SAP
Minutes
No.
2005­
03
A
Set
of
Scientific
Issues
Being
Considered
by
the
Environmental
Protection
Agency
Regarding:

A
COMPARISON
OF
THE
RESULTS
OF
STUDIES
ON
PESTICIDES
FROM
1­
OR
2­
YEAR
DOG
STUDIES
WITH
DOG
STUDIES
OF
SHORTER
DURATION
MAY
5
AND
6,
2005
FIFRA
Scientific
Advisory
Panel
Meeting,
held
at
the
Holiday
Inn
­
Rosslyn
at
Key
Bridge,
Arlington,
Virginia
2
of
23
NOTICE
These
meeting
minutes
have
been
written
as
part
of
the
activities
of
the
Federal
Insecticide,
Fungicide,
and
Rodenticide
Act
(
FIFRA),
Scientific
Advisory
Panel
(
SAP).
The
meeting
minutes
represent
the
views
and
recommendations
of
the
FIFRA
SAP,
not
the
United
States
Environmental
Protection
Agency
(
Agency).
The
content
of
the
meeting
minutes
does
not
represent
information
approved
or
disseminated
by
the
Agency.
The
meeting
minutes
have
not
been
reviewed
for
approval
by
the
Agency
and,
hence,
the
contents
of
these
meeting
minutes
do
not
necessarily
represent
the
views
and
policies
of
the
Agency,
nor
of
other
agencies
in
the
Executive
Branch
of
the
Federal
government,
nor
does
mention
of
trade
names
or
commercial
products
constitute
a
recommendation
for
use.

The
FIFRA
SAP
is
a
Federal
advisory
committee
operating
in
accordance
with
the
Federal
Advisory
Committee
Act
and
established
under
the
provisions
of
FIFRA
as
amended
by
the
Food
Quality
Protection
Act
(
FQPA)
of
1996.
The
FIFRA
SAP
provides
advice,
information,
and
recommendations
to
the
Agency
Administrator
on
pesticides
and
pesticide­
related
issues
regarding
the
impact
of
regulatory
actions
on
health
and
the
environment.
The
Panel
serves
as
the
primary
scientific
peer
review
mechanism
of
the
EPA,
Office
of
Pesticide
Programs
(
OPP),
and
is
structured
to
provide
balanced
expert
assessment
of
pesticide
and
pesticide­
related
matters
facing
the
Agency.
Food
Quality
Protection
Act
Science
Review
Board
members
serve
the
FIFRA
SAP
on
an
ad
hoc
basis
to
assist
in
reviews
conducted
by
the
FIFRA
SAP.
Further
information
about
FIFRA
SAP
reports
and
activities
can
be
obtained
from
its
website
at
http://
www.
epa.
gov/
scipoly/
sap/
or
the
OPP
Docket
at
(
703)
305­
5805.
Interested
persons
are
invited
to
contact
Myrta
R.
Christian,
SAP
Designated
Federal
Official,
via
e­
mail
at
christian.
myrta@
epa.
gov.

In
preparing
the
meeting
minutes,
the
Panel
carefully
considered
all
information
provided
and
presented
by
the
Agency
presenters,
as
well
as
information
presented
by
public
commenters.
This
document
addresses
the
information
provided
and
presented
by
the
Agency
within
the
structure
of
the
charge.
3
of
23
CONTENTS
PARTICIPANTS...........................................................................................................
5
INTRODUCTION.........................................................................................................
7
PUBLIC
COMMENTERS............................................................................................
8
SUMMARY
OF
PANEL
DISCUSSION
AND
RECOMMENDATIONS...................
8
PANEL
DELIBERATIONS
AND
RESPONSE
TO
CHARGE.................................
10
4
of
23
SAP
Minutes
No.
2005­
03
A
Set
of
Scientific
Issues
Being
Considered
by
the
Environmental
Protection
Agency
Regarding:

