements for Non-Food Use
Pesticides, William L. Burnam, Deputy Director, Health Effects Division.
Memorandum dated 7/5/91: Suggestions for Analytical Determinations of Dosing
Solutions for Mutagenicity Testing, Kerry L. Dearfield, Ph.D, Geneticist, Health Effects
Division.
177
-------
Attachment 1
* A i UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
" 1 WASWNQTON, D.C. 20460
OFFICE OF
MEMORANDUM PESTCCES AND TOXIC
^^ SUBSTANCES
SUBJECT: Response to NACA Comments on EPA's Rejection Rate
Analysis of Toxicology Studies: 81-1 Rejection Factor
# 2 and 81-2 Rejection Factor * 4
TO: Deborah McCall, Biologist
Section III, Toxicology Branch II
Health Effects Division (H7509C)
FROM: Pamela M. Hurley, Toxicologist **
Section I, Toxicology Branch I
Health Effects Division (H7509C)
THRU: Roger L. Gardner, Section Head
Section I, Toxicology Branch I
Health Effects Division (H7509C)
The following paragraphs contain responses to two comments
submitted by the National Agricultural Chemicals Association
(NACA) on the Rejection Rate Analysis of Toxicology Studies.
Guideline 81-1t Acute Oral Toxicitv
Re-i action Factor 2: Inadequate dose levels to calculate an LD50
value.
The NACA comments state that there is a great deal of
confusion with respect to the number of animals and doses
required. Specific guidance needs to be provided with respect to
the use of a single sex, the limit dose and the other selected
dose levels. NACA further states that there is a conflict
between the Guidelines, the Phase III Guidance Document and the
Agency's Revised Policy for Acute Toxicity Testing (memorandum
from Victor Kimm, dated 9/22/88). The Guidelines contain the
strictest requirements, stating that the data should be
sufficient to produce a dose response curve, and where possible,
permit an acceptable determination of the LDjQ. The Victor Kimm
memo states that "We [the Agency] plan to accept only newly
generated industry data that conform with our revised guidance
unless an adequate rationale accompanies the submission." The
Phase III Guidance Document indicates that the doses tested
should be sufficient to determine the Toxicity Category or use a
limit dose. Finally, NACA suggests the possibility of dosing
-------
animals at the label cutoffs such as 2000 mg/kg, 500 ng/kg and 50
mg/kg body weight.
The Agency agrees with NACA that there is confusion
concerning the number of animals and the doses required in the
acute oral toxicity test. At the time that the Revised Policy
for Acute Toxicity Testing memorandum was released in 1988, the
Agency informed the staff to follow not-only the guidance given
in the memorandum but to also follow the same guidance that has
been suggested by NACA, that retesting was not to be required if
the Agency can place the material in a label category (i.e.
Toxicity Category I, II, III or IV). This encompasses the
example provided in the NACA comments, that if 1/5 animals/sex
dies at the limit dose, then additional testing is not required
if a Toxicity Category can be estimated.
Guidance on the use of a single sex was provided in the
Revised Policy memorandum mentioned above on page 6 which states:
"The Agency emphasizes that parallel assays on male and female
animals to determine an approximate estimate of acute toxicity
need not be routinely determined, since male and female animals
of the same strain generally show only slight and insignificant
differences in susceptibility to toxic agents. However, for some
chemicals, one sex may be somewhat more sensitive than the
other...Therefore, consideration should be given to limiting
studies to the more sensitive sex. Previous history on the class
of chemical being evaluated would be helpful in making this
determination. For confirmation, a few animals of the other sex
should also be tested." The Agency accepts acute studies using
only one sex when the Registrant has provided sufficient evidence
that the product was tested on the more sensitive sex. This
evidence can include a single dose (often a limit test) in which
both sexes were tested or even subchronic or chronic toxicity
testing data. Using only one sex for acute toxicity testing
without evidence as to which is the more sensitive sex, as
indicated in the OECD testing guidelines is a topic that has yet
to be resolved by the EPA Test Guidelines Harmonization
Committee. It is noted here that an analysis of the HED 1-liner
data base for acute oral and acute dermal toxicity tests
indicated that neither sex was uniformly more sensitive than the
other. Sometimes the female was more sensitive and sometimes the
male was more sensitive (see attached document).
The statement from the 1988 memorandum that "He plan to
accept only newly generated industry data that conform with our
revised guidance unless an adequate rationale accompanies the
submission" is confusing in light of the fact that the Agency is
still accepting newly generated studies that were conducted
according to the classical methods. The Agency will continue to
accept the classical studies until all Registrants are officially
made aware of the new policy. In addition, the Agency will also
-------
continue to accept data from classical studies that were
conducted prior to the new policy.
The NACA suggestion of testing animals at label cutoffs is
an interesting one and will be referred to the Agency Test
Guidelines Harmonization Group.
Guideline 81-2: Acute Dermal Toxieity
Rejection Factor 4i Improper number of animals tested per dose
group.
NACA does not feel that guidance on this issue is clear and
is asking if the policy statement provided in the 1988 memorandum
is still in effect.
As stated above, the policy statement provided in the 1988
memorandum is still in effect. The use of alternative methods
for acute toxicity testing are being accepted, particularly the
new fixed-dose method that has recently been discussed and
revised by the United States and by the OECD and the Up-and-Down
Method. It is of note that the Agency receives extemely few of
these tests and thus the toxicology reviewers are not familiar
with them and may reject them without prior instruction. The
present policy is if the Agency can adequately determine a
Toxicity Category from the data, then the study will not be
rejected on the basis of too few animals.
-------
COMPARISON OF HALE AMD FEMALE RAT ORAL AND DERMAL LDSQ VALUES
IN OPP'S ONE-LINER DATABASE
Praparad for:
Office of Paaticida Profrau
U.S. Ettvtronmantal Protactlon Agancy
1921 Jaffaraon Davia Highway
Arlinfton, VA 22202
Prtparad by:
Clamant Intaraational Corporation
9300 Laa Hljhiray
Fairfax. VA 22031-1207
Daeaabar 2, 1991
Contract Nuabar: 68D1007S
Work Asaifmant Ruabar: 1-23
Projact Offlcar: Mr. Jim Scott
-------
Contract Number: 66D1007S
Work Assignment Nuaber: 1-23
Deces&er 2, 1991
COMPARISON OF MALE AMD FEMALE RAT ORAL AND DERMAL LD.Q VALUES
IN OPF'S ONI-LINER DATABASE
Prepared by: v
/"• i '*
/ ,- • ,»
Carrie Rabe, Ph.D. Signature; '/>.-..,
Principal Author . . ,
Clement International Corp. Date; /?_. •// :i
Sharon Segal. Ph.D. Signature:.
Independent Reviewer and
QA/QC Manager Date:
Clement International Corp.
Approved by:
Robert Zendzian, Ph.D.
Science Analysis and
Coordination Branch
ReregUtration Section
-------
SUlOUftY
Hal* and faul* LD,0 values from acute oral and daraal studies in the
rat were extracted from ch« Offlea of Paatlclda Programs' (OFF) One-liner
Database and compared to determine vhethe'r one sex was uniformly more
sensitive In these types of tests. Results from 125 acute oral and 8 acute
dermal studies on technical grade material or metabolites were analyzed.
Comparison of the LDSO values found only 3 male LD^ values that were at least
K of a log greater than the corresponding female LDgg value and 1 male U>30
value that vas at least K of a log less than the corresponding female LD,Q
value. Comparison of the 95Z confidence Intervals for the LD^ values shoved
that In 14 cases no overlap of the confidence limits existed. In 11 of the 14
cases, the confidence Interval of the male LD^g value vas greater than the
confidence Interval of the female LD^ value, and In the remaining 3 cases,
the male confidence Interval vas less than that of the females. However,
comparison of the distribution of the malt and female LD,0 values revealed no
significant differences. These data do not support the selection of either
sex as a "uniformly most sensitive sex* for use in acute oral and dermal
toxlclty testing.
-------
For most chemicals, acute oral and tonal toxic Ity tests are required
•
for registration under the Federal Insecticide, Fungicide, and Rodenticide Act
(FIFRA). Only those manufacturing or end-use products chat are highly
volatile or corrosive substances that cannot be administered orally or
dermally are exempted. Acute oral-and dermal toxicity tests provide
information on the health hazards associated with short-term oral and dermal
exposure, give some information on the mechanisms underlying toxicity. and
provide information useful for the design of longer-term studies. The results
of these tests also serve as the basis for regulatory decisions such as
whether to require use restrictions or. special packaging or labeling.
Guidelines for acute oral and dermal testing have been developed by the
Office of Pesticide Programs to provide registrants with information on the
standards by which test results submitted to OPP for the purpose of
registration under FIFRA will be evaluated.
The Health Effects Division of OPP is currently reevaluating and
revising the pesticide assessment guidelines. As part of this process, public
comment has been solicited. One issue that was raised during the public
comment period was the possibility of further reducing the number of animals
required for these testa by identifying a most sensitive sex and conducting
acute oral and dermal toxicity tests only, on that sex.
In order to evaluate the potential impact of single-sex testing, LDj0
data from acute oral and dermal toxicity tests in OPP's One-liner Database
were examined. OPP's One-liner Database contains a compilation of toxicity
test results from over 30,000 studies on over 950 chemicals submitted to OPP
over the past 7-12 years to support pesticide registrations under FIFRA. As
2
-------
such, the database contain* * typical cross section of the rang* of acute oral
and dazaal toxicity test results likely to ba submitted to OFF in th* future.
METHODS
OFF'* One-liner Database vaa searched and all acute oral and denial
toxicity study test results were extracted. The search was limited to studies
on technical grade materials and metabolites. Proa this, male and female rat
oral and denial LDj0 values (with their 9SX confidence limits) from studies
with core grade evaluations of minimum or guideline were extracted (Tables 1
and 2) and analyzed for sex-based differences. Only those studies with LDM
values for both males and females ware used. In addition, only LDjg values
expressed as discrete numerical values were used. LD^, values expressed as «
or a a given number were not used. A study was not excluded if the 95X
confidence interval was not presented. Statistical analysis of the data for
differences between male and female LDjg values was performed using the
Vilcoxin Rank Sum Test.
USULTS AHD DISCUSSI01
A total of 125 paired acute oral LDyg values and 8 paired acute dermal
LDjg value* for male and female rats were extracted from the One-liner
Database. Seventy-seven of the mala and female oral LDjg values and 2 of the
male and female dermal LDg, values were accompanied by their respective 95X
confidence limits. The most direct approach for analysing for potential
differences between male and female LD^ data would have been to determine the
number of chemical* for which th* sale LD^ value for a chemical was
significantly different from th* female U^ value for that chemical.
3
-------
However, the One-liner Database did not contain this information. Therefore,
the paired mala and female LDjg values were examined for differences using a
number of criteria. The first criteria used was to determine those male LDj0
values that differed from the corresponding female LDjg values by # of a log
or greater. A total of 4 out of 133 male LDjg values differed from the
corresponding female LDjg values by this amount (Table 3). All 4 of the
values were oral LDjg values. Three of the male oral LDjg values were K of a
log greater than the corresponding female oral LDjg values and one was K of a
log less.
The next criteria used for analyzing the LDjg data was to determine the
number of male LD,0 values with 9SX confidence limits that fell outside the
range defined by the 95X confidence limits from the corresponding female LDjg
values. A total of 14 out of 79 male LDjg values had 9SX confidence limits
that met this criteria (Table 4 and Figure 1). All of these were from oral
studies; In 11 cases, the rang* defined by the 9SX confidence limits of the
male value was greater than the range defined by the 95X confidence limits for
the female LDjg value. In the remaining 3 cases, the range defined by the 9SX
confidence limits of the male LDjg values was less.
Finally, the distribution of male and female oral and dermal LDj0 values
was examined for differences. Figures 2-4 demonstrate the frequency
distribution of extracted male and female LDjg values from oral and dermal
studies and the combined oral and dermal data. Although males had slightly
more high LDjg values than females, statistical analysis of the data showed no
significant difference (pX).3796) between the distribution of male and female
values.
-------
These results demonstrate that neither MX can be Identified as the
uniformly Boat aenaieive aax for uae in acute toxlclty testing of rata. In
addition, the data examined auggeat that the aexea are not equally aenaitive
to all of the chemicals teated. Analyaia of the overlap of 95X confidence
limits for paired mala and female LDjQ valuea auggeata that in aoae caaea
•alee were more aenaitive than feaalea and in other eaaea the reverse was
true. In approximately 14Z (11/79) of the results, female rata appeared to be
more aenaitive than male rata, and in 4Z (3/79) of the results, males appeared
to be more aenaitive. This finding indicatea that the choice of a single sex
as representative of both aexea would alao be unreliable. Thua, the propoaed
uae of a a ingle aex in acute toxicity teata, either becauae one aex ia more
aenaitive or becauae both aexea are equally aenaitive, cannot be aupported by
the data currently in the One-liner Databaae.
-------
TABLE 1. RAT ORAI, Life DATA*
LOWER 95Z UPPER 95Z
LOWER 9SX UPPER 95X
flRID N0.b
241253
40504833
258740
99807
249878
4072242
71466
246070
246070
265147
247193
70894
70894
MALE CONFIDENCE CONFIDENCE
CHEMICAL NAME LD«, fclHIT LIMIT
Acaphata tach 971 1400.00
Nathylthioacatata 99.21
(structural analog)
Flucythrlnata
Acatochlor MOM 097
NON-4620 technical
•thiozln tach (90X pura)
Batch 5-25-0023D
KUG 0519 (Baytan) Tach (92.71)
Bia(tri-n-butyltin)oxida (95X)
Bla(tributyltin)oxida
(Alkyl-aourcad) (95X)
Boric acid (100Z)
Bronopol
(2-brow»«2-nltro-l,3-
propanadlol) Tach.
Buctrll
Broamxynil octanoata (Buctrll)
426.00
33.00
3712.00
8762.00
1115.00
689.00
193.00
180.00
5280.00
307.00
782.00
720.00
NO*
349
24
2794
4764
571
136
130
4630
596
596
.00
.00
.00
.00
ND
.00
.00
.00
.00
ND
.00
.00
523
47
5297
12760
831
250
230
6020
1026
1026
ND
.00
.00
.00
.00
ND
.00
.00
.00
.00
ND
.00
.00
FEMAU
tfiso-
1000
519
29
2018
6395
59
752
123
150
5830
342
793
793
B CONFIDENCE CONFIDENCE
LIMIT LIMIT
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
ND
420.00
21.00
ND
5691.00
ND
647.00
97.00
130.00
4690.00
ND
500.00
500.00
750
41
7099
874
149
160
7230
1258
1258
ND
.00
.00
ND
.00
ND
.00
.00
.00
.00
ND
.00
.00
-------
CaMJB 1. (Continued)
LOWER 95Z UPPER 9SZ
LOWER 95X UPPER 95X
HRID MO.
148500
4570701
244164
247692
41662409
73530
41206105
41206104
72932
259425
159371
261127
248166
71466
MALE CONFIDENCE CONFIDENCE
CHEMICAL NAME LDw, LIMIT LIMIT
Caroaryl (99. OX)
Mevinphoa Tech.
Chloro-M-creaol Technical
OCA- 1223 tech (93+X)
SAN 582H Tech. (91.4X a.i.)
DPX-Y6202 (99. IX)
NC-302 (Levo ainua S compound)
(97X Aaaure)
MC-302 (Dextro plus R c«pd)
97X (Asaure)
Anlllno acid (98r6X)
Cupric hydroxide (77X)
Cuprlc hydroxide (77X)
Coppar oxychlorlda (94. IX)
Coaan 145 Tech. (SOX a.l.)
KWG 0519 (Baytran) tech (92.71)
302.60
3.50
5129.00
118.68
2139.80
1670.00
1088.00
1209.56
424.00
1330.10
2500.00
1537.00
1950.00
689.00
272.00
NO
ND
99.23
1444.90
ND
ND
ND
382.00
1001.10
1714.00
1319.00
1620.00
ND
336.50
ND
ND
141.95
3168.90
ND
ND
ND
471.00
1768.00
3360.00
1791.00
2420.00
ND
FEMALE CONFIDENCE CONFIDENCE
LD*« LIMIT LIMIT
311.50
2.30
3636.00
48.21
1296.80
1480.00
870.00
1181.75
346.00
682.60
2200.00
1370.00
1620.00
752.00
280.50
1.00
ND
40.94
899.00
ND
ND
ND
310.00
332.90
1497.00
1138.00
1270.00
ND
345.90
3.60
ND
56.77
1871.50
ND
ND
ND
385.00
1399.60
3234.00
1649.00
1990.00
ND-
-------
TABLE 1. (Continued)
LOWER 9SX UPPER 95X
LOWER 951 UPPER 95K
MRID NO.
40345406
72008
41235004
41776115
99855
41563908
40607713
249937
249937
401264
263861
-?1££1
CHEMICAL MAMK
UnlcoiMXol* (97.21) I E/Z -
96.3/3.8; BS/ER - 79.2/20.8)
Cyfluthrin T«ch.
Haxacinon* t«eh (98X pure),
whit* .olid; A3674-207
me 56701 T.ch.
(CypcnMthrln S; 88.11 «.i.)
CypanMthrln T«ch.
53:47 cl«-tr«M
CGA 163935 T«ch. (96.61)
Cyproconazol* c«ch (95. 7X)
P«npropathrin (91. 81)
F«npropathrin (97.31)
DTEA (2-0«cylthio«ttuuM aain«)
(99. 8X)
DicMriia (3,6-dichloro-o-*ni»lc
•eld) ... T«ch.
nnu.AAAn/A.nirhloroAcatvL-1-
MALE CONFIDENCE CONFIDENCE
LDsft LIMIT LIMIT
2020.00
869.00
1100.00
134.40
247.00
4613.00
1020.00
70.60
164.00
3940.00
3299.80
2800.00
1740.00
ND
810.00
100.40
187.00
ND
ND
53.70
115.00
3164.00
1849.60
ND
2340.00
ND
1800.00
168.50
329.00
ND
ND
92.70
234.00
5556.00
5867.20
ND
FEMALE CONFIDENCE CONFIDENCE
LD^ LIMIT LIMIT
1790.00
1271.00
1200.00
86.00
309.00
4212.00
1330.00
66.70
107.00
2272.00
3604.00
2400.00
1490.00
ND
1000.00
45.70
150.00
ND
ND
50.60
69.80
1361.00
3021.30
ND
2150.00
ND
2000.00
126.30
500.00
ND
ND
87.90
164.00
3362.00
4299-. 00
ND-
oxa-4-«EMplro{4.S)decan«) ...
(94.97Z)
00
-------
TABLB 1. (Continual*
LOWER 9SX UPPER 951
MALE CONFIDENCE CONFIDENCE
LOWER 95X UPPER 95Z
FEMALE CONFIDENCE CONFIDENCE
LIMIT LIMIT
rt»Tf»MQ. CHEMICAL NAME
251863 Diallata EC
(S- (2 , 3-Dlchloroallyl)
Mi i •Aoratwl thiocarbABAta)
150953 Dlchlococyanurata, aodiun
aalt tach.
253099 laopropylaatar of 2, 4-D...
Tach.
41164301 SoditM aalt of 2, 4-D
128854 2,4-DB (98X)
73192 RO 15-197/000 (99X pure)
41062506 Quinclorac (BAS 514 H Tach)
Rag. * 150 732
5467 DDVP tach.
146179 Diazol Tach. (Dlazlnon)
246501 Diio4lo»athyl-para-tolyl-
aulfona
246798 Matacil 180 oil flowabla
40583901 Dlawthyl focwuiida tach (99
UhO &
1256.00
2094.00
640.00
594.30
2.33
3095.00
3060.00
80.00
775.00
15400.00
148.00
.11) 477.50
inn *j
961.00
1555.00
500.00
488.90
1.45
1990.00
ND
ND
583.00
ND
131.00
ND
i&raa> m
1642.00
2636.00
829.00
722.50
3.76
4436.00
ND
ND
967.00
ND
168.00
ND
865.00
1671.00
440.00
449.70
1.54
2864.00
2190.00
56.00
499.00
15400.00
162.00
387.50
417.00
1423.00
275.00
354.00
1.14
1519.00
.ND
ND
363.00
ND
137.00
ND
1149.00
1962.00
704.00
571.3d
2.08
4033.00
ND
ND
635.00
ND
190.00
ND-
-------
TABLE 1. (Continued)
MRID NO.
243414
256258
40280101
261098
248349
70652
71259
40042106
40042106
249155
157590
255690
72165
CHEMICAL NAME
Methyl parathion tach
(after 1 year atoraga)
NIRAN M/8 (BOX) (ABML-05001)
Axinphoa-Mthyl tach (85X)
Bldrln (dicrotophoa) tech.
(88. 3X a.i.)
Dlodlna (98. 9X)
EL- 9 19
laouron (94.41)
MALE
-J&O
14.00
10.00
9.00
11.00
1931.00
7.20
613.00
lHBia(4-fluorophenyl)Bethyl- 1110.00
ailyl)Mthyl)-lH.l,2.4-trlazole
(97X)
INN- 6573 tech (97X) Batch f
3, 5»DlbroMo-4-hydroxy-
benzonltrile (94. OX);
Inerta (6X)
Ethion tech (purity 98. 8X)
FMC 67825 (94.91) (in corn oil)
Cycloate Tech. (98. OX)
1110.00
81.20
191.00
47.50
3200.00
LOVER 95X
CONFIDENCE
LIMIT
11.02
ND
7.20
ND
ND
6.70
ND
1008.00
ND
NO
ND
40.30
2717.00
UPPER 95X
CONFIDENCE
LIMIT
17.78
ND
11.40
ND
ND
7.70
ND
1222.00
ND
ND
ND
54.70
3769.00
FEMALE
!&o
18.50
15.00
6.70
8.00
1117.00
9.30
484.00
674.00
674.00
93.30
21.00
30.10
2275.00
LOVER 95X
CONFIDENCE
LIMIT
11.21
ND
5.60
ND
ND
8.88
ND
563 i 00
ND
ND
ND
26.50
2066.00
UPPER 95X
CONFIDENCE
LIMIT
30.53
ND
7.90
ND
ND
9.72
ND
765.00
ND
ND
ND
33.80
2505.00
-------
TABLE 1. (Continued)
LOWER 95X UPPER 95X
MRID NO.
254690
261729
41379716
248473
265046
40700917
253165
263525
257431
41013703
72853
CHEMICAL NAME
ButyUte Tech. (98. OX)
Lot I CCC-0301
EPIC tech
Flucyc loxuron
(PH 70-23 llq 25)
PMC 54800 Tech. (91. 4X)
FlutrUfol Tech. (93X)
Batch P10.D2518/75
HUG 1608 (97. IX a.l.)
(Tertmconazole )
Folpet tech (91. 2X a.i.)
(cod* SX-1346)
Hexaconazole (PP523)
(92. 3X a.l.)
3-Iodo-2-propynyl butyl
carbonate (99X)
Chlorprophan Tech.
(SX-1817) (99. 7X pure)
S- (l.l-dlnethyl) -o-ethyl-ethyl
MALE CONFIDENCE CONFIDENCE
LD$4 LIMIT LIMIT
4850.00
1465.00
4061.00
70.10
1140.00
4264.00
43800.00
2189.00
1795.00
4100.00
3.90
ND
1290.00
ND
57.07
880.00
3952.30
35000.00
1076.00
1437.00
0.00
3.20
ND
1663.00
ND
83.13
1470.00
5330.20
55600.00
4083.00
2243.00
7000.00
4.60
LOWER 95X UPPER 95X
FEMALE CONFIDENCE CONFIDENCE
LIMIT LIMIT
4785.00
1065.00
2.10
ND
6071.00 2283.00
ND
ND
1712.00 1324.00 2214.00
4585.00 ND ND
53.80 48.88 58.72
1480.00 1090.00 1980.00
3352.00 2341.40 3977.50
19500.00 7500.00 51000.00
0.00
783.00 1329.00
4800.00 2900.00 7100x00
ND—
phosphorothioftte Tech. (93X)
-------
TABU 1. (Continued)
HALE
LOWER 95X UPPER 95Z
CONFIDENCE CONFIDENCE FEMALE
MR1D NO. CHEMICAL NAME
LOWER 95X UPPER 95X
CONFIDENCE CONFIDENCE
LIMIT
263461
245474
364390
264268
72962
253414
247582
248688
(521001
1521001
243412
248286
1883711
1667411
Butoxyethyl ••tar of 2 ••ethyl -
4-chlorophenoxyacetic acid
(93.31)
Vydate (97. IX)
Inert. (2.9X)
Mathyliaothiocyanata (97X)
Zectren Tech. (90. 5X *.i.)
HOE 39866 (92. IX e.l.)
MAK-1654 tech (97. 2X pure)
1-Sodlu* nepthyl acetate
(95X)
Paclobutracol (97X pure)
p-Dlchlorobenzene
p-Dichlorobenzene
Parathlon Tech. (In corn oil)
Pentachlorobenzene (99X)
Fortress (86X a.i.)
XRD-429 (Lot f AGR- 185781)
1000.00
3.10
82.00
8.51
2000.00
85.00
1350.00
1954.00
3863.00
3863.00
10.80
1125.00
4.80
3.20
MP»*B* • l^ll III J
NO
2.60
43.00
ND
1600.00
69.00
1120.00
1147.00
3561.00
3561.00
6.75
1015.00
4.40
ND
»«»••* •
ND
3.50
155.00
ND
2490.00
101.00
1640.00
4985.00
4153.00
4153.00
15.12
1247.00
5.30
ND
p.*p^aj : <
785.00
2.50
55.00
9.12
1620.00
87.00
930.00
1336.00
3790.00
3790.00
2.52
1080.00
1.80
1.10
M»r»» » !!•••
ND
2.40
12.00
ND
^190.00
69.00
630.00
837.00
3425.00
2425.00
1.33
ND
1.70
ND
ND
2.70
99.00
ND
1740.00
106.00
1380.00
1969.00
4277.00
4277.00
4.76
•ND
2.00
ND
(98.81 purity)
-------
MRID MO.
73280
248855
40974507
72896
259842
259805
264268
73203
256581
252599
246326
261401
251666
LOWER 95X
HALE CONFIDENCE
CHEMICAL NAME LD., LIMIT
Pyrldata T«ch. (90.31 a.l.)
Sulfaqulnoxallna Tach. (99.51)
RE-45601 tach (SX-1688)
(83. 3X)
RH-53.866 Tach. (Lot f 83159-5)
(91.91 pur.)
Cokllaht tach (93. 6X)
Karata (92.61 & 96 X)
Zactran tach (96. 5X a.i.)
Cyttalothrin • 94X pyrathoid.
97X cla-iaoaar
Trophy tach
Captafol Tach. (98.31)
Captafol (80X)
PP93 tach
Dazoaat (99X)
5993.00
1370.00
1630.00
1600.00
318.00
79.00
9.77
243.00
2479.00
6780.00
5600.00
21.80
596.00
3164.00
940.00
ND
MD
219.00
ND
ND
183.00
ND
ND
4000.00
ND
ND
•»
UPPER 95X
CONFIDENCE
LIMIT
33610.00
1860.00
ND
ND
463.00
ND
ND
312.00
ND
ND
7700.00
ND
ND
LOWER 95X UPPER 95X
FEMALE CONFIDENCE CONFIDENCE
LJ^o LIMIT LIMIT
3544.00
1600.00
1360.00
2290.00
419.00
56.00
12.00
144.00
2283.00
6330.00
3800.00
34.60
415.00
871.00
1140.00
ND
ND
281.00
*
40.00
ND
100 ..00
ND
ND
2400.00
ND
ND
8848.00
2100.00
ND
ND
624.00
78.00
ND
320.00
ND
ND
6100.00
•ND
ND
u»
-------
t-4i
MRID NO.
246892
247279
244531
41127501
163854
150959
242367
73463
249422
71364
252512
CHEMICAL NAME
0,0,0,0-tetrapropyldithio-
pyrophoaphate (90X)
Inorta (10X)
Thlabenduole (98. 5X)
[2- (4-thlazolyl)ben*lMidazola]
2- (4- thlaEolyDbaslBidazola
(98. SX) (43410-T)
A0159 tech ineectlcide (98. OX)
(2H-1 , 3-thiacine- tetrahydro-2-
nitroMthylene)
Thlraa) tech (99.41)
Trichlorocyanurate Tech.
Trichlopyr tech
(Dow233) intubation in
acetone/corn oil (1:9)
TrlfluBlzole tech
Landrln tech (in corn oil)
Triphenyltin hydroxide tech
Triphenyltin hydroxide (96X)
atava
MALE
-L&M
2800
5070
3970
285
3700
787
729
1057
125
165
165
LOWER 95X UPPER 95X
CONFIDENCE CONFIDENCE
LIMIT LIMIT
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
2314.00
3982.00
2920.00
ND
ND
585.00
515.00
863.00
ND
113.00
ND
3388.00
6389.00
5400.00
ND
ND
1059.00
1127.00
1297.00
ND
230.00
ND
LOWER 95X UPPER 95X
FEMALE CONFIDENCE CONFIDENCE
LDjQ LIMIT LIMIT
740
4734
3540
314
1800
868
630
1780
134
156
156
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
623.00
3371.00
2140.00
192.00
ND
622.00
450.00
1369.00
ND
115.00
ND
879.00
6541.00
5850.00
398.00
ND
1114.00
629.00
2314.00
ND
208.00
ND-
-------
TABLX 1. (Continued)
LOWER 95X UPPER 95X
MALE CONFIDENCE CONFIDENCE
LOVER 95X UPPER 95X
FEMALE CONFIDENCE CONFIDENCE
MRID NO.
71811
71811
71811
71811
71811
71811
71811
71811
248139
251418
72330
CHEMICAL NAME LDt*
Larvin tech
Larvin tech
cellulose)
Larvin tech
cellulose)
Larvin tech
cellulose)
Larvin tech
cellulose)
Larvln tech
cellulose)
Larvln tech
cellulose)
Larvln tech
cellulose)
U56215 Tech.
(In
(in
(in
(in
(in
(In
(In
(In
corn oil)
•ethyl
•ethyl
•ethyl
•ethyl
•ethyl
•ethyl
•ethyl
Vltuin D3 tech
SY-83 (L(+)Lactlc acid)
84
82
96
51
74
46
129
68
9098
352
LIMIT
.10
.70
.10
.60
.80
.50
.00
.90
.00
.00
61
65
59
46
59
33
89
56
263
4936
LIMIT
.50
.70
.90
.30
.90
.40
.60
.60
ND
.00
ND
115.00
104.00
154.00
57.50
106.00
64.70
186.00
83.80
ND
484.00
ND
LD«4 LIMIT LIMIT
50.00
50.80
57.40
36.70
72.00
50.90
59.10
39.10
7652.00
619.00
3543
34.. 90
39.30
39.80
28.60
49.20
46.10
40.70
29.40
ND
495.00
ND
71.70
65.70
82.80
47.20
102.00
56.20
86.00
52.10
ND
782*. 00
ND
-------
TABLB 1. (Cootlnuad)
LOWER 95X UPPER 95X LOVER 95X UPPER 9SZ
MALE CONFIDENCE CONFIDENCE FEMALE CONFIDENCE CONFIDENCE
MRID NO. CHEMICAL NAME _ LD^ _ LIMIT LIMIT LJ£, _ LIMIT LIMJT
248258 Haloxyfop methyl (99. OX) 393 339 465 599 453 874
248473 IMC 57020 Tach. (88. 8X a.i.) 2077 1976 2358 1369 1127 1611
(DlMthaxona)
•Data praaantad In
*MRID Mo., Maatar Racord Identification Nuabar. A unlqua Identifying nuabar aaalgnad to aach docuaent
aubalttad to tha Offica of Paaticida Program. Tha nuabara 11*ted Identify tha report of tha Acute
Toxlclty Study froa which tha coafxmnd-related data were extracted.
*Mo Data
i
_*
i
-------
TABLE 2. RAT DERMAL LDjc DATA*
LOWER 95X UPPER 95Z LOWER 95X UPPER 95Z
MALE CONFIDENCE CONFIDENCE FEMALE CONFIDENCE CONFIDENCE
MRID NO. *»
261971
40504836
261971
40364203
5467
261098
259805
261401
CHEMICAL NAME
Methylthioacetata (SX-1500)
(991 pure)
Nethylthloacetata (99.21)
(contaminant)
Methyl thioacetata (SX 1500)
(99X pure) (contaminant)
Benazolin tech (97. 6X)
Batch CR16/343/3
DDVP Tech.
Bidrin (dicrotophos) tech
(88.31 a.i.)
Karate (92.61)
PP993 Tech.
LDfl, LIMIT LIMIT
1590.00
1920.00
1590.00
2100.00
107.00
876.00
632.00
316.00
ND6
1550.00
ND
ND
ND
ND
300.00
ND
ND
2390.00
ND
ND
ND
ND
900.00
ND
LDcA LIMIT LIMIT
1580.00
1410.00
1580.00
2100.00
75.00
487.00
696.00
177.00
ND
1140.00
ND
ND
ND
ND
309.00
ND
ND
1760.00
ND
ND
i
«-4
ND
ND
1169.00
ND
•Data presented in
•MRID No., Master Record Identification Nueber. A unique identifying nueber assigned to each document
submitted to the Office of Pesticide Programs. The nusbers listed identify the report of the Acute
Toxicity Study fro* which the compound-related data were extracted.
cNo Data
-------
TAIL! a. nmrrcatiB mem MAU AMD PBULE u^ VALOBS DZVTBKDM vr ORiatn nun 1/2 uxr
LOWER 95X UPPER 95Z LOWER 9SZ UPPER 95X
MRID H0>
40042106
157590
243412
246892
MALE CONFIDENCE CONFIDENCE FEMALE CONFIDENCE CONFIDENCE
CHEMICAL NAME LD^ LIMIT LIMIT LD^ LIMIT LIMIT
l|(Bi«(4-fluorphenyl)»ethyl- 1110.00 1008.00 1222.00 674.00
•ilyl)pethyl)-lH,l,2.4-trUzole
<97X)
Bthion t«ch (purity 98.81) 191.00 HP ND 21.00
Parathlon T«ch (in corn oil) 10.80 6.75 15.12 2.52
o.o,o,o-t«trapropyldlthlopyro 2800.00 2314.00 3388.00 740.00
phosphate (90S); Imrts (10X)
563.00 765.00
ND ND
1.33 4.76
623.00 879.00
•Data pr«««nt«d in
INRID Mo., Naatar Rocord Identification Nuabar. A unique identifying nuabar assigned to each document
aubeiittad to the Office of Pesticide Prograaa. The nuejbera liated identify the report of the Acute
Toxicity Study frosi which the coopound-related data were extracted.
•Mo Data
00
I
-------
.TABLB 4. CHWXCALS WITHOUT OVnLAPPIMQ MALI AMD fBULS LOj* 95% COMPXDKNCB LIMITS*
LOWER 95X UPPER 951
LOWER 95* UPPER 95Z
MRID N0.b
247692
70652
40042106
255690
72165
248473
257431
243412
40883711
246892
73463
71181
MALE CONFIDENCE CONFIDENCE
CHEMICAL NAME LDc. LIMIT LIMIT
CCA- 123 tech (93+X)
EL- 919
1( [Dia(4-fluorophenyl)Mthyl-
ailyl|Bethyl]-lH,l,2,4-triazole
(97X)
FHC 67825 94.91 (in corn oil)
Cycloete Tech (98X)
FMC 57020 Tech. (88.81 a.i.)
(DiaethaEone)
3*Iodo-2*propynyl butyl
carbaaata (99X)
Parathion Tech (in corn oil)
Fortreaa (861 a.i.)
0,0,0.0- tetrapropy Idi thlopyro
phosphate (901); Inerta (10X)
Tifluaizola tech
Larvln Tech. (in aethyl
118.68
7.20
1110.00
47.50
3200.00
2077.00
1795.00
10.80
4.80
2800.00
1057.00
129.00
99
6
1008
40
2717
1976
1437
6
4
2314
863
89
.23
.70
.00
.30
.00
.00
.00
.75
.40
.00
.00
.60
141
7
1222
54
3769
2358
2243
IS
5
3388
1297
186
.95
.70
.00
.70
.00
.00
.00
.12
.30
.00
.00
.00
FEMALE CONFIDENCE CONFIDENCE
LDjQ LIMIT LIMIT
48
9
674
30
2275
1369
1065
2
1
740
1780
59
.21
.30
.00
.10
.00
.00
.00
.52
.80
.00
.00
.10
40.
8.
563.
26.
2066.
il27.
783.
1.
1.
623.
1369.
40.
94
88
00
50
00
00
00
33
70
00
00
70
56.77
9.72
765.00
33.80
2505.00
1611.00
1329.00
4.76
2.00
879.00
2314-.00
86.00
«o
I
cellulose)
-------
TABLB 4. (Continued)
MR1D MO. CHEMICAL HAMB
LOWER 95X UPPER 9SX LOVER 95X UPPER 9SX
HALE CONFIDENCE CONFIDENCE FEMALE CONFIDENCE CONFIDENCE
LIMIT LIMIT Uto LIMIT LIMIT
71181 Larvin Tach. (In Mthyl
calluloaa)
251418 Vitamin D3 Technical
68.90
352.00
56.60
263.00
83.80
39.10
484.00 619.00
29.40
495.00
52.10
782.00
•Data praaantad in a)g/kg.
No.. Naatar Racord Idantiflcation Hiaibar. A unlqua idantlfylng nuafcer aaaignad to aach dociwent
aubailttad to tha Offica of Paaticlda Prograaa. Hia nual»ara liatad Identify the report of the Acute
Toxic Ity Study froa which the compound-related data were extracted.
-------
-21-
Flgur* 1
Comparison of Overlap of 95%
Confidence Limits of Oral and Dermal
LDso Values
70
60 -
SO -
40 -
30 -
20 -
10 -
NoOvtrtap
M>F
MOvwiaptF
NoOvwtap
F>M
Ovtrtap of 95% Conftfenca Unite
-------
-22-
Figuz« 2
LD50 Frequencies. Oral Dosing
40 -
•A _
Mi m
20 -
10 -
1771 ueiM
(T^ -FcmalM
^ r.
£E3 . » . V
~ ^ ^ 4
I
I
1
I
1
1
1
7
y
y
y
X
v
x^
y
V
y
y
x
y
y
X
s
kxxxxxxxxxv
P«
\
N
\
X
X
X
X
\
\
\
V
X
V
V
V
V
V
V
V
V
V
V
v
fe^
<1
<100
LDSOWtM
-------
-23-
Flgur* 3
LD50 Frequencies, Dermal Dosing
5 -
4 -
3 -
2 -
1 -
1771 MOIM
I
\
\
1
/
y
y
y
y
/
V
/
/
KXXXXXXXXX
. s
X
\
\
\
\
\
\
\
\
x
NS
V
V
V
V
V
V
ftl
<100
-------
-24-
Figur* 4
LD50 Frequencies, Combined Dosing Data
60
SO -
40 -
30 -
20 -
10 -
1771
^
/^
I
I
m
XN
®
<1
-------
Attachment 2
AVERAGE BODY WEIGHT (g) OF THREE DIFFERENT STRAINS OF RATS
It is suggested that test animals should be in the 6 - 10 week range when received
from the supplier and no older than 8-12 weeks of age at the initiation of dosing for acute
studies.
Age in
weeks
4
6
8
10
12
Sprague Dawley*
M
68
163
255
313
362
F
68
144
191
215
240
Wistar*
M
78
168
262
334
378
F
69
146
190
225
251
Fischer 344C
M
61
123
188
223
230
F
57
98
136
156
178
**b>c For each strain, mean body weights from three different supply laboratories (Charles
River, Harlan, and Hilltop) were utilized.
-------
Attachment 3, Section 1
DRAFT
MEMORANDUM
SUBJECT: Interim Policy for Particle Size and Limit Concentration Issues in Inhalation
Toririty Studies.
FROM: John E. Whalan, D JLB.T. Toxicologist
Section 1, Toxicology Branch I
Health Effects Division (H7509Q
and
John C Redden, M&, Toxicologist
Section 3, Toxicology Branch I
Health Effects Division (H7509C)
TO: Penelope A. Fenner-Crisp, Director
Health Effects Division (H7509C)
THRU: Karl P. Baetcke, Chief
Toxicology Branch I
Health Effects Division (H7509Q
Backptmmk
Particle size and limit concentration requirements have been a major point of
misunderstanding and contention within the Agency, and with registrants and laboratories
who must meet our requirements. A recent examination of the OFF. Guideline Rejection
Factors revealed that of 617 acute inhalation toxicity studies reviewed between 1985 and
1991,131 (21%) were rejected. Of those rejected, 40% oiled to meet the criteria for having
25% of the particles <1 jun. There was an unknown number of other studies that also failed
to meet this criteria, but were nevertheless accepted by the reviewers. In addition, 21% of
the. rejected studies could not define an LC^ and/or failed to achieve a limit concentration.
In 1991, HED requested public comments on the Inhalation Guidelines. These critiques,
along with the results of recent interviews with several inhalation lexicologists (attached),
indicate these issues are universal concerns. Of the 9 responses, 9 considered the 25%
criteria for particle sizes to be unrealistic, and 6 considered the 5 mg/1 limit concentration
to be excessive (3 had no comment). Their responses are siimmarired below:
-------
RocomiMfMtod Partfcto Sfa»
1 iua
MMAD
2-3 nm
MMAD
3 ^m
MMAD
1-4 pm
MMAD
Acute Studies
h
h
e
*W*!U
Sobchrnnfe start
e/f
•W
Relax
Requirement
Define by
Specks
e
tt
e
d
d
S^fc^ J.J-.—.—. ~.~.^^*M • !••&•£
mcofnnwncMa umn
2a|/l
5mi/l
5 flM/1 tt
b,d
a - National Agricultural Chemical Association (NACA)
b - Harry Salem, Aberdeen Proving Grounds, US. Army
c • Jellinek, Schwartz, Connolly & Freshman, Inc.
d « wa Research Laboratories, Inc.
e - Id Agricultural Products
f. OBA-GEIGY
g • LOly Research Laboratories
h « Hsu-Cbi Yen, Lovelace Inhalation Toxicology Research Institute
i * Kenneth Nitschke and Richard Coriey, The Dow Chemical Company
OBA-Geigy recommended that an MMAD of up to 10 Mm is acceptable in
those situation* where the physical characteristics of the test material prevent
reducing particle size any further.
t CIBA-Oeigy suggested that having >90% partide mass less than 5 Mm and
50% less than 3 Mm is more appropriate.
-------
There are three documents available to registrants, laboratories, and HED reviewers which
describe the conduct and interpretation of inhalation toxitity studies:
1. Subdivision F Guidelines (1984)
2, Hazard Evaluation Division Standard Evaluation Procedure: Inhalation Taricity
Testing, (EPA-540/09-88-101; August 1988) written by Stanley B. Gross and Frank
J.Vocci.
3. A memorandum from Stanley B. Gross (April 18, 1989), which added, "...some
historical clarifications concerning particle testing sizes and the limit testing which
have apparently caused some confusion with testing requirements."
These three documents convey the following guidance regarding aerodynamic particle sizes
and limit concentration testing:
Atrodynamte Partteto Sizes
Subdivision F Guidelines (1984):
The Guidelines do not offer any direction on particle sizes.
HED Standard Evaluation Procedure;
"It would seem appropriate that at least 25% of the partide distribution used
in these studies should be in the submicron range for acute and repeat
exposure studies."
"When studies are carried out using large partide distributions (median
diameters greater than 3.0 Mm), judgment is necessary in determining whether
the study should be repeated using a smaller partide size range. If the
chemical proved quite toxic (Toricfty Category 1,40 CFR 162.10), no further
acute testing is necessary as the chemical wfll already require the strictest
labeling. If the test results show minimal toxidty by the inhalation route while
showing figpjfif**1* tootidty via other routes, then the acute inhalation testing
should be repeated using smaller partide sizes." [Page 15, paragraphs 3 & 4]
Memorandum from Stanley B. Gross:
."If the mass median aerodynamic diameter reported in a study is larger than
1 jim, we can accept the study if at least 25% of the particles are 1 Mm or
less. If the laboratory is having difficulty in achieving the required diameters,
the study needs to indicate what they did and why they were unable to
provide the small particles." [Page 2, item B]
-------
Submicron particles had been required in the past because HED defined a respirable
particle as having an aerodynamic diameter < 1 MEL, both in humans and laboratory animate
Larger particles are more likely to be captured in the upper airways. Since most acute
inhalation toririty studies of pesticides have MMAD's >1 Mm, the 25% criteria has been
applied to nearly every study.
Limit Concentration Testing
Subdivision F Guidelines (1984):
"If a test at an exposure of 5 mg/1 (actual concentration of respirable
substances) for 4 noun or, where this is not possible due to physical or
chemical properties of the test substance, the mMi'mnm attainable
concentration, produces no compound-related mortality, then a full study
using three dose levels might not be necessary." [Page 51, item (3Xg)]
HED Standard Evaluation Procedure:
None
Memorandum Iron Stanley B. Gross:
"The limit test usually applies to the acute 4 hour inhalation test This limit
is set at the Toricity Category IV in which the material would be considered
to have minimal advene effects during an acute exposure.*
"In order to favor a reduced use of animals during tenacity testing, the Agency
has suggested ibe use of limit test (when such a test seems appropriate). If
deaths are seen during the limit test, t mil LCSO test as described in the
Guidelines is still required. However, a number of registrants have used the
limit test as the only test, as a 'yes/no test' and usually at levels below the
5 mg/L concentrations. This does not fulfill the testing requirements for this
guideline. *
"Further, the limit test can be carried out at the maximum attainable
concentration, A number of registrants have reported test results from a Umit
test at concentrations below 5 mg/L which did not cause any deaths. The
concentration was reported as a mf*1'"1^*" attainable concentration without
any documentation to supportthlt conclusion. This has not been accepted.
In order to declare the concentrations as the «M«i'fM™it attainable, the
registrant needs to indicate what efforts were made to reach the 5 mg/L
concentrations, what problems were encountered and, if possible, try to
explain why higher concentrations were not achievable." [Page 2, item C]
-------
This guidance acknowledges that submicron particles and limit concentrations cannot always
be achieved. Although some contingencies are provided, experience has shown that a hard
line has generally been applied by HED in judging the adequacy of studies, perhaps because
the guidelines were not fully understood. This has complicated the performance and
interpretation of inhalation toxicity studies for the following reasons:
• Studies have been rejected if the limit concentration were not attained, or if only
10% of the particles were 5 mg/L From a regulatory
standpoint, and in the interest of avoiding wasting of life, a concise LC^ is not
needed provided t Toxicity Category can be assigned.
• Many studies have been rejected because no mortality was seen at the maximum
attainable concentration, simply because that concentration was less than S mg/l.
• The limit concentration for t formulation his never been defined If, for example,
a formulation contains 10% active ingredient dissolved in 90% xylene vehicle, the 5
-------
nag/1 limit concentration could conceivably be for the active ingredient alone, or for
the active and inert ingredients combined. The fallacy in measuring only the active
ingredient is that the test would mimic a study of the technical, except that the test
animals would additionally be exposed to a huge quantity of the vehicle. This could
make an inherently less toxic formulation appear more toxic than the technical
• The limit concentration bears no resemblance to human exposure. A concentration
of 5 mg/l results in a cloud so dense that in-chamber observations may be
impossible. During whole-body exposure, the animal's fur will be coated with dust
or soaked with liquid. Airways may be physically clogged Death by suffocation may
be misconstrued as toritity. In the unlikely event a human were accidentally exposed
to such a high concentration, it would probably be for a matter of seconds • short
enough to hold one's breath. The animal exposure persists about 1000-times that
long. A rat cannot bold its breath for 4 hours.
On September 6, 1991, a contingent representing the Society of Toxicology (SOT),
Inhalation Toxicology Specialty Section, and the National Agricultural Chemical Association
(NACA) Toxicology Roundtable met with HED representatives to present a consensus
position paper entitled, SOT Inhalation Specialty Section Position Paper • Recommendations
Jbr the Conduct of Aaae Inhalation Limit Tests. This document was written by the Technical
Committee of the Inhalation Specialty Section (GX. Kennedy, JJJ. Morris, M.V. Roloff, R
Salem, CE. Ulrich, R. Valentine, and RJL Wolff), and approved by the Executive
Committee. This paper was recently published as a Commentary. The mil text of the
commentary is attached. The abstract is reproduced below:
Recommendations for the Conduct of Acute Inhalation Limit Tests, Prepared by
the Technical Committee of the Inhalation Specialty Section, Society of
Toxicology. Fundamental and Applied Toxicology. Volume 18. Pages 321-
327. 1991
"This paper reviews the scientific issues related to exposure concentration and
particle sizes used in acute inhalation limit tests. The current United States
Environmental Protection Agency (USEPA) recommended exposure concen-
tration for such tests is 5 rag/liter, while this level is very high, it is often
achievable. On the other hand, its toxkological relevance is questionable.
The USEPA recommendation that 25% of the particle distribution be less
than 1 Mm is a more difficult issue to address. Physical laws for aerosol
particle generation and behavior limit the minimum size of particles in an
exposure atmosphere at a concentration of 5 mg/Uter. Particle size also
influences deposition site in the respiratory tract Since damage to any region
of the respiratory tract can produce lethality, and it is not possible to predict,
a priori, the most responsive region of the tract or the most harmful particle
size of an untested agent, acute limit testing should employ particles in a size
range that deposits throughout the entire rodent respiratory tract Particles
-------
a priori, the most responsive region of the tract or the most harmful panicle
size of an untested agent, acute limit testing should employ particles in a size
range that deposits throughout the entire rodent respiratory tract. Particles
.between 1 and 4 Mm mass median aerodynamic diameter (MMAD) are well
suited for such studies. It is, therefore, recommended that the limit test
concentration should be the highest concentration (up to 5 rug/liter) that can
be achieved while still maintaining a panicle size distribution having an
MMAD between 1 and 4 Mm.
Based on findings by Mauderly er of, 1987; Raabe er a/., 1988; and USEPA, 1982, SOT has
recommended accepting acute studies with MMAD's of 1-4 Mm for the following reasons
(page 322, last paragraph):
"-.inhaled particles between 1 and 4 Mm MMAD will deposit within all
regions of the rat respiratory tract Within this size range, nasopharyngeal
and tracheobronchial deposition increase as panicle size increases, but
pulmonary deposition remains relatively constant Based merer/ on
pulmonary region deposition efficiencies, 1 Mm panicles offer no distinct
advantage over 4 Mm particles. Thus, because 1-4 Mm parades will likely
deposit in all regions of the respiratory tract this size range is highly desirable
for acute limit testing."
As panicles are inhaled, a portion is retained in the airways, and the balance is exhaled.
A particle' s aerodynamic diameter determines where it will be deposited in the respiratory
tract Pesticide aerosols are polydisperse panicles (geometric standard deviation 6, >L2)
which can be deposited to varying degrees throughout the respiratory tract The deposition
site for each panicle is determined by its aerodynamic diameter.
Studies of radiolabelled monodisperse panicles have been conducted to correlate deposition
sites to aerodynamic particle sizes. Raabe «t aL (1977, attached) performed such a study
using '"Ytterbium-labelled monodisperse spherical aluminosilicate particles with
aerodynamic diameters of 0.2.0.52,1.04,109, and 3.0S urn. Anesthetized Long Evans rats
and Syrian hamsters were placed in pletbysmographs to monitor their respiration during
dosing.
Figure 1 depicts respiratory deposition in the rat (generated from Raabe et aL, Table 3).
Tracheal and laryngeal deposition, which are very low, have been combined for simplicity.
Bronchial deposition is also very low. The primary lines of interest are for pulmonary,
nasopharyngeal, and total deposition.
This graph shows that the majority of inhaled panicles having aerodynamic diameters <2
Mm are removed with the exhaled air (only about 15-30% of the particles are deposited in
-------
Mean Deposition in Rats of Inhaled
169-Yb-Labeled Aluminosilicate Aerosols
III
I
III
oo
0.2
0.52 1.04 2.09
AERODYNAMIC DIAMETER (MICRONS)
3.05
PULMONARY
NASOPHAHYNGEAL
BRONCHIAL
TOTAL
TBACHEAUIARYNGEAL
-------
0.2 04 04
PHYSICAL DIAMETER. i«n
M M0.7
I-
1* 2,0 ^b~
AERODYNAMIC DIAMETER IOMI.,
f
10
4.0 SJ)
DefMwiionor inhaled monodispene aerowb of fused aluniinosilicMe spheres in smai rodents showing the deposition in the
the irachedbronchial region
-------
the respiratory tract). Total deposition decreased from 32% at 0.2 Mm, to 14% at 0.52 and
1.04 Mm, then increased to 28% at 109 Mm, and 48% at 3.05%. The major deposition site
for the larger panicles is in the nasopharyngeal region. Pulmonary deposition decreased
from 21% at 0.2 Mm to a plateau of 5-10% for particles 20.52 urn. The line plots on this
graph bear a striking resemblance to those on the graph depicted in the SOT paper (page
323). Figure 2 is a copy of this graph. These data demonstrate that we should expect to
have minimal deposition and toxiciry with submicron; particles unless, as is rarely the case
for pesticides, the panicles are <0.2 Mm.
The lungs are protected from large panicles because nasopharyngeal deposition sharply
increases with panicle size. Rodents are far more efficient than humans at removing small
particles in the upper airways. Nearly 100% of panicles having aerodynamic diameters >5
Mm ace deposited in the nasaopharyngeal region in rats (compared to >10 Mm in humans).
Unless these large panicles are absorbed in the upper airway, they can be eliminated by
sneezing, mucociliary transport, and swallowing.
As particle size increases beyond 1 Mm. the number of particles reaching the pulmonary
region decreases, but these larger panicles have a greater toxic potential. For example, a
3 Mm parade has 27-times the mass of a 1 Mm parade, and thus 27-times the toxic potential
As a result, the overall percentage of mass deposition in the pulmonary region is nearly the
same whether the particles are 0.52,1.04, 2.09, or 3.05 MOL
In order to maximize the percentage of parades reaching the alveoli, HED has always
requested submicron particles. Parades with aerodynamic diameters <1 Mm are able to
reach the alveoli in humans and rodents because they are small enough to avoid inertia!
impaction in the turbulent air of the upper airways. Unless these particles are readily
absorbed through the alveolar walls, they are slowly removed by macrophages, and alveolar
lesions may ensue.
According to Dr. Hsu-Chi Yeh of Lovelace Inhalation Toxicology Research Institute
(personal communication), the pulmonary deposition curve is biphasic in rats, with a major
peak at 0.05 MRU and a minor peak around 2J Mm (see Figure 2). Minimal deep lung
deposition occurs between 0.3 and 0.7 MOL Only when the parade size is <02 Mm does
pulmonary deposition begin to increase. Cascade impactors can only measure panicles > 1
Mm. so smaller partides are collected on an absolute filter without any means of measuring
aerodynamic particle size. This has never been a problem for HED since submicron
panicles are usually a small fraction of the total. Thus, HED has been requesting a panicle
size that minimizes pulmonary deposition. In order to maximize deep lung deposition for
pesticides, Dr. Yeh has recommended using an MMAD of 2-3 Mm.
The SOT recommendation to accept panides with MMAD's of 1-4 Mm means that studies
will be accepted having particles ranging from submicron size to approximately 7 Mm. As
the MMAD increases, so does the percentage of nasal and total respiratory tract deposition.
The SOT proposal would provide the following benefits and disadvantages:
10
-------
"For inhalation toxicity evaluations with rodents, the test aerosols should
typically have a MMAD of 3 Mm or less with a geometric standard deviation
of no greater than three to maximize alveolar deposition. Aerosols with even
a smaller MMAD of 1 to 2 Mm and a tighter size distribution, Len a geometric
standard deviation of less than two, would be preferable. However, this is
sometimes not possible due to the nature of the material required in large
quantity for chronic toxicologic study. For hygroscopic aerosols a smaller size,
approximately 1 Mm MMAD, is preferred recognizing that the particle will
enlarge in the humid environment of the respiratory tract."
"Some materials for which toxicity or cartinogenicity testing evaluations are
desired may normally occur as panicles with sizes greater than a few
micrometers and perhaps as large as several hundred micrometers. Such
powders may be physically altered to produce aerosols with MMAD's of a
few micrometers to facilitate the experimental study of their toxicity."
The SOT recommendation to accept particles with MMAD' s of 1-4 Mm means that studies
will be accepted having particles ranging from submicron size to approximately 7 Mm. As
the MMAD increases, so does the percentage of nasal and total respiratory tract deposition.
The SOT proposal would provide the following benefits and disadvantages:
B«n«ffts
It will be easier to perform an inhalation study for troublesome chemicals because
heroic efforts to achieve submicron particles will no longer be necessary.
The lower concentrations used in repeated exposure studies should make it easier to
generate finer particles than in an acute study.
Very few studies will be rejected for failure to achieve the desired particle size.
It will be easier to achieve a limit concentration.
If larger particles are used, more of the inhaled chemical will be retained throughout
the respiratory tract, especially in the nasophaiyngeal region. This should increase
toxicity and thus yield a lower LC,, than for submicron particles. This will be more
protective from a regulatory standpoint
It will be more difficult for a registrant to justify a waiver because "real-world"
exposure involves larger particles.
The use of larger particles will more dosery resemble human exposure.
11
-------
Disadvantages
If the requirements for particle sizes are relaxed, there will be a tendency for
laboratories to generate larger parades (around MMAD of 4). The majority of
particle deposition will then be in the nasopharyngeal region rather than in the
pulmonary (lung) region.
An unknown portion of the particles retained in the upper airway will be eliminated
by sneezing. The balance will contribute to systemic toaririty via absorption through
the mucous membranes, or by swallowing of contaminated mucus (Le. oral exposure).
Pulmonary absorption may be only a minor contributor to the overall toxicity seen
in a study.
It is reasonable to expect particle sizes to be smaller hi long-term studies because the
physical constraints are not as demanding as in acute studies performed at high
concentrations. The SOT stated that, "Chronic respiratory tract toxicity often results from
the accumulation of insoluble particles within the pulmonary region. The use of particle
sizes to man'mme deposition in this region may be desirable for assessing chronic effects but,
...may not be ideal for acute testing because the use of small particle sizes to maximize
pulmonary region deposition minimizes nasal deposition, enhancing the possibility of Ruling
to detect potential nasal toxicity.11
Hie SOT, commenting on particle sizes for acute and long-term studies, explained that,
"Because acute limit tests are designed to provide only an approximate index of toxicity, and
because nasal effects can be of considerable ""por^w*. upper particle ffet cutoffs need not
be so stringent as recommended for chronic inhalation toxicity studies." The SOT did not
recommend a panicle size range for long-term studies. HED concurs with SOT' s rationale,
and proposed that the MMAD range be 1-3 tun for a long-term study in order to maximize
deep lung deposition, and n»«i™v» nasopharyngeal deposition. In later discussions, the
SOT concurred with the 1-3 Mm range, provided it can be adjusted based on future
evaluation.
HED recommends the following interim guidelines be used in conducting and evaluating
inhalation toxicity ftudify
1. Aerodynamic particle sizes are acceptable if MMAD's are 1-4 Mm in aa acute study,
and 1-3 Mm in a long-term study (the range for subchronics may be adjusted in the
future based on additional information). • It is expected ttiat long-term studies are
worm the commitment to attain a panicle j*rf that wfll «**•»*«•••»£ deep lung
deposition. Hie study report should include panicle size distribution data, MMAD
values, and a description of the generation methods and equipment. If the MMAD
12
-------
guidelines cannot be met, the study report should explain why. A reasonable effort
to meet these guidelines is expected, but heroic measures are not required.
2. The analytical limit concentration for aerosols, gases, and vapors in an acute study
is 2 mg/1 (based on the recommendations of the SOT Inhalation Specialty Section,
and several other inhalation lexicologists, see table on page 2). This concentration
is generally achievable for aerosols. Although gases and vapors can often be
generated at much higher concentrations than aerosols, they are inherently more
bioavailable, and potentially more toxic, than aerosols. In most cases, a
concentration of 2 mg/1 will be achievable, but if not, the maximum attainable
concentration should be used, and the study report should provide reasons why a
higher concentration could not be attained. A reasonable effort to achieve a limit
concentration is expected, but heroic measures are not required. The Toricity
Category Criteria will be changed as follows:
Current Categories:
Hazard
Indicators
Inhalation LCg,
(analytical
concentration •
4-hour exposure)
Catecory I
**m»mgv*j m
£0.05 mg/1
Category II
>0.05 - 0.5 mg/1
Caietory III
wvw^wij mmm
>O5-5mg/l
Category IV
>5mg/l
Revised Categories:
Hazard
Indicators
Inhalation LQo
(analytical
concentration;
4-hour exposure
to aerosols,
gases, & vapors)
Category I
SO.Q5 mg/1
Category II
>0.05 - 0.5 mg/1
Category HI
>0.5-2mg/l
Category IV
>2mg/l
3. It is recommended, but not required, that nose-only or head-only exposure be used
for aerosol studies in order to minimize oral exposure due to animals licking
compound off their fur. Individual housing must be used during whole-body exposure
to prevent filtering by the fur due to animals huddling together.
13
-------
4. Whenever the test article is a formulation, the analytical concentration must be
reported for the total formulation, and not just for the active ingredient (aJ.). If, for
example* a formulation contains 10% aJ. and 90% inerts, a chamber Analytical limit
concentration of 2 mg/1 would consist of 02 mg/1 of the active ingredient. It is not
necessary to analyze inert components provided the mixture at the animals'
breathing zone is analogous to the formulation; the grounds for this conclusion must
be provided in the study report If there is some difficulty in measuring chamber
analytical concentration due to precipitation, nonhomogeneotis mixtures, volatile
components, or other futon, additional analyses of inert components may be
uy.
These recommendations were designed to be realistic in the laboratory environment, and
to satisfy regulatory requirements. As such, they reflect the current state of the science.
14
-------
Stanley B. Gross and Frank J. VoccL Hazard Evaluation Division Standard Evaluation
Procedure: Inhalation Tojdcity Testing. EPA-540/09-88-101. August, 1988.
Stanley B. Gross Memorandum. Comments on Standard Evaluation Procedure. Inhalation
Toxicology Testing (SEP/Inhalation). April 18,1989.
Bruce Jaeger. Pesticide Assessment Guidelines, Subdivision F, Hazard Evaluation: Human
and Domestic Animals. Revised Edition. November, 1984. U.S. Environmental Protection
Agency. Pages 47-54.
Trent R. Lewis, P.E. Morrow, R.O. Mcdellan, O.G. Raabe, Gl~ Kennedy, B.A. Scbwetz,
TJ. Goehl, J.H. Roycroft, and R.S. Chhabra. Establishing Aerosol Exposure Concentrations
for Inhalation Toadcity Studies. Toxicology and Applied Pharmacology. VoL 99,1989. Pages
377-383.
Otto G. Raabe, Hsu-Chi Yen, George J. Newton, Robert F. Phalen, and David J. Velasquez.
Deposition of InruiMMonodupmeAercwls in SmaO Rodents. Inhaled Particles IV. Edited
by W. H. Walton. Pergamon Press, New York. 1977. Pages 3-21.
National Agricultural Chemical Association (NACA). January 2, 1991.
Dr. Harry Salem. January 16, 1991.
Jeilinek, Schwartz, Connolly & Freshman, Inc. January 31, 1991.
Wil Research Laboratories, Inc. January 15, 1991.
ICI Agricultural Products. January 16, 1991.
CIBA-GEIGY. January 15, 1991.
Lilly Research Laboratories. January 15, 1991.
Telephnne Copigrmtinr* with Tnhalfttinn
Hsu-Chi Yen, Lovelace Inhalation Toxicology Research Institute. September 17, 1992.
Kenneth D. Nitschke and Richard Corley, Dow Chemical Company. September 15, 1992.
15
-------
Poblk Comments Pennant to Subdivision F Revision for Guidelines 81-3 Acute
Inhalation ToxteHy, and 82-4 90-Day Inhalation Toxkity
Docket Control Number "OPP-36175" (January 2,1991). Only
specific reference to particle size and limit concentration are
included. Unless indicated otherwise they are direct quotes.
a • National Agricultural Chamteai Association (NACA)
Comments on FIFRA Subdivision F Guldallnaa
January 2, 1991
882-4 A halatJon
1.
(I) fohjftttiM!* to he tested (3) *~aero$ol particles may have to be reduced to sizes
which are inhalable for the animal being tested considering the entire respiratory
system of the animal"
EPA Memorandum (4/18/89) Regarding Inhalation SEP • "-at least 25% of the
particles are 1 Mm mass median aerodynamic diameter (MMAD) or less."
If possible, aerosol particles should have an MMAD of 4 pm or less. However, an
MMAD of up to 10 Mm is acceptable in those situations where the physical
characteristics of the test material prevent reducing particle size any further.
Justification:
The main objective of acute inhalation tozkity testing is to assess the spectrum of
toxic effects resulting from short-term exposure of laboratory animals to inhaled
materials. The ultimate fate and toxidty of these inhaled particles is highly
dependent upon the deposition site (a function largely of aerodynamic particle size)
and any biotransformarion and transport processes. Given the spectrum of possible
responses to inhaled particles, it is extremely important to evaluate particles which
deposit throughout the respiratory tract and not just those which selectively deposit
in the pulmonary or other anatomical regions.
The criteria, however, for establishing an appropriate particle size range which
encompasses inspirabflity (a size capable of entering the nose), inhalabfliry (a size
capable of entering the hing) in laboratory animals, such as the rat, have not been
well defined. While acknowledging that some particle size standard is needed for
consistency in acute inhalation toricity studies, the current guidance for 25% of the
particulates with MMAD of 1 Mm or less is considered to be inappropriate. This
16
-------
distribution is considered to favor selective particle deposition in the pulmonary
region and minimizes deposition in the upper respiratory tract of the rat In
addition, it has been noted by several laboratories that accurate size determinations
for submicron particles are difficult and that the stated distribution is impossible to
achieve in many instances despite sincere effort
The available data show that particles with 1-4 and 10 Mm aerodynamic equivalent
diameters are reasonable and appropriate size limits regarding respiratory tract
deposition and inspirability, respectively, in rats. The selection of these cut-points
was based largely on experimental deposition data for rats and general consensus
among several inhalation toricologists. Information on the deposition of 03-10 Mm
monodisperse aerosols in rodents has been published by Raabe (1977, 1988). These
studies with Fisher 344 rats show that pulmonary deposition decreases from a
of approximately 13% with 03 am particles to a mhrfrniim of 1% with 10
particles; pulmonary deposition for 1 and 3-4 Mm particles is comparable,
averaging approximately 9 and 6%, respectively. Bronchial deposition is also
comparable for 1 and 3-4 Mm particles with average values, of approximately 7-8%.
Trachea! and laryngeal deposition generally increase with increasing particle size;
deposition at either site is less than 9% for 4 Mm particles but less than 2% for 1 Mm
aerosols. Nasopharyngeal deposition generally increases with particle size and
accounts for over 90% deposition for particles larger than 5 Mm. An important
observation is that while the total deposition of particles greater than 3 Mm within
the respiratory tract is approximately 100%, total deposition of 03-1 M particles is
typically 25%; that is approximately 75% of the aerosol mass with panicles less than
1 Mm would not deposit within the respiratory tract and would simply be exhaled.
This discrepancy arises from the comparatively low pulmonary deposition efficiency
of fine particles (Le. less than 1 Mm). A practical consequence of this observation
is that for systemic toxicants, monodisperse submicron aerosols might produce less
lethality than the same concentration of a larger aerosol since less of the fine aerosol
is actually deposited within the respiratory tract and available for absorption, Indeed,
work by Tsuda et ai (1986) has shown that the acute LCSO of chtorofenvinphis was
approximately 5-fold lower for a 7.8 Mm MMAD aerosol compared to 0.66 Mm
MMAD aerosol, demonstrating the higher toxitity of t test material with a larger
particle size.
Overall, a 4 Mm MMAD cut-point is a reasonable compromise for respirability since
these particles have pulmonary deposition comparable to submicron particles, yet
also hive appreciable deposition in the upper airways and nose. While 4 Mm
particles cannot be generated (based on the presence of surfactants, dispehants, the
physical characteristics of the test material, or other factors which decrease aerosol
stability and favor agglomeration and particle growth), a 10 Mm MMAD limit offers
a reasonable compromise for inspirability in rats and the practical aspects of
generating and mfiintai'fl*ng stable atmospheres containing these more rapidly sealing
particles.
17
-------
(g) nmhTtett "If a test at an exposure of 5 mg/1 (actual concentration) for 4 hours,
or where this is not possible due to physical or chemical properties of the test
substance, the maximum attainable concentration, using the procedures described for
this study, produces no compound-related mortality, then a full study using three dose
levels will not be necessary."
"If a test at an exposure of 2 mg/L.produces less man a 50% compound-related
mortality, then a full study using three doses will not be necessary."
Jmrifiration
This would demonstrate that the LCSO is greater than the "limit** test cutoff and
could potentially reduce the number of animals required for testing of compounds
with minimal acute inhalation tenacity.
The rationale for a timh test conducted at an aerosol concentration of 5 mg/1 has not
been described in the test guidelines, standard evaluation procedures or other
documents from the EPA. This represents an exceptionally high concentration for
particulates which would not nicely be achieved in an occupational setting for any
appreciable period of time. There appears to be no scientific basis for the use of
such a high dust concentration in acute inhalation toxkity tests.