A
COMPARISON
OF
THE
RESULTS
OF
STUDIES
ON
PESTICIDES
FROM
1­
OR
2­
YEAR
DOG
STUDIES
WITH
DOG
STUDIES
OF
SHORTER
DURATION
MAY
5
AND
6,
2005
FIFRA
Scientific
Advisory
Panel
Meeting,
held
at
the
Holiday
Inn
­
Rosslyn
at
Key
Bridge,
Arlington,
Virginia
Myrta
R.
Christian,
M.
S.
Stephen
M.
Roberts,
Ph.
D.
Designated
Federal
Official
FIFRA
SAP,
Session
Chair
FIFRA
Scientific
Advisory
Panel
FIFRA
Scientific
Advisory
Panel
Date:
July
13,
2005
Date:
July
13,
2005
5
of
23
Federal
Insecticide,
Fungicide,
and
Rodenticide
Act
Scientific
Advisory
Panel
Meeting
May
5
and
6,
2005
A
COMPARISON
OF
THE
RESULTS
OF
STUDIES
ON
PESTICIDES
FROM
1­
OR
2­
YEAR
DOG
STUDIES
WITH
DOG
STUDIES
OF
SHORTER
DURATION
PARTICIPANTS
FIFRA
SAP
Chair
Stephen
M.
Roberts,
Ph.
D.,
Professor
&
Program
Director,
University
of
Florida,
Center
for
Environmental
&
Human
Toxicology,
Gainesville,
FL
Designated
Federal
Official
Myrta
R.
Christian,
M.
S.,
FIFRA
Scientific
Advisory
Panel,
Office
of
Science
Coordination
and
Policy,
EPA
FIFRA
Scientific
Advisory
Panel
Members
Janice
E.
Chambers,
Ph.
D.,
William
L.
Giles
Distinguished
Professor
&
Director,
Center
for
Environmental
Health
Sciences,
College
of
Veterinary
Medicine,
Mississippi
State
University,
Mississippi
State,
MS
Steven
G.
Heeringa,
Ph.
D.,
Research
Scientist
&
Director
for
Statistical
Design,
University
of
Michigan,
Institute
for
Social
Research,
Ann
Arbor,
MI
Gary
Isom,
Ph.
D.,
Professor
of
Toxicology,
School
of
Pharmacy
&
Pharmacal
Sciences
Purdue
University,
West
Lafayette,
IN
FQPA
Science
Review
Board
Members
Michael
L.
Cunningham,
Ph.
D.,
Toxicologist,
National
Institute
of
Environmental
Health
Sciences
(
NIEHS),
Research
Triangle
Park,
NC
Joseph
J.
DeGeorge,
Ph.
D.,
Vice
President,
Safety
Assessment,
Merck
Research
Laboratories,
West
Point,
PA
Karen
Ekelman,
Ph.
D.,
Leader,
Animal
Feed
Safety
Team,
U.
S.
Food
&
Drug
Administration,
Center
for
Veterinary
Medicine,
Office
of
Surveillance
&
Compliance,
Rockville,
MD
Wanda
Haschek­
Hock,
BVSc,
Ph.
D.,
Professor,
Veterinary
Pathobiology,
College
of
6
of
23
Veterinary
Medicine,
University
of
Illinois,
Urbana,
IL
A.
Wallace
Hayes,
Ph.
D.,
Visiting
Scientist,
Department
of
Environmental
Health,
Harvard
School
of
Public
Health,
Boston,
MA
Ernest
E.
McConnell,
D.
V.
M.,
M.
S.,
President,
Toxpath,
Inc.,
Raleigh,
NC
Douglas
B.
McGregor,
Ph.
D.,
Consultant,
Toxicity
Evaluation
Consultants,
Scotland,
UK
Lynn
O.
Post,
D.
V.
M.,
Ph.
D.,
Director,
Division
of
Surveillance,
U.
S.
Food
&
Drug
Administration,
Center
for
Veterinary
Medicine,
Rockville,
MD
Nu­
May
Ruby
Reed,
Ph.
D.,
Staff
Toxicologist,
Department
of
Pesticide
Regulation
California
Environmental
Protection
Agency,
Sacramento,
CA
Paul
W.
Snyder,
D.
V.
M.,
Ph.
D.,
Associate
Professor,
Department
of
Veterinary
Pathology,
School
of
Veterinary
Medicine,
Purdue
University,
West
Lafayette,
IN
7
of
23
INTRODUCTION
The
Federal
Insecticide,
Fungicide,
and
Rodenticide
Act
(
FIFRA),
Scientific
Advisory
Panel
(
SAP)
has
completed
its
review
of
the
set
of
scientific
issues
being
considered
by
the
Agency
pertaining
to
a
comparison
of
the
results
of
studies
on
pesticides
from
1­
or
2­
year
dog
studies
with
dog
studies
of
shorter
duration.
Advance
notice
of
the
meeting
was
published
in
the
Federal
Register
on
March
10,
2005.
The
review
was
conducted
in
an
open
Panel
meeting
held
in
Arlington,
Virginia,
on
May
5
and
6,
2005.
Dr.
Stephen
M.
Roberts
chaired
the
meeting.
Myrta
R.
Christian
served
as
the
Designated
Federal
Official.