It should be noted mat while the objective of acute inhalation testing is to provide
an LCSO, it has been stated that the "LCSO is only a rough estimate of relative
toxidty for container labeling~"(Gross, 1986). In tins regard, the EPA labeling
requirements (CFR 40, Fan 156.10; revised July 1, 1989) list category m (LCSO
between 2 and 20 mg/L) and category IV (LCSO greater than 20 mg/L) materials at
ranges which are greater than the aerosol limit test of 5 mg/L. According to these
classifications, there is no difference in the labeling requirement! for a limit test
conducted at 2 or 5 mg/L Rather than expend toe resources and can animals to
determine if t material has an LC50 between 2-5 mg/L ft would be more reasonable
to set the hmit test at 2 mg/L More frequently than not, inhalation lexicologists are
limited by maximum attainable exposure concentrations (of respirable particles)
rather dim the current "omit" value. Consequently a greater number of
formulations would pass the 2 mg/L limit reqtirenient; therefore, reducing animals
required for testing products with minimal fa***1?^ toxkity.
It has been toe experience of most investigators mat the generation of a 5 mg/L
aerosol while t"*{t|»««trf"g a submkron aerosol size are mutually exclusive
conditioning [conditions]. After 70 separate trials at one laboratory (I) in the last 5
years with both liquid aerosols and solid dusts, the mMinmm attained atmospheric
18
-------
concentration with a submicron MMAD for either dusts or liquid aerosols was
approximately 02-1 mg/L. .At the limit test concentration of 5 mg/U the smallest
aerosol attained with a dust ranged from approximately 23-33 jim MMAD
(corresponding to approximately 5-15% of the aerosol mass less than 1 M); the
situation is somewhat better with liquid aerosols as the smallest size attained was
approximately 1.5-2 Mm MMAD (corresponding to approximately 5-30% of the
aerosol mass less than 1 Mm). Thii observation was based on numerous attempts at
grinding test materials prior to generating, optimizing aerosol generation procedures
and incorporating size selective elutriators to increase the amount of submicron
aerosol The data supporting this conclusion is included in Fiqures 1 and 2 (1).
While acknowledging that 5 mg/L limit test for aerosols is an arbitrary classification
limit, achieving this concentration with submicron aerosol is not realistic A "limit"
of 2 mg/L would increase the fraction of submicron particulates achievable in test
atmospheres. A MMAD of 4 Mm would further ensure deposition throughout the
respiratory tract for test conducted at this proposed limit
&R3-4 Siihrhrnnie Inhalation toricitv: 90>dav Studv
(C)(3) . "Inhalable Diameter".
4.
Change! Include in this section something more than a description of this
term. For example, what size ranges are considered to be "inhalable" or
"respirable" for commonly used test species such as rats and mice?
Justification! This would help in setting goals for optimum generation parameters
and would assist in the review process. For example, justification for a mass median
aerodynamic diameter of 4 Mm has been presented for the rat for inhalation tests,
section 81-3(0<3)- This size range would also seem preferable for subchronic testing
in the rat
(e) finfrjrt«nr* to he tested, "...aerosol particles may have to be reduced to sizes
which are inhalable for the animal being tested considering the totte respiratory
system of the animal".
Change to "When possible, aerosol particles-." Also, explicitly
state suggested size distributions for commonly used test species.
Jmtifeation: Many materials have high viscosity, surfcce tension and/or other
restrictive physical characteristics that prevent achieving the desirable paracle-size
19
-------
distributions. Also, see comment 2 of this section. Hie SEP is aiming toward a
distribution that contains sizes which deposit primarily in the respiratory
compartment of the lung (at least 25% of the mass must be less than 1 Mm). This
requirement is neither consistent with a view toward the jnjufi respiratory system nor
supported by current scientific knowledge (Snipes, M&, Long-Term Retention and
Clearance of Particles Inhaled by Mammalian Species, Critical Reviews in Tex, vol.
20, pp. 202 [1989]). There is ample evidence that a criterion of >90% particle mass
less than 5 Mm and 50% less than 3 Mm is more appropriate. It must further be
stressed and emphasized that elevated test concentrations (e.g. 5 mg/1) and maximum
respirabflity are in general mutually exclusive (also see comments 1 and 2 for the 81-
3 guidelines).
b - Dr. Harry Satom
Commants on FIFRA Subdivision F Guidallnaa
January 16,1991
"...the current requirement of submicron particle size and a concentration of 5 mg/L are
almost mutually exclusive. Aerosol physicists have calculated a theoretical minimum particle
size of 1.0 to 1.1 micrometers MMAD for a concentration of 5 mg/L» assuming a
monodisperse aerosoL Many inhalation toxkologistt, including myself; support conducting
inhalation toxicity testing with aerosols of a size that deposits throughout the respiratory
tract rather than those mat target one region such as the alveoli."
e • Jalllnak, Schwartz, Connolly & Fraahman, Inc.
Commants on FIFRA Subdivision F Guidallnas
January 31,1991
81-3 Acute Inhalation TtaicJtY
. EPA Memorandum dated April 18,1989, regarding the inhalation SEP requirement
that "~at least 25% of the particles are 1 Mm mass median aerodynamic diameter
(MMAD) or ten,"
Consideration should be given to relaxing this requirement In some cases it is
impossible to generate particles of this size. After a good faith, documented effon
(not heroic) has been made, a study using the smallest particles that reasonably can
be generated should be acceptable.
20
-------
d • Wil Rssaarch Laboratories, Inc.
Comments on FIFRA Subdivision F Guidallnss
January 15,1991
Inhalation
Proposed Change
•Define the acceptable aerodynamic diameters for each species which may be used and the
percentage or proportion of generated panicles which must be of the indicated aerodynamic
diameter."
Justification
•Requirements regarding aerodynamic diameter of particles are vague and should be
clarified**
f Jmit Tktt
•Clarify concentration requirement in regard to *respirable substances'.*
• • ICI Agricultural Products
Comments on FIFRA Subdivision F Guldalinaa
January 16, 1991
8 81-3 Acute Inhalation Toandty Study
•^however, the particle size criteria contained within the SEP are considered as overly
restrictive by the majority of inhalation lexicologists, based on both on experience and
scientific evidence in the published literature.*1
"-.the attached document, which discusses these issues, recommends a mean aerodynamic
size of 3 /im as being appropriate for inhalation studies in the rat."
21
-------
f - CIBA-GEIGY
Comment* on FIFRA Subdivision F Guidelines
January 15, 1991
8 81-3 AflltP ^n^a^atiffiTl Toricitv
Propose
If possible, aerosol particles should have an MMAD of 4 Mm or less. However, an MMAD
of up to 10 Mm is acceptable in those situations where the physical characteristics of the test
material prevent reducing panicle size any further."
Justification
"The main objective of acute inhalation toricity testing is to assess the spectrum of toxic
effects resulting from short-term exposure of laboratory animals to inhaled materials...while
acknowledging that some particle size standard is needed for consistency in acute inhalation
toxicity studies, the current requirement for 25% of the particulates with MMAD of 1 Mm
or less is considered to be inappropriate. This distribution is considered to favor selective
panicle deposition in the pulmonary region and mininwft deposition in the upper
respiratory tract of the rat In addition, it has been noted by several laboratories that
accurate size determinations for submicron particles are difficult aad that the stated
distribution is impossible to achieve in many instances despite sincere effort
"The available data show that particles with 1-4 and 10 Mm aerodynamic equivalent
diameters are reasonable and appropriate size limits regarding respiratory tract deposition
and inspirability, respectively, in rats~~ Overall, a 4 Mm MMAD cut-point is a reasonable
compromise for respirabflity since these partides have pulmonary deposition comparable
to submicron partides, yet also have appreciable deposition in the upper airways and nose.
Where 4 Mm partides cannot be generated (based on the presence of surfactants,
dispersants, the physical characteristics of the test material, or other factors which decrease
•»m«nl ttahili'ty »nA fmvnr
a reasonable compromise for inspirability in rats and the practical aspects of generating and
maintaining stable atmospheres containing these more rapidly settling particles—. There is
ample scientific evidence that a criterion of 100% of mast less than 5 Mm and 50% less than
3 Mm is more appropriate. High test concentrations in the air (ej* 5 mg/1) and maximum
respirabflity are physically mutually exclusive of each other."
f Jtnft
Proosed
"If a test at an exposure of 2mg/Uproduces less than 50% compound mortality, then a full
study using three dose levels will not be necessary.11
22
-------
"...there appears to be no scientific basis for the use of such a high dust or liquid
concentration in acute inhalation toricity tests.... According to these classifications [tox
categories], there is no difference in the labeling requirements for a limit test conducted at
2 or 5 mg/L.... Consequently, a greater number of formulations would pass the 2 mg/L
limit requirement, therefore reducing animals required for testing products with minimal
inhalation toritity."
8 82-4 Subchronjg Inhalation Toxicilv! QfcDay Study
"Many materials have high viscosity, surface tension and/or other restrictive physical
characteristics that prevent achieving the desired particle-size distributions. The SEP is
aiming toward a distribution that contains sizes which deposit primarily in the respiratory
compartment of the lung (at least 25% of the mass must be less than 1 pm). This
requirement is neither consistent with a view toward the sonic respiratory system nor
supported by current scientific knowledge [Snipes, MB., Long-Term Retention and
Clearance of Panicles Inhaled by Mammalian Species, Critical Review* in Tmrieoiogv. 20,
202 (1989)]. There is ample evidence that a criterion of >90% particle mass less than 5 Mm
and 50% less than 3 Mm is more appropriate.*1
g • Lilly Research Laboratories
Comments on FIFRA Subdivision F Guidelines
January 15,1991
"It would be useful to change the SEP guidelines for acute inhalation studies as follows
- The upper limit for particle size should be based on the largest size panicles which
are inhalable by rats. Available data and models suggest that this value is
approximately 6 pm MMAD. Input from a scientific panel would be useful on this
issue.
•It should not be a requirement to carry out exposures at a limit concentration of 5
mg/L. If a well conducted study with no mortality is conducted at a concentration
of 0.5 • 1 mg/L with a particle size of 4 Mm MMAD or smaller then a category III
category is warranted. It should be an option to conduct other studies at 5 mg/L, or
as high as achievable, not a requirement"
23
-------
Conversatio halation To«gnlogists
h - HaihChi Yth, Lovtlaet Inhalation Toxicology Research Institute
September 17, 1992
MMAD values in the rat should be 5 Mm are
captured in the nose (compared to >10 Mm in the human). He recommended an MMAD
of 2-3 Mm if we want to have maximum deep lung deposition. He considered <3 to 4 Mm
MMAD to be a reasonable range for acute studies because higher concentrations are
needed than in subchronic studies. Although submicron particles would be more likely to
avoid nasopharyngeal capture, they would also be more readily eliminated in the exhaled
air, thus resulting in a lower pulmonary exposure. The exception to this rule is extremely
small particles (those affected by brownian motion) which have increased pulmonary
I • Kenneth D. NKachke and Richard Corley, Dow Chemical Company
September 15,1992
Particle size should be 1-4 MOL The limit concentration should be for the whole
formulation, including the vehicle, and not just for the active ingredient A reasonable limit
concentration is 2 mg/1, unless the product is "like rocks.*
24
-------
Other Items for Df*en««iAf».
1. The label requirements for about 10,000 products are being revised by OREB in
accordance with the Respirator Decision Logic now under development The labels
for all Toricity Category I and U products will require respirators. Respirators will
not be required for Toriciry Category m and IV products because they are not
considered to be an inhalation hazard. Thus,-we may no longer need a inhalation
Toxicity Category beyond ffl. Consider the following Toxicity Category Criteria for
aerosols, gases, and vapors:
Option 1:
Hazard
Indicators
Inhalation LCg,
AEROSOLS:
(analytical
concentration -
4-hour exposure)
Category I
S0.05 mg/1
Category II
>0.05 • 0.5 mg/1
Category HI
>0.5- lmg/1
Category IV
N/A
Option 2:
Hazard
Indkators
Inhalation LCm
AEROSOLS:
(analytical
concentration -
4-hour exposure)
Category I
50.05 mg/1
Category U
>0.05 • 0.5 mg/1
Category III
>0.5mg/l
Category IV
N/A
NOTE: The inhalation Toxicity Category Criteria in the CFR has been incorrect for the
past 10 yean. The concentrations are based on the old 1-hour exposures. We have been
using 4-hour exposures since 1982. Unfortunately, many registrants do not realize the error,
and have been choosing their study concentrations based on this erroneous information.
Many HED reviewers are likewise uninformed, so incorrect tcoticity categories have been
assigned to many products.
.2. A limit concentration has not been defined for subchronic inhalation tojririty studies.
In keeping with the 5-fold difference in oral limit concentrations for acute and
subchronic studies, the subchronic inhalation limit concentration could be one-fifth
that of the acute limit The acute and subchronic limit concentrations for aerosols
25
-------
should be lower than that for gases and vapors to reflect the inherent difficulty in
generating aerosols.
Inhalation concentrations in long-term studies have typically been based on technical
"feasibility, but recent studies suggest that the degree of pulmonary burden due to deposited
particles affects particle clearance (Trent R. Lewis, et at, 1989)
26
-------
Attachment 3, Section 2
\ UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINOTON, O.C. 10440
I 8 'S89 ornci •»
*•• r*«ic
SUBJICTt Comanto on Standard Evaluation Proeoduro.
inhalation Toxicology Tooting (S»/Inhalation)
TO; Xocipiontm of SIP/Inhalation
mans Stanloy •• Orosa, Ph.D., OABT, —
sanior Toxicologitt/lnduatrial Hygioaist
_im (H7SOfC)
you hava or ar« nov r^aiving a copy of oar final
(valuation OlvUion Standard Ivaluation Frocadura.
Taating." m-S40/Oflol( Auguat iftt.
T wiAh to add hara aomo hlatorical clarificationo concarning
iarti^la taatina oisao and tha limit tooting which navo apparantly
c^od ^ eSfSi" ™tnto^ing rooMromonta., to*i f**^
iSformatioT waa diaeuaaod in tha chaptar (Oroaa, Iff?) oitad on
pago 20 of tha SIP.
Tha If 7i Fill* ^^^lixt^nitM limit, for
t partiolao in a product I that U, taatlng
fpUe^t of
sag .r^SJJSJKS^SS^SKSHIS
aoroaol aisaa aro to bo taatod.
-------
laboratoriee have submitted data based on particle aisea of 10 or
15 urn. assuming thsss sizss, which ars inhalable for man, are the
particle eisaa to be teated in animala. This is not the caae.
For humane* "Inapirabla particles" capable of entering the
noee of man have been identif led «• IIS urn. aerodynamic diametera.
"Znhalable" particlee which can pass through th« trachea have been
identified ee 15 m for BUI. "lUspirmble" p4rticl««, small •nough
to roach tho d«op lung or alveoli of man, ara conaidarad to ha i,
5 or 10 m or laaa, dapanding on diffarant litaratura aourcaa.
comparabia diamatara for animala had not baan aatabliahad at tha
tima tha Ouidalinaa wara finalisad, nor hava thay baan at thia
tima. Howavar, 1 urn diamatar had baan racommandad for rata by a
numbar of inhalation toxieolofiata. Thia numbar (1 m mama madian
aarodynamie diamatar) ia atill baiag racommandad for rodant
inhalation atudiaa. If tha maao madian aarodynamie diamatar
raportad in a study ia largar than 1 UB» va can accapt tha study
if at laaat 25% of tha partielaa ara 1 urn or laaa. If tha
laboratory ia having difficulty in achieving tha; racuirad
diamatara, tha study naads to indicate what thay did and why thay
wara unable to provide tha raall partielaa. See the f If/Inhalation
Texieity Tasting for more discussion ea this matter.
Z need to note that many of the particle aiia data submitted
a registrants ara submitted am •optieal" diameters or "sieve"
•meters, rather than aarodynamie diameters. Data en particle
site is to be measured and raportad am •aarodynamie" diameters.
As noted in tha Sit, it ia tha aerodynamic diamatar that determines
where a particle is likely to be deposited in the respiratory
tract.
C. ' *.<«*
The limit test (section g, page si of the guidelines, PMC-
lOSfSS, *ov.'S4) saysi "Zf a teat at an exposure of S mg/L (actual
concentration of respireble substances) for 4 hours or, where this
is not possible due to physical or ffheminsl properties ef the teat
substance, the mswlmnm attainable concentration, producee no
compeund«related mortality, then s full study using three doee
levels might net be necessary. • Tho limit toot usually appliae to
the acute 4 hour inhalation test. This limit is set at the
Toxicity Category ZV in which the material would be considered to
have minimal adverse ef foots during en scute expoeure.
In order to favor s reduced use of animals during toxicity
testing, the Agency has suggested tho use of limit toot (when such
•a test seems appropriate). Zf deaths are seen during the limit
test, a full LCSO test as described ia tho Guidelines is still
required. However, s number of registrants have used the limit
teat as tha only test, as a "yes/no test* and usually at levels
below tho S mg/L concentrations. This doss not fulfill the testing
requirements for this guideline.
-------
further, the limit teat can be carried out at the maximum
attainable concentration. A number of registrants have reported
test results from a limit test at concentrations below s mg/L which
did net cause.- any deaths. The eoneantratien was rapertad as a
maximum attainable concentration without any docuBentation to
support this conclusion. This has not been aeesptsd. Zn order to
declare the concentrations as the aaxiaua attainable, the
registrant needs to indicate what efforts were sade to reach the
5 Bg/L concentrations , what problems were encountered and, if
possible, try to explain why higher concentrations were not
achievable.
D. Data Hapertino?
These Guidelines are or vill be available from the National
Technical Information iervice (MTZf)s
• PISTZCZOB AMISfNBIT QVXDBZiZVIf. Acute and tubchronio
Toxieity Testing." by Stanley 1. Oross, fm.D. DA1T, CZ>. Addendum
• to Subdivision ft Hatard tvaluationt Human and Domestic Animals
(IPA 340/f-M-007) , VTZ« |H«t-ia4077.
Please contact me if you have any further questions on these
issues (or any other inhalation toxicity testing matters) at 337-
cc: M. Copley
J. Hauswirth
April it, i*at
sep.mmo
-------
Attachment 3, Section 3
UNTOO ITATIS tNVmONMIJHTAL PflOTICTlON AQCNCY
WASHINGTON. O.C. 20460
MEMORANDUM
«tncioo AW rate
•UMTANCCS
SUBJECT: Policy on Acute Inhalation Tojririty Data Waivers
FROM: Penelope J0 Tenner-Crisp, Director .,
Health EffcctTDfrision (H7509Q l*j*|1<
TO: Anne E. Lindsay, Director
Registration Division (H7505Q
This is in response to your memorandum of October 25,1991 concerning the determination
of a vapor pressure regulatory limit for the waiver of acute inhalation tenacity studies. HED
has considered this issue and has concluded that low vapor pressure is not a key factor in
determining whether an acute inhalation study for a pesticide formulation should be waived.
Rather, the possibility of the generation of respirable particles or vapors is the more
relevant criterion. Vapor pressure plays only a minor rote in this determination.
The issues involved in the decision to waive an acute inhalation toricity study are discussed
in the attached interim policy paper. Because the issues are complex and dependent on the
specific composition of each product, decisions on whether to waive these studies must be
made on a case-by-case basis. The principles embodied in the examples described in the
attached policy paper form the criteria on which these data waiver decisions should be
based.
We have discussed this policy with Tom EUwanger of your staff. He agrees in principle with
the policy, but wants to discuss it further with the personnel that will be affected by it
Please let m§ know if any additional clarification of this policy is needed.
Attachi
cc R. Schntitt, HED
K. Baetdce, TOXI
R, Gardner, TOX I
J. Redden. TOX I
M,VanGemert,TOXn
W. Burnam, SACB
-------
INTERIM POLICY
WAIVER CRITERIA FOR INHALATION STUDIES
CFR $158340, Note 16 states that for technicals and all formulations an acute inhalation
study is, "Required if the product consists of, or under conditions of use will result in, an
inhalable material (e.g. gas, volatile substances, or aerosol/paniculate).11 The three forms
of inhalable materials can be more simply stated as gases, vapors, and aerosols. For clarity,
a few definitions are offered:
A fas is a substance which normally occurs in the gaseous state at standard
temperature and pressure (e.g. nitrogen, freon).
A vapor is a substance which normally exists as a liquid or solid at standard
temperature and pressure that is dispersed in air in its gaseous state (e.g. methanol,
iodine).
An aerosol is a suspension of solid particles (dusts, fumes, smoke) or liquid particles
(mists, fogs) in a gas.
Vapor pressure is the pressure exerted by the gaseous state over the normal liquid
and solid states. If a chemical's vapor pressure and molecular weight are known,
the tpygfaimiii obtainable vapor concentration in mg/1 can be calculated as follows:
Maximum obtainable vapor concentration (ppm) •
vapor PT9ISUT9 (an Her) x (1 x 10*)
760 an Hg (Atmospheric frewure)
Concentration in mg/1 -
Concentration (PCO) x Moltcultx Mtighfc
24.5 x 1000
AstodiMBk peitkk ste is a measure of particle size that takes into account the
diainear and mass of a particle. Tne distribution of aerodynamic particle sixes is
measured with a cascade impactor, and is reported in micrometers Gun) with a
geometric standard deviation (0A These data are used to estimate where the
particles might be deposited along the respiratory wet Although the term "particle
size" is commonly used in inhalation studies, it always reto to aerotyiiiiiic parade
size and bears no resemblance to optical or sieve peitide stti*wm^ measure only
particle diameter. Aerodynamic particle sizing is often done for vapor studies
because vapors tend to condense and form aerosols.
-------
lahalabl* materials include all gases and vapors, and aerosol particles fine enough
to enter the nose and mouth. Although the current HED cutoff criterisvfor
inhalabflity is 15 Mm, much larger particles can be inhaled. Most large particles
(approximately >8 Mm in human, >2 Mm in rodents) are captured in the nasal region,
where soluble materials may be absorbed and insoluble materials are cleared
Smaller particles pass deeper into the respiratory tract
Respirable materials are capable of entering the lung alveoli While these particles
can range up to lOpm, most large particles are deposited along the respiratory tract
before reaching the lung. A respirable particle in man and rodent is defined as
having an aerodynamic particle size of il Mm.
Acute toxitity studies are important because most serious pesticide poisonings involve acute
exposure. All reasonable efforts should be taken to perform acceptable acute inhalation
studies in order to characterize the toxidty of the technical and demonstrate the potential
inhalation hazards fhyt could result from use of the formulation. Waivers should be granted
sparingly and judiciously. Altn«jgh it is preferabte to have mhalatic« toxidty dau for ^
the technical and the formulation, it may not be possible to do both.
Acute inhalation study waiver requests are orratfonalry submitted to the Toxicology
Branches. These waivers can be granted provided the Registrant adequately demonstrates
that inhalation exposure will not occur under conditions of use, and/or an inhalation study
cannot reasonably be performed. Some pesticides are by their nature impossible to generate
in inhalable form and thus pose no inhalation hazard. These include some waxes, resins,
high viscosity liquids, micro-encapsulated products, and non-friable granules.
Low vapor pressure cannot be used exclusively to justify t waiver because there are too
many other variables involved, faH™«««g method of application, """"P"* attainable
concentration, and overall toricfty of the substance. A chemical with a low vapor pressure
may still be an inhalation hazard. Thus, it is impossible to designate a definitive cut-off
vapor pressure to use in resolving waiver issues.
AUptttttoMfcrwalvtfiaMMtbtcoiiidertdosiaeaiibyctMbasia, It is not possible to
develop a policy to cover all contingencies. The following hypothetical waiver requests with
discussions of their merits and options are provided at guidance in the decision process:
1.
Most acute inhalation study waivers are far products applied at dusts or sprays with
large particle sizes. These waivers are usually denied because it is tiktry that some
of the particles may be inhalable (capable of entering the upper respiratory tract).
and possibly even respirable. The Registrants are urged to nffl solids into fine
powders, and to use nebulizers that yield the smallest partick size possible.
-------
te a low v
that cannot be Mm
3.
Rather than granting a waiver in this case, the test substance for the study should be
a solution of the technical If the technical is water soluble, it should be mixed with
water to lower its viscosity and facilitate nebulization. If it is non-polar, it could be
mixed with a non-polar vehicle such as alcohol in order to reduce its viscosity, but
this will introduce vehicle toxicity. Current EPA Guidelines do not require a vehicle
control group unless the toxicity of the vehicle is unknown (this policy may change
when the Guidelines are revised). If a study can be performed using the end-use
product, a waiver should be granted for the unwieldy technical
AMIICA f\f • »*
T I « I
iw> itwiifffefent vannr can he venerated tr>
4.
The fact that a technical 's low vapor pressure may preclude generating sufficient
vapor does not rule out testing with an aerosol (the maiiimim obtainable vapor
concentration can be calculated using the formulas listed under the definition for
vapor pressure). Any chemical which cannot be generated as an aerosol and has a
low vapor pressure is an excellent candidate for a waiver.
For example, if a formulation containing one or more low vapor pressure insecticides
is sprayed onto baseboards, it should be tested because of the potential hazard of
exposure to fine aerosol particles. If the same insecticide is applied as a waxy
formulation, there may be no inhalation exposure during application.
Plastic collars and ear tags are designed to provide slow-release dermal exposure to
pesticides. Because there is minimal inhalation exposure, and because it is
impractical to grind up collars and tags to produce inhalabk particles, waivers should
be granted in most cases. If the technical is expected to pose an inhalation hazard
due to high volatility and toxicity (LeJow !£»), exposure testing will be required.
Exposure to vapors it not applicable to toridty data waivers.
.
fgMnfed far inhalation tnridtv tejrint^arfcnilgye tea
d. however the
product fa mixed with diesel fuel Hid ttTlYM
Thii liifaniM k potetitiillv inhalahle
An acute inhalation toxicity study of the end-use product should be performed using
the diesel fuel solvent A vehicle control group should be considered in the study
-------
protocol This is the only reasonable way to assess the toxidry of the pesticide by the
inhalation route, and it tests the only use pattern that poses an inhalation hazard.
6. The gnd-UM formulation is a non-friable granule.
As long as the granules remaint intact, there is no need for an inhalation study of the
formulation because the granules are not inhalable. The Registrant must demon-
strate that the granules do not produce fine particles when subjected to shipping and
fry pilling, A simulation is required followed by particle «
7. Tiig ctid-uM forrnulttioD is
Some micro-encapsulated products are prime niytidiff^ for inhalation study waivers
because they tend to be large (ie. generally not inhalable), their shells are nontcotic
plastic, and they are difficult or impossible to generate as inhalable particles.
Nevertheless, each must be considered individually. Capsules that are readily
fractured or dissolved, time-released, leaky, or small in size should be subject to
inhalation studies, probably using homogenized or dissolved capsules.
Prepared by John E Whalan
November 27,1991
-------
DEPOSITION OF INHALED MONODISPERSE
AEROSOLS IN SMALL RODENTS
OTTO C. RAAS*, HSU-CHI YEH, GIORGI J. NEWTON,
ROBERT F. PHALENf and DAVID J. VELASQUEZ
Inhalation Toxicology Rf starch Institute,
foundation, Albuquerque, N.M., U.S.A.
Reprinted with permission by the U.S. Environmental Protection Agency
XAwew—Aerosols of five different tarn varyiai from about 0.05 tan to 3 tan ia
aerodynamic equivalent diameter wen pnpand from spherical alumiaosilicate panicles
(ft m 12 i/tnt*) labelled with >«Yb. Fifty hamsters and fifty no wen exposed, white
anacsthetind. to UMW monodispam aerosols under controlled laboratory cooditiona
to dctarmim quantitatively deposition in the respiratory airways with respect to panicle
size and volume of air inhaled. Tea animals of each special wan exposed to each of the
particle sixes. Five animate were exposed simultaneously for 20 mia ia aa inhalation
exposure apparatus which provided automatic neordinp of respiratory rates and
volumes uaiai individual whole body ptethysmoaraphs; three wera sacrificed immedi-
ately after exposun aad the rcmaiaiaf two altar 20 h had elapsed for radioanalysts of
selected tissues to determine deposition quantities aad orpa distribution. Detailed
results an fivea for the observed dtpositioo fractions for the different aerosols ia the
various para of the respiratory tract The lowest fractional total deposition occurred for
particles between 0.3 *m and I *im »attodynanu^equrvale«alameier. Relative deposi-
tmainoni the various luiiilobeasltt»eU i smarlrihhm^^
sue. Based upon therasultaof morphomatric measuremeata of replica casta of luafiof
the two specaea, iaterpretatioaa an made nlatiag tat i
to i
INTRODUCTION
Many iovcstifaton ust rodrats of various typa* iadudinf rats and Syrian buniuri
in inhalation toxicity studies with polydisptrst atrosols. THOMAS (1969) reported
results of detailed deposition studies of the exposun of nice, rats and guinea pip to
various polydisperse aerosols. In such studies the deposition and clearance of the
particles in the respiratory tract and the pattern of insult and nsuliinibiotofical effects
may vary markedly for partidea of different sizas. Thett variation ait difficult to sort
out and evaluate with data from polydisperse aerosols when panicles of widely
different SUM an present. With the availability of carefully pnpand and characterized
•Current addraas: IUdioM«loay Laboratory, School of Vesariaary MiHrini. University of
California, Davis, California, UAA.
t Current address: Department of Community and Earironmental Mediriae. Uarvcntty of
California. Irviaa, CaWcnua, UAA.
-------
4 O.G. RAAll«ftff.
monodispene aerosols of insoluble forms labelled with radionuclides, the behaviour or
specific sizes of panicles can be studied. Although some studies have been performed
with rodents and monodisperse aerosols, such as those reported by HOL.MA (1968)
these were limited in scope and involved only larger particles.
In the research reported herein, a detailed investigation was conducted of the
deposition of well characterized monodisperse aerosols of aluminosilicate spheres
labelled with lMYb. The respiratory parameters describing the breathing of each
animal during each exposure were carefully measured and recorded. By measurement
of the radiolabci in various pans of the respiratory tract and other organs it was
possible to determine the fractional deposition in the major regions of the respiratory
tract of each panicle size for each animal. Morphometric measurements of replica
casts of the respiratory airways of the lungs of each species were made to provide data
on the typical path characteristics for each lung loot for evaluation of the effects of
airway geometry on deposition pheaoi
The results of these studies help clarify the effects of panicle sin on regional
deposition and interlobar deposition of inhaled particles ia rodents. The panicle sin
providing a minimized fraction of bronchial deposition was observed to be. about
0.5pm aerodynamic equivalent diameter. The eject of partide size on lobar deposition
was found to be consistent with the airway morphometry of the different lobe*.
MATERIALS AND METHODS
The assumed practical definition of a monodisptrse panide size distribution is a
distribution of particles whose coefficient of variation (ratio of standard deviation to
•the n»ean) bless than (U (20 %)wreconimencM by FUCHS and SUTUGW (1966). Fora
sin distribution which is adequately described by a log-normal function, this is about
equivalent to t geometric standard deviation *t ) the slip
correction used to correct Stokes* Law for tht behaviour of small particles.
Tin TASK GUOUP ON two DYNAMO (1966) has recommended an aerodynamic
(equivalent) diameter for use in inhalation studies denned as the -diameter of a unit
density sphere with tht same settling velocity as tht pejtfcte to question''. Hence tht
unit density aerodynamic (equivalent) diameter, Dm ia given by:
-------
Deposition of Inhaled Monodispene Aerosols in Small Rodents 5
with C(D.J the slip correction for a panicle of geometric diameter equal to £M. For
DM values greater than 0.3 ^m corresponding Du values are about Q.lpm larger.
Preparation of Aerosols
The monodisperse aerosols used in these studies were prepared by separating and
collecting monodisperse fractions of a polydisperse aerosol of fused aluminosilicate
spheres which were labelled with "*Yb, a radioactive isotope of ytterbium. The poly-
disperse aerosol was prepared as described by RAAII tt at. (1971). Basically this
procedure consisted of treating montmonllonite clay samples with concentrated
H:02 to degrade organic impurities, decanting the finely-divided clay suspension.
Fw. 1. Etoctroo miooenph* of aeroiot samples shadowed with Ghromium vapour showiat (A)
aerosol pmrtfri by nebuluacie« with a Lov«iaee Nebuloer of aa aqueous luipemicm of mont-
morillonite day, (•) aerosol of alumiaotilicate sphena created by heatinf of the day eeroioi to
1130*C aad (Q moaodttpene panidet collected with the spiral duct eeatriftifli «ed to prepare
monoitopem aerotols for 'nMlvfra studiea.
-------
*~ O. C. RAAM it at.
treatment -with sodium chloride solution to load the ion-excnann
" v°^',dialy$i$ Wlth deion"* **« to remove excess sodium
W"h
oum *•
YbCl, wuh the final clay suspension to allow for ion exchange of u»v! *nd m'""l
the clay. The resulting suspension of "'Yc-labelled clay was filtered V C*tion$ ""°
de.on.zed water to remove unexchanged '«»Yb. The polydisoenT *°dw«*fcd *uh
*as generated by nebulization of an aqueous suspension of Mm/*!0101 (Fil- 1A)
pai»ed through a quartz tube furnace at 1 150'C to fuse the clay oan i ^ ml and
silicate spheres (Fig. IB) in which the "»Yb label was entrapped I$int0alum"»o-
F». 1 Scbmuk dnwinf of aa exploded view at the U*tea Aemol Nniek Scpww«r
(LAPS), UM spiral duct enttiAip uMd m this study to stparaie aad cotai taoaodiiptm «a
flnctioM of fUMd alumioetilkit* pvticta l^eiled with 'MY* (from Konum * 4.. I9U by
Pirn
This l**Yb-!tbtJI«d ahimiaosilicitt atrosol wu umpled with • Lovelace Aerosol
Ptnidt Scptrator. LAPS (KOTKAPTA tod LIGHT, 1972). • spiral duet aerosol centrifuge
(SrOm and FUCMSIART, 1969), which operates as an atrosol spectrometer (Fig. 2).
Aerosols drawn into the LAPS art separated and collected according to their aero-
dynamic properties under the influence of centrifugal and corioits forces as (he duct
rotor is spun at t constant 60 Hz with an induction motor. The aerosol tampk
(0.3 1 ./coin) is drawn into the spiral duct through aa inlet slit at the aais of the rotor sod
-------
Deposition of Inhaled Monodispene Aerosols in Small Rodents 7
flows parallel to t laminar stream of clean air (4.7 l./min). The aerosol particles move
to the outer wall Of the duct with speeds depending upon their respective aerodynamic
diameters and deposit upon a 46.2 cm long, 3.1 cm high and 0.01 cm thick stainless-
steel foil lining the outer wall of the spiral duct. Particles with aerodynamic (resistance)
diameters, Dut larger than 0.6 pm are collected on the foil and the smaller particles
are collected on a 3 x 3 cm glass fibre filter at the end of the LAPS duct After codec*
(ion of a satiable quantity of particles with the LAPS, the foil was removed and cut
into twenty-three segments providing different size groups of monodispene particles
which were subsequently suspended in water for nebulization to yield the desired
monodispene aerosols for the inhalation exposures.