The
FIFRA
SAP
met
to
consider
and
review
a
comparison
of
results
from
1­
or
2­
year
dog
studies
on
pesticides
with
dog
studies
of
shorter
duration.
Under
the
current
CFR
Part
158
toxicology
data
requirements,
a
13­
week
and
a
1­
year
non­
rodent
(
dog)
study
(
guidelines
82­
1
and
83­
1)
are
required
for
all
food
use
pesticides
and
for
pesticides
with
nonfood
uses
if
use
of
the
pesticide
product
is
likely
to
result
in
repeated
human
exposure
to
the
product
over
a
significant
portion
of
the
human
life­
span.
Over
the
last
three
decades
the
Agency
has
received
the
results
of
a
large
number
of
dog
studies
in
support
of
the
registration
of
pesticides.
The
Agency
has
conducted
a
retrospective
analysis
of
dog
studies
that
provided
the
basis
for
the
selection
of
reference
doses
(
RfDs)
in
order
to
determine
whether
the
requirement
for
both
a
subchronic
and
a
chronic
dog
study
continues
to
be
justified.
The
analysis
involved
a
comparison
of
the
results
of
13­
week
studies
and
1­
or
2­
year
studies
or
a
comparison
of
interim
data
(
13­
weeks
or
less)
from
1­
year
dog
studies
with
the
data
from
1­
year
in
the
same
study.
The
Agency
solicited
comments
from
the
FIFRA
Scientific
Advisory
Panel
on
this
retrospective
analysis
of
the
results
of
dog
studies
and,
specifically,
on
whether
the
analysis
supports
the
continuation
of
the
requirement
for
both
subchronic
and
chronic
dog
studies
or
whether
consideration
should
be
given
to
a
modification
of
the
current
requirements
for
dog
studies.
The
agenda
for
this
SAP
meeting
involved
an
introduction,
background,
and
detailed
presentations
of
the
issues
related
to
a
comparison
of
results
from
1­
or
2­
year
dog
studies
on
pesticides
with
dog
studies
of
shorter
duration
provided
by
Dr.
Karl
Baetcke
(
Health
Effects
Division,
Office
of
Pesticide
Programs,
EPA).
Dr.
Tina
Levine
(
Acting
Director,
Health
Effects
Division,
Office
of
Pesticide
Programs,
EPA)
offered
opening
remarks
at
the
meeting.

In
preparing
these
meeting
minutes,
the
Panel
carefully
considered
all
information
provided
and
presented
by
the
Agency
presenters,
as
well
as
information
presented
by
public
commenters.
This
document
addresses
the
information
provided
and
presented
at
the
meeting,
especially
the
response
to
the
Agency's
charge.
8
of
23
PUBLIC
COMMENTERS
Oral
statement
was
presented
as
follows:

On
behalf
of
ZEBET
­
National
German
Centre
for
the
Documentation
and
Evaluation
of
Alternative
Methods
to
Animal
Experiments:

Rainer
J.
Box,
Ph.
D.

Written
statement
was
provided
by
or
on
behalf
of
the
following
group:

U.
S.
Food
and
Drug
Administration
David
Jacobson­
Kram,
Ph.
D.,
DABT,
Center
for
Drug
Evaluation
and
Research,
FDA
SUMMARY
OF
PANEL
DISCUSSION
AND
RECOMMENDATIONS
The
Panel
commended
the
Agency's
effort
in
conducting
this
comparison
of
dog
studies
and
encouraged
the
Agency
to
continue
their
analyses
to
better
inform
the
Agency's
final
decision
regarding
the
length
of
dog
studies
that
is
appropriate
for
reference
dose
(
RfD)
determinations.
On
the
basis
of
the
current
analysis,
the
Panel
was
unable
to
reach
consensus
regarding
the
strength
of
the
scientific
evidence
supporting
elimination
of
one­
year
dog
studies.
Some
Panel
members
thought
that
the
analysis
conducted
by
the
Agency
was
adequate
to
justify
eliminating
the
requirement
for
a
oneyear
study
in
dogs.
Other
members
of
the
Panel
were
not
ready
to
support
this
conclusion,
citing
concerns
regarding
the
Agency
analysis,
as
discussed
in
more
detail
in
the
responses
to
individual
charge
questions.
The
Panel
discussed
two
possible
paths
forward
while
the
Agency
addresses
the
limitations
in
their
analyses:
1)
retain
the
oneyear
dog
study
and
2)
eliminate
the
one­
year
dog
study
but
employ
an
additional
uncertainty
factor
until
sufficient
confidence
to
indicate
otherwise.
In
addition,
the
Panel
encouraged
the
Agency
to
engage
the
international
scientific
community
on
this
issue
(
e.
g.,
through
international
meetings
working
along
the
lines
already
pursued
by
the
pharmaceutical
regulating
agencies)
to
establish
equivalent,
internationally
harmonized
requirements
for
pesticides.

Even
though
some
Panel
members
agreed
that
the
general
approach
taken
by
the
Agency
appears
adequate
to
answer
the
question
of
whether
a
13­
week
study
is
sufficient
to
assess
the
toxicity
of
pesticides
in
dogs
for
determining
an
RfD,
they
identified
limitations
in
the
Agency
data
review.
The
following
recommendations
for
addressing
these
limitations
were
made:
1)
increase
the
robustness
of
data
analysis
by
including
dog
study
datasets
that
were
not
used
for
the
RfD
determination,
2)
conduct
an
analysis
more
representative
of
a
prospective
comparison
through
delineating
the
13­
week
NOAELs
(
No
9
of
23
Observed
Adverse
Effect
Level)
and
LOAELs
(
Lowest
Observed
Adverse
Effect
Level)
independent
of
the
1­
year
study,
and
establish
data
review
criteria,
3)
consider
data
analysis
for
separate
classes
of
pesticides,
4)
include
additional
background
information
on
RfD
that
provides
better
perspectives
for
reviewing
the
Agency
position
paper,
and
5)
revise
the
title
of
the
Agency
position
paper
to
better
reflect
the
purpose
of
the
data
analysis.