The basic idea of aerodynamically separating insoluble particles into monodispene
size groups using a spiral duct aerosol centrifuge and subsequently generating mono-
disperse aerosols from these separated particles was suggested by KOTEAPFA and Moss
(1971). The equipment and methods used in this study wen essentially as described by
RAAM er aL (1975) and NIWTON «r aL (1976).
Figure 3 shows the D* calibration of the LAPS, which was performed using poly-
styrene latex panicles (Dow Diagnostics. Indianapolis, Indiana, U.&A.), the equivalent
geometric six* of the clay aluminotUkate particles (p -12 g/cm3) used in this study,
j e
OftTANCEOOWN
Fio. 3. Calibration for the Umtoea Acreeol Partide Separator (LAP1) used*
coitet aluminosilieaie penidei into monodispene frictions showma the —— •
dyaamieCnrtittiiceldiaineter.^verswUiepoiiiio
down eteimeL The data is teed^P»"«««aitis, douUeia"J-:---.--•-^ ta^,-,... *.
. jk •— S^^^AAlA^^**^^^ ^«>^^^^^^^ A ^Mh A^MMMM flB A BUMB InOfl^BKlBfl UiB
MKMM^kff^^H HW (v^A fl^AAU^BCKIB^^i 9 IHBBWnCwMOw Q^^BuWv* f^U9 VBvWvl IV v mnv »w ••^^^^^^^•^^^p
poneufe dieimur of ftjsetf day ilumimxilirm partidei (»-12a/cea9 eallested in the
-------
S O.G.
and tht positions of the twenty-thnt segments of the collection foil. Figure 1C is an
electronmicrograph showing a sample shadowed with chromium vapour of one size of
monodisperse aluminosilicate particles.
Monodispene sizes used in this study had aerodynamic (resistance) diameters. Du,
of 3.15 Mm, 2.19 pm, 1.14 pm and 0.61 pm which corresponded to aerodynamic
(equivalent) diameters, />„, of 3.05 ftm, 2.09 pm, 1.04 pm and O.S2 urn, respectively,
at the local conditions (21*C, barometric pressure 83 kPa). These aerosols were
generated by nebulization of dilute aqueous suspensions with the concentration chosen
to provide an aerosol with a suitably high proportion of single particles as opposed to
aggregates that are formed when two or more monodisperse particles are aerosolized
in the same droplet from the nebulizer. RAAM (1968) provided the criteria by which
these concentrations were determined. The pH of each suspension was adjusted to 10
with NH| gas for stabilization. The Lovelace nebulizer (RAAK, 1970) was used to
generate thaw aerosols for the exposures; it provided about 50*4 of droplets in
1.5 l./min of air with volume median diameter of about 5.1 M«(«f» 1.IX Each aerosol
was mixed with about lOL/min of data, dry air to dry. the droplets and passed through
a 2 mCi "Kr discharger (Thermo Systems, St Paul, Minn.) to reduce the aerosol par-
ticle charge distribution to ff^*»p**««««i equilibriuai prior to inhalation by the •"«•*• u
In addition to the four monodispent aerosols, § fifth aerosol of smaller panicles
was prepared which turned out to be two slightly different near monodisperse aerosols
(9, - IJ) with activity median diameter about OJpa0tf (tttpm^Jia one case
and about 0.1 pa £„ (0.05 pm 0 J in the other. These wen generated by nebulr
tion of dilute silicic acid solutions with l**Yb tad conversion to silicate panicles »/
heat treatment at 11SO*C When required, these smaller particles were delivered
directly to tht exposure apparatus without intermediate LAPS separation; the "Kr
discharger was used.
Tht 1<§Yb label used in this study was entrapped in tht insoluble fused aluminosili-
cau particks and its radioactive emissions wen used for quantitative measurement of
tht aerosol concentration and regional deposition in tat animals. The l**Yb decays by
electron capture with a half-life of 31.3 days, ft emits t host of highly detectable gamma
tad X-ray photons including 310 keV tad 200 ktV gamma photons in 11 and 49%,
respectively, of the disintegrations. It is available with specific activities exceeding
1000 G/g (General Electric Co., Valkcitot, California).
Aerosol simples wen assayed using a Nal crystal and multichannel analyser. Tissue
samples wen assayed using large volume liquid srintillifion well counters (Packard
Co* Inc, Downers Grove, Illinois) tad, in addition, the smaller organ
samples wen assayed with t Nal auto-famma counter (Nuckar-Oucago, Inc..
Chicago, Illinois).
Syrian hamsters tad rats wen exposed individually ia tat five-port inhalation
exposure apparatus with associated whole body plethysmographs which was dcscri >
by THOMAS tad La (19*3). A sixth port ia tat apparatus wit used for obtaining
-------
Deposition of Inhaled Monodisperse Aerosols in Small Rodents 9
aerosol samples including seven-stage cascade impactor samples (Mticn it al., 1970)
for verification of the size and monodispenity of the aerosols, filter samples for
measurement of "»Yb concentration during the exposure and electrostatic precipitator
samples (MORROW and MERCER, 1964) for electron microscopic evaluation of the
aerosols. Figure 4 is a schematic drawing of the. exposure system used in this study;
components were housed in three stainless-steel and glass glove boxes for radiological
safety. Each plethysmograph had one animal in an airtight plastic chamber with the
head protruding through an airtight latex rubber collar. In this apparatus the aerosol
was drawn continuously past each animal's head. Pressure changes in the plethysmo-
Fio. 4. Schematic raprcMotatioa of the appantui we* in d* study tor i
ft* (odeatt to oooodbpcm acraofc of i« Yb-tabelhd njeid tlumiDO»dk»M
afrtMp^fWwfaote-bcdypteUty^^
awv« ooMitnomuiiin«quipai •»MM um MWhwim www«— ~— - - — ••••-«-: •--
of aluminouUcau sphra «« atmeUMd with (he eebulte optftMd by eoapraMd av (UunA
i. iiii mm. tn iiiiiniiiii mil mi-111 f --1 •''* - • r^""*•*"* ** •"•••>M•'^^
SViTaeiuiied. Tbe acraeei WM flm diluted (to provide for drop* evmpentioa)
HK * a fiher aod i«^
v*ej ac( aouawi. the air which wmmlly p«ert rtwu^ihtaebutor
WM dincMd to tht dilutiof air UM m a vmlv« at SV (which had twa adtjusMd to provide a flew
tqual in voiuxmtric we to the aormal atbulter output rue). Tbe aeieeol peaied ihrouih a at prawn |Jv«« by vtcuum pufai V1-V4 (aot
^
^
mai coaditkMt: 21'C aad »%
-------
10 O. G. RAAII ft al.
graph chamber were used to monitor respiratory activity. Such respiratory data were
both plotted on a strip chart for analysis by hand and recorded on magnetic tape for
analysis by analogue computer to determine tidal volumes, minute volumes and
breathing rates.
Prior to exposure, each animal was anaesthetized with sodium pentobarbital given
iruraperitoneally to render it unconscious for about 1 h. In a few cases, individual
animals succumbed to the anaesthetic and were not considered in analyses of the
results. Each animal was then placed in a plethysmograph unit, a leak test was made
and the front of the plethysmograph chamber was inserted into an exposure port. The
exposure began after all chambers were inserted. Exposures were normally 20 mm in
duration.
The Syrian hamsters used in this study were non-inbred golden Syrian hamsters.
.Vfoocrictfitf aurana. Sen: (SYR) obtained from ARS Spragut Dawley. Madison.
Wisconsin, U.S.A. The black and white hooded rats used in this study were Long
Evans Mack and white hooded rats, Jtamrmnreftaa. Bin: (LE)BR obtained from
Black Spruce Farms Inc., AUamont, New York. U.&A. Ten Loot Evans black and
white rats and ten golden Syrian hamsters wen exposed to each of the five particle
sixes, four mooodispent and one near monodisptnt. la all, fifty animals of each
species were exposed. Each sex was represented by half of the animals exposed to
each aerosol. Twenty separate exposures of five animals each were conducted. For
each exposure three animals were immediatriy sacrificed and dissected for radioassay
of the tissues and two wen placed in excreta-coUectioa cages to be sacrificed 20*
after the exposure. Each animal was canfutty dissected after sacrifice and tissu*
samples were placed in preweighed containers for weighing and radiation counting.
The individual tissue samples for each animal win (a) right apical lobe of lung, R. A.;
(b) right cardiac toot of lung, R.C; (e) right diaphragmatic lobe of tang, R.D.; (d)
right intermediate lobe of lung, RX; (e) complete left lung (both left apical, L.A., and
left diaphragmatic, L.D., lobes); (0 trachea below larynx including carina and portion
of major bronchi; (g) larynx and associated tissue; (h) skull; (i) G.L tract; (j) pelt;
(k) paws and tail; 0) carcass including spleen and kidney; (m) liver; (n) head skin and
(o) misceOaneous remainder including paper towels and dissection wipes. Excreta
samples wen also counted for animals sacrificed 20 h after exposure.
Deposition data for the lung lobes of the immediate sacrifice and 20 h sacrifice
animals wen compared for each aerosol panide size to evaluate the lung clearance
that occurred during the first 20 h altar exposure. Material cleared from the lung
during these 20 h was assumed to have beta deposited on the ciliated airways of the
bronchial region, and the remainder was assumed to be pulmonary deposit The
fraction,/?, of the total respiratory tract deposit which was represented by the bronchial
deposit is given oa the average by:
*-«L-«L
with £• and lie the observed lung burdens of deposited aerosol and T* and F20 the
total respiratory tract deposition in the immediate and 20 hstcsito animals, respec-
tivery, and ( )„ indicating tat average for aU animals; r* includes all observ
internally deposited activity including excreta and is equivalent to TV The fraction.
-------
TAMJ I. UMMI WHOM? Omnmmu* N» *MX AND Wmre Huomu RAT AND SVMAN HAMKUH
Syrian fcMMtaY
S.D.
MCM S.D. %T.L. MOM
27-Mafc 27-
S.D. %T.L. MOM
34
114 12
• 2M
11.7
•.424 AIM H.t
•JS2 «JMO 11.7
UA
•.71ft AIS2
Xltt
7t 24
2DI SS
2S7 17
114 12
• IS2 «j012 10.2
•.211 »JM7 14.2
•.427 «JM1 21.7
0.114 •Jttl I2.S
• 114 MI2 M.S
1.410
SO l»
III 41
1)2 I)
140 10
0.0027 OuM2 10.1
0.0944 OJ019 12.4
. '/.T.l.
I4f 3J
O.IOS OttMk 11.2
O.IIS 0.01S 12.)
•X21)
OJ059
0.244
•.744
Of
45
•AM
•J02S
0453
34
22
305
11.2
341
0.20)
0.114
0322
0.9)9
49
40
0.040
O.OJI
O.UNS
24
20
30.1
12.1
34.3
-------
12 O. G. RAABI tt al.
/L, of the initial lung-burden which is represented by the bronchial denoiir ft •
the average by: FWI1« »vea on
"-"©„
Airway '\forphomttry
Replica casts of lungs of rats and hamsters were carefully prepared and moron
metric measurements made. Silicone-rubber castings were made in lungs that we°"
intact and still within the thorax by an in situ method that has been described bv
PHALEN */ al. (1973). After the injected rubber had cured, the lungs were removed
from the thorax, and the lung tissues were digested in sodium hydroxide. Each flexibl
replica was carefully trimmed to leave only that portion of the tracheobronchial tree
down to and including terminal bronchioles. Measurements on casts, performed by
five trained persons, included branching angles, lengths, diameters and inclinations to
gravity for individual bronchial tubes. Morphometry was parfonne4 on casts from
one rat and one hamster. A "typical path" model of the tncheobronchiat airways
down to and including terminal bronchioles was developed for each lobe from com-
puter analyses of the morphometrie data to provide a simple scheme for relating
morphometry to panicle deposition phenomena.
RESULTS
Anatomical and Pkysiohfical DOM
Basic characteristics of the animals used in this study (and some extras) are sum-
marized in Table 1, including the averages and standard deviations for body weight,
age and lung toot weights at necropsy. Although the absolute weights of each lobe
varied among species and sexes, their relative weights expressed as percent of total
lung weight wen nearly the same irrespective of sex and species. This property simpli-
fied the interpretation of relative deposition in each lung lobe.
Deposition Data
The values of/rand ft and the calculated standard errors observed in this study for
both hamsters and rats are given as percentages in Tabit 2 for the various particle
2. taoNCHUL (20Ht cuAiao) Dumnajn AS moKTAOS or
INITIAL TOTAL Dtrosrr w Rawaunav TaACT./i». AND AS
or boriAL Una iwuMN. /t (wire CALCULATBD STANOAID wees),
OJ2 1J04 109 3.05
7r *7±2 5 ±7 II ± 9 I«± 7
i 37 ±2 «±9 34±15 51 ± 15
{£
•*t ffT »±3 4±2 IS± I 5± • W±,J
*** \Z 24±4 5±3 24±13 II ± 31 5I±23
-------
Deposition of Inhaled Monodispene Aerosols in Small Rodents 13
sizes. Although quite variable, they show basic trends with the bronchial fractions
being minimum for the O.S2 urn particles.
Using the calculated avenge values of/1 and/r. the observed lung burdens of "»Yb
were divided into appropriate bronchial and pulmonary portions. Each deposition
quantity was converted to the fraction of the total inhaled aerosol which was associated
with that quantity by dividing by the product of animal minute volume, the exposure
time and the aerosol concentration. The skull and gastrointestinal activity for the
immediate sacrifice animals was taken to represent the nasopharyngeal deposit. The
results are summarized in Table 3. Other internal organs were free of IMYb.
TASLI 3. MKAN DvonnoN PwctxrAOU AND Ovnvio STANDARD EMOM or INMAUO
MoNooorasi i*Yb-utacuio ALUMNOOLICATS Anosou
Aerodynamic dianeter 00) D* (equtatat unit deadly tpeen)
<04 0.52 1.04 109 3.05
KAMBms
f Pulmonary 134 ± 1.9 114 ± 1.7 5J ± 0.7 5.5 ± U 9.« ± 13
L^\tfOMfaiaJ 7.9 ±1.1 0.15 ±0.09 3.0 ±0.4 7.9 ±0.7 3.4 ±04
Trachea OJ ±0.1 0.10 ±0.03 0.09 ±0.02 0.4 ±0.1 0.6 ±0.3
Laryes 0.9 ±0.4 04 ±0.1 1.0 ±OJ 44 ±19 44 ± 1.1
:T~^ fSkufl 6.9 +1.1 17 ±OJ 4.« ±OJ 24.1 ± 5J 214 ±55
*, . /*«« ^ .^ ^ i^ oj ±7±0.04 04 ±0.1 04 ±04 0.7 ±OJ 3.4+0.01
'^^ rSkufl 4J ±0.« 10+0.4 4J ±OJ 17.9 ±4J I7J + II
Maw phftf)U^o L 0.1+0.1 OJ ± 0.1 OJ ± 0.« 0.04 ± OM 17.3 ± 6.4
Total of above values 313 + IJ 14.0 ±1.4 14.0 ±1.4 2U ±4.9 4IJ + 74
To3SbM(V«0 31.1 ±1« 14.1 ±1.7 144 ±1.7 41.1 ±44 511 ± 3.1
Tablet 4 tod 5 show the relative lobar burdens of the immediate (0-h) and 20-h
stcrulct rttt, retpe«ivtiy; and Tablet 6 and 7 show the inimediate and 20-h burdris,
nsptetivtiy for the hanisters.Coniparison of the datt for the two sacnto tunes for
eachspecies shows that the rtlativt amounts in the lobee varies littlt up to 20 h post-
exposure. This impliei that clearance rates wen about the stint for til lobes.
-------
0. G.
RfUTTVt LOIA* DWOSmON OP ««»Yb UMLUO AiWCNOWJCATI AttOKXJ I* BUCK AND
WHITI HOODED RATS AT O-K Posr-tx*osuM. mctsr or TOTAL LCNC Be ROIN
(mean and standard deviation)
Aerodynamic diameter Oim) 1>M (equivalent unit density sph*rt)
Lobe
3.05(5)'
2.09(4)
1.04(7)
0.52(6)
< 0.2(6)
Numbv of animals to
ILA.
R.C.
R.D.
R.L
LefUuttf
13.77 ± 13J
12.07 ± 5.06
25.86 ± 1.42
10.44 ± 164
37.M ± 5.19
14.77 i 3.52
12.27 = 0.64
28.14 z 3-22
10.31 ~ l.?6
33.82 = 141
11.41 ± 1.73
14.19 ± 1.31
28.19 ± 117
11.31 ±0.89
34.80 ± 141
11.35 ± 1.13
14.17 ± 1.97
25.23 ± 177
1142 = 1.14
36.83 ± 117
10.95 ± 1.73
13.23 r 1.49
26.99 ± 1.56
1168 - 1.49
36.15 t 1.81
12.20
13.35
26.98
11.54
35.95
TAMJ 5. RuATivt LotAa DuounoH or
WIBTI HboMD RATI AT 20m
(
men* or TOTAL LuNoButMM
deviation)
AND
Lota
3.0S(4)«
109(4)
1.040)
OJ2(4)
Av.(19)
1121 ± 1.97
14.05 ± 1.07
26.60 ± « 29
I3J» i 1.39
34.14 ± 3J3
11.79
13.98
25.92
10.99
37.32
•Numbvof
TAHI 6. RnAim LOBAI DtmaiuM or t«Yb LAMUIO
HAMRIU AT
-------
Deposition of Inhaled Mooodispene Aerosols in Small Rodents 15
TAHS 7. RttATTvi LOBA» DmemoN or «*»Yb UMUO ALUMNOBUCATI Asueou IN SVBUN
HAMSHU AT 20>H Pojr.txfosuu. naciMT or TOTAL LUNO ButotN (HMO tod suodtrd deviation)
Aerodynamic diameter (Mm) Dm (equivalent unit dnsity sphere)
Lobt 3.05(4)* 2.09(4) 1.04(4) 0.52(4) < 0.2(4) Av.(20)
R.A.
R.C.
R.O.
R.L
Left lung
17.93
17.62
21.7S
1.34
34.36
±5.69
±3.41
±3.76
± 1.93
±1.10
io.S2 r
15.62 =
23.19 ±
10.24*
40.12 ±
3.17
2.04
7.02
0.12
10.07
14.24
11.52
28.62
10.70
34.90
±2.51
± 1.02
±2.43
±1.93
±3.61
U.M
1166
25.26
1241
3S.OO
± 1.3S
±2.33
±3.27
±113
±2.01
10.51
9.52
27.60
12.26
40.10
± 1.95
± 1.96
±4.19
±3.01
±7.63
13.0S
13.39
2321
10.75
37.50
• Number of animals in parenthesis.
The relative lobar activity percentages at both 0 and 20 h were combined and divided
by the appropriate weight percentages of total hug to yield Table S for rats and Table 9
for hamsters. The values in these tables are equivalent to relative lobar concentrations
of deposited IMYb labelled particles; In almost every case, the right apical lobe had
the highest concentration (except for the case of <0.2 pm in hamsters). For the remain-
ing lobes, the relative deposition changed with panicle size. The relative deposition
RttATtvf LeeAa GOHCSMBATKN or i«Yb
UCATI PAftTKUB IN HACK AND WMRI HOOOCO
PUCSNT or TOTAL Una lutoex/maxT or TOTAL LVNO WBmrr
Lobe 3.05 109 1.04 0.52 < 04 Mat*
ILA. Ul U2 1.12 145 1.10 1.15
R.C OJI 0.95 140 142 0.97 OJI
OJ4 0.99 OJf OJf OJJ OJ2
0.75 0.11 OJ5 142 1.04 OJ4
Ul 0.97 141 14f 141 1.05
0.99 1.02 141 141 142 141
S.O. ail 049 041 041 043 045
TABU 9. RBVAHVI LOCAK CoNCBrnuTBN or *•*¥*
Aunes«oaucATS PAMICLSS w Fun Cuv w SVBIAM HAMRSM
or TOTAL LUNO •wmoeN/ftAONT or TOTAL LUNO Wkmir
Lobe I4J 241 144 OJ2
-------
W O. G. RAAU
-------
Deposition of Inhaled Moaodispene AerosoU in Smill Rodents
17
-.
•H J*» • " I- •<*•
Fn.
of luxptomcric data iUurtntiat th»
oMphotoay oa
-------
TAM* II. TVTICM. PAIN
or AMWAVI or MAT MM EACM Lam
UNUIN (U UUMTIUI U). HUANTM
YD DMKIMJM
-------
12. TVMCM. PAIN ItfoMk off AMWAVBW HAOTU M» R
tff,*)AM»A»«MJJOG«AVnV(4*)R»V
tOM MOWING SMMMT LMOTH (t), DlAMtTU fc/). BRANCH ANULK
HUM IfcACMiA CO
ILL
R.D.
LA.
L.D
N*
T
2
2
4
S
4
7
f
10
II
12
11
IS
17
It
L
24
4J
47
0.19
IJ4
1.70
OJI
0.90
0.79
4XS4
OJ4
*««
2J
IS 25
IJ 40
LSI 11
1.25 II
OJt 14
•Jt 22
4XS9 24
0.41 21
OJt 22
till 41
1.
24
4J
4.5
S.4
O.I
IS
•44
IJI
•97
OJI
•42
0.47
4
14
IS 25
19 IS
IJ 40
1.4 0
IJ 25
IJ If
OJI 4
•71 12
•59 II
•51 It
•41 24
•11 2S
L
24
4J
4.S
IJ
IJ
IJ
IJI
OJt
1.42
•74
•Jt
•57
•4S
•51
^*
2.4
IS 25
19 IS
1.1 0
11 SO
IJ 40
1.4 0
IJ 30
•94 0
•74 IS
•42 21
•51 24
OJI 14
•42 4
•27 19
L 4
(M*) (MO *l
24 2.4
4.2 IS 25
45 19 IS
IJ 1.1 0
25 IS 10
2.4 11 0
2.51 1.04 1
•20 1.74 0
OJt 1.70 2
141 1.41 10
IJI Lit 17
OJI IJI 10
I.M OJI 22
•42 0.75 II
0.90 OJO 21
0.79 0.44 25
•41 0.39 11
0.41 OJ7 40
L
(M)
24
1.0
S.I
0.72
LSI
•94
IJI
OJ2
•SI
•24
) 'I
2.4
11 20
1.2 40
IJ4 11
•94 4
•44 14
•51 27
•44 41
4X11 44
t
(Ml)
24
MO
1.0
OJ
1.45
0.44
•51
1.24
•51
I.M
1.41
1.44
•M
0.71
0.41
0.47
«/
(mm) «i
24
2.1 20
2.9 10
2.2 0
2.05 I
1.94 0
I.4S 0
1.75 15
1.57 II
1.24 9
1.04 II
••2 24
0.42 24
0.49 21
O.M 17
O.M 47
*
105
90
Oft-90
2590
10-91
T«al 10J
44.5
41.4
SOJ
43.0
51.2
-------
20 O.G.
0.52 pin and 1.04 pm panicles with from 2% to 4.6% deposition on the averaie.
These results can be explained by the deposition caused by the inenial propenies of
panicles larger than I ?m and the deposition caused by diffusion*! propenies of the
panicles smaller than 0.5 pm. The data suggest that inhaled ultrafinc panicles (about
0.01 /«m) may exhibit greater than 20% deposition in the pulmonary region.
For the very small panicles the relative lobar deposition tends to be proportional to
the lung lobe weight which may be explained as being proportional to the relative
aeration of the lobes. However, pulmonary deposition of larger panicles appeared to
be greater in those lobes having shorter average path lengths from trachea to terminal
bronchioles. For example, the typical path from trachea to terminal bronchioles for
the right apical lobes of the hamster and rat is shorter than for other right lobes and
the relative concentration of 3.05 urn panicles deposited in these apical lobes was
also greater.
_ -Valuable imtructtooa aad edvtae ware provided by Mn Read! Lie Thome*
concemiaf tbe calibration aad operation of the five-port e^neure appantai (Tanus aad La.
1963) and concerning animal care aad rtiiertinni TT» author ate indebted for aaaunce in MaMfr
of morphometric data to Mr O. Mkhaet Sebum: for skiDed technical eerinince to Mr Robert
Yarwood. Miss Dotom Esnana and Mn Donna Duncan; for preparation of iUuttratiooj to Mr
Enenea Goff: for assiataaee in data aaslyea aad radiation eouatrng to Mr William GrOta aad
Mn Phyllis Petenoo aad for mamacript preparation to Mn Judith Millar. Tte work wes sopponed
by the National Institute of Environmental Health Sdencea via UJ. Eaav RaMich aad Develop*
neat AdniaatraUea contract E09-1M013 aad conducted ia snimal rare faoMtiei fuily acoedited
by tbe American Anociition '
HEFEUBNCES
FuaB,N.A.aod$tfTOO»«, A. O.(lMQ*N«r&*Mt (edUed byDAV«,
Houu. B. (1961) Afdaantf. Oik 17. 171-171.
. aad Uovr. M. E. (1972) **. ***. tmawm. 4X 1101
. aad Mo* O. R. (1971) flb*«aj* tt, 511-S3S.
WUEMOH. C 1. aad Bom, H. A. (1972) Bbk P*H. H, W7^4J.
Mncn. T. T^ Tbamv. M. L and NIWIOM. O. J. (1990)/. A***ScL l, 9-IS.
NtwTOH, O. J^ fjuae. O. O. YAawooo, R. U aad KAMAIUY. O. M. (197O ia Urn Prtiein
Y. H. Acadaaie Pieav H.Y.
4.. _
fjuaa, O. O. (1910) flrtefcffcit CkeMafawft (•*»* by KAMM. M. O. Jr. J*™» '
Gttjamr,J.IL)op. J2J-lUUJ.Aiott*EttafwC(XBmi«ioaDiv«oa<>CT«aaicaaiflf<»rma
Oak RkbaxTeaaawe, UJ^L
ILuat, O. On §0¥0, H. VL. ICAiuftAY. O. R. WbBMON, C J. aad Nfwidi, O. I. (197S) l»*
KAKA^W. a It aad l*wm
(edited by WALTON. W. H. ) vet i pp. J-17. Uevto
O.
Sieeaa, W. aad FuoaaAar. H. (19C9) finftm Sd. 4 TtdmtL J, 12IO-12M.
TASE Oaow ON LWHO OVWAMGI (19*1) tt* fhf. 12. 17MW.
THOMM, IL O. aad La. IL (19fS) ***** a* IWP*^*
aad
-------
Dtpoiition of Inhaled Monodiiptne Atroaob in Small Rodems 21
DISCUSSION
G. BOUUY: Did you tad any limiftnn' differences in retention between the mm at equal body
weights? Wt havt observed sexual differences ia toot microbe iafccUoas duf to differences ia
respiratory volume.
Or RAAM : The variability was very great, to there was no statistically significant difference between
the sexes. This may be because there were so few animals involved—five of each sex for each group
exposed to each panicle sue.
D. J. Flaw: How does the deposition in rats and hamsters compare?
Dr RAAM: Again, for the tea animals used for each partide sia, the data could not show any
statistical difference. Ia fact, they lead to agree quite well for the two species.
D. J. Fiam: The difference between 0-h and 20-h deposition seems to be small. Caa you comment
on this?
Dr RAAU: Our definition—an arbitrary one—for bronchial deposition is "those particles which
were removed from the luni during a 20-h period alter exposure.- By compariaa the fraction of the
total body burden (including the excreta for the 20-h animals) ia the lung both for the animals that
were sacrificed initially and for those that wen maintained for 20 h, we were abUw find the fraction of
the lung burden which to bronchial deposition; this is gvea to Table 1 Ttt teoadual dearaace was
normally small, but for the 2 and 3 *m panides it approaches «0% ia the two taeessa. la maay iaa>
vidual cases the difleftace was about 30% tor the larger particle* but it wes usually lees than 10%
for the 0.5 itm neirirlei.
W. T. Uuoa: Table 3 shows a sharp increase (4J-1?.9% for rats) in the deaositioa between 1.04
aadlO>^imdia.paBfciesiatl>eBieopharyB»(skuflXWhitcai«Bitl^?C^
Dr RAAH: We would expect the larger sixes to have greater inertkl depeeitioa due to fanpactioaia
the smell passages of the nose; this probably expiaias the higher values for the two larger SIMS of
-------
AMD AffUID TOXJCOtOQY It, J21-J2T (1992)
COMMENTARY
Recommendations for the Conduct of Acute Inhalation Limit Tests
Prepared by the Technical Committee of the Inhalation Specialty Section, Society of Toxicology
Reprinted with permission by the U.S. Environmental Protection Agency
Mai** Octobtr II. 1991; MOM* Oootar II. 1991
i Paper was drafted and approved by the Executive
Commentary—Recommendarioai for the Coodoct of Acute Committee of the ISS.
Inhalation Limit Tern (Prepered by the TeehoJeal CoawitMe Acute toxicitvunmtesaotlMaauDew limit for exBOwn
of the bhatadM Special* Seafat, Sod* of ToxMocr) nnnmtt^Vtm mOTn^ii" SriflMiSTSiin
(1992). A«4sM4»*< •**•*• II, 321-127. «wios^8»ad«^ionilaxp0iuraim required Guidebaes fbt
TWs peper reviews the sdantJae awes relaied to exposure the coodua of acuttiahalctioolunh tests havt beta pctparec
coacentratkMsaAdptttkksiatiMadBaaiieialulatiMllsak by a number of rcfulatocyaecocies, with those of the U^
testa. The correat Usted Scales Einriroaatattl PrBtecitoa Eaviromnantal Protection Aesacy(1916,19tt)betnt«moa|
A^ney(USEjPA)neomiiisiiilad V.pwwV,'T^!!Vri_^?? *• •»« comprshensivt. The USEPA |utdeUae related te
exporart coocentntion (U4. Environmental
. «uD.tt«thepufpooi
liiiui iiM to ill leiiiMMiiii iieii Ihioe on irnctatint aerosols at hiari
ThefukteluMi far acute inhalabofi toxicity tarn, as pro- u»ocan«tioni. an coMtdarad in this paper. Engag>*
itftlProieeojonAemey piawd o« exposure of rats, auea this is th« specm most
^^^ *• _. -» .. ••_
(US&AKcoer|Bdataa^itMflfcofMenaao««aMO> oftsa toed ia acute toxfcity
bmc/U«lnhalatio« Specialty Seeiioa(B$) of the Society _
ofT«ticolofy(SCT)omitepf«i>wyeMi,TteT€dmic^ CXP08UU CONCTNTIUTION
Committee of tht ISS was. therefore, eharaad with nvkwiof
of S mi/lhaT oinaatiy ' recpm-
databate oo this issue, and with de- Iht Itanit
bcoctnbuooabotbtheCc^mitmn^^ aoyoiaiPcot
tad other asetnbm of thaSpadatey Section, a o»sei»is him ooewawtioos ait rarety achieved in reaWift occupa-
-------
322
COMMENTARY
uonal or ambient environmental exposures. Furthermore,
exposure to dusts at levels near this concentration has been
reported to cause physical blockage of the airways and out-
right suffocation (Scott ft */.. 1982). Nevertheless, such high
concentrations achieved in laboratory tests do serve to max-
imtze exposure, and numerous data are available from ex-
posures conducted at S mi/liter. As it is often possible to
attain this concentration, it may represent a reasonable upper
limit Tor an acute toxicity test It is. however, imperative to
recognize that problems may exist with the generation of
some aerosols which can make achievement of this concea-
tration impossible. If so. the highest practically achievable
itration must represent the limit for that specific agent.
It must also be recognized that due to aerosol coagulation,
the median particle size increases as total aerosol concentra-
tion increases. Thus, de facto imposition of particle size cri-
teria for limit tests imposes upper limits on achievable ex-
posure concentrations while maiaraiflipg that particle size.
PARTICLE SOI
Deposition Patum and Motion toAcuu Toxieay
The deposition of partides having a polydisptm distri-
bution is usually best predicted by the MMAD (Mono* it
aL. 1966X50% of the nu*s of U^ partide distribution is less
than the MMAD, aad 50% of the mass is greater than the
MMAD.
Both the total respiratory net and the regional deposited
dependant i
exposures to materials of unknown toxicity, h is most a
propriate to use a panicle size distribution that win result in
significant deposition within all regions of the respiratory
net Such an approach would maximize the potential for
producing and detecting toxic effects (see Appendix C). In
this regard, an important distinction between acute and
chronic testing should be noted. Chronic respiratory tract
toxicity often results from the accumulation of insoluble
panicles within the pulmonary region. The use of partide
sizes to maximize deposition in this region may be desirable
for assessing chronic effects (Lewis «r al.. 1989) but for the
reasons cited above, may not be ideal for acute testing because
the use of small partide sizes to maximize pulmonary region
deposition minimizes nasal deposition, enhancing the pos-
sibility of failing to detect potential nasal toxicity.
Differences between laboratory tests in rats and actual hu-
man exposures should be recognized when attempting to use
acute inhalation data in aliening hazards. Both particle size
distribution iJa the real world compared to that used in the
laboratory and rat-human deposition differences need to be
i deposited in the i _ ^_
dual, or pulmonary (alveolar) airways may be absorbed di-
rectty into the Mood stream, and/or translocated to the gas-
trointestinal net where absorption nay occur. Thus, the
systemic effects of soluble agents (that ait not locally dam-
afing) would be expected to depend on the total amount
deposited aad to be more independent of partide sot than
afeno which cause effects at the initial deposition site. In
contrast, deposition site is of primary imponinm fee i
which act locally, si
faridt Siu md Deposition fauna
Particles of interest in inhalation toxicology generally
penetrate beyond the upper respiratory net into thoracic
airways (Phalcna 4.1988) and, thas> may deposit in the
nchaohmnrhial and pulmonary regions of the lungs (Mor-
row « 4.1966). Whan inhaled through the nose, the na-
lopharyngeal deposition of such partides is also significant
(Cham « ei. 1990). For humans, partides op to 10
MMADnwyexhibittfaortckdepoaitioawith
aad particles up to IS »m MMAD with mouth breathing
(U&&PJL. 1912; Jarabek it aL 1919; Phalen et oL. 1988).
Species differences exist in regional partide deposition
toxjc €o CB& jmopcmyttgjBgM* i
regions (Dahi« aL. 1988; Feroo and Boated. 1989). Local
effects and/or lethality can result from injury to any of these
regions (Amdur. 1958; Morris and Smith. 1982; Rottahettt
ctdL. 1963).
Because of the complexity of the respiratory net and its
responses to inhaled materials, h is impossible to predict a
priori whether or not greater toxicity would be expected from
small or large particles ofa given material Exposure to targcr
size particles of some agents produces greater toxicity than
exposure to smaller size parbdes of the same material
whereas for other agents the opposite is observed (Amdur.
1958; Salem and CuUumbina, l961;Tsuda«at. 1986; Wolff
naL. 1979; Oberdorster et aL. 1990). Therefore, for .acute
position man do humans (Raaba * aL 1988; SchlesingeT.