The
Agency
clearly
presented
the
analysis
of
data
for
the
19
chemicals
where
NOAELs
or
LOAELs
were
lower
in
the
chronic
dog
studies
than
in
the
13­
week
dog
studies.
The
Panel
agreed
that
the
cases
where
NOAELs
or
LOAELs
appeared
to
differ
between
the
13­
week
and
1­
year
dog
studies
detract
from
the
overall
conclusions
regarding
the
equivalent
value
of
the
1­
year
and
13­
week
dog
studies;
in
addition,
these
cases
serve
to
identify
weaknesses
in
the
existing
study
designs.
However,
opinions
differed
among
Panel
members
regarding
the
practical
significance
of
a
small
difference
in
the
NOAEL
relative
to
the
100­
fold
default
uncertainty
factor
commonly
used
in
calculating
the
RfD.

The
comparison
between
similar
data
analysis
for
pesticides
and
pharmaceutical
products
was
informative.
However,
the
different
objectives
of
toxicity
testing
for
the
two
classes
of
chemicals
rendered
the
comparison
of
results
and
conclusions
of
limited
utility
to
address
the
questions
posed
to
the
Panel.
Nevertheless,
a
harmonizing
approach
similar
to
the
International
Conference
on
Harmonization
data
review
for
pharmaceuticals
should
be
considered
for
pesticides.

Some
Panel
members
concluded
that
there
was
insufficient
data
for
determining
whether
an
additional
uncertainty
factor
should
be
considered
for
deriving
the
RfD
in
relation
to
a
13­
week
dog
study
duration.
It
was
held
by
some
that
the
studies
used
in
the
RfD
determination
should
be
adequate
to
address
the
hazard
identification
(
i.
e.,
toxicity
endpoints)
as
well
as
the
dose­
response
relationship
(
e.
g.,
NOAEL
and
LOAEL).

Several
Panel
members
agreed
that
refinements
of
the
13­
week
dog
study
would
add
confidence
and
decrease
uncertainty
in
the
calculation
of
RfDs.
Recommendations
for
refinement
were:
1)
optimize
the
study
design
by
using
all
available
data,
2)
increase
the
number
of
dose
groups
to
reduce
the
uncertainty
associated
with
the
NOAEL
and
LOAEL
determination,
3)
collect
toxicokinetic
data
for
interspecies
comparison
(
e.
g.,
between
rats
and
dogs),
4)
conduct
in
vitro
metabolism
studies
for
interspecies
extrapolation,
and
5)
establish
triggers
for
requiring
a
1­
year
dog
study.

During
the
Panel
discussions
the
Agency
clarified
that
the
purpose
of
the
analysis
as
presented
was
the
identification
of
NOAELs
and
LOAELs
for
the
derivation
of
an
RfD.
This
clarification
was
important
to
some
Panel
members
for
understanding
the
intent
of
particular
charge
questions.
In
addition,
some
Panel
members
suggested
that
the
Agency
consider
revising
the
position
paper
title
to
avoid
the
apparent
discordance
between
the
title
of
the
Agency's
position
paper,
the
approach
to
data
selection,
and
the
Agency's
10
of
23
conclusion
from
the
analysis.
One
suggested
title
is
"
A
Comparison
of
the
Results
from
1­
or
2­
Year
Dog
Studies
with
Dog
Studies
of
Shorter
Duration
in
Cases
where
they
have
been
used
for
the
Determination
of
an
RfD."

PANEL
DELIBERATIONS
AND
RESPONSE
TO
QUESTIONS
The
specific
issues
addressed
by
the
Panel
are
keyed
to
the
Agency's
background
documents,
references,
and
the
Agency's
charge
questions.

Questions
Issue
1:
Comparison
of
the
results
of
13­
week
and
chronic
dog
studies
OPP
has
concluded
that
there
is
evidence
that
little
qualitative
and
quantitative
value
is
added
by
requiring
both
a
13­
week
dog
study
and
a
1­
year
dog
study
to
support
the
establishment
of
chronic
oral
RfDs
for
pesticide
residues.

Question
1
Please
comment
on
the
adequacy
of
the
approach
used
and
the
comparisons
made
regarding
the
results
of
dog
studies
of
different
durations.
OPP
would
also
appreciate
recommendations
from
the
Panel
on
improving
the
analysis
(
e.
g.,
figures
and
tables)
or,
if
needed,
discussion
of
additional
analyses
that
would
elucidate
the
validity
of
the
conclusions
made.