1985). Deposition in the nchanhffmrhial and pulmonary
regions also (Us off mom rapidly with partide v» in rats
compared to humans. Unfortunately, the rodent data are
not sufficiently complete to altar precise estimation of an
upper size cutoff for partides which deposit in the thoracic
The data presented in Fig. I. at wefl at those from other
sources (Maudedy «oL 1987; Raabe et aL. 1988X indicate
that inhalsdparidssbstween I and 4 mm MMAD wffl de-
posh wttluaattrepOM of the rsi respiratory tract WUhin
this tJrt range, aasopharyngeal and nchsobronchial de-
position increase as partide aaa increase!, bat pulmonary
deposition remaimrdativery constant Based merely on pul-
nonary region iftfTti'trn efficiencies. I i*m partides offer
00 (K*iiu* nttummtmm* MM* A ««• narrirlM TIltM. **•"•"«• I
4 urn panadas wiB tiksty deposit in aft* regions of theresot-
-------
COMMENTS
OJ
1
•AT
A
— O
1 °
1 1 1 1 1 1 1 1 1
/I
MAMTIN / /
A anuTMOiucK- / /
• ntACMOMONOMM. A. /—
• »UUMNAJIV /
y «
/
/
OJ 04 04 04 OJ
PMVSWAl OUatfTM. |M
(TortX (FM USIPA. IM2)
ntocy tract, this sue naae is hfetty desirable for acute limit
Najopharynteal depoaitiofl in: nun substantially in ro-
dents for larat ptrtidea, aad is the dominant dcposinoo tin
(br ptrtida Uryer than 4 urn. Btcunt aom limit ton «t
desicMd to provide ooty aaappradaaft tadn of toxitity,
aad bocauM aaal ofloos cu bt of<
The dofrot of polydisptmty is not a major concern, be-
cattM ptnidt diuributiooi haviat nomotric standard de-
viatioaa bamnn 1.5 and 2,5 will provide widespread dt-
poation tfareuiheut the nspintocy tract
i used fcr the limit ten require
panide soe eutoA oead oot be ao sthnpat at ne-
ommeaded (at duonk inhalatioo Mxkiiy itudiev
The data of Riabt * aL (19M) are wtaatty an that ait
available oa the depoeiboo oflatfar peridot ia the rodem
feipintocy tact Tfcty aa«aic that, far 6 «un MMAD par-
tickt, pulaoMiy dapoaitioei a arnaU, while for !0*ffl
MMAO oanidaa, k • oafUpWe. Modatiot by Schom aad
Yeh (1980) aad by Jarabek * aL (1919) ladkaw that tn-
cfaMbtoacaiai aad pvlaomvy depoaitioB decnam rapidly
Theh*hi
extttaaty larp numbsn of parocks per unit voiume of air.
Under tbi
aad results in
19I2X Coojequendy, there are physical
particle sne whfle maintatiiina hisjh exposurt
is provided ia Appendix A; experimental data demoostratini
tht cxtfleocB of coafutoboo ia iahaiaaoa chambers are provided
ia Appeodtx B. At 5 mt/Uiar, the minimum theoretical attain-
able an fcr onitdctuity partidei ii I »m (Appendix A), bated
above S «un MMAD.
aate oa the aawmpaoa that parodes art infinitely small when oof-
of 4 IUB MMAD is aa appropriate upper cutoff for acute
iohalatioa toxkity temasj; at this sue there ia substantial
pulmonary depoeidoa. while eaaal depoaitioa has not be-
come tht complctety dooiaaat flsetor. Furthermore, than
is cJfBiftcaat depoaitioa ia all nsjioaa of the rat respiratory
• Iheorcticaity impoaibio to aeaerate an
a concentration of 5
o*/Via*aadaparo«kwleBithaalitfnMMAD.
la practice, it hat not beea possible to achieve the theo-
rcoeal limit withia a metor of two (i*. 23% of panicle mass
<1 nav «fecommended by tat U.SE.PA.) at concentra-
-------
324
COMMtKTAHY
tions of 5 ma/liter (Appendix B). Vmuilly all aerosol ten-
eratioo methods, and particularly dry powder methodologies
for generating solid particles, use finite size pamdes as nan-
mi matenals fGrassd. 1976: Hinds, 1979; McCldlan and
Henderson. 1989: Phalen, 1984). U is often very difficult to
reduce the size of starting materials below 2-3 urn because
of the kinetic energy needed for deagglomention (Cheng a
al.. 1985). Generation of aerosols of materials designed to
be "dust-free" or "sticky" is also extremely difficult
The production of liquid aerosols also has physical limits.
Finite droplets are initially generated, and then dried to
evaporate the solvent The minimum particle size is achieved
when all solvent has evaporated and only solute remains.
Because the evaporated solvent also enters the exposure
chamber, potential toxic effects of the solvent can signifi-
cantly confound the results.
The above problems increase as attempts are made to re-
duce panicle size. As stated previously, a parade sue of I
urn offers no distinct advantage over 4 urn particles for acute
limit tests. Thus, the expenditure of considerable effort to
approach this theoretical minimum of I *m & without sig-
nificant toxicologjc value.
Parade
CONCLUSIONS
itration and soe are
termuung the depoeited dose and the uttu
resulting from inhalation exposure. A penide cooeeatniion
of 5 tag/liter can be attained in exposure chambers, albeit
TABLE 1
Coeccaffatioas tad Particle Sizes Achieved
When Using a Cyclone*
CoaceatmiM
(ma/Uur)
I.I
1.2
2.S
MMAD
(*•)
0.4
0.6
2.9
% < 1 Mffl
61
62
16
with difficulty. Exposure to such a concentration may. how.
ever, be of limited value due to: (1) lack of lexicological or
real-world exposure relevance, and (2) the possibility of air*
way obstruction and suffocation. As it may be impossible to
predict the most sensitive respiratory tract target or the most
haraJul particle sat oft particular agent, acute bait teats
should use a particle sne distribution that results ia dcpo»
shioo throughout the entire rodent respiratory tract Particle
distributions having MMADa betweea 1 aad 4 ma should,
therefore, be cocjidered a* acceptable for the needs of acute
inHltVt "*** *••*?.*
APPENDIX A
TheomktJ Miaimum Partiek Size Cafcmlations
The ihinmiral minimum particle ate wffl occur ia co-
withnttximumpartide concentration. Tae anal*
ysa of Miidi(19t2) stows that tlieoitticBl maiimum pattick
iare0.9-Ux M>» pattides/cm1 for typical
jof4-5min.(Ftg.2).The
i partide ate ten he caJculated from the equation
below far spherical parades of unit density. The mass in
individual particles is multiplied by the partide coocentra-
aad the equation
Massc>in 3 and 4 show that it is rarely possible to generate
-------
COMMENTARY
o
<
FIG. 3. Outnbuuon of numbtf of aeun inhUmoa unueitv ttuAm n fune* fat (MMAP ma) fee
(nwu). Espown eeacntnoooi »tft ia ibt 0.5-5 rni/Unr nap. Oau frtxa 244 acuw ""^Utiffi rin
-------
326
COMMENTARY
I
7
6
S
Z
2
Eieesare
i far
••and 10 prod1-
tide mass associated with parades lev than 1 *m, and the
highest concentration at which this occurred was 1.4
mg/liter.
Coagulation effects arc also evident from Pig. 5. which
shows data from studies in which multiple concentrations
were generated for the same compouada. Because the gen-
eration method was constant for each compound, initial
particle tifft IMCTT fi^iltn the initial BOTibtr CTtictHtfaiioni
were greater at the higher mass concentrations. This figure
dearly shows the increase in particle ate that ocean with
tncressug parudc cot
feature at these high i
The argument might be made that met
vices, such as cydooes, canoe used to reduce particle size.
Although this is achievable, exposure concentrations will be
reduced (Table 1 and Appendix Q. However, the aerosol
would still be subject to the physical laws of coagulation.
Table I snows that it is possible to produce aerosols with
MMAD's < I JOB at concentration! of approximately I mg/
liter. At a concentration of 2.5 mg/liter the particle size in-
creases to* 2.9 MA MMAD. Thus, even when using some of
the best available aerosol technology, it was hnpossibk to
approach the theoretical minimum of 1 urn MMAD at a
concentration of 5 mg/liter.
APPENDIX C
Effects of Cydone Use in Inhalation Study
Given below is an example in which the use of a cydom
reduced the panide size at the expense of also reducing the
exposure concentration.
With Without
cydonc cyclone
% of parade mass < 1 urn
Chamber concentration (mg/liter)
MMAD Gun)
GSD
30.0
0.54
1.5
2.0
9.7
1.7
3.0
1.7
Calculations of the mesa concentration of particles less
than 1 »a ia diameter fix both conditions are
OJO X 0.54 mf/titer » 0.16 mg/liter
0.097 x 1.7 mg/liter • 0. 17 mg/liter.
without cydonc:
Despite the smaller fraction of submkrometer particles when
a cydonc was not used, the mass concentrations of tubmi-
crometer partides were almost identical under both coo/
tions. Interestingly, than was greater mortality in aninu-.
exposed to an aerosol produced without a cyclone, for which
-------
COMMENTARY
32
t higher dose of tea material presumably would have been
Deposited in the upper respiratory tract These data under-
score the fact that larfe parades can contribute significantly
to acute lethality (Tiuda et al.. 1986)
ACKNOWLEDGMENTS
Memben of the Technical Committal wtrt On. G. L. Kennedy. I B.
Moms. M. V. Rololf. H. Salem (Chair). C. E. Ulnch. R. Vakaunc. and
R K. Wolff. The assistance of On. R. Hendenon. J Mauderly. and R 8
SchJesinter as well ai the comments provided by *aneu« metnben of the
Specialty Section, at- ejsufuily ackaowkdotd.
REFERENCES
Atndur. M 0. (1951). The
ratory
i of guinea pip 10 sulfunc
aod mm. AMA Aid*, lad HtaUh It. 407-414.
Cheat Y. 1. MwihalL T. C.. Headenon. R. f. and Newton. C. J. (1913)
Utt of • j« mill for diapen&f dry powder far mhaUuoa stud**. Am.
M*K>f.Xaoe. 44,449-454,
Cat*, Y. i, HUM. a K- So. Y. P, Yak, H. C, aid Morpa. K. T.
(1990). Depeaiooe, of uhrrtot smote iani ami moid*. Tonerf. <4**V.
Mamon*. 104 222-233.
Data. A. R~ Bood. J. A, PeuiOou rurhar. J.. Sebouiia. f. J.. and Whiter.
t J. (1911). Efccta M the mpimory ma, «f iofcated material, front
<4*e* P*om*nL 93,4|4-»92.
Fcroa. V. J.. and Boatood, M. C (Ed*.). (1919). A'asof Cvnnorowru M
CrttMi. E, E. (I97«. Awoaol ftaemiM
u«ni$u. Lewi:
Pubbstan. Chctsee, ML
Raata. O. O, AtBtyaft. M. A, Teafue, 1V, aad Itatok. A. (19UX RJt«oa
i of iahated iiuiuilupiw ceane aa^ ii
171-lli. Aoa After Sciaaei Pub, Aaa After.
Hiads. W. C (19121, Awotei fecfewfep. pa. 233-239. Witty. New York.
Janbtk. A. M, Meatcst, M. O, Ownoo, J. It, Oovnoo. M. U aad
Mifler. F. I (1919). lafcatooa reJertoadoac Ai appfacooa of uMsrspeaai
doamctry medtUiai for risk aoMancu of iatolubst pamcks. Htatth
«* SKSapat I), 177-113.
Lewd. T. IL Morrow. P. P. Miriafm R. O. Raobt, O. C. Ktootdy.
0> U Sd>^> ^ °°^L T J" "y**J "•' ^ Gubn> ^ *
stttdiA roacei 4«pl ^mMfo/ **. 377.311
Mtuderty. J. L. Kot. a E. Carpaaser. R. U. Kaoaoa. R. L. Henderson.
R. P, /ones. R. It, McOeoja. IL O. tad Wo« R. K. (1917). Effects of
4Mwekiahalit»oa«pe««rtofftiiiooyslukduaa«ddieaelejil«u«Lla
Oca*
Roataaoftt. M. J, Canoe, T. R. Weeks, M. H, WiUoaki. P, Port. O. P.
aad Otam. P. W. (1963). A toaico pathoteok study ia aaiauto after bnr
itohydrooM fluonde. Am. In* H*. Aaee. / U.233-
261.
Sakm. H. aad CuUuwbine. H. (1961)
poUuttOO. Ank. Cnvtfen. Httttk 1641-647
r. R. B. (IMS).
ofiohakd
it
landhttsftaaxAftview./. Taaeat. Emm «•((>
II 197.214.
SduuB, M. aad Yen. H. COMOX
fiat 42,1-IS.
Scott. J. IX MaeMkilL 1 M^ lalia. P. 1, Scan B. 0.. and Z'
evaluaaoa of aerosol d»
toot urwtyi Bull Matk
dust
39-60.
of ponkst an* oo iohatoboai touchy e/cUorfoviapooi./pn./. Ktt So.
UJ Co •uniiiaiiiirTnianinn ipng tlTITl rftnK-r~^— «-'-••*-
Oudea aad Panculam CPA40IVt42429e, 11-2B-11-32.
U A Cunronmennl Proascaoa Aotocy. (I9M). Ponode iiaeiimentiuide-
P. htard avuwtiea; Hamaa aad
(RevtaedX- NTS Repon PM»>lOt9SI. Waribaejoa. DC
Ui CiminaJMhiiirmiarrtni fijnnr (ITll) Hiardmtuatioadivisioa.
c^^t^rf ^^IOMW^ •~~4.^. .«K«IMW>. UMiatv M«IM- NTTS r
PH9.I00366, Wsotaatioo, DC
Wolt R. IL. SilbausA. 1 A, Brawnsuia. O. 0.. Carpmter. R. L. and
Maudarty. J. L (I979X Toaseicy of 0.4 aad O.I »a wlfune aod acroteb
ia tta two* p«. / Tuaeoi. £/mron. Htati* 5.1037-1047
-------
INHALED PARTICLES IV
Edindby
W. N. WALTON
PERGAMON PRESS OXFORD AND NEW YORK 1977
-------
Attachment 4
NATIONAL AGRICULTURAL CHEMICALS ASSOCIATION
COMMENTS ON THE USE OF MAXIMUM TOLERATED DOSE
FOR CARCINOGENICITY BIOASSAY
The controversy over the use of the Maximum Tolerated Dose (MTD). in carcinogenicity
bioassays revolves around a desire for maximizing sensitivity of the bioassay on one hand
and minimizing the identification of carcinogens which are not relevant human health
hazards on the other hand. Hence, we believe the National Academy of Sciences
Committee on Risk Assessment Methodology (CRAM) should address the policy and
philosophical questions of the appropriate degree of sensitivity and the relevance of the
results obtained at the MTD.
WHAT IS AN MTD?
Some of the MTD controversy has been generated by inconsistent definitions or
interpretations. -In fact", some scientists and regulators prefer the concept of "Minimally
Toxic Dose" rather than "Maximum Tolerated Dose." In 1984 the International Life
Sciences Institute listed 27 different published recommendations for dose selection in
carcinogenicity bioassays which differed relative to the definition of the highest dose to be
tested"1. Similarly there is a clear lack of harmony between the European regulatory
community (OECD/EEC) and the U.S. EPA Office of Pesticide Programs concerning the
1
-------
criteria for the highest dose level. The OECD/EEC guideline for carcinogenicity studies
states:
The highest dose level should be sufficiently high to elicit signs of minimal
toxicitv without substantially altering the normal life span due to effects other
than tumors. Signs of toxicity are those that may be indicated by alterations
in certain serum enzyme levels or slight depression of body weight gain
fless than 10 percent).*" (Emphasis added)
In contrast, the 1987 EPA position document on MTD from the Office of Pesticide
Programs states:
The highest dose to be tested in the oncogenicity study should be selected
below a level which resulted in significant life-threatening toxicity in the
subchronic study. The level should not be selected too far below a life
threatening level because the highest dose tested in the oncogenicity study
should elicit significant toxicity without substantially altering the normal life-
span of the test species from effects other than tumor formation.ffl (Emphasis
added)
There is also a Jack of consistency between U.S. offices and agencies. The Office of
Science and Technology Policy (OSTP) in a 1985 review of chemical carcinogen science,
listed several basic principles for tests of cancer induction. Relative to dose-level selection
they stated:
-------
It is appropriate to use test doses that generally exceed human exposure
levels in order to overcome the inherent insensitivity of the traditional design
of the long-term animal test. The highest dose should be selected after an
adequate prechronic study and after evaluating other relevant information,
as necessary, to determine the highest dose consistent with predicted
minimal target organ toxicity and normal lifespan, except as a
consequence of the possible induction of cancer.™ (Emphasis added)
These may appear to be only subtle differences in guidance for dose selection. However,
the consequences of selecting maximum sublethal doses vs minimally toxic doses are
very significant not only to the results and interpretation of the study, but also to the
expenditure of regulatory resources, the development of new technology, the relevance
of the data to human health hazard and public perception of the regulatory process.
These inconsistent views on the definition and interpretation of the MTD have led to
rejection of studies by EPA which were determined to be acceptable by European
regulatory agencies. European regulators are concerned about the difficulty of
interpretation and the relevance of testing at excessively high doses'51 while EPA rejects
some European" studies because they have failed to attain an MTD. These situations have
led some companies to suggest that two different bioassays be conducted with the same
material- one for submission to European regulatory authorities, conducted at a
"minimally" toxic dose and another to meet EPA's MTD philosophy. This is an
unnecessary waste of resources and animals. These fundamental differences clearly
point to the need for a consistent single definition of the highest dose to be used in
3
-------
carcinogenicity bioassays. That definition should be uniform in wording and in
interpretation internationally.
Equally important is the practical problem of estimating the MTD. Historically, in the
absence of other significant toxicity, a decrement of 10% in body weight in a subchronic
study has been the most commonly used endpoint for estimating the chronic MTD. In
1987, the EPA suggested that a 10 -15% body weight decrement was necessary.131 The
EPA has pointed out that a more thorough assessment of various parameters, such as
clinical chemistries, organ weight changes, hematologic effects and histopathology should
be used in conjunction with decrement in body weight gain in selection of the MTD. We
agree, however in practice the Agency uses these endpoints only when effects are quite
severe. We further believe that pharmacokinetics and metabolism data should be used
to select the MTD. Thus, decrement in body weight is only one of many parameters
which can be used in MTD selection. Toxicologically significant changes in hematology,
clinical chemistry, organ weight, histopathology, metabolism and pharmacokinetics should
all be considered in estimating the MTD.
WHY USE A MINIMALLY TOXIC DOSE ?
The omphasis in the phrase "maximum tolerated dose" should be placed on the word
"tolerated". Dose levels which produce physiologically or metabolically compromised
animals are doses which are, in essence, not tolerated, and thus are much less relevant
to hazard identification. In addition, doses which exceed pharmacokinetic saturation
levels, are not relevant for hazard identification or dose-response extrapolation to the
much lower dose levels at which humans may be exposed. Such levels can only be
4
-------
justified when human exposure is similarly high. Subchronic or chronic exposure of
humans to pesticides never attains such levels. The criteria for MTD should focus on
identifying the maximum dose level relevant to assessment of risks at human exposure
levels. In practice this means identifying the maximum level at which the test animal's
general health is not greatly compromised by exposure to the test substance, and one
which is in the linear pharmacokinetic range. This has been emphasized in a report of a
joint meeting of FAO/WHO experts on pesticide residues.
// is valid to extrapolate animal data to man only if the biotransformation
pathways of the chemical are identical or very similar between species and
if the doses do not exceed the capacity of the pathways being compared.®
SENSITIVITY
The Committee needs to address the question of the appropriate degree of sensitivity for
the carcinogenicity bioassay because this is the primary justification for testing at a
maximum tolerated dosage. If maximum sensitivity is not needed, then testing at some
multiple of maximum human exposure or some fraction of the MTD would be sufficient.
The philosophical difference is between "blindly" using a rodent MTD, which is believed
by some to be useful for identifying human carcinogens, and carefully selecting a highest
dose that is significantly above human exposure levels and is likely to provide dose-
response information that is relevant to human risk at human exposure levels. Even if
assays conducted at a minimally toxic high dose fail to detect a carcinogenic response,
an upper bound on carcinogenic response could be estimated if necessary.
-------
If the purpose of the carcinogenicity bioassay is to demonstrate the carcinogenic potential
of a chemical under any set of circumstances, without regard to human relevance, and
to create lists of potentially carcinogenic substances, then the use of the MTD is
appropriate. This apparently is the philosophy of the National Toxicology Program (NTP)-
- to make sure they don't "miss" anything. However, if the purpose of the bioassays is to
identify a chemical likely to be a carcinogenic hazard for humans, and to provide relevant
dose response information, the use of the MTD may be misleading and wasteful of public
resources. For example, over 1000 chemicals have been tested for carcinogenicity.
Several analyses suggest that 50-60% of these chemicals are animal carcinogens.(8l78) A
small number of these animal carcinogens are known to be human carcinogens because
reliable human data are available. Conversely, a small number of these animal
carcinogens are considered unlikely to be human carcinogens because of their mode of
action or conditions of exposure. However, the vast majority of animal carcinogens have
unknown human carcinogenic activity and are considered to be at least suspect human
carcinogens. A positive response rate for the carcinogenicity bioassay at approximately
50% raises the legitimate question whether the sensitivity is too high. Human cancer
statistics certainly do not reflect such high rates of chemically induced carcinogenesis.
We believe this across-the-board testing in rodents at excessive dose levels typified by
MTOs as currentl used in the U.S. is-
diverting scarce resources from significant human health problems and
devoting them to research and protracted controversies on carcinogenic
risks which may be trivial.
Needlessly alarming the public.
6
-------
contributing to public apathy by identifying too many carcinogens and failing
to distinguish serious from trivial human health risks-- "if everything is
dangerous then nothing is dangerous".
Clearly, it would be preferable to have a carcinogenicity bioassay give a positive response
only to significant human carcinogenic hazards.
A 1985 review of National Toxicology Program (NTP) studies revealed that 2/3 of
chemicals judged positive were considered carcinogenic based on effects observed solely
at the highest dose (MTD) level.* Thus about 2/3 of the chemicals considered positive
would have been considered negative if the maximum dose tested would have been 1/2
MTD. We believe many results at the MTD are "false positives" and/or not relevant to
human hazard. Our view, which is shared by many others in the scientific community,
is that it makes sense to alter the carcinogenicity bioassay by testing at doses lower
than an MTD (as currently defined in the U.S.) when justified on the basis of human
exposure considerations. This would allow greater focus on more significant human
health issues. Apostolou (1990) has presented data which suggests that most of the
known human carcinogens can be detected in animals a* dosage levels well below the
MTD."01 There is therefore some reason to believe that a reduction in the highest doses
tested would be sufficient to detect important human health hazards and at the same time
not detect trivial carcinogenic hazards.
-------
Since the MTD is chosen empirically, controversy over its definition and interpretation is
difficult to resolve. However, no matter how the MTD is defined, when a carcinogenic
response occurs in rodents at levels of significant toxicity, the relevance of that response
to human risk assessment is very questionable. This is especially true when a
carcinogenic response is not detected at dose levels significantly below tne MTD. When
a carcinogenic response occurs only at toxic doses, the finding should be considered
irrelevant to human hazard identification and risk assessment. Accordingly, any definition
of MTD must include an interpretive statement which renders such a finding irrelevant to
human risk.
8
-------
Attachment 5
DEVELOPMENTAL HISTORICAL CONTROL INFORMATION
-------
DEVELOPMENTAL HISTORICAL CONTROL. INFORMATION REQUESTED BY NACA
83-3 fal - Rat
Historical controls needed for the incidence of supernumerary ribs
Historical controls needed on fetal anomalies - unossified hyoid, reduced thoracic centra
Historical controls on ecchymosis
Historical controls needed for renal pelvic cavitation and mean fetal weights
Historical controls needed on non-ossified metacarpal #5, dam's liver weights, anophthalmia, microphthalmia,
bent hind limb, diaphramtic hernia
Historical controls needed on situs inversus , .
Historical controls needed on skeletal anomalies - delayed ossification and/or abnormal bones, missing
sternebrae, xiphisternum absent or single vertebral centra, and absent vertebral arches
Information on the historical controls lists studies numbered 6-10 in the report, but where are 1-5?
Historical controls needed on skeletal and visceral anomalies
Historical controls needed on unossified sternebrae # 5 and 6
Historical controls needed on dumbbell-shaped thoracic vertebrae _,,,,., . ^ .
Historical controls needed on all fetal anomalies (including testicular and skeletal variants)
Historical controls needed on skeletal parameters with maternal and fetal statistics
Historical controls needed on skeletal broken down into individual bones
83-3 (Bl - Rabbit
Historical controls needed on the incidence of hydrocephaly and lung agenesis
STSiLt. when data was collected and from what laboratory
Historical controls needed on SSiiJSd'SJh'.SS^^ and gall bladder agenesis
Historical SSSS" a^set needed due to high mortality rate probably due to inappropriate dose selection
Historical controls needed on delayed ossification and rudimentary sternum bones
Historical controls needed on the litter incidence for various parameters
Historical controls needed on resorptions, runting, live young ratio
Historical controls needed on fetal resorptions and fetal body weights
Historical controls needed on fetuses and litters with lumbar and sacral vertebral abnormalities
Historical controls needed on all skeletal abnormalities of Table VI of the study report
* Some of the effects occurred more than once
-------
Attachment 6
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHMOTON.O.C. »*0
0 5 JUL1891
tUMTANCCt
MEMORANDUM
SUBJECT: Suggestions for Analytical Determinations of
Dosing Solutions for Mutagenicity Testing
FROM: Kerry L. Dearfield, Ph.D
Geneticist
Science Support and Special Review''Section
Science Analysis and Coordination Branch
Health Effects Division (H7509C)
TO: Marcia van Gemert, Ph.D.
Chief
Toxicology Branch ZZ
Health Effects Division (H7509C)
THRU: Reto Engler, Ph.D.
Chief
Science Analysis and Coerdinatioi
Health Effects Division (H7509C)
A meeting was convened on June 13, 1991 to discuss appropriate
measures to be taken concerning the analytical determinations of
dosing solutions for mutagenicity testing. .Kerry Dearfield, John
Chen, Irving Mauer and Marcia van Gemert of RID met with
representatives of the Office of Compliance Monitoring to discuss
this issue.
Zt appears that the current GLP Standards (Part 160 of 40 CFR)
were originally written for long term bioassays and do not
specifically address short term mutagenicity tests. A concern that
has arisen-it when to perform analytical measurements of dosing
solution* W"^- short term tests. The concern arose when contract
laboratory'pasjagement queried the OPP about when such analysis is
required. Over the years, some mutagenioity studies lacking such
analyses were being judged unacceptable by OPP reviewers whereas
other mutagenicity studies were not declared unacceptable.
Contractors (and registrants) wanted clarification of the OPP
policy when to perform analytical determinations of dosing
solutions.
The group discussed this problem and determined that OPP
genetic toxicologists should provide their ideas to be incorporated
into a position to be sent to the Office of Compliance Monitoring
for final discussion. This memo contains suggestions for dealing
-------
with this issue.
The major issue deals with the Good Laboratory Practice
Standards (Part 160, 40 CFR) Subpart F, Test and Control
Substances. Section 160.105 deals with test and control substance
characterization. it was agreed in the above mentioned meeting
that this information if required for all tests, not just
mutagenicity testing. The identity; strength, purity, and
composition are important elements in the characterization of test
articles. The stability and storage life of the chemicals are also
important, as is the stability of both test and control chemicals
under the actual assay conditions.
The major problem of interpretation involves Section 160.113
paragraph (a). It requires that for mixtures (when a test or
control substance is mixed with a carrier), analytical testing
shall be done to determine the uniformity of the mixture and,
periodically, the concentration of the mixture. This paragraph
also requires that the stability of the mixture be determined. How
this relates to short term mutagenicity testing is the expressed
concern(s) of the contractors and registrants. In most instances,
mixtures or solutions of test and control substances for
mutagenicity testing are prepared only once and employed only once.
Also, several stock solutions are made up, usually from a top
concentration or stock solution. Should all dilutions from a top
concentration solution be analyzed? What are the differences
between in vitro tests and in vivo tests in regards to dosing
solutions? These were questions needing resolution.
In vitro testing
For in vitro testing, dosing mixtures or solutions are usually
prepared only once, then applied (usually only once) to the test
organism (e.g. Salmonella or other bacteria, or mammalian cells in
culture). Several concentrations are also prepared, usually via
dilutions from the top concentration or from a stock solution. If
a registrant wants to perform analytical analysis for every
experiment, it is recommended that analysis be performed on the top
concentration solution. If the test and control compounds are
soluble in aqueous solution, then the analysis would be performed
with the carrier that is the same as the aqueous medium the test
organisms, are usually grown in. However, in many instances, the
test and control substances are dissolved in an alternative carrier
(e.g. DMSO or acetone) and are then applied to the aqueous
environment that the test organisms are grown in. In this case,
the analytical analysis should be performed on the top dosing
mixture found in the test situation.
If analytical analysis is not performed with every experiment,
there are some biological signs that can be used to satisfy the
concern about any interaction of the test and control substances
with the test organisms. If a positive mutagenic response is
observed, it will be presumed that the response is due to the test
or control substance. Analytical analysis will not be necessary
-------
(unless there may be reasons to suggest a possible contaminant (s)
is responsible for tha effect). If tha mutagenic response it
negative, avidanca of cytotoxicity may ba uaad to imply tha
interaction of compound and taat organism . Thaaa considerations
point to tha importance of performing a preliminary dose-ranging
experiment to daf ina tha cytotoxicity curve for each taat aubatance
and tha ranga of concentrations to use in tha actual mutagenicity
assay. Information from this preliminary experiment may provide an
indication of a positive response in tha main assay (a.g. a
toxicity experiment with TA100 for tha Salmonella assay may also
reveal an increase in ravartants) . Also, tha axistanca of a
cytotoxicity curve (or other toxicity signs such as call cycle
delay in a cytoganatics aaaay) may ba demonstrated in the
preliminary dose-ranging experiment. If these parameters are
evident, than analytical analysis in tha subsequent definitive
mutaganicity taat may not ba necessary (tha concurrent cytotoxicity
still needs to ba determined with tha Bain assay) . However, if
there is neither a positive indication nor cytotoxicity aaan at
concentrations up to tha dosa limits, than a decision to perform
analytical analysis with tha top concantration aolution from the
definitive mutaganicity taat should ba made. Without tha analysis
in tha absence of biological parameters (a.g. positive response or
signs of cytotoxicity) , tha study may ba judged unacceptable.
In vivo testin
vivo assays may require a slightly different approach to
analytical maaaurements of dosing solutions. Whole animal taat ing
can involva exposure to test and control substances only onca (and
therefore dosing solutions are prepared only onca) or aavaral
times, parhapa over several days or weeks (a.g. as with subchronic
exposure in a dominant lethal assay) . Zf a mixture or solution is
prepared fresh before a singla axpoaura to tast animals, than
analytical analysis need ba performed only onca with tha top dosing
solution. Such analysis should ba performed on tha tast and
control substances in tha carrier that will ba used as a vehicle
for animal exposure. Zf a subchronic exposure regime is employed
with samples prepared frash for each exposure, than analysis needs
to be performed only onca, at tha first dosing time if the
stability of tha tast substance remains constant over tha antira
exposure pariod. Zf tha same tast substance sample ia to ba used
during an entire subchronic exposure pariod (i.a. not praparad
frash before) aach exposure), than analysis will naad to ba
performed periodically, as given in tha CLP Standards for long-term
assays.
As with In vitro tests, biological signs can ba used with In
vivo tests to determine any interaction of tha taat and control
substances with tha tast organisms. This information can ba used
to dacida whether analytical analysis is necessary. Zf a positive
mutaganic response is observed, it will ba presumed that the
response is dua to tha tast or control substance. Analytical
analyais will not ba necessary (unless there may ba reasons to
suggest a possible contaminant (s) is responsible for tha affect) .
-------
If there is not a positive mutagenic response, than evidence of
toxicity (e.g. clinical signs and/or cytotoxicity at the target
tissue or organ (e.g. change in ratio of PCEs to NCEs in a
micronucleus assay)) nay be used to imply the interaction of
compound and tast animal. it is as important h«r« that a
preliminary dose-ranging experiment be performed to define the dose
range for each test substance. Information from this preliminary
experiment may provide an indication of .test substance interaction
with the test organism. If these parameters are evident, then
analytical analysis in the subsequent definitive mutagenicity test
may not be necessary (the concurrent cytotoxicity still needs to be
determined with the main assay). However, if there is no
biological indicator of test substance interaction at doses up to
the dose limits, then a decision to perform analytical analysis
with the top dosing solution from the definitive mutagenicity test
should be made. Without the analysis in the absence of biological
parameters (e.g. positive response or signs of cytotoxicity/
toxicity), the study may be judged unacceptable.
It is hoped that these considerations will provide some
guidance as to when analytical analysis of . test and control
substances should be performed when conducting short-term
mutagenicity testing. Please contact Kerry Dearfleld if further
discussion is necessary.
cc: Irving Mauer
John Chen
-------
Attachment 7
CLINICAL PATHOLOGY TESTING RECOMMENDATIONS FOR
NONCUNICAL TOXICITY AND SAFETY STUDIES
AACC-DACOASVCP JOINT TASK FORCE
KURT WHNGAND, Chairman, JOHN BLOOM, MICHAEL CARAKOSTAS, ROBERT HALL,
MARIA HELFRICH, KENNETH LATIMER, BARRY LEVINE, DOUGLAS NEPTUN,
ALAN REBAR, KATHERINE SITTZEL, AND CATHERINE TROUP
Made in United State* cfAmerica
Reprinted from Toxicouxac PATBOLOCT
VoL2aNo.3(P«2).1992
Copyright 01992 by the Society of Toxicolojic PMhofefists
Reproduced with permission from Toxicologic Pathology Vol. 20, No. 3 (Part 2), 1992
-------
TOXKOLOCTC PATHOLOGY lSSNH)192-6233
CopyrightC 1992 by the Society of ToxkxOoik
Volume 20, Number 3 (Part 2), 1992
Printed in US jL
Clinical Pathology Testing Recommendations for
Nonclinical Toxicity and Safety Studies*
AACC-DACC/ASVCP JOINT TASK FORCE
KURT WETNOAND, Chairman, JOHN BLOOM, MICHAEL CARAKOSTAS,
ROBERT HALL, MARIA HELFRICH, KENNETH LATIMER, BARRY LEVINE,
DOUGLAS NEPTUN, ALAN REBAR, KATHERINE STTTZEL, AND CATHERINE TROUP
ABSTRACT
Clinical pathology testing in nondinical toxicity and safety studies is an important part of safety assessment
In recent years, clinical laboratory testing has rapidly expanded and improved. Some government regulatory
agencies provide guidelines for clinical pathology testing in nonclinical toxicity and safety studies. To improve
these testing guidelines and the resultant safety assessments, the American Association for Clinical Chemistry's
Division of Animal Clinical Chemistry and the American Society for Veterinary Clinical Pathology formed
a joint committee to provide expert recommendations for clinical pathology testing of laboratory species
involved in subchronic and chronic nonclinical toxicity and safety studies. These recommendations include
ffrhniryi recommendations on blood collection techniques and hematology, serum chemistry, and urinalysis
tests.
Keywords. Serum chemistry; hematology; coagulation; urinalysis; blood sampling; rodents; toxicity test-
ing; safety studies
INTRODUCTION
Clinical pathology testing of laboratory animals
in nonclinical toxicity and safety studies is an im-
portant part of safety assessment for new food ad-
ditives, drugs, and chemicals in development Some
government regulatory agencies provide guidelines
for clinical pathology testing in nonclinical toxicity
and safety studies. Some of these testing guidelines
have contained technically inappropriate and im-
practical recommendations.
It has become increasingly recognised that clinical
pathology testing of laboratory animals is a distinct
scientific subspecialty that requires qualified pro-
fessional personnel for technical execution and data
interpretation (4,10). In an effort to improve clinical
pathology testing guidelines for safety assessment,
the American Association for C*«n'c«' Chemistry's
Division of Animal Clinical Chemistry (AACC-
DACQ and the American Society for Veterinary
Clinical Pathology (ASVCP) formed a joint com-
mittee in the fall of 1991 to provide expert rec-
ommendations for clinical pathology;testing~of lab-
• Adcfam correspondence to: Dr. Kurt Wemaand, The Procter
& Gamble Co., Miami Valley Laboratories, P.O. Box 398707.
Ohio 45239.
oratory species involved in subchronic and chronic
nondinical toxicity and safety studies.
In general* these clinical pathology testing rec-
ommendations are applicable to most mammalian
species used in safety evaluation. Because at least 1
species of rodent is commonly used in toxicity test-
ing and comparative safety assessment, some of the
technical and testing recommendations provided are
specifically intended for use with rats.
BLOOD SAMPLING
A ftan^i|ixli7fd feeding protocol it*imfdiat
-------
540
AACC-DACC/ASVCP JOINT TASK FORCE Towoouxnc PATHOLOGY
blood volume obtainable firom weanling an-
imals and is not necessary for study conduct if an-
imals are obtained from a quality supplier. Evalu-
ation of test compound effects by statistical
comparison of group means to respective prestudy
values in subchronic and chronic safety studies is
, because a number of clinical pa-
thology test value* change with age. This is es-
pecially apparent in young growing rats in which
hematologic and serum chemistry values change
within a few weeks time (19, 27, 31). Test value
means for each gender should be compared statis-
tically between the groups treated with test com-
pound and the concurrent study control groupfc) at
each time period of clinical pathology testing (14,
29). The use of statistical analyses in the interpre-
tation of clinical pathology data firom toxichy and
safety studies should be limited to pattern recog-
nition of effects caused by treatment with increasing
doses of the test compound (4). Furthermore, ap-
propriate scientific judgment should be applied to
clinical pathology data interpretation for determi-
nation of toxicologic effects caused by the test com-
pound (4).
Clinical pathology testing of mice is limited by
blood sample volume. Because of mis limitation,
interim (in-life) study blood samples should not be
collected routinely from mice. Blood samples firom
mice for clinical pathology testing are collected op-
timally at terminal sacrifice only. Furthermore, it is
often not possible to conduct both hematologic and
serum chemistry analyses with the small blood vol-
ume obtained from mice, so separate groups of an-
imals for hematologic and tfmft chemistry testing
are sometimes necessary. Because of blood volume
limitations, cttnical pathology testing of mice should
be limited when similar information is available
from rats.
The following tnuiimmn standards are recom-
mended for chronologic blood sampling from rats
and nonrodent species in nonctinical toxichy and
safety studies. For 2 4 week studies, hiood samples
should be collected within 7 days after study initi-
ation and at study termination. For 13-week studies,
Mood samples should be collected at approximately
4 and 13 weeks, if 2-4-week study information at
similar test compound doses is concurrently avail-
able. Without dose-comparable 2-4-week study in-
formation, an fl^ffitvyal blood sample collected
within 7 days of study initiation is optimal for the
13-week study protocol For chronic toxkfty,«ar-*
dnogenicity, and combined studies, blood samples
should be collected at approximately 13, 26, and 52
weeks after study initiation. Clinical pathology test-
ing after 52 weeks in rodents is often affected by the
and seldom yields meaningful information concem-
fag tfst CT>inpo""4 effects. The increase m test value
variability that accompanies naturally occurring
disease results in decreased test sensitivity for de-
tection of test compound-induced toxicity. For this
reason, guontifcztfvedinical pathology testing during
the latter half of chronic (104 weeks) toxicity and
safety studies involving rodents is of minimum val-
ue and not f»**Mnfti»n
-------
VoL 20, No. 3 (Part 2), 1992 NONCLINICAL TOXICITY AND SAFETY STUDIES
541
plastin time, and platelet counts are the recom-
mended laboratory tests of hemostasis for nonclin-
ical toxicity and safety studies (28, 32). Platelet
counts are routinely measured by most automated
hematology analyzers, and measurement is recom-
mended with each quantitative hematologic anal-
ysis in toxicity and safety studies. Whole blood clot-
ting time is an imprecise laboratory test ofhemostasis
that provides little meaningful information and is
not recommended as a routine screening test in non-
clinical studies (9,24). Activated clotting time is a
different test, and its use in toxicity and safety stud-
ies is limited by the large quantity of blood required
and nonspecificity of test results.
Blood samples for plasma coagulation time mea-
surements should be collected by venipuncture us-
ing a needle and syringe (9). There are some tech-
nical limitations concerning blood volume and site
of collection in rodents that influence this recom-
mendation for coagulation testing. Blood samples
collected from the orbital venous plexus of rats
should not be used for plasma coagulation time
measurements, because the coagulation times are
often prolonged with this technique (8). It is also
sometimes difficult to obtain the quantity of blood
needed from the orbital venous plexus of rats for
all hematologic, serum chemistry, and coagulation
tests. Because of these limitations, interim study
coagulation testing of rats is not currently practical
or recommended until an acceptable method for
survival venipuncture is developed for routine use.
At a minimum, it is recommended that plasma sam-
ples for coagulation testing be collected by needle
and syringe venipuncture from all species at study
termination of subchronic (2-13 wk) studies. Co-
agulation testing in chronic (104 wk) toxicity and
safety studies does not provide additional infor-
mation to that obtained from subchronic studies
and is not recommended.
SERUM CHEMISTRY
The core serum chemistry tests recommended for
nonclinical toxicity and safety studies include glu-
cose, urea nitrogen, creatinine, total protein, albu-
min, globulin, calcium, phosphorus, y*ft""y po-
tassium, chloride, and total cholesterol
A variety of serum chemistry tests for evaluation
of liver health and function are appropriate for use
with laboratory •nitnaU. For evaluation of hepa-
tocellular health, measurement of a minimum of 2
scientifically appropriate serum tests is recom-
mended (e.g., alanine aminotransferase, aspartate
aminotransferase, sorbitol dehydrogenase, total bile
acids) (5, 15, 25, 26, 30). For evaluation of hepa-
tobiliary health, measurement of a minimum of 2
scientifically appropriate serum tests is recom-
mended (e.g. alkaline phosphatase, gamma-glu-
tamyltransferase, S'-nucleotidase, total bile acids)
(6,15,18,25, 26).
Serum creatine kinase, lactate dehydrogenase, or-
nithine decarboxylase, and omithine carbamoyl-
transferase activities should not be used routinely
as screening tests in nonclinical toxicity and safety
studies. Serum levels of creatine kinase have min-
imal utility for detecting chronic striated muscle
damage, because they are increased only for a short
period of time after injury (7). Furthermore, the
serum activity levels of creatine kinase and lactate
dehydrogenase are highly variable and lack speci-
ficity as indicators of major organ toxicity in animal
species (1,11). The analytical requirements and test-
ing performance for serum omithine decarboxylase
(3) and omithine carbamoyltransferase (5) make
these tests impractical for use as routine screening
tests in nonclinical toxicity and safety studies.
Fractionation of serum proteins by electropho-
resis is not recommended as a routine screening test
in nonclinical toxicity and safety studies. There is
currently very little scientific information available
concerning the effects of toxicity on serum protein
fractions in laboratory animal species to support
meaningful interpretation of such data in nonclin-
ical toxicity and safety studies.
URINALYSIS
Uroanalytic tests are often inaccurate due to ar-
tifactual changes induced in vitro by the suboptimal
conditions under which timed urine samples are
commonly collected from animus in metabolic cag-
es (12,20,21, 23). Urine sample contamination with
food, drinking water, and feces should be mini-
mized. Pooling of urine samples from multiple an-
imals is not recommended.
Uroanalytic testing is recommended at least once
in each subchronic or chronic nonclinical study. The
recommended time for urinalysis in subchronic
studies is at study termination. The recommended
time for urinalysis in chronic studies is between 25
and 52 wk after study initiation. It is optimal to
conduct and injerpret Uroanalytic tests with con-
current serum chemistry and hematology tests (1).
The core urinalysis tests recommended for non-
clinical toxicity and safety studies include a timed
(16-24 hr) urine volume and a measurement of renal
concentrating ability (e.g., specific gravity, osmo-
lality). These 2 screening tests do not require tem-
perature-controlled (cooled) urine, collection con-
tainers. Urine reagent strip (dipstick) testing,
microscopic urine sediment examinations, and uri-
nary mineral and electrolyte excretion are not rec-
ommended as routine screening tests in nonclinical
studies. If a test compound is suspected to cause
-------
542
AACC-DACC/ASVCP JOINT TASK FORCE TOHCOLOOIC PATHOLOGY
kidney toxicity, measurement of additional stien-
tifically appropriate urine tests is recommended from
a 16-24-hr urine sample collected in a temperature-
controlled (cooled) urine collection container (e.g.,
microscopic sediment examination, AT-acetyl-0-i>
glucosaminidase activity, urine protein excretion
rate, mineral and electrolyte excretion rates) (2,13,
16,22).
CONCUUSIQNS
This paper provides conical pathology testing rec-
ommendations for nonclinical toxicity and safety
studies. These recommendations represent mini-
mum testing standards based on current technical
and scientific information. Additional rifafaai pa-
thology tests g?« be used in nondinical toxicity and
safety studies if appropriate information concerning
valid analytical use and test interpretation is avail-
able for the species of interest Our recommenda-
tions against the use of certain tests as routine
screening tests should not preclude appropriate spe-
tialized use of these tests if indicated. As a general
rule, it is optimal to interpret concurrent hemato-
logic, serum chemistry, and urinalysis test results
collectively.
Recommendations for routine clinical pathology
testing in nonclinical toxicity and safety studies are
long overdue. These recommendations represent a
consensus opinion of a select group of experts di-
rectly involved with clinical pathology testing in
nonclinical studies conducted at universities, pri-
vate laboratories, and drug and chemical companies
in the United States. Because nonclinical studies are
conducted and submitted to government regulatory
agencies throughout the world, international har-
monization of clinical pathology testing recommen-
dations is needed. The authors hope this paper will
be a catalyst for international harmonization of clin-
ical pathology testing fw this purpose.
REFERENCES
1. Basel DL,VilkneuYe DC,andYagmuasAP(1990).
Oinkal chemistry. In: Handbook cf hi Vno Toxicity
7efl/n«.DLAnMdd,HCGike,andDRKrewski(eds).
Academic Press, San Diego, California, pp. 509-534.
2. Bovee KC (1986). Renal function and laboratory
evaluation. Tax. PathoL 14:26-36.
3. OrakostasMC (1988). What is sennnomhhinede-
carboxylase? Clin. Chem, 34:2606-2607.
4. CarakosusMCandBaner)eeAK(1990).Inteipretin«
rodent clinical laboratory data in safety assessment
studies: Biological and analytical components of vari-
ation. Fund. AppL ToxicoL 15:744-753.
5. Carakostas MC,GossettKA, Church GE, and Cfcgn-
om BL (1986). Evaluating toxin-induced hepatic in-
jury in rats by laboratory results and discriminant
analysis. Vet. PathoL 23:264-269.
6. Carakostas MC, Power RJ, and Banerjee AK (1990).
Serum £ nucleotidase activity in rats: A method for
automated analysis and criteria for interpretation. Vet.
Gin. PathoL 19:109-113.
7. Cardjnet GH (1989). Skeletal muscle function. In:
Clinical Biochemistry of Domestic Animals, 4th ed.,
JJKaneko(ed). Academic Press, New York, pp. 462-
495.
8. Dameron GW, Wemgand KW, Dierckman T, Dud-
erstadt JM, Odioso LW, Schwecke W, and Baran K
(1992). Effect of bleeding site on clinical pathology
testing of rats: Orbital venous plexus vs posterior
vena cava. Lab. Anim. Set. 42:299-301.
9. Dodds WJ (1980). Hemostasis and coagulation. In:
Clinical Biochemistry of Domestic Animals, 3rd ed.,
JJ Kaneko (ed). Academic Press, New York, pp. 671-
718.
10. Dootey JF (1979). The rote of clinical chemistry in
chemical and drug safety evaluation by use of labo-
ratory animals. Clin. Chem. 25:345-347.
11. Evans GO (1991). Biochemical assessment of cardiac
function and damage in animal species. J. AppL Tax-
icol. 11:15-21.
12. Evans GO and Parsons CE (1986). Potential errors
in the measurement of total protein in male rat urine
using test strips. Lab. Anim. 20:27-31.
13. Penman MJ (1987). Evaluation of the usefulness of
routine microscopy in canine urinalysis. J. Am. Vet.
Med. Assoc. 190: 892-896.
14. Gaylor DW, Suber RL, Wolff GL, and Crowell JA
(1987). Statistical variation of selected clinical patho-
logical and biochemical measurements in rodents.
Proc. Soc. Exp. BioL Med. 185:361-367.
15. Hoffman WE, Kramer J, Main AR, and Torres JL
(1989). Clinical enzymology. In: The Clinical Chem-
istry of Laboratory Animals, WF Loeb and FW Quim-
by (cds). Pergamon, New York, pp. 237-278.
16. KluewWM (1981). Renal function tests as indicators
of kidney injury in subacute toxicity studies. Ttodcol.
AppL Pharmacol. 57:414-424.
17. Leissing N, Izzp R, and Sargent H (1985). Variance
estimates and individuality ratios of 25 serum con-
stituents in beagles. Clin. Chem. 31:83-86.
18. LeonardTB,NeptunDA,andPoppJA(1984).Senim
gfTBTna glutamyl transferase as a specific indicator of
bik duct ksicms in the rat liver, ^m./. PathoL 116:
262-269.
19. MdnfassnerJG" and SchmookFP (1990). Reference
values for CrbCD (SD) BR rats. Reference Paper, VoL
3, fl. Charles River Laboratories, Wilmington, Mas-
sachusetts.
20. Osbome CA and Stevens JB (1981). Handbook of
Canine and Feline Urinalysis. Ralston Purina, St
Louis, pp. 57-120.
21. Ragan HA (1989). Markers of renal function and in-
jury. In: The Clinical Chemistry of Laboratory Ani-
mals, WF Loeb and FW Quunby (eds). Pergamon,
New York, pp. 321-343.
22. Stonard MD. Gore CW, Oliver JA, and Smith IK
(1987). Urinary enzymes and protein pattterns as in-
-------
VoL 20, No. 3 (Part 2), 1992 NONCLINICAL TOXICTTY AND SAFETY STUDIES
543
dicaton of injury to different regions of the kidney.
Fund. Appl. ToxkoL 9:339-351.
23. Taylor DM and Neal DL (1982). False negative hy-
perglucosuria test-strip reactions in laboratory mice.
Lab. Anim. 16: 192-197.
24. Theus R and Zbinden G (1984). Toxicological as-
sessment of the hemostatic system, regulatory re-
quirements, and industry practice. Reg. Ttaacol.
Pharmacoi 4: 74-95.
25. Thompson MB, Neptun DA, Garvey LK, and Popp
JA (1984). Effects of three hepatocarcinogens, alpha*
naphthylisothiocyanate (ANIT), lithocholic acid
(LCA) on serum bite acids in the tat Proceedings of
the American College of Veterinary Pathology, San
Antonio, Texas, p. 79.
26. Thompson MB, Blair PC, Morris TW, Neptun DA,
Deyo DF, and Popp JA (1987). Validation and ap-
plication of a liquid chromatography/enzymatic as-
say for individual bile acids in the serum of rats. Clin.
Chem. 33:1856-1862.
27. Turton JA, Hawkey MG, Hart MG, Gwynne J, and
Hicks RM (1989). Age-related changes in the he-
matology of female F344 rats. Lab. Anim. 23:295-
301.
28. Varg«ftigBB,G>naidJ,andSamamaM(1979).Blood
coagulation and platelet function. Pharmacoi Ther-
ap. 5:225-227.
29. Weil CS (1982). Statistical analysis and normality of
selected hematologic and clinical chemistry mea-
surements used in toxicology studies. Arch. Toxicol.
5(nippl.): 237-253.
30. WeingandK,DameronG,DierckmanT,Duderstmdt
J, Odioso L, and Bruner H (1991). Serum chemistry
of acute hepatotoririty in rats. Proceedings of the
American College of Veterinary Pathologists, Orlan-
do. Florida, p. 206.
31. Wolford ST, Schorer RA, Gallo PP, Gobs FX, Bro-
deck M, Falk HB, and Ruhren R (1987). Age-related
changes in serum chemistry and hematology values
in normal Sprague-Dawley rats. Fund. Appl. Toxicol
8:80-88.
32. Zawidzka ZZ (1990). Hematologic evaluation. In:
Handbook of In Vivo Taxicity Testing. DL Arnold,
HC Grice, and DR Krewski (eds). Academic Press,
San Diego, California, pp. 463-508.
-------
Attachment 8
Mutation AcMrcft. 258 (1991) 239-:3J
C 1991 Elwvwr Science Publishers B V All rtthu reserved 0165-1 IIO/9I/SOJ.50
MUTREV 07307
Considerations in the U.S. Environmental Protection Agency's testing
approach for mutagenicity
Kerry L. Dearfield ', Angela E. Auletta 2, Michael C. Cimino : and Martha M.
Moore3
' HttUt Elftca anam. Offlet efPaacidt fnt**u. ** * Mm** tut EiuinMHaiul Ktiitw Duitioii. Offkt of Taac
U.S. CiummKiMtl ftoMcno* Afptef. NtoftnfM* DC2MM4M* -' G0w* Tuacotov OKWON. MM** Effects Arsra/r/i
Lebanon. U.S. &KVWMMMM/ Aotsowi <4j0tt* AOMIC* Tn**t ftrtc. NC277/7
(Received 12 Pebnaiy 19*1)
(Accepted MMey 1991)
U.S. Enwwuntnul fnxectioa Afenqr. Muu»t«icity test procedure*
Svmmaiy [[[ 25*
Introduction [[[ 2W
Office o* Pesticide Protraim(OPP) [[[ -fi2
Badcground ......................... • [[[ -"^
Revised initial bettery .................................................. • .................... -M
Fufdwr tatent beyond tesanc w «««»l bwtefy [[[ X*
Office of Tone Substances (OTS) [[[ -67
-. ........................... 267
frintir [[[ ^
Funtar actioai beyond testiiif in initial tier .................... ............. - .................... -^
Testiaff (protocol) twdeltees 2
Appnda&Scieetifle Adrimy NM! t^ puMk CMMMM PRMW for Off re«iMd iMdeKM ...................... 274
Of?
This ptpn provides the ratiotule and support for the decisions the OPP will make in requiring and
reviewing nutafenkity information. The regulatory requirement for mutagenicity testing to support a
pestiddt rtjittnTfrM is found in the 40 CFR Part 1S8. The guidance as to the specific mutagenicity
testing to be performed is found in the OPP's Pesticide Assessment Guidelines. Subdivision F. Hazard
Evaluation: Hunan and Domestic Animals (referred to as the Subdivision F guideline).
TtoMMOfpt has ben renewed by the Office of Pesticides
-------
A revised Subdivision F guideline has been presented that becomes the current guidance for
submitters of mutagenicity data to the OPP. The decision to revise the guideline was the result of close
examination of the version published in 1982 and the desire to update the guidance based on
developments since then and current state-of-the-science. After undergoing Agency and public scrutiny,
the revised guideline is to be published in 1991.
The revised guideline consists of an initial battery of tests (the Salmonella assay, an in vitro
mammalian gene mutation assay and an in vivo cytogenetics assay which may be either a bone marrow
assay for chromosomal aberrations or for micronudei formation) that should provide an adequate initial
assessment of the potential mutagenicity of a chemical. Follow-up testing to clarify results from the
initial testing may be necessary. After this information as well as all other relevant information is
obtained, a weight-of-evidence decision will be made about the possible mutagenicity concern a chemical
may present. Testing to pursue qualitative and/or quantitative evidence for assessing heritable risk in
relation to human beings will then be considered if a mutagenicity concern exists. This testing may range
from tests for evidence of gonadal exposure to dominant lethal testing to quantitative teats such as the
specific locus and heritable transkxarion assays. The mutagenicity assessment will be performed in
accordance with the Agency's Mutagenicity Risk Assessment Guidelines. The mutagenicity data would
also be used in the weight-of-evidence consideration for the potential carcinogcnicity of a chemical in
accordance with the Agency's Carcinogen Risk Assessment Guidelines. In instances where there are
triggers for carcinogcnicity testing, mutagenicity data may be used as one of the triggers after a
consideration of available information.
It is felt that the revised Subdivision F guideline will provide appropriate, and more specific, guidance
concerning the OPP approach to mutagenicity testing for the registration of a pesticide. It also provide* a
clearer understanding of how the OPP will proceed with its evaluation and decision making concerning
the potential heritable effects of a test chemical.
OTS
As a result of recent information in the field of mutagenicity (the WOliamsburg meeting, its
precedents,and sequelae), a modification to the Section 4 test scheme has been proposed. In the
proposal, test schemes for gene mutation and chromosomal aberrations are combined. The revision
comprises three tests in the mitial tier, the Salmonella assay, an in vitro assay for gene mutation and an
in vivo assay for chromosomal effects which may be either a bone marrow assay for chromosomal
aberrations or the micronudeus assay.
Since submission of this article for publication, the proposed OTS test scheme has been approved by
OTS management The old two scheme, four test scheme is no longer in effect, and will no longer be
used in future TSCA Section 4 Test Rules.
Under Section 5 of TSCA ("new" chemical, or Premanufacture Notice (PMN)), mutagenictry data are
used for three purposes: (1) as pen of exposure based testing; (2) to assess the potential of the PMN
chemical to induce heritable genetic effects; and (3) as part of the weight-of*«vidence that a chemical
may be a potential carcinogen.
As part of the exposure beted testing program, the USEPA has required testing of certain high
volume chemicals vfch a two test battery of the Salmonella assay and a mouse mkroaudeus essay. In
.supporting a concern lor potential carcmogeniciry of a PMN chemical, the USEPA generally cites data
on an analogue whkfc is known to be carcinogenic (Le.. demonstrated tumor forming ability in one or
more animal studtesX In such instances, mutagenicity data on the PMN chemical or on the analogues are
used to lend support to the case for potential cartinogenicity. Where there is no analogue of the PMN
chemical which has been tested for caidnogcnictty, mutagem^w datt ak»e are generally not considered
sufficient to support a concern for potential cartinofenicity. Regulatory action under Section 5 is seldom,
if ever, taken on the basis of muragftitrity data alone, especially on the basis of in vitro mutagenicity
data.
-------
Introduction
Periodically, the U.S. Environmental Protec-
tion Agency (USEPA or Agency) reviews its mu-
tagenicity testing guidelines to assess their effec-
tiveness for evaluating genotoncity data based on
the current state of the science. The Office of
Pesticide Programs (OPP) and the Office of Toxic
Substances (OTS). which are both under the aus-
pices of the USEPA's Office of. Pesticides and
Toxic.Substances (OPTS), are now recommend*
ing changes in their current guidelines for mutt-
genicity testing. The science of mutagenicity test-
ing has undergone considerable rethinking in the
yean since the current mutagenicity guidelines
were proposed and initiated. In part, this rethink-
ing has been the result of efforts by the USEPA's
Gene-Tox Program to evaluate short-term tests
for mutagenicity and related endpoints (Waters
and Auletta. 1981) and the National Toxicology
Program's studies of the ability of short-term in
vitro tests to predict carcinogenkity (Tennant et
al., 1987). Also, meetings such as the one spon-
sored by the USEPA in Williamsburg, VA in
January, 1987 (Workshop on the Relationship
Between Short-Term Test Information and Car-
cinotenkity) have spurred reconsideration of the
role of short-term mutagenicity tests in toxkaty
testing (Auletta and Ashby, 1988; Kier. 1988).
Many articles have been published over the pest
several yean reflecting a myriad of opinions and
viewpoints on the roMs) of mutagenicity testing
(e.*, Ashby, I986aj>; Gamer and Kirkiaad, 1986;
Gatehouse and Tweets, 1986; Lave aad Ouena,
1986; Ami et al., 1988; Brodnnan and De Mariai.
1988; Ennever and Roseaknaz, 1988; Legator
and Harper. 1988). Many of these have dealt with
short-term teats at predkton of carcinogenicity
although some have emphasfred heritable muta-
tions as aa endpoint of concern.
In 1986, the Agency published its Mutagenicity
Risk Assessment Guidelines (USEPA. 1986a).
These guidelines deal with heritable mutation as
a regulatory endpoint They present a weight-of-
evidence scheme for determining if aa ageat may
be a potential human germ cell mutagea. Empha-
sis is placed upon a chemical's intrinsic rauta-
genic potential, its ability to reach the foaad and
interact with germ cell DNA. and its ability to
induce heritable mutations in a mammalian
species. The categories of evidence for chemical
interaction in the gonad and of evidence that
contributes to the weight-of-evidence evaluation
for potential human germ cell mutagenicity are
summarized in Appendix A. It is clear from the
Mutagenicity Risk Assessment Guidelines that
the USEPA intends to regulate chemicals based
on risk of an advene heritable effect in human
germ cells. This is consistent with the evolution of
Agency policy as detailed for example in USEPA
(1978) and Hill (1979).
It is noted that the USEPA has historically
wt4 •mutagenicity information as part of its
weight-of-evidence approach to discern the po-
tential for human carcinogenicity. This is still a
major use of awiateaidty data as detailed ia the
Agency's Carcinogen Risk Assessment Guide-
lines (USEPA. 1986b). However, it is emphasized
that this is not the sole use for mutagenicity data
and that the Agency will evaluate mutagenicity
data ia terns of heritable risk.
It is the policy of the Agency to periodically
reevaluate its guidelines ia accord with the cur-
rent state of the science. By revising its muta-
genicity guidelines at this time, the USEPA will
be ia concert with other countries aad interna-
tional bodies which have recently revised or are
ia the process ofrevisinf their mutagenicity
guidelines <«4t Canada (Health and Welfare
Canada. 1986), the United Kingdom (Diggle and
Fielder, 1989). and the EEC). The U.S. Food and
Drag Administration is also revising its muta-
genicity guidelines found ia the "Red Book"
(Toricotofical Principles for the Safety Assess-
ment of Direct Food Additives and Color Addi-
tives Used in Food; V. Dunkel aad D. Benz.
personal communication).
By revising both the OPP aad the OTS guide-
lines at the same time, the USEPA intends to
narnx*"™ mutagenicity testing requirements of
the two offices to the extent possible within the
statutory M«"iMrinM of both the Federal Insecti-
cide. Fungicide, and Rodeaticide Act (FIFRA)
aad the Tone Substances Control Act (TSCA).
The OPP aad the OTS are also ia the process
of htramfrfag their existing testing guidelines in
the areas of chemical fate aad health and ecologi-
cal effects. It is anticipated that the end result of
-------
26:
this effort will be a set of OPTS testing guidelines
that will be used by both offices. Differences
which are necessary to meet' the statutory re-
quirements of each law will be detailed within
each guideline. Once this effort is complete, the
OPTS and the Organisation for Economic Coop-
eration and Development (OECD) guidelines will
be compared to identify and reconcile differences
to the degree possible. The thrust of this effort is
to reach harmonization of testing methods and
thereby ensure mutual acceptance of. data both
within the USEPA and between the USEPA and
the international community.
This document will present the requirements
for mutagenicity testing for submission of data to
the U.S. Environmental Protection Agency's Of-
fice of Pesticide Programs and the Office of Toxic
Substances. The rationale and support for the
revised guidelines will abo be presented and
questions about the revision process and the
guidelines themselves will be addressed.
Oflktef Pesticide
(OF?)
Background
Section 3 (Registration of Pesticides) of the
Federal Insecticide, Fungicide, and Rodenticide
Act (FIFRA) permits the Administrator of the
USEPA to publish guidelines specifying the kinds
of information required to support the registra-
tion trf a pesticide with the OPP. It abo states
that revisions should be made from time to time
(see Section 3 (cX2) of FIFRAX Pan 158 of the
Code of Federal Regulations (40 CFR - Protec-
tion of Environment) is the regulation that deter-
mines- what toxicity testing must be performed
and when to perform such tests in support of a
pesticide registration under FIFRA, The OPFs
Hazard Evaluation: Human and Domestic Ani-
mals (subsequently referred to as Subdivision F
guideline; present version from USEPA. 1962)
provides guidance on how to implement the Pan
158 requirements. This paper will discuss the
revisions to the OPFs Subdivision F rautagenicity
guideline and the rationale for moat revisions.
The OPP is revising Pan 158 to allow for the
generation of data more appropriate for making a
sound regulatory decision. The revisions to Sub-
division F would therefore also be timely in antic-
ipating upcoming revisions to Part 158.
Part 158 - Data Requirements for Registra-
tion - specifies the types and minimum amounts
of data required in order to make regulatory
judgments about the risks of pesticide products.
The current requirements concerning mutagenic-
iry testing are detailed in Part 158. It states that
mntagenkity testing is required for all general
use patterns of pesticides (including terrestrial.
aquatic, greenhouse, forestry, domestic outdoor
and indoor for both food crop and nonfood uses).
The technical grade of the active ingredient is the
test substance to be used for testing. One test in
each of three categories (gene mutation, struc-
tural chromosomal aberrations, and other geno-
toric effects) is minimally required. It does not
appear that the upcosMg revisions to Pan .158
will change the requirement that mutagenicity
testing be performed for all general use patterns
of pesticides. However, the general requirement
for testing in the three categories mentioned
above will be eliminated in favor of testing in
specified assays as detailed in the Subdivision F
guideline revision (sec bctowX
The OPP mutagenicity guideline is found in
Series 84 of the Subdivision F guidelines. It is
divided into two sections: (1) 84-1: Purpose and
General Recommendations for Mutagenicity
Testing and (2) 84-2: Mutagenicity Tests. Section
84-1 is fairly setf-exptanatory in describing the
purpose and general recommendations for muta-
genicity testing, such as which substance to test,
standards for metabolic activation, and v«MLUv4
materials. Section 84-2 states that a mmrnium of
three mutagenicity tests be performed, one each
in the categories of gene mutations, structural
chromosomal aberrations and other gcaotoric ef-
fects (the last including numerical chromosome
anomalies and direct DNA damage and repair).
Lists of representative tests that would satisfy
each category arc provided from which particular
tests may be chosen. Although no specific proto-
col guidelines are provided, references to some
standards are mentioned, such as those of the
Gene-Toe Program from the USEPA's OTS. Cur-
rent practice by the OPP, however, is to utilize
the protocol guidelines issued by the OTS that
are found in the 40 CFR Part 798 - Health
-------
Effects Testing Guidelines (for discussion, see
section on Testing (protocol) guidelines).
Reused initial battery
A major criticism of the 1982 published Subdi-
vision F guideline is that it provides little guid-
ance regarding the choice of tests to be per-
formed for a particular chemical or class of chem-
icals. The OPP agrees with this assessment and
believes it would be appropriate to provide more
specific guidance as to what tests should be per-
formed. To that end. it has been suggested that
chemicals to be submitted to OPP for registration
purposes should be tested in a defined battery of
mutageniciry tests (Dearfield. 1989). The results
of such testing would provide the USEPA with
information to be used in assessing the potential
mutagenic hazard of chemical agents subject to
regulation under FIFRA.
The Subdivision F guideline revision has un-
dergone rigorous Agency and public examination
(for details and discussion of various issues relat-
ing to this guideline, see Appendix B). The deci-
sions and rationale that culminated in the current
revised Subdivision F guideline are .presented
here and in Appendix B. The revised Subdivision
F guideline is projected to be officially published
in 1991. Submitters of mutigeniciry data to the
OPP are advised to take these new guidelines
into account when planning future submissions to
the OPP.
The tests that will be included in the OPP
initial battery are as follows (also seen in Fig. 1):
(1) Salmonella typhimunum reverse mutation as-
say.
(2) Mammalian cells' in culture forward gene mu-
tation assay allowing detection of point muta-
tions, large deletions and chromosomal rear-
rangements, such as:
(a) mouse h/mphoma LS178Y cells, thymidine
kinase (tk) gene locus, maximizing assay condi-
tions for small colony expression and detection:
or.
(b) Chinese hamster ovary (CHO) or Chinese
hamster lung fibtoblast (V79) cells, hypoxanthine-
guanine phosflMriDMy* transferase (hgprt) gene
locus, accompanied by an appropriate in vitro test
for clastogeniory; or,
(c) Chinese hamster ovary (CHO) cells strain
AS52, xanthine-guanine phosphoribosyl trans-
ferase (xpn) gene locus.
(3) An in vivo assay for chromosomal effects
using either.
(a) metaphase analysis (aberrations); or.
(b) a micronudeus assay.
For an initial assessment of mutagenic activity.
the in vivo assays are. most often performed in
rodent bone marrow. Use of other species or
CURRBfT OPP MUTAGEUOTY TEST QUDBJNE
Sabnofwla + In Vitro
G«* Mutation •
In Vivo Bon*
Marrow Cytogtntttot
•Aberrations
or
•Mteronuctou*
• in Vlto Gene Mutation (ehoiee):
(a) MOUM tympt«ma LSlTgV eeia. tk toeui, amai and
(b) ChineM heineMr ovary eela strain AS82
(e) CNneee hamiter ovaiy (CHO) or Chines* huneter
lung fibrobieM (v78| ee«a + appropriate in vteo
ten foe cieetoojtmcty
Fij. 1. Current OPP muugemcify ttst guidtliat.
-------
other target organs should be discussed with the
OPP prior to testing.
According to the Agency's Muugenicity Risk
Assessment Guidelines (USEPA. 1986a). there
are several muugenesis endpoina of concern in-
cluding point mutations (Le~ submicroscopic
changes in the base sequence of DNA) and struc-
tural or numerical chromosome aberrations.