Response
Some
Panel
members
agreed
that
the
general
approach
taken
by
the
Agency
appears
adequate
to
answer
the
question
of
whether
a
13­
week
study
is
sufficient
to
assess
the
toxicity
of
pesticides
in
dogs
for
determining
an
RfD.
To
answer
such
a
question
one
would
typically
survey
what
is
known
from
Agency
pesticide
submissions
and
if
available,
compare
the
results
of
that
evaluation
with
what
has
been
seen
with
other
such
analyses.
Both
of
these
approaches
have
been
included
in
the
Agency
analysis.

The
two
major
objectives
of
using
animal
studies
for
risk
assessment
of
pesticides
are
to
determine
potential
toxicity
and
to
determine
a
dose
below
which
such
an
event
would
not
occur.
The
degree
of
correlation
in
the
data
for
both
endpoints
from
13­
week
and
1­
year
studies
is
far
greater
than
one
would
have
expected
and
gives
credibility
to
the
conclusions
of
the
review.
To
have
a
difference
in
RfDs
of
less
than
1.5X
between
the
two
time
endpoints
is
quite
impressive
and
supports
the
conclusion
that
13­
week
results
are
adequate
for
establishing
credible
RfDs.
When
studies
are
conducted
multiple
times
on
the
same
chemical
using
an
identical
protocol
in
the
same
laboratory,
the
NOAEL
or
11
of
23
LOAEL
derived
from
the
different
studies
would
be
expected
to
differ
to
this
extent
based
on
chance
alone.
In
contrast,
one
would
expect
essentially
a
one­
to­
one
correlation
with
regard
to
the
target
tissue
as
has
been
demonstrated
in
this
review,
i.
e.
if
a
change/
lesion
was
observed
at
13­
weeks,
it
would
also
be
expected
to
be
found
at
1­
year,
albeit
one
could
envision
that
additional
lesions
might
be
observed
because
of
the
chronicity
of
exposure.
However,
as
shown
in
the
German
studies,
no
additional
target
organs
were
identified
in
the
chronic
dog
studies
that
were
not
identified
in
the
shorter
dog
studies
or
in
rodent
studies.
It
was
noted
that
the
German
review
may
be
problematic
as
additional
supporting
documentation
because
of
the
possible
overlap
of
studies
with
those
used
by
the
Agency
(
see
below).

The
comparison
with
the
German
reviews
[
Gerbracht
and
Spielmann
(
1999)
and
Spielmann
and
Gerbracht
(
2001)]
may
be
appropriate
and,
if
so,
would
give
confidence
in
the
Agency
results
because
they
essentially
arrive
at
the
same
conclusion,
i.
e.
that
studies
shorter
than
1­
year
appear
to
be
adequate
and
1­
year
studies
are
not
needed
to
establish
a
credible
RfD.
However,
the
German
reviews
only
identify
their
chemicals
by
"
number"
rather
than
providing
the
actual
chemical
name
as
in
the
Agency
document.
Because
of
the
global
nature
of
pesticide
manufacturers,
it
is
likely
that
both
the
Agency
and
German
reviews
are
reporting
findings
from
many
of
the
same
studies.
To
understand
the
extent
of
corroboration
provided
by
the
German
studies,
it
will
be
important
to
determine
the
degree
to
which
they
rely
upon
different
studies.
If
they
are
using
different
sets
of
data,
this
certainly
supports
the
Agency's
position
to
a
greater
degree.
However,
if
data
are
from
the
same
studies,
the
German
results
would
be
expected
to
be
similar
to
the
Agency's,
reducing
greatly
their
corroborative
value.

Limitations
in
the
Review:
Some
Panel
members
expressed
concerns
with
the
Agency
review
of
the
supporting
data
as
follows:

1)
Comment
on
the
Adequacy
of
the
Approach
Used
by
the
Agency
The
available
data
for
the
study
included
304
pesticide
studies.
The
EPA
chose
to
focus
its
analysis
on
77
of
these
studies
in
which:
a)
testing
in
dogs
was
used
to
establish
the
final
RfD
(
116
of
304
studies);
and
b)
where
13­
week
or
interim
data
were
available
to
compare
to
the
results
of
the
1­
year
chronic
study
(
77
of
the
116).
Of
the
77
studies
meeting
these
criteria,
42
provided
data
for
independent
13­
week
and
1­
year
studies.
The
remaining
35
studies
provided
interim
data
(
e.
g.
13­
weeks)
from
animals
that
were
continuing
on
the
1­
year
chronic
study.
The
13­
week
and
chronic
results
for
these
latter
35
are
therefore
correlated
statistics.
If
the
final
assessment
for
the
1­
year
studies
were
blind
to
the
corresponding
interim
results
and
there
was
no
significant
measurement
error
at
either
time
point,
the
intra­
individual
correlation
in
these
latter
studies
should
actually
improve
the
precision
of
the
contrast
between
the
13­
week
and
chronic
study
results.
However,
unless
explicitly
established
by
the
testing
protocol,
it
is
more
likely
that
the
assessor
of
the
1­
year
study
had
access
to
the
13­
week
results.
If
this
is
the
case,
the
degree
of
correspondence
between
the
two
sets
of
observations
in
the
same
dogs
will
be
12
of
23
greater
than
if
the
comparison
were
based
on
independent
studies.