Structural aberrations include deficiencies, dupli-
cations, insertions, inversions, and translocattons
of pans of chromosomes, whereas numerical
aberrations are fains or tosses of whole chromo-
somes (e.s>. trisomy. monotony) or sets of chro-
mosomes (hapkxdy, polyploidy). The revised OPP
battery will identify those agents which induce
point (or gene) mutations and/or structural chro-
mosome aberrations. Since there is no assay that
is currently validated and routinely available for
testing agents that may induct numerical chromo-
some aberrations (e JM aneuptoidy events; Del-
larcoiet aL, 1986 J. such an assay is not included in
the proposed battery. Until such time as a reli-
able test is developed and validated, the Agency
will deal with agents which nay induce numerical
chromosomal aberrations on a case-by-case basis.
Despite this obvious limitation, the OPP believes
that the revised battery will provide sufficient
information about chemical activity to allow a
reasonable assessment of potential mutagenic
hazard.
The first test in the revised OPP testing bat-
tery is the Siifn*>fwMi/'"*innn!ii*1 tpfer^t^mal
assay. The Salmonella assay is a revcne natation
assay which employs several specific taster strains
for the detection of point muiaiions Since die
genetics of each tester strain have been well-de-
fined, it i» possible to identify the specific type of
genetic effect (i*. bast subuirution. fnmeshift
mutation) induced by apnti which are active in
this system (Ames et at, 1973; Maron and Ames,
1983). In addition to genetic characterization, the
Salmonella assay bat several other advantages
which make it a logical choice for inclusion in a
testing battery. This assay is easy to perform and
is used routinely throughout the world and in
most laboratories which perform genetic toxicol-
ogy testing (Farrow et aL 1984). As a result of
this wide-spread use. it is both well-validated and
has an extensive data base of tested chemicals
(Kier et at.. 1986: Auletta and Kier. in prepara-
tion). It is extremely useful for detecting intrinsic
mutagenidty of many classes of biologically active
chemicals, especially ones that appear to act via
an electrophilic mechanism (e.g.. Rinkus and
Legator. 1979: Ashby and Tennant. 1988). Al-
though there are disagreements about the assay's
precise ability to accurately predict chemical car-
cinogenicity, it nonetheless provides useful infor-
mation in predicting carcinogenic, as well as mu-
•tagenic, potential of chemical agents. Both Kier
(1988) and Auletta and Ashby (1988) recommend
inclusion of the Salmonella assay in any screening
program to set priorities for further testing by
identification of potentially mutagenic and car-
cinogenic chemical agents. The Salmonella assay
is included in the OPP battery for all of these
reasons.
Although the Salmonella assay is a primary
test for gene mutations, it was felt that it should
not be the sole test for this endpotnt The Agency,
according to its Guidelines for Mutagenidty Risk
Assessment (USEPA. 1986a). places greater
weight oa tests conducted in eukaryotes than in
prokaryotes and in mammalian species rather
species in conducting a
hazard evaluation of a chemical. Major differ-
ences between mammalian ceUs and bacterial
cells, such as membrane structures. DNA repair
capabilities and the organization and complexity
of mammalian genomes, suggested that it was
necessary to have a """""fl'tfi system included
in the battery (Fox, 1988). Furthermore, there are
chenkais which give negative results in the
Salmonella assay which are mutagenic when
tested in a mammalian cell culture assay for gene
nutations (Arlett and Cole. 1988). Finally, the
results of the NTP study on the use of short-term
tests as predictors of carcinogenidty (Tennant et
aL 1988: Auletta and Ashby. 1988; sec also Be-
nigni. 1969, for duster analysis of NTP data)
show that the L5178Y mouse lymphoma cell as-
say and the sister chromatid exchange, assay ap-
pear to be sensitive to a different subset of chem-
icals than the Salmonella assay and the in vitro
cytogenetks assay. The OPP concluded, there-
fore. that the combination of the Salmonella as-
say and a mammalian cell culture assay for gene
mutation would provide more information than
-------
that obtained from the Salmonella assay alone.
This additional information may provide a better
idea of the mechanism of mutagenic activity, a
refinement of possible mutagenicity concern.
and/or a basis tor further testing.
In selecting a mammalian celi culture assay for
inclusion in the OPP battery, primary considera-
tion was given to the L5178Y mouse lymphoma
assay. This has generated a great deal of com-
ment during the Subdivision F Agency and public
examination period (see Appendix B). In select-
ing an in vitro mammalian gene mutation assay,
an important consideration was the ability of the
chosen assay to provide maximum information on
the genotoxiciry of the test chemical Recent ad-
vances in the understanding of the types of ge-
netic events detectable by mammalian cell assays
indicate that two assays, the L5178Y/TK*'"
mouse lymphoma assay and the CHO AS52 as-
say, detect chemicals capable of inducing both
point mutations and chromosomal events (Hsie.
1987; Stankowski and Tindall 1987; Moore et at.
1989: Appiegate et al, 1990). Use of an assay
capable of detecting a broad range of genetic
events provides the most information and is thus
preferred.
The mouse lymphoma assay is the better char-
acterized of the two assays with regard to the
types of genetic damage detected. By using a
combination of molecular analysis and banded
karyocype analysis, research has shown that tk
mutants appear to include presumed point muta-
tions (no visible alteration in karyocype or South-
ern Mot pattern), tool tic gem deletions, mitotic
nondisjunctkm. translocaooo. homologous mi-
totic recombination, and gene conversion (Ap-
piegate et at, 1990). In humans, genetic lesions
including rearrangements, additions and dele-
tions of genetic material result in heritable disor-
ders and an associated with neopUsia. The ASS2
assay, a relatively new modification of the stan-
dard Chinese hamster ovary (CHO)/hgprt assay.
appears capable of detecting dastogenic events
not detected by the standard CHO assay (Hsie.
1987; Stankowski and Tindall. 1987). Because of
the ability to detect other genotoxk effects in
addition to point mutations, both of these assays
are appropriate for use in the initial assessment
of the mutagenicity of a test compound.
An in vivo cytogenetics assay was chosen as the
assay of choice for determining effects on chro-
mosomes. This may either be metaphase analysis
for chromosomal aberrations (excluding SCE for-
mation) or a micronucleus assay, both conducted
in rodent bone marrow. These tests have been
performed routinely for many yean and each has
a substantial data base of tested chemicals (Pre-
ston et al.. 1981: Heddle et at.. 1983: Mavournm
et al.. 1990). Other organ or tissue sites may be
considered (e.g.. liver, lymphocytes, spleen), par-
ticularly if knowledge about the test chemical
provides support for the selection of other or-
gans/tissues (e.g.. in George et al.. 1989. the rat
liver carcinogen 2-nitropropane was negative in
the bone marrow micronucleus assay, but positive
in a liver micronucleus assay; similarly, Schmezer
et al.. 1990. show that AMutrosodibenzytainine
induces mkronudei in rat liver, but not in rat
bone marrow). Since the Agency's Guidelines for
Mutagenicity Risk Assessment place greater
weight on results from in vivo tests than in vitro
tests (USEPA. 1986*). it was felt that the chro-
nal aberration assay should be performed in
vivo, thus allowing for such factors as intact in
vivo metabolism, repair capabilities, pharmacoki-
netic factors (e.g.. biological half-life, absorption.
distribution, excretion), and target specificities
(see. e.g., Legatorand Harper. 1988).
This initial battery was designed for use with
chemical* of unknown genotoxk potential This
battery should provide much useful information
about mutagenic potential without imposing un-
duly burdensome testing requirements upon the
registrant. It is designed to satisfy the minimum
regulatory requirement for initial mutagenicity
testing. However, alternative tests may be pro-
posed if such testing is based on knowledge of the
test chemical. The OPP intends to be responsive
to unique testing considerations for specific
chemicals or classes of chemicals. If other tests or
special testing conditions appear more appropri-
ate for specific chemicals, then the submitter or
the OPP may propose a discussion of these points
before testing is initiated.
These overall changes in the mutageniciry test-
ing scheme were made in response to the criti-
cism that the current guideline, with its lists of
representative tests, was too broad in scope to
-------
provide guidance for many submitters. By propos-
ing this initial battery of tests, the OPP accom-
plishes two purposes: (i) it provides a defined set
of mutagenicity tests to be performed on chemi-
cals to be submitted for registration, and (ii) it
ensures the generation of a body of data on
which to base decisions about either the need for
further testing and/or the degree of concern
about the potential mutagenicity of the test agent.
In addition to the initial battery, other provi-
sions are provided to enhance the information
submitted to the OPP for registration of chemi-
cals. If other tests for endpoints that may be
predictive of mutagenicity are performed in addi-
tion to the initial battery, these results must also
be submitted to the OPP. A reference list of all
studies/papers known to the submitter concern-
ing the mutagenicity of the test chemical is also to
be submitted. This does not need to be a totally
exhaustive effort but one that reasonably cap-
tures most of the relevant studies on the test
compound. Registrants most likely perform this
function in their own deliberations (or should)
and it should not be an additional burden. Sub-
mission of other relevant data is encouraged (e-g^
metabolism, distribution ttwtift. reproductive
studies — which in many cases are already re-
quired by the Part 158 Toxicology Data Require-
ments). This additional information may provide
better insight into the •*§*•**»••*•> md interpreta-
tion of mutagenicity test results, and would greatly
facilitate the OPFs effort to provide a timely and
more accurate i
icaL
Funher action btfond toting in MnW Aenery
Confirmatory Hating or other relevant infor-
mation may be raqmnd to provide clarification
of equivocal or dkwfdaat results among the tests
initially submitted to the OPP. This would pro-
vide information that may further darify the po-
tential genotoxfe hazard of the test chemical. For
example, additional in vivo cytogenetics testing
may be required to address such concerns as
target tissue/organ or species specificity, differ-
ences in metabolism or distribution, or
structure-activity relationship (SAR) considera-
tions.
Results from the initial battery and confirma-
tory testing (if performed) are reviewed along
with all other available relevant information be-
fore decisions on subsequent regulatory action
are made. The purpose of the initial testing is to
assess inherent mutagenicity for the purpose of
hazard identification, (f no mutagenicity hazard is
identified from the available information, further
action may not be necessary. However, if addi-
tional information becomes available which M.J-
gests a mutagenicity hazard, then the decision
take no further action may be reconsidered.
Further testing to determine if a chemical may
induce heritable mutation in mammals will be
performed, if necessary, in accordance with the
Agency's Guidelines for Mutagenicity Risk As-
sessment (USEPA. 1986a). When evaluating 'a
chemical for the potential &» induce heritable
mutations, all available data including mutagenic-
ity test, data: exposure data; SAR considerations:
studies of mechanism of action; pharmacokinetic
and metabolism information; the results of test-
ing for reproductive effects, target organ speci-
ficity, subchronic and- chronic effects; and the
ability of the chemical to reach the germ cell and
interact with gonadal DNA will be considered.
Once the available da.ta have been reviewed, the
Agency may decide that no further testing is
warranted. -However, if jhe weight-of-evidence
further >f«»«"g, that ttniin may involve
IB SBttfBBwOflQfilA elflfl/OT
speiHiatocytes of rodents, dominant Jethal test-
ing, or tetting for titiw evidence of fttfttnfrtl
interaction with mammalian germ cells. Results
of such tens, if positive, would provide evidence
that the chemical in question has the potential to
KB O
My including humans). The need for additional
testing to support a quantitative risk assessment
would depend upon both available mutagenicity
data and other relevant considerations such as
human exposure levels, chemical use patterns and
release to the environment
When the qualitative evidence using this ap-
proach "'gg**1* a potential hazard for heritable
imitagenic effects, appropriate tests for quantify-
ing heritable risk shall be performed. Currently,
these tests are the specific locus test (cither visi-
ble or biochemical) and the.heritable transtoca-
-------
tion test, both performed in rodents. Upon com-
pletion of appropriate tests for quantifying heri-
table risk, a quantitative risk assessment will be
performed. It is recognized that quantitative mu-
ugenicity risk assessment is still a fledgling area.
but previous etfons to perform quantitative risk
assessment will be examined (e.g.. see UN-
SCEAR. 1977: Selby. 1979: BE1R. 1980: Ehling,
1988: Rhomberg et al.. 1990) in order to identify
and apply appropriate methods to quantify the
risk due to the test chemical under examination.
Mutagenicity test results will also be consid-
ered in decisions about the carcinogenicity of the
test chemical. If a chemical has been tested for
carcinogenicity. available mutagenkity data will
be used along with the carcinogenicity testts)
results as pan of the weight-of-evidence ap-
proach for classifying the chemical in accordance
with the Agency's Guidelines for Carcinogen Risk
Assessment (I986b). When carcinogenicity testing
is conditionally required in accordance with the
Pan 158 Toxicology Data Requirements, evi-
dence of chemical mutagenicity may provide the
basis to require a carcinogenicity study for that
chemical.
OfBctofTexk
(OTS)
Background
The Toxic Substances Control Act (TSCA)
provides the USEPA with the authority "to regu-
late commerce and protect human health and the
environment by requiring testing and necessary
use restrictions on certain chemical substan-
ces..." This authority supplements other existing
laws, such as the Federal Insecticide, Fungicide
and Rodenticide Act (FIFRA). the Clean Air
Act. the Clean Water Act and the Occupational
Safety and Health ACL TSCA is designed to fill
the gap in die government's authority to test and
regulate chftmre'* In TSCA. the term "chem-
ical" encompasses a wide variety of organic and
inorganic substances manufactured or imported
for industrial uses, such as dyes, pigments. lubri-
cant additives, chemical intermediates, synthetic
fibers, structural polymers, coatings — essentially
any commercial chemical except those used as
drugs, food additives, cosmetics, pesticides and
certain other uses.
TSCA has two main regulatory features: <1)
acquisition of sufficient information by EPA to
identify and evaluate the potential hazards from
chemical substances: and (2) regulation of the
production, use. distribution and disposal of such
substances when necessary.
Section -Ha) of TSCA gives the USEPA au-
thority to require the testing of chemicals if un-
reasonable risk to health or the environment is
suspected. To require testing. EPA must find
that: (I) the chemical may. present an unreason-
able risk or a significant potential for exposure:
(2) there are insufficient data available with which
to perform a reasoned risk assessment; and (3)
testing is necessary to generate such data (and is
not already underway). A testing requirement is
promulgated in^a Tes^ Rule, which must: (1)
identify the substance to be tested and the tests
to be conducted: (2) provide (or reference) guide-
lines for performance of the tests: and (3) specify
a reasonable period of time for completion of
testing.
In the past, the OTS required that chemicals
which were subject to mutageniciry testing under
Section «'M cells in culture.
A positive in either assay triggered a Drosophila
sex-linked recessive lethal assay. A positive in the
sex-linked recessive lethal assay triggered a visi-
ble specific locus assay. The chromosomal effects
test scheme (Fig. 3) began with an in vitro cytoge-
netics assay. A negative in vitro assay triggered an
in vivo bone marrow cytogenetics assay (either
metaphase analysis for aberrations or micronu-
clei). A positive in either assay triggered a rodent
dominant lethal assay. A positive rodent domi-
nant lethal assay triggered a rodent heritable
translocation assay.
The first four assays in these two test schemes
(Salmonella, in vitro gene mutation, and in vitro
and in vivo cytogenetics assays) were selected to
optimize the detection of intrinsic mutagenic po-
tential. If all were negative, no further testing was
-------
FORMER OTS GENE MUTATION TEST SCHEME
^5 ^k^^^ktf^flh^hJI^h
ovnonwu
InVtoo
Gtna MuUtfon
DrosttpNtaSLR
KI» 2. Forattt OTS fMM •Mittw tttt
requited. The next two tests (Drosophila sex-lin-
ked recessive lethal and rodent dominant lethal
assays) demonstrate the ability of the chemical to
reach the aonad of the intact organism and to
interact with term cell DNA. If both assays were
negative, no further testing was required. Agents
which were positive in the sex-linked recessive
lethal assay were to be tested further in the
mouse visible specific locus test: those which were
positive in the rodent dominant lethal assay were
to be tested in the rodent heritable translocation
assay. Agents which were positive in either of
these two assays (specific locus and/or heritable
translocation assays) would be presumed to be
FOflMER OK CHflOMOeOME MUTATION TEST
biVfto
Trigger
Fig. 3. Fonwr OTS chnwrnpflw mittiioa tc*t KtaM.
-------
potential human mutagens as outlined in the
USEPA's mutagenicity risk assessment guide-
lines.
Several factors entered into the choice of tests
in this scheme. First, it was deemed necessary to
test independently for both gene mutation and
chromosomal effects in the event that agents ex-
isted which were specifically gene or chromoso-
mal mutagens. Second, it was decided to use only
those tests which measure defined genetic end-
points. Tests such as the sister chromatid ex-
change or unscheduled DNA synthesis assays,
which either do not measure a defined genetic
endpoim or for which the specific endpoint of
concern is not known, were not included in the
generic test scheme. Such tests have been in-
duded in specific Test Rules if existing data
suggested that they might be sensitive indicators
of genotoxicity for the chemical in question.
The OTS also uses short-term muiagenicity
test data as an indicator of potential cartino-
genkity. In 1982. certain key tests in the Section
4(a) test schemes were selected as triggers to a
2-year carcinogenicity bioassay. At that time, a
positive response in certain tests served as a
trigger to a cancer bioassay:
(1) A positive response in both the Salmonella
and the Drosophila sex-linked recessive lethal
assays; or
(2) A jtngir positive response in:
(a) the •"•••«"«•"•• cell culture assay for gene
(b) the in vitro assay for chromosomal aberra-
tions; or
(c) the in vivo assay for chromosomal effects
(either chromosomal aberrations or micronudeus
formation).
To daM, the OTS has not required a 2-year
bioassay based solely on the results of short-term
mutagenkfcy tests.
Section S of TSCA requires that manufactur-
ers and importers of "new" chemicals submit a
Premanufacturing Notification (PMN) to the
USEPA 90 days before beginning the manufac-
ture or import of the new substance. By "new"
, TSCA means those chemicals not ap-
Section S does not require that submitters
conduct toxiciry testing prior to submission of the
PMN, only that they submit such data which are
either in their possession or readily obtainable by
them. Therefore, the OTS has developed tech-
niques for mutagenicity hazard assessment which
can be used in the presence of few or no test data
on the substance itself. This approach involves
the following three components:
(1) Evaluation of available toxiciry data on the
PMN chemical, if any.
(2) Evaluation of test data available on sub-
stances which are analogues of the PMN chemi-
cal, or of data which are available on key poten-
tial metabolites or analogues of the metabolites.
(3) Use of knowledge and judgment of scientific
assessors in thf interpretation and integration of
the information developed in the course of the
In general, mutagenicity data are used for three
purposes under Section S: (1) as part of exposure
based testing; (2) to assess the potential of the
PMN chemical to induce heritable genetic ef-
fects; and (3) as part of the weight-of-evidence
that a chemical may be a potential carcinogen.
Recently, as part of the exposure based testing
program, where the USEPA requires testing of
certain high volume chftmcitt which reach a trig-
ger for occupational'or consumer use exposure,
the requirement has been for a two test battery of
the Sshnoire11* assay and a mouse mkronudeus
pearing on the Inventory of Existing Commercial
Chfmkrslf, which was originally compiled in 1977
and is continuously updated.
Because of the nature of the PMN assessment
process, which relies heavily upon the use of
analogue data, and because of limitations in the
sue of the data base of chemicals tested for
heritable genetic effects, concern for a chemical's
ability to induce heritable gene or chromosomal
mutations is rarely supportable under Section (S)
Therefore, the principal use of mutagenicity data
under Section S has been as part of the weight-
of-evidence for carcinogenicity.
In supporting a concern for potential carcino-
genicity of a PMN chemical, the USEPA will
generally cite data on an analogue which is known
to be carcinogenic (Len demonstrated tumor
forming ability in one or more animal studies). In
such intTun^i mutagenicity data on the PMN
chemical or on the analogues) are used to lend
-------
support to the case for potential carcinogenicity.
Where there is no analogue of the PMN chemical
which has been tested for carcinogenicity. muta-
genicity data alone are generally not considered
sufficient to support a concern for potential car-
cinogenicity. Regulatory action is seldom, if ever.
taken on the basis of mutagenicity data alone.
especially on the basis of in vitro mutagenicity
data.
Mutagenicity data have been required as part
of several Section 5(e) notices. Initially, these
requirements were primarily for Salmonella test
data. For certain classes of chemicals, the re-
quirement was for in vitro gene mutation data:
specifically for data from the L5178Y/TK*'-
mouse lymphoma system. More recently, the
USEPA has begun requiring certain manufactur-
ers to submit data from a battery of tests which
has included the Salmonella assay and an in vivo
assay for chromosomal effects, especially the mi-
cronudeus assay.
Where an appropriate carcinogenic analogue
has been tested in the same assays as those
required for the PMN chemical, this agent is
generally-included in the test as a positive control
chemical. In some cases, where activity of the
analogue chemical in short-term testa was not
known, the USEPA has required the simultane-
ous testmg of ooth the M^ chemical and the
analogue. Positive results for both the PMN
chemical and the analogue an used to support
the weignt-of-cvidtnc* that the PMN chemical
may be a carcinogen. In these instances, a 2-year
bioassay of the PMN chemical or the use of
protective equipment to limit cnosura arcjencr-
ally required.
Negative results for the PMN chemical in the.
face of positive results for the analogue, arc taken
as an indication that the PMN chemkal is praba-
My nofMaranopaie; or • cases when the anal-
ogy may have been doubtful that the analogue
was not appropriate. In cither case, concern for
potential carcinogenicity is lessened as a result of
mutagenicity data and a 2-year bioassay is gener-.
ally not considered necessary.
In those instances when the analogue chemi-
cal is inactive in short-term tests for mutagenicity.
negative results for the PMN chemical do not
alleviate concern for potential carcinogenicity.
Because of the cost of conducting a long-term
cancer bioassay, such a requirement is often in-
terpreted by the regulated industry as a de facto
ban. Chemicals subjected to the requirement for
a long-term cancer bioassay are often withdrawn
by the submitter because they cannot support the
cost of testing. The USEPA hopes that judicious
and reasonable use of short-term testing will in-
crease in the Section S process and that this
increase in the use of short-term testing will
reduce the number of chemicals subject to a
Rttixd teaatf taftcry
As stated above, in 1962 the OTS designated
certain key tests in the TSCA Section 4 muta-
genicity test schemes as direct triggers to a cancer
bioassay. Since than^jdata ^rom the Agency's
Gene-Tot Program and the NTP Testing. Pro-
gram have stimulated discussion on the predictive
ability of some short-term tests. These data were
the subject of discussion at an Agency sponsored
Workshop on the Relationship Between Short-
Term Test Information and Carcinogenicity ("the
WOliamsburg meeting") (Auletta and Ashby.
1968: Kier, 1968). As a result of data presented at
this workshop and subsequently pubiishtd in the
scientific literature (Tennant et at, 1967). the
OTS hat proposed revving its mutagenicity test
schemes and the tests which'serve as triggers to a
cancer bioassay. These revisions are shown in Fig.
4.
The major change found in the revised test
scheme is in the first tier when ft is proposed to
combine the test schemes for gene mutation and
chromosomal aberrations. The revision includes
three tests for the first tier the Salmonella assay.
an in vitro assay for gene mutation and an in vivo
assay for chromosomal effects which may be ei-
ther a bone marrow assay for chromosomal aber-
rations or for nucronudci formation.
Under the provisions of TSCA Section 4 the
USEPA has the right to require a cancer bioassay
based upon "may present an unrea-
sonable risk" criteria such a* structure-activity
relationship (SAR) information and/or produc-
tion or release data. If accepted after proposal
and review of pubUc comment, the revised scheme
would be used in instances where the USEPA
-------
makes a policy decision to trigger carcinogenicity
testing from muugenicity test data. In these in-
stances, the situation vis-a-vis the use of short-
term tests to trigger a bioassay would be as fol-
lows.
A positive response in all three first tier tests
or a positive in the Salmonella assay and the in
vivo assay for chromosomal effects or a positive
response in the in vitro gene mutation assay and
the in vivo assay for chromosomal effects would
lead directly to a 2-year bioassay. Although there
would still be an automatic trigger to a bioassay,
it would be dependent upon a minimum of two
positive responses, at-least one of which must be
in an in vivo assay.
Any other combination of responses, including
a Jtflgfe positive response in any one assay, or a
positive response in both the Salmonella assay
and the in vitro assay for gene mutation, would
result in a "data review." The data review, which
would occur before a decision was made to re-
quire further testing, would consider all available
information including other test results, struc-
ture-activity relationships, production volume
and exposure figures.
The in vitro cytogenetics assay would no longer
be pan of the test scheme.
At one time, the OTS considered removing the
in vitro assays for gene mutation from the test
scheme as well. However, that position has been
reconsidered in the light of unofficial comment
on that proposal and as a result of work which
has been performed since the Williamsburg meet-
ing. For now. these assays would remain pan of
the test scheme but they would no longer serve as
single test triggers to a bioassay.
The Drosophila sex-linked recessive lethal as-
say would be replaced by other assays to assess
interaction with mammalian gonadal DNA. These
other assays include some combination of un-
scheduled DNA synthesis, chromosomal aberra-
tions, sister chromatid exchange and alkaline elu-
don. ail in testicular cells, and the dominant
lethal assay.
The Drosophila test would no longer serve as a
trigger to a bioassay.
The OTS proposed revisions to the first tier of
the Section 4 mutagenkity test schemes are not
substantially different from the OPP initial bat-
tery. The OTS has chosen not to designate a
CURRBff OTS MUTAGEMOTY TEST
m vitro
+ InVh/oBor*
Marrow GytOQanatica
•Abarrattona
or
•Mfcronudaus
Dominant Lathal
1
Haritabla Tranatocation
•Viaibla
or
• Biocharnical
Ftf. 4. Currtm OTS nmtutaicity wai
-------
specific test as the assay of choice for in vitro
gene mutation studies preferring to make a deci-
sion about use of a particular assay at the time of
promulgation of a Section 4 test rule. However.
the OTS agrees that for routine screening pur-
poses, chemicals should be tested in either the
L5178Y mouse lymphoma or the CHO ASS2
assays.
It is anticipated that no further testing would
be required for the majority of chemicals which
are negative in all three first tier tests. However.
where exposure data, structure-activity relation*
ships or other factors indicate it is warranted.
these agents may also be subject to a data review
and subsequent testing in a cancer bioassay.
Further actions beyond toting in initial tier
Once intrinsic mutagenicity has been identi-
fled, the revised scheme requires assayis) that
assess the ability of the chemical to interact with
DNA in the gonad. This agrees with the princi-
ples presented in the USEPA Mutagenicity Risk
Assessment Guidelines (see Appendix Ah i.e.. in
vivo tests are weighted more heavily than in vitro
tests, mammalian systems note heavily than non-
mammalian systems, and gem cell assays more
heavily than somatic cell assays. Examples of as-
says that assess interaction with gonadal ONA
are in vivo unscheduled DNA synthesis (UDS).
alkaline elution (AETX sister chromatid
or.chromosomal aberration assays in testicular
tissues, or the rodent dominant lethal assay.
There are presently no practical assays to eval-
uate potential gene mutagenicity in vivo in the
mammalian gonad. Results m testicular UDS and
AET assays show good correlation with the spe-
cific locus assay, based upon review of the USEPA
Gene-Tox data bate (Bendey et aL, 1991). This
scheme has already been used on one occasion in
the Test Rule praam It ha been presented to
the scientific community on several occasions in
open meetings, tad has met with general ap-
proval. Removal of these assays from the test
scheme would necessitate an immediate trigger to
a specific locus assay from a positive response in
an in vitro assay (Salmonella assay or mammalian
In addition, although present evidence indi-
cates that chemicals positive in the heritable
translocation assay are positive in the specific
locus assay as well, there is no assurance that all
gonadal chromosome mutagens will be gonadal
gene mutagens. There is evidence from testing in
somatic cells that such agents may exist (e.g.,
acryiates: Moore et aL. 1989).
At present, there are no changes in the third
tier of the scheme, except that the gene mutation
scheme now includes a mouse biochemical spe-
cific locus test as an alternative to the mouse
visible specific locus test Agents which are posi-
tive in the second tier gonadal DNA assayU) may
be tested in either specific locus assay; the choice
will be left to the regulated industry. It is antici-
pated that a data review, similar to that men-
tioned above for the cancer bioassay, will be
performed for those*lgents which may require
testing in either the specific locus assay or the
rodent heritable translocation assay.
cells in culture), a requirement deemed to be too
costly in the absence of some evidence of gonadal
effect.
One of the most frequent inquiries the OPP
receives addresses the lack of specific guidance
on how the OPP expects the mutagenicity tests to
be performed. It is not the purpose of the Subdi-
vision F guideline revision effort to formulate
specific protocols for eachjnutagenicity test Such
guidelines already exist, drafted by the OTS
(USEPA, 1985, 1987). The OPP felt it is appro-
priate at this time to provide specific guidance to
registrants for the conduct of mutagenicity tests.
Thus, the OPP now recognizes the published
OTS protocol guidelines as ones the OPP will
follow in reviewing submitted studies. Adoption
by the OPP of the OTS protocol guidelines allows
harmonization between the two offices? a com-
mon set of protocol guidelines for mutagenicity
testing is therefore provided in response to muta-
genicity testing requirements.
Guidance for the performance of mutagenicity
testing is found in the 40 CFR Part 798 - Health
Effects Testing Guidelines, Subpart F - Genetic
Toxiciry (issued annually in the CFR; also pub-
lished in USEPA. 1985. 1987). These guidelines
have already undergone extensive public review
and are periodically revised when appropriate to
-------
reflect the current state of the science for each
test. Submitters therefore should be aware of
updated protocols before initiating testing for
submission to the OPP and the OTS. Where no
specific guideline is given, submitters are advised
to discuss with the Agency proposed methods for
the chosen test to ensure suitability of the test
and acceptability of data.
All testing submitted to the Agency needs to
be performed under the requirements of the Good
Laboratory Practice (CLP) Standards. The CLP
standards for TSCA requirements are found in
the 40 CFR Part 792 - Good Laboratory Practice
Standards (USEPA. I989a). The GLP standards
that satisfy FIFRA requirements are found in the
40 CFR Pan 160 - Good Laboratory Practice
Standards (USEPA, 1989b>. The GLP standards
are harmonized between both offices.
We would like to express our sincere thanks to
Drs. Irving Mauer, Richard Hill. David
Jacobson-Kram. Reto Engler. Penelope Fenner-
Crisp, Vicki Dellarco. Lawrence Valcovk and
Michael Waters for their many contributions to
this effort and for critical reading of this material.
AppaadU A: Swuuiy USEPA
risk
The USEPA published its Mutagenicity Risk
Assessment Guidelines in 1986. These guidelines
provide guidance for assessing evidence for chem-
ical interaction in the gonad and evidence that
would contribute to the weight-of-evidence for
potential human germ cell mutagenicity. They are
briefly summarized here. For full discussion, re-
fer to the published guidelines (USEPA. 1986a).
According to the guidelines, there are two
categories of evidence for chemical interaction in
the gonad:
(1) Sufficient evidenct of chemical interaction
is given by the demonstration that an agent inter-
acts with germ cell DNA or other chromatin
constituents, or that it induces such endpoints as
unscheduled DNA synthesis (UDS), sister chro-
matid exchanges (SCE), or chromosomal aberra-
tions in germinal cells.
(2) Suggestive evidence includes the finding of
adverse gonadal effects such as sperm abnormali-
ties following acute, subchronic or chronic toxic-
ity testing, or findings of adverse reproductive
effects such as decreased fertility, which are con-
sistent with the chemical's interaction with germ
cells.
There are eight categories of evidence that
contribute to the weight-of-cvidence for potential
human germ cell mutagenicity. These are. in or-
der of decreasing strength-of-evidence:
(1) Positive data derived from human germ cell
mutagenicity studies.
(2) Valid positive results from studies on heri-
table mutational events (of any kind) in mam-
malian germ ceUs.
(3) Valid poftOve resflts from mammalian germ
cell chromosome aberration studies that do not
involve transmission from one generation to the
next.
(4) Sufficient evidence for a chemical's interac-
tion with mammalian germ cells, together with
valid positive mutagenicity test results from two
assay systems, at least one of which is mammalian
(in vitro or in vivo). The positive results may both
be for gene mutation or both for chromosome
aberrations; if 'one is for gene mutations and the
other for chronioeome, aberrations, both must be
from tnarnmt*'"* systems.
(5) Suggestive evidence for a chemical's inter-
action with minrnMJJM gem cells, together with
valid positive 'mutagenicity evidence from two as-
say systems as described under 4, above. Alterna-
tively, positive mutagenicity evidence of less
strength than defined under 4, above, when com-
bined with sufficient evidence for a chemical's
interaction with mammalian germ cells.
(6) Positive mutagenicity test results of less
strength than defined under 4. combined with
suggestive evidence for a chemical's interaction
with m»«!HM|iM germ cells.
(7) Although definitive proof of non-mutagen-
icity is not possible, a chemical could be opera-
tionally fUttifif* as a non-mutagen for human
germ cells if it gives valid negative test results for
alt endpoints of concern.
(8) Inadequate evidence bearing on either mu-
tagenicity or chemical interaction with mam-
malian term cells.
-------
Appendix B: Scientific Advisory Ptntl and public
comment process far OPP revised guideline
Before a revision to an OPP guideline is final-
ized and issued as the guidance for testing under
FIFRA. it must be presented to the OPP's Scien-
tific Advisory Panel (SAP) and the public tor
comments. The SAP is an external peer review
group that meets periodically to provide expert
analysis and opinions on scientific decisions the
OPP will make. Once the SAP and the public
provided comments on the proposed guidelines.
the comments were analyzed and addressed be-
fore issuing the final Subdivision F guideline. A
proposed mutagenkity guideline was presented
before the SAP. on September 29. 1989 and the
public comment period was opened with the is-
suance of a Federal Register notice that an-
nounced both the SAP meeting tune and the
availability of appropriate documents for exami-
nation (USEPA. 1989c). The. Federal Register
notice also provided for a public comment period
from August 25,1989 to September 12,1989: this
was subsequently extended to October 31,1989.
The proposed revision to the Subdivision F
guideline was rigorously discussed during the SAP
meeting and in comments received from the pub-
lic. These considerations weighed heavily in the
formulation of the final Subdivision F guideline
revision. The single issue that generated the most
discussion was the original proposal to rely exclu-
sively on the mouse tymphoma assay as the in
vitro tMiMMfaii gene mutation assay. As noted
in the final revised guideline and in response to
the SAP -and the public comments, this prefer-
ence had been changed to allow a choice to
satisfy .this testing requirement. Other aspects of
the Subdivision F guideline wen also discussed
and are detailed Mow.
The report of the SATs iteomBendations ad-
dressed nine tenant awes on the FIFRA pro-
posed revised mutagenirity testing guidelines (re-
port issued via the SAP Executive Secretary. Oc-
tober 16,1989; to obtain copies, address inquiries
to the SAP Executive Secretary. Office of Pesti-
cide Programs). The nine issues wen:
(1) LS178Y as the preferred assay for gene muta-
tions in nwTHUftitn cells.
(2) Omission of the in vitro cytogenetks assay.
(3) Dosing limits.
(4) Basis for positive controls.
(5) Three tests versus two tests.
(6) Negative (nonsotvent) controls in in vitro tests.
(7) Aroctor as the preferred enzyme inducer.
(8) Mouse peripheral blood mkronudeus (MN)
test
(9) Test results to support proposed test scheme.
A total of 14 public commentors submitted
written comments on the proposed mutageniciry
testing guidelines. In addition to the nine general
issues raised above, the public comments raised
six additional issues for consideration. These
were!
(1) Requirement for confirmation of in vitro as-
says.
(2) Discontinue "other genotoxk effects" cate-
gory from current scheme.
(3) Provide more guidance oaf test protocols.
(4) Guidance to assist in cardnogenicity classifi-
cation and mathematical modelling
(5) Use of biochemical specific locus assay.
(6) Tuning of the review of genetic toxiciry tests.
Specific responses and comments to the SAP
and public comments are presented below.
LSlTSYas tht prtftmd assay for jew mutations
in mammalian ceflt
The preference for the mouse tymphoma assay
to satisfy testing with thejnammalian gene muta-
tion assay was the issue that generated the great-
est amount of comment from the public as well as
from die SAP. Among the points brought for-
ward, as discussed by the SAP, were:
(a) "...theassay responds to certain chemicals
that an not recognized at this time to be
mutagenic in other systems. For this reason,
responses to chemicals or conditions of un-
known or unverified mutagenkity in LS178Y
cannot be concluded, with a sufficient degree
of certainty, to be .evidence of mutageniciry or
of potential hazard."
(b) "...the SAP foniscs difficulties in estab-
lishing the assay in new laboratories and ob-
taining consistent results."
(c) "Studies of chemicals that induce an in-
crease in TFT resistant clones nay yield im-
-------
portant information on the mechanism by
which such clones arise, but until a better
understanding of this range of mechanisms is
achieved, the L5178Y assay is not recom-
mended for EPA's preferred test for mutation
in cultured mammalian cell."
(d) "It is recommended that EPA not include
this assay in its testing guidelines; however, in
the future, if an adequate scientific rationale
can be developed for its inclusion, it should
appear with alternative mammalian assays
listed before it."
The OPP feels that all of these issues can be
adequately addressed in supporting a preference
for the mouse rymphoma assay in testing for
mammalian gene mutations. It is noted that in
the public comments, there was some support for
retaining the mouse lymphoma assay preference.
The support for this preference is detailed here.
It is well established that the mouse lymphoma
assay responds to known mutagenic agents and
that thfluorothymidine (TFT) resistant clones
identified by this assay may arise as the result of
genetic changes induced by interaction between
such agents and the genetic material (dive et ah,
1983; Mitchell et aL, in preparation). It has also
been demonstrated that some chemicals that ap-
parently should not be mutagenic have produced
activity in this assay (e.g^ set Gfone et aL. 1987;
Wangenheim and Bokstoldi. 1988X In many in-
stances, this activity in this assay could be at-
tributed to extreme test conditions. e.g« effects of
pit osmotic imbalances, and concentrations of S9
mix used. However, this is not unique to the
mouse lymphoma assay. Other in vitro assays
experience similar problems when non-physio-
logical coaditioas an employed (high osmotic
pressure, pH, etc); for example, increased chro-
mosomal aberrations due to increased osmotic
pressure (Gateway et aL, 1987).
There are two additional considerations. (1)
Test results must be carefully evaluated with
common sense with regard to what is occurring in
the assay under the test conditions. This consid-
eration should apply for all assays, but historically
seems to be miniff'Tf** when evaluating data from
the mouse tymphoma assay. In some instances,
the evaluation of test data considered a result
"positive" when in actuality it may have not have
been a biologically significant increase. This is a
matter of interpretation that is being directly
addressed in upcoming publications from leading
experts in the mouse Jymphoma assay and should
provide guidance for consistent evaluations in the
future (Gene-Tox Phase III evaluation of the
mouse lymphoma assay. Mitchell et aL in prepa-
ration; CItve et al.. in preparation). This guidance
will be available to the OPP when mouse lym-
phoma results need to be evaluated. (2) The
decision on interpreting a dear positive response
in the mouse lymphoma assay when there are
negative responses in other assays will be dealt
with on a case by case basis. In these instances.
all available data will be considered before a
decision is reached on the potential genotoxicity
of the test chemical. There is no reason to dismiss
as irrelevant a dear positive response in the
mouse lymphoma assay when there are negative
responses in other assays.
The SAP commented on the difficulty in estab-
lishing the assay and obtaining consistent results.
The mouse lymphoma assay is one of the major
mammalian gene mutation assays in routine use
(Farrow et al.. 1986) and is performed in labora-
tories all over the world. As with any assay being
newly established in a laboratory, it is not un-
usual to experienet problems with start-up and
routine performance. Also, assay systems are con-
tinually evolving and questions about proper
methodology for many of the "standard" assays
may arise over time. Similarly, techniques for the
performance of the mouse lymphoma assay are
evolving and being refined for a better characteri-
zation of activity by test chemicals. However, the
need for consistency in the performance of a
"routine" assay is understandable and efforts
have been made for optimizing the performance
of this assay (Give et al.. in preparation). The
Health Effects Testing Guidelines will include a
new. individual guideline for the mouse tym-
pnoma assay which will be published in the near
future (Gmino and Auletta. 1990).
The types of genetic damage detected in the
mouse lymphoma assay have been well character-
ized; the USEPA itself has funded a major re-
search effort towards this end. As detailed above.
-------
tk mutants appear to include presumed point
mutations (no visible alteration in karyotype or
Southern blot pattern), total tk gene deletions.
mitotic nondisjunction. translocauon. homolo-
gous mitotic recombination, and gene conversion
(Applegate et al.. 1990V As new techniques in
this area become available, it is expected that the
range of genetic effects detected by the mouse
Ivmphoma assay will be further defined. There-
fore, in comparison to other mammalian gene
mutation assays, the mouse lymphoma assay may
detect a wider range of genetic effects, thus mak-
ing it a more powerful tool for assessing genetic
activity by a test chemical. The genetic locus used
in the mouse lymphoma assay may account for its
ability to detect a broad range of effects. In
contrast, in the standard CHO/hgpn assay, the
hgpn locus is found in a hemizygous state (on the
X chromosome). The nature of this locus may
limit- the recovery of multi-locus deletions and
also prohibit the induction of homologous mitotic
recombination. Agents which exert such effects
would therefore not be detected. The tk locus is
heterozygous (located on an autosome) and thus
should be able to tolerate multi-locus deletions.
It also has been demonstrated that mitotic re-
combination and/or gene conversion can be de-
tected in L5178Y cells (Appkgate et at, 1990).
Moore et aL (1989) have reviewed the literature
and made direct comparisons of the two assays.
This analysis dearly demonstrates the inefficiency
of the CHO/hgpn assay for detecting a number
of chemicals that are detected by the mouse
rymphoma assay (e*. chemicals acting via a das*
togenic mechanism). The CHO ASS2 assay, a
relatively new modification of the standard
CHO/hgpn assay, appears capable of detecting
some of the genetic events not detected by the
standard CHO assay OWe, 1987; Stankowski and
Tindall, 1987X The OPP recognizes this assay as
a possible alterative to the mouse lymphoma
L5178Y assay. Another gene mutation assay, the
TK-6/tk assay, also appears capable of detecting
a range of events similar to that detected in the
mouse lymphoma assay (Little et aL, 1987). How-
ever, it has not been used for chemical screening
and thus is not recommended as a pan of a
routine test battery at this time.
The SAP and public comments suggested that
if the mouse lymphoma assay is retained in the
initial battery, it should not be the preferred
mammalian gene mutation assay. The rationale
behind the. selection of the mouse lymphoma
assay as the preferred assay was its ability to
detect a wider range of genetic effects than the
other "routine" mammalian gene mutation as-
says, such as the CHO or V79 assays using the
hgprt gene locus. For this reason the OPP stresses
detection of small colonies. This will be reflected
in the proposed Health Effects Testing Guide*
lines which will specify conditions for optimum
detection of small colonies. It appears that small
and large colonies induced by test chemicals rep-
resent different genetic effects (Moore et al.,
1985aj>) and these should be quantified to take
full advantage of the assay. The standard*
CHO/hgpn assay ha*4een ihtwn to not detect
many dastogenk test chemicals (Moore et aL,
1989) and would not provide as complete infer-
mation on the genetic activity of many test chemi-
cals. Then was a consistent recommendation
made in the public comments that the
CHO/hgpn assay be acceptable if it is
nied by an in vitro assay for chromosomal aberra-
In the revised Subdivision F guideline, the
OPP accepts the SAP recommendation that the
mouse lymphoma assay not be the exclusive assay
for the •*••"•"•"•• gene Imitation assay. The
OPP is providing a choice for this data require-
ment, but with the understanding that there will
not be a loss of the *f|
-------
277
In response, the OPP felt that the Salmonella
test plus the mammalian gene mutation assays as
described above arc adequate to detect in vitro
mutagenk activity. The OPP believes it is impor-
tant to supplement the prokaryotic Salmonella
test with an eukaryotic gene mutation test. It is
recognized dut the in vitro cytogenetics assay can
provide valuable information concerning the mu*
tagenic potential of a test chemical However,
there are several considerations that led the OPP
to prefer the mammalian gene mutation assays
(mouse tymphoma or CHO ASS2) as the in vino
test The genetic damage induced in
an in vitro cytogenetics assay may not be compati-
ble with cell survival. The gene mutation assays
count colonies that consist of viable cells. Muta-
tions compatible with cell survival may ultimately
be more relevant to risks that can be passed on to
future generations. Additionally, the in vitro cyto-
genetics assay may not be as quantitative as the
gene mutation assay; for example, up to 200 cells
are scored for cytogenetics versus 10* for gene
Furthermore, there is a test for structural
ul aberrations in the initial battery.
the in vivo cytogenetics assay. This was selected
for the reasons provided above (in OPP section).
With the in vivo test in the initial battery, it was
felt that the in vitro cytogenetics test would not
be at necessary at the other tests in the initial
battery. Abo. historically, whenever there hat
been a positive result in the a vtoo cytogenetics
test, an in vivo cytogenetks assay was performed
to attest the relevance of the a vtao result to the
in vivo situation. Therefore, it was decided to
perform an invivo cytogenetics assay in the initial
battery. This preference hat been voiced else-
where (Legatee and Harper, 1988; Shelby, 1988).
Concent have been iiynmsd •bout organ/tis-
sue spedfidry • vivo aad the possibility that only
limited infirmr**4"* from the in vivo cytogenetics
assay would be obtained However, decisions to
perform additional in vivo testing with other tar-
get organs or tissues can be made using all avail-
able information on the test chemical.
^\^h^BM^ JS^MJVA
UOtotf Uinta
The SAP states "In both in vitro and in vivo
assays, excessive dote levels may kad to re-
sponses that are not the result of a direct interac-
tion between the test chemical and the genetic
material of the test organism. For this reason, it
may be advisable for the guidelines to recom-
mend upper limits of test concentrations." Exam-
ples such as toxkity parameters and moltrity
considerations are provided by the SAP. In the
absence of any other limiting signs, arbitrary lim-
its of 5000 Mg/ml in vtoo and 5000 mg/kg in
vivo are figitfi*^ as common upper limits.
It is agreed that excessive dose levels and
concentrations may lead to responses which are
not due to direct genetic effects by the test chero-
icaL The suggested limits are believed to be rea-
sonable. DOM Imits ote u other factors such as
toxiciry are discussed for each genetic test in the
Health Effects Testing Guidelines (see Testing
(protocol) fttjdjpncs above). For this reason, it is
not necessary to reiterate them hi the Subdivision
F guideline (which is not a protocol guideline).
Btta far tattoo* controls
The SAP statet -the section of the guidelines
dealing with positive controls needs to be written
more dearly in order to explain the purpose of
such controls and to distinguish between recom-
mendations for in vitro and in vivo tests. The
rationale for recommending that positive controls
use the sane solvent at the test agent and the
SUM route of exposure should be dearly stated."
It it agreed that this section (positive controls in
Subdivision F guideline could be made clearer
and a extinction made between in vtoo and in
vivo testing circumstances. However, it should be
remembered that these are general recommenda-.
tions and specific guidance for positive controls
for each genetic test it found in the Health Ef-
fects Testing Guidelines (see Testing (protocol)
above).
Positive control compounds should be selected
to demonstrate the sensitivity of the test system
and, for m vitro assays, the functioning of the
metabolic activation system. In several instances.
the raviMd Health Effects Testing Guidelines for
mutagenkity wifl recommend that a positive con-
trol be selected to ensure the detection of a
minimal response. This would test not only the
functioning of the teat system, but also the ability
-------
of the investigator to perform the assay and ana-
lyze the resulting data. For. in vitro assays, the
positive control compound is usually adminis-
tered in a solvent consistent with the properties
of the compound. Ideally, but not necessarily, the
solvent for the positive control chemical will be
the same as that used for the test chemical. The
positive control may be selected in accordance
with the performing facility's historical data base
to allow comparison with previous performance
of the assay.
For in vivo assays, where it is feasible, the
positive control should be administered in the
same vehicle and by the same route as the test
chemical However, it is recognized that there are
where this would not be feasible
and positive controls administered with a differ-
ent vehicle and by a different route would be
ibte.
Three ttsu rersui two usa
There were several public
request-
ing informatioo on the use of three tests in the
OPP initial test battery instead of the use of only
two tests (Salmonella and m vivo cytogenetics
being the two most often mentioned). The SAP
preferred the three test option: "The Panel dis-
cussed the adequacy of irrnmrnf ndmg only two
tests (Salmonella plus in vivo chromosomal aber-
rations) as being sufficient for peel control prod-
ucts. While there is some evidence that a priori
thtte tftti may be ffrWtfit to identify nunt
mutagenic fhrmicats, they do not provide suffi-
cient assurance, given the postiMfcy of human
exposure, that such chemicals art not mutagenk.
Accordingly, a «"'»fr"""« of two in vino aad aa in
vivo test provide a more appiopmt* level of
nffirincff of a lack of nftWitial hazard." This
recommendation b reflected in the Subdivision F
guideline
jVcfoai* (noniokmt) cowofr fc m vitro too
The draft Subdivision F guideline in its Gen-
eral Recommendations section wnmnwruls that
assays include both a solvent control and, where
applicable, a aonsotvent negative control The
SAP states that "although useful information is
sometimes obtained from negative controls, fully
adequate tests need not include such a control It
is. therefore, recommended that negative (non-
solvent) controls not be included in the guide-,
lines." It is agreed that a negative (nonsotvent)
control is not necessary for a fully adequate test.
Arodor at tht prtftmd tnzymt inductr
The draft Subdivision F guideline in its Gen-
eral Recommendations section states that "a
metabolic activation system should be incorpo-
rated into any test system that does not provide
adequate metabolic capabilities.'' It suggests that
rat liver extracts have the greatest usage and
provides an example of such an activation system,
e.g* post-mttochondrial fractions prepared from
Arodor 1254 induced rat livers. Generally, it is
necessary to induce various enzyme activities in
order to maiinine thg^osstbi|£ty of converting a
compound requiring metaboGsm into potential
genotoric products, especially in genetic tests that
are of short duration where an underestimation
or a lack of possible metabolic activation by en-
dogenous processes may occur. The SAP states.
with regard to the inducing agent, "it is known
that the species of origin and concentration of
liver hoffKMMMfot as well as the ThTiHtral used as
an enzyme inductr, can influence the mutagenk
response of in vitro tests. It is recommended that
the EPA indicate that these variables should be
optimized where possibk
-------
priate justification and validation would be al-
lowed.
Mouse peripheral blood micronudeus (MN) test
It was suggested in the public comments (and
has been suggested to the OPP in the past) that
results from mouse peripheral blood micronu-
deus tests performed as pan of a subchronk
study (e.g.. 30-40 days exposure) be used to sat-
isfy the requirement for an in vivo cytogenetics
assay. The SAP states "the majority of available
data support this proposal but in view of the
limited data base, the SAP considers it premature
to include this proposal in the guidelines.- White
the Agency agrees that the technology to perform
the mouse peripheral Wood nucronuckus tests
exists, it also recognizes that the data base is
limited. Whereas the current protocol guideline
in the Health Effects Testing Guidelines (see
Testing (protocol) guidelines above) does not
covtr the periDberalbkxxi assay, a onjposed revi-
sion to the micronucteus test guideline will pro-
vide guidance on the performance of this test
However, in keeping with the current state of the
science for this assay as reflected in the SAP
reservation, the protocol guideline revision for
the mkronudeus test will emphasize the bone
marrow as the prime target tissue. The develop-
ment of a larger data base for the peripheral
blood mkronudeus assay is encouraged.
Test raults to support propoead test scheme
The SAP suggests that an analysis of the avail-
able data from short-term genetic oodty ten.
germ cell mutagenicity tests, and rodent carcino-
genkity tests should be performed to support the
revised Subdivision F guideline. For the OPP to
perform such • detailed analysis on results from
short-term genetic toricity tests, germ cell muta-
genkiry teats end rodent cerdnogenkicy tests
would he thatt aad resource prohibitive. Numer-
ous analyses of this nature have already been
performed (e*. the USEPA Oene-Tox Program
and the NTP efforts*. The OPP has considered
.these analyses in formulating its current guide-
line.
Rtquirtment for confirmation of in vitro taays
Good scientific practice suggests that every
experiment should be verified in an independent
repeat assay. The Subdivision F guideline, how-
ever, does not deal with this question directly.
The issue of repeat assays for both in vitro and in
vivo tests is addressed in the protocol guidelines
for each particular genetic test (see Testing (pro-
tocol) guidelines).
Discontinue "other genotoae effects" category from
current scheme
It is recognized that the category "other geno-
toxk effects" covers a broad category of genetic
endpoints and that no specific guidance is given
for choke of tests in this category. This category,
as well as the other two categories, gene muta-
tions end chromosomal aberrations, are being
disconrinoed in favor of more specific guidance
for ttft selection. For full discussion, see Revised
initial battery (OPP secfion).
Provide more guidance on test protocols
The Subdivision F guideline is not a protocol
guideline; it provides guidance as to the types of
tests necessary to support a pesticide registration.
when ten should be performed relative to each
other, and how the OPP will use mutagenicity
data towards a decision for heritable risk. It is
recognized that the OPP in the pest has not
provided specific protocol guidance on how to
perform mutsgmkify -tests. With this guideline
revision, the OPP formally adopts the Health
Effects Testing Guidelines, Subpert F. Genetic
Toxicity (in Code of Federal Regulation 40, Part
79* also in USEPA, 19tS, 1967) as the protocol
(see Testing (protocol) guidelines
for fufl discussion).
Guidance to taut m cardnogenicity classification
and mathematical modelling
The use of mutagenicity data in the weight-of-
evidence approach for classifying the carcino-
genkity of a test compound is outside the purview
of the Subdivision F guideline. White this guide-
line nates tht* mutagenicity data will be used in a
classification dtcision. tptfific guidance on the
use of mutagenicity data in the classification and
modelling of caidnogenkhY test data is found in
the US. Environmental Protection Agency's
Guidelines for Carcinogen Risk Assessment
-------
(USEPA. 1986bK The OPP follows these guide-
tines. The Carcinogen Risk- Assessment guide-
lines themselves are periodically examined for
revisions: comments such as this would be appro-
priate at those times.
Use of the biochemical specific bats assay
One of the public comments expressed a reser-
vation about the use of data from the mouse
biochemical specific locus assay (MBSL) for
quantitative risk assessment. The specific locus
test (Russell et at, 1984) is one of the tests that
could be required if information suggests that a
chemical may present a heritable genetic risk and
if the Agency believes quantitation of that risk is
necessary for regulatory action. Historically the
Agency has ircommtndcd the mouse visible spe-
cific locus assay (MVSL) (Russell et aU 1981) for
this purpose. However, the MVSL is not widely
available. Also, because of the relatively small
number of chemicals targeted for testing in the
MVSL. laboratories have not set aside the re-
sources to develop the assay and to establish a
historical control data base. At a viable alterna-
tive to the use of the MVSL, the USEPA exam-
ined the use of the MBSL to detect chemicals
which elicit gene mutations in mammals (USEPA.
1988). With both tests, mutations can be detected
at two important states of development in the
male sperm cell, the spermatofonial stem cell
state and the postspermatogonial state. The
USEPA hat concluded that the MBSL is as ac-
ceptabk at the MVSL for determining the poten-
tial of a chemical to elicit heritable gene muta-
tions in mammals (for rod rationale, see USEPA,
1988, 1990). A testing guideline for the MBSL
has been published and incorporated into the-
Health Effects Testing Guidelmes (USEPA.
1990). Although the testing guidelines for both
specific locus eeMyt do not specify the use of
more than one dote level, it thould be noted that
for the purposes of qusnritarion for risk assess-
ment, more than one dote level win be required.
ITmvig of the mifw of emetic ttcfcfiy tests
The proposed guideline describes a decision
point after review of available genetic toxiciry
tests. At that time, the OPP will determine if
there is a concern for genetic toriciry and/or if
additional testing is necessary. Some commemors
suggest "... that EPA review the initial battery of
genetic toxiciry studies as soon as the reports are
available. This would allow OPP to review the
data while long-term feeding studies are being
performed and if additional genetic toxicology
studies are required, then they could be per-
formed concurrently with other studies." They
recommend that the OPP -review the genetic
toxicology results as soon as they are available
rather than waiting for an entire registration
package to be completed"
It is outside the purview of the Subdivision F
guideline revision effort to set times and dead-
lines for the •Mlri— {M and review of genetic
toxicology tests. The times for submission of ge-
netic toxicology tests, at well as for all toxicity
testing, are determineiLby FlfRA, Registration
Standard! for each chemical, or specific regula-
tory actions taken by the OPP, at in the registra-
tion of new rhemirah. Special Reviews, and Data
Call-in Notices. For further information on the
timing of tnhnitf^^T, one should contact the
appropriate OPP DMsioa (ej« Registration Di-
vision. Special Review and Raregistration Dhri-
sionX Once the genetic toxicology tests have been
submitted, the OPP win review the results and
according to in mandates Deci-
sions from this review process should be timely
and not cause undue delay in the regulatory
BK. FJX Lao art VX-Damt (1*73) Aa •t-io U
imapai art camaopai, Hoc. Nad. Acrt. So. (USA.).
*.7l2-m
ataaM. ML. MJtf. Moor*. CB. Birtar. A. BantU. C.
Jake. KJ. toiict. F.-F. LM. A. Wadhan art J C
(1990) Motacskw dkMctioa of •watioM at tha
to
cab. fioe. Nad. Acrt. Set. (U1A.K 17. SUSS.
AritttCF, art J. GoMlttMTfai rote of aiiailiin call
tof. Mataaaaaaa, 3,4SS-4SI.
Ani. P.. J. Aikfty. 1 CanaUteo. O. faaiftanlr. BA. Har-
taU. «-AJ. Fitaoa art WJ. BoMtek (19M) Ajaaaneot
of tba poiaatial jam caB mttftftairity of irtamial «nd
pteM pratactiM ctaBfcab ai pan of aa ianemad nudy
of naoiortriiy • vtoo art « wa Mutttioo Ra*. 1Q3.
177-114.
-------
\shby. J. . lfll!> il, •. Myhr. V. Ray. A.E.
Aulana, J. Cote. P. Proy. R. Combu, K. Dearftetd and J.
HarbaaV •avaaHwadaooas for tha partonaaace and eval-
uaooa of oa UlTIY/if'- - ft''' awiai tymphoma
assay. Ma«a»rta< ai
OearfiaU. U. (l«l> •otaadal EPA/OTP mataaenidty last-
ing imHauan - gwdaNaaa raviama. .Eavitoa. MoL
Mutagasu 14 (S«aat 15X47.
OeUarco. V U p E. Voyiek aad A. Ifniiiailir (Ed*.) (19M)
Aneuptoidr- Etiology aad tintiiiiaa Bask Life Set-
ences. VoL 36. Ptenam. Haw York aad London.
Diggw. D.R. and D.R. FWdar (1989) Revised gvideliacs of
UK Committee on Mutagaaioty (COM) 1989. Enviroa.
MoL Mutaaaa, 14 (SuppL 15X 49.
Ehling. UH. (1988) Quantification of the genetic risk of
environmental muttgens. Risk Anal.. 8. 43-57.
Ennever. Fit. and H.S. Rosenkraiu (1988) Influence of the
proportion of carcinogens on the coat effectiveness of
short-term tests. Mutation Res.. 197. 1-13.
Farrow. M.C.. N.E. McCanoll and A.E. Auletta (1986) 1984
Survey of genetic toncotogy testing in industry, govern-
ment and academic laboratories. J. Appl. Toucol.. 6. :i 1 -
233.
Fox. M. (1981) The case for retention of mammalian cell
mutaganiciiy assay*. Muugenesis. 3. 439-461.
Galloway. S.M.. O.A. Deasy. CL. Bean. A.R. Knynak. M.J.
Armstrong and M.O. Bradley (1987) Effect* of high os-
mouc strengtli oa chroaneome aberrations. sister*chro-
matid •*r*t"ia** and DNA strand breaks, and the relation
to toricny, Mutatioa Res^ 189,15-23.
Garnar. R.C. aad DJ. KirkUad (1986) Reply. Mutagenesis.
1.233-235.
Gatehouse, D.G., aad DJ. Twean (1986) Utter to tha Editor.
Mnttataaaa. 1.307-308.
George, E- & Barfflata and£. Gatahouse (1989) Geaotonc.
ity of 1- aad 2-mnopraoaae ia the rat Carcinngiiiawj. 10.
232?r2334.
Health aad Welfare Canada (1986) Guidelines oa tha aw of
mutagaiisciryie*tsuithetc«k»lc«icalevahiatioaofc5emi-
cak. A report of the DNH ft W/DOE Eavironmeatal
runnmiasati Advisory Committee oa Mvtaganesis. Pub-
lished by Awhoriiy of the Mtaietar of National Health and
Wetfua and the Miaieter of the Eavironment. Ottawa.
Heddte. JJL. M. Htt. & Kirkhait K. Mavourwn. J.T. Mac-
Gtaaar. O.W. Newell aad MJ. Tiliainii (1983) The
i of miaoaadss at a meaiare of genotoooty. A
of tha UJ. Eavtroaejeatal Proiactioa Agency
Oeaa-Toi PrograB. Maniioa Rau 123.61-118.
HO. IL (197W lauedwam ia: VJL McBaaay aad S. Abra-
i (Eds.X Biatam Rapon 1 - Aiasseiag Oteaucal
Taa Riak » Hamgeji. Cold Senas Harbor
Ubomory. Cold Settag Harbor. NY. pa. 1-5.
Hate. A.W. (198DTBa>iee of thehgpTtven» gat lotus for
DM- Oa MarM PJ. da Serres aad K.R. Tiadatt (Eds.).
Baabary Report 28. Mammafaa CeU Matapaeiis Cold
Spriaf Harbor Laboratory. Cold Spnag Harbor, NV. pp.
37-46.
Kicr. L.D. (1988) Comments aad perspective oa the EPA
workshop oa -The lUlstioashiB Between Shan-Term Test
Informatioa aad Caniacasaicinr. Eaviroa. MoL Muta-
gasu U. 147-157.
Kiar. U>, DJ. Braga*. A£ Aastaa. ES. voa Hal*. M.M.
VJ». Semam. V. Daatal J. Mcdaa. X. Monel-
TJL y*1?-"!?--^.1^.-1?1"Th-
repart of *• UA Ea^voaaHatal Pruaectioa Agency
Lam LB, aad OA
short*
29-34.
(1986) Cost-eflactiveneu of
r. Namia (Loedon). 32^.
-------
:s2
Lejator. M S. ind B.L Harper (1988) Muia«emciry screen-
ing, m vuro testing - the end of an era: animal and human
studies - the direction for the future, m: C. Maltoni and
I J. Sehkoff (Eds.). Living m a Chemical World. Occupa-
tional and Environmental Significance of Industrial Car-
cinogens. Annals of the New York Academy of Sciences.
Vol. 534. The New York Academy of Sciences. New York.
pp. 837-444.
Little. J.B.. D W Yandell and H.L. Liber (1987) Molecular
analysis of mutations at the tk and *fpn loci in human
cells, m: MM. Moore. D.M. De Manni. FJ. de Serres and
K.R. Tindall (Eds.). Banbury Report 28. Mammalian Cell
Mutagenestt. CoM Spring Harbor Laboratory. Cold Spring
Harbor. NY. pp. £5-236.
Maron. DM., and B.N. Ames (1983) Revised methods for the
Salmonella mtttageniciiy test Mutacioa Rcs~ 113.173-215.
Mavournin. K.H.. O.H. Blakey. M.C Omina M.F. Salaaoae
and J.A. Heddte (19W The ia wo bone marrow rmcnxm-
cleus assay in mammaliaa boat marrow and peripheral
blood. A report of the U.S. Environmental Protection
Agency Geite-Toi Program. Mutation Res, 239. 29-80.
Mitchell. A.D- A.E. Autetta, O. dive. P.E. Kirby. M.M.
Moore. B.C Myhr and ICH Mtwjunin. The L5I78Y
mouM rymphoma tk*'~ cctt forward nutation assay. A
phase III report of the UJ. Emiroaaiiaul Protecnoo
Agency Geae-To* Program. Manuscript ia preparation.
Moon. M.M, D. Oh«. B.E. Howard. A.G. Baaoa and N.T.
Turner (198Sa) In sin analysis of trigaorothymidine-re-
sistaat (TFT") tmtaats of LS178Y TK"' mouse Km-
phona cells. Mutation Rat, 151.147-159.
Moore. M.M_ D. CXva. J.C Hoaar. B.E. Howard. A.G.
Batson. N.T. Tanm and J. Sawyer (I985b) Aaalyw of
trifhiuiuiliiiiiiiJiM nsnuim mutants of U1T8Y/TK<*/-
movta tynpaoatt cask. Mutation Rat, 151.161-174.
Moore. M.M. K. Hiiilagam Brn<>, CL Doart and ILL
DaarfiaU (1989) Dtfteaatial mwant qjianrirartoa at the
394-403.
Preston. RJ- W. Au. MA Bender. J.O. Brewen. XV. Car-
rand J.X Heddte. XP. Mrfea. 1 Wottaad IS. Wassom
(198t) I iinMniliaa In vivo and • vido ryMpaiiir«"»«
a report of the UA EPA's Gana-Toi Prafna. Mutation
Rat, 87.143-188.
Rhomberg. L, V L Deaarco. C Tiinal Tmrr. KJ. OaaraaU
and 0. Jacobeon-Kna U99» A «MMStatt*a eetisaation of
the aenetic .risk atMnBatf wtt tka i
16.104-125.
Rinkuf. SJ, an
tion of 465
, —•••»:riwi
COffTSanUIWI
' Rat, 39.3289-3318.
aadL.
RioattL L&. P.B. Safty. E. »on Haaa, W.
Vakavtc (1981) The mettM specifie-loaii
otnar than radiation: interpretaooa of daa and
mandaisoai for futwe work. Mutation Rat. 86. 329-354.
Ru$s«ll. L.B.. C.S. Aaron. F de Serres. W M. Ceneroso. K.L
(Carman. M. Shelby. /. Springer and P Voytek (1994) A
report of the U.S. Environmental Protection Agency
Gcne-Tox Program. Evaluation of mutageniciiy assays for
purposes of genetic risk assessment. Mutation Res.. 134.
U3-157.
Schmezer. P.. B.L. Pool. PA. Lefevre. R D. Cailandcr. F.
Ratpan. H. Tinwell and J. Ashby (1990) Assay-specific
genotoxKuy of .V-nirrosodibenzylamine to the rat liver m
viva Environ. Mol. Muttgen.. 15. 190-197.
Selby. P.B. (1979) Induced skeletal mutations. Genetics. 92.
S127-S133.
Shelby. M.D. (1988) The genetic toncity of human carcino-
gens and its implications. Mutation Res, 204. 3-15.
Stankowsta. L.F. Jr.. and K.R. TindaU (1987) Cbanctarizatioa
of the ASS2 cell line for use in mammalian cell mutagene-
sis studias. UK M.M. Moon. D.M. Da Manni, FJ. de
Serres and K.R. TiadalMEdi.). Banbury Report 28. Mam-
malian Can llnnpniiii, CaU Sprttuj Harbor Laboatory.
Cold Sprint Harbor. NY, pp. 71-79.
Tennant R.W.. B.H. Marnotin. MJ>. Shefcy. & Zataar. J.K.
UMA—^bA^ f C^AlriilM^ ^If frn^^m*^ ^M B^^AA^k C
najeoMM. j. apanMsa, w. ^anpary. HI. Kesasca. *-
Sta>iewkx.&And^nonandR.MmOT(19rnPredictiooof
< hemnsal cafononasnccty ui lodentt from in vitro asnetic
tonaty assay*. Science. 236.933-941.
UNSCEAR (United Nations Scientific Comnmtaa on the
Effects of ANNasc Radiation) (1977) lonoiag radiatioa:
levels and eflsca. Report to the General Assembly, United
Nations. New York. VoL 2. p. 8.
USEPA (1978) AddendM m - Criteria far cvataating the
Rea>. 43.37400-37403.
USEPA (1982)
F. Haaard
DC EPA.540/9424B5.
USEPA (1915) Put m - Health Eflaca Tesong OwdatiMa.
Swbpan P - OynatttTopciry,Fed. fjii.m 394S5-39451
USEPA (19Ha) Owidaiinai fir nMtasjenseln/ riak
Fed.Rea-51.3400»-340l2,
USEPA (19MM OuHiHaai far careaonn risk
Fed. R«f. 51.33992-34008.
USEPA (I9tn Revwion of TSCA test gairtiliaii Fed
32.19078-19MI.
USEPA (1988> MOWN visMa specific lorn taat requirements:
51856.
USEPA (1989a) Toric 5»lm«rai Control Ad (TSCA); Good
Labomory Practice Standard* Ptanl Jtala. Pad Reg. 54.
34034.34049.
USEPA <19HW Petfaral laiicrirlifci, renniriiii and Rodanti-
ode Act (FmAk Good Laboraeory Practka Standards;
Fuul Rate. Pad Re*. 54.340S2-34074.
-------
I'SEPA (I'We) F1FRA Scientific Advisory Panel: Open L5l'8Y thymidine kmase locus issay of 50 compounds.
Meeting. Fed. Reg.. 54, 35387. Mutagenesis. 3, 19I-:OJ.
L'SEPA (I TO) Mouse visible specific locus test requirement; Waters, M.D.. and A. Auletu (1981) The Gene-Tox program:
final imendmem m test rules. Fed. Reg.. 55. i:639-i:6*». genetic activity evaluation, i. Chem. Inf. Comput. So.. :i.
Wingenneim. J. and G Bolcsfoldi (1988) Mouse lymphoma 35-38.
-------
PAGE NOT
AVAILABLE
DIGITALLY
------- |