The
criteria
by
which
studies
were
selected
for
analysis
was
discussed
at
some
length
by
the
Panel.
Of
particular
interest
was
exclusion
from
the
analysis
of
dog
studies
for
pesticides
where
chronic
rodent
toxicity
data
were
the
primary
basis
for
RfD
determination.
Exclusion
of
these
dog
studies
could
be
justified
if
the
question
to
be
addressed
is
simply
whether
or
not
the
absence
of
1­
year
dog
study
data
would
have
made
a
difference
in
the
RfDs
for
pesticides
included
in
the
analysis.
In
answering
this
question,
it
makes
sense
to
focus
the
analysis
on
pesticides
where
dog
study
data
were
critical
for
RfD
derivation.
However,
the
Panel
thought
that
a
different
question
is
also
germane
 
whether
a
chronic
RfD
derived
from
dog
studies
can
be
adequately
predicted
based
on
a
13­
week
study
alone
(
i.
e.,
without
also
conducting
a
1­
year
study).
This
question
relates
to
the
issue
of
whether
the
contribution
of
dog
studies
to
overall
RfD
development
can
be
adequately
achieved
by
requiring
a
study
of
only
13­
weeks
duration.
For
this
question,
all
studies
offering
toxicity
data
from
both
13­
weeks
and
1­
year
have
potential
value,
and
much
of
this
information
is
lost
using
the
Agency
study
selection
criteria.

The
exclusion
of
these
studies,
plus
the
exclusions
made
because
comparisons
of
dog
studies
of
different
durations
were
not
available
(
or
were
unreliable)
reduced
the
original
respectably­
sized
database
of
304
pesticides
to
a
less
robust
one
of
77.
Inclusion
of
additional
studies
would
not
only
allow
more
comparisons
of
NOAELs/
LOAELs
from
the
different
study
durations,
but
also
more
test
cases
to
determine
whether
new
effects
observed
in
a
1­
year
study
might
lead
to
application
of
a
different
uncertainty
factor.
The
title
of
the
analysis
produced
by
the
Agency,
"
A
Comparison
of
the
Results
of
Studies
on
Pesticides
from
1­
or
2­
year
Dog
Studies
with
Dog
Studies
of
Shorter
Duration,"
suggests
that
the
question
is
not
limited
to
comparison
of
13­
week
and
1­
year
studies.
Studies
shorter
than
13­
weeks
could
be
included
in
the
analysis.
Some
members
of
the
Panel
considered
it
appropriate
to
include
data
from
studies
of
less
than
13­
weeks
as
surrogates
for
13­
weeks,
if
the
studies
are
long
enough
to
elicit
a
response
and
that
response
was
also
observed
in
the
1­
year
study.
Since
there
were
only
2
chemicals
of
less
than
4
weeks
dosing
duration
(
one
of
3­
days
and
one
of
12­
days)
out
of
the
35
in
this
group
and
since
both
elicited
responses
similar
to
1­
year,
their
data
could
also
be
used,
depending
upon
how
the
question
being
addressed
by
the
analysis
is
defined.

2)
Comparisons
Made
Regarding
the
Results
of
Dog
Studies
of
Different
Durations
A
full
comparison
of
the
results
from
13­
week
dog
studies
with
the
results
from
studies
of
1­
year
or
longer
duration
was
possible
for
only
42
of
the
77
pesticides
analyzed.
Analysis
of
the
remaining
35
pesticides
relied
on
comparing
the
chronic
studies
with
studies
of
just
4
weeks
duration
or
(
more
commonly)
on
intra­
study
comparisons
of
interim
and
chronic
results
of
blood
chemistry
and
hematology
observations
without
the
possibility
of
comparing
histopathology.

There
are
two
important
differences
between
the
group
of
42
pesticides
and
the
13
of
23
group
of
35:
(
a)
histopathological
data
exists
for
both
the
13­
week
and
the
1­
year
or
longer
studies
for
the
42
pesticides
but
not
for
the
shorter
term
or
interim
data
used
for
the
35
pesticides
and
(
b)
there
is
a
precise
correspondence
of
the
doses
between
the
comparisons
made
for
the
group
of
35
pesticides
that
is
not
observed
among
the
42
pesticides.
Thus,
the
full
comparison
of
all
77
pesticides
contains
a
bias
introduced
by
the
inclusion
of
the
35
pesticides
for
which
no
histopathological
data
are
available,
although
some
Panel
members
did
not
think
this
was
a
serious
bias
if
the
same
blood
chemistry
and
hematology
observations
were
used
for
determining
both
the
interim
and
1­
year
or
longer
RfDs.

The
differences
between
the
groups
of
42
and
35
pesticides
suggest
that
their
data
should
not
be
pooled
so
that
the
bias
can
be
avoided.
In
addition,
more
analysis
of
the
groups
of
42
and
35
pesticides,
and
their
differences,
is
necessary.
When
the
groups
are
separated,
the
chronic
studies
seem
to
provide
no
additional
toxicological
data
for
the
determination
of
the
RfD
for
the
group
of
35
pesticides.
However,
the
chronic
studies
do
provide
additional
toxicological
data
relevant
to
the
RfD
for
19
pesticides
from
the
group
of
42.
Of
the
19,
there
are
only
3
pesticides
where
the
chronic
dog
study
potentially
is
more
appropriate
than
the
13­
week
study
for
the
selection
of
NOAELs
and
LOAELs.
If
one
accepts
the
various
arguments
put
forward
by
the
Agency
regarding
the
dose
selection
and
spacing,
inter­
experimental
variability,
and
other
limitations
relevant
to
16
of
these
19
pesticides,
then
there
is
reasonable
concordance
in
the
results
from
the
different
study
durations.

The
Panel
provided
the
following
recommendations
for
additional
analysis:

1)
The
comparative
analysis
of
the
significant
(
i.
e.
greater
than
1.5X)
differences
in
NOAEL
and
LOAEL
levels
established
from
13­
week
and
1­
year
dog
studies
is
a
retrospective
analysis
of
a
database
compiled
from
over
20
years
of
pesticide
exposure
testing.
As
is
the
case
in
any
comparative
retrospective
study
(
e.
g.
case­
control),
inferences
drawn
from
the
analysis
assume
that
all
factors
that
bear
on
the
observations
are
either
random
with
respect
to
the
outcome
or
can
be
controlled
through
statistical
adjustment.
The
decision
by
the
Agency
to
restrict
its
analysis
to
only
pesticides
for
which
the
dog
data
were
used
to
set
the
RfD
requires
the
assumption
that
the
historical
decision
to
use
rodent
data
to
set
the
RfD
was
independent
of
the
results
for
the
tests
conducted
in
dogs.
This
is
a
strong
assumption,
though
not
necessarily
an
unreasonable
one.
It
is
an
assumption
that
the
Panel
agrees
should
be
tested.
The
Panel
recommends
that
the
Agency
extend
its
comparison
of
NOAEL
and
LOAEL
values
for
13­
weeks
and
1­
year
dog
studies
to
include
a
random
sample
of
those
pesticides
for
which
dog
studies
are
available,
but
the
RfD
was
based
on
the
rodent
study
data.
The
purpose
behind
this
additional
work
would
be
to
test
the
null
hypothesis
that
the
level
of
agreement
in
13­
week
and
1­
year
studies
results
for
NOAEL
and
LOAEL
in
this
previously
omitted
set
of
studies
is
comparable
to
that
established
in
the
77
studies
that
are
reviewed
in
the
current
paper.
The
Panel
suggests
that
this
sample­
based
investigation
be
restricted
to
those
pesticides
for
which
tests
were
performed
under
GLP
standards.
The
required
sample
size
14
of
23
can
be
computed
using
standard
sample
size
calculation
programs.
The
sample
size
calculation
will
require
the
Agency
to
set
an
objective
criterion
(
effect
size)
for
determining
if
the
relationship
between
13­
week
and
1­
year
NOAEL
and
LOAEL
values
for
pesticides
with
RfDs
derived
from
rodent
studies
is
equivalent
to
the
relationship
that
has
already
been
observed
and
reported
for
pesticides
with
RfDs
based
on
the
dog
studies.

2)
Currently,
13­
week
studies
may
not
have
been
reviewed
to
the
same
level
of
rigor
as
for
the
1­
year
dog
studies.
Without
the
knowledge
and
support
from
the
1­
year
study,
early
response
seen
in
a
13­
week
study
(
e.
g.,
minor
changes
in
liver
enzyme
levels)
may
be
recognized
as
indicators
of
exposure
rather
than
adverse
endpoints
suitable
for
risk
assessment.
Thus,
without
the
1­
year
study,
the
NOAEL
and
LOAEL
from
a
13­
week
study
may
be
higher
than
as
presented
in
the
Agency
position
paper.
A
more
realistic
data
comparison
between
the
13­
week
and
1­
year
dog
studies
is
to
review
the
13­
week
studies
independent
of
the
1­
year
chronic
studies
(
i.
e.,
as
if
they
do
not
exist)
for
establishing
an
RfD,
and
contrast
this
evaluation
with
the
information
from
1­
year
studies.
By
conducting
this
exercise,
the
Agency
could
also
begin
to
formulate
a
set
of
criteria
for
evaluating
the
short­
term
studies
(
e.
g.,
13­
week
study)
for
establishing
the
NOAEL/
LOAEL.

3)
The
Agency
analysis
might
benefit
from
analyzing
the
major
classes
of
pesticides
separately,
i.
e.
herbicides,
insecticides,
fungicides
and
others,
as
was
done
in
the
German
review,
just
to
make
sure
that
one
of
these
classes
is
not
an
outlier.
This
can
be
determined
by
searching
the
tables,
but
a
simple
summary
table
using
the
number
of
chemicals
by
class
would
be
informative.

4)
The
following
background
information,
if
included
in
the
Agency
position
paper,
can
provide
a
more
thorough
understanding
and
perspective
for
the
reader.

 
A
description
of
the
types
of
dog
data
that
are
submitted
to
the
Agency,
e.
g.
dose
selection
criteria,
number
of
dose
groups,
number
of
animals
per
dose
group,
major
toxicity
endpoints
evaluated
in
a
study.

 
A
description
of
the
nature
of
the
"
peer
review"
conducted
for
studies
submitted
to
the
Agency.

 
A
definition
of
an
"
adverse"
effect
and
its
distinction
from
treatment­
related
effects
that
are
not
adverse.
Are
there
some
adverse
effects
that
are
of
greater
significance
than
others?
If
so,
provide
some
examples.
How
do
such
adverse
effects
impact
upon
the
determination
of
an
RfD?

 
A
description
of
how
the
RfD
is
calculated
and
used.
Specifically,
in
describing
the
approach
to
calculating
the
RfD,
include
the
concept
of
using
"
the
most
sensitive
endpoint
in
the
most
sensitive
species"
and
"
uncertainty
factors."
A
brief
explanation
of
the
relationship
between
RfD
and
the
often
used
expression
15
of
23
"
Margin
of
Exposure"
(
MOE)
could
provide
perspectives
on
how
the
RfD
is
used
in
pesticide
risk
assessment
and
regulation.

5)
To
avoid
the
apparent
discordance
between
the
title
of
the
Agency's
position
paper,
the
approach
to
data
selection,
and
the
Agency's
conclusion
from
the
analysis,
it
was
suggested
that
the
Agency
consider
revising
the
position
paper
title
to
better
reflect
the
apparent
purpose
of
the
analysis
as
presented.
Based
on
the
selected
number
of
datasets
included
in
the
analysis,
one
suggested
title
is
"
A
Comparison
of
the
Results
from
1­
or
2­
Year
Dog
Studies
with
Dog
Studies
of
Shorter
Duration
in
Cases
where
they
have
been
used
for
the
Determination
of
an
RfD."

Issue
2:
Examples
where
NOAELs
or
LOAELs
were
found
to
be
lower
in
chronic
dog
studies
than
in
13­
week
dog
studies
Nineteen
examples
(
Table
A3,
Appendix
on
Agency's
position
paper)
were
initially
identified
where
NOAELs
or
LOAELs
were
found
to
be
lower
(>
1.5X)
in
chronic
dog
studies
than
in
13­
week
studies.
Further
analyses
showed
that
for
11
of
these
examples,
the
observed
differences
seem
to
be
associated
with
differences
in
dose
selection
or
interexperimental
variability
(
interim
data
from
1­
year
studies
indicated
the
same
NOAELs/
LOAELs
could
be
identified
at
13­
weeks).
For
the
remaining
8
pesticides
(
Table
1,
main
text),
there
are
rat
studies
that
would
provide
comparable
NOAELs/
LOAELs
to
the
chronic
dog
study
(
2
pesticides)
and,
for
3
pesticides,
data
were
insufficient
for
an
in­
depth
comparison
of
the
results
of
13­
week
and
chronic
dog
studies
or
the
data
were
inconsistent
with
current
biological
understanding.
Finally,
three
pesticides
were
identified
where
results
of
the
chronic
dog
study
seem
to
be
more
appropriate
for
selection
of
NOAELs/
LOAELs
that
would
be
used
for
derivation
of
a
chronic
RfD.

Question
2
Please
comment
on
the
clarity
and
soundness
of
the
evaluations
presented
on
the
19
pesticides
where
NOAELs
or
LOAELs
appeared
to
differ
between
the
13­
week
and
chronic
studies.
Specifically,
do
these
cases
detract
from
the
overall
conclusions
regarding
the
value
of
1­
year
dog
studies
in
RfD
selection?

Response
The
Agency
presented
clearly
the
analysis
of
data
for
the
19
chemicals,
where
NOAELs
or
LOAELs
were
lower
in
the
chronic
dog
studies
than
in
the
13­
week
dog
studies.
The
Panel
agreed
that
the
cases
where
NOAELs
or
LOAELs
appeared
to
differ
between
the
13­
week
dog
studies
and
chronic
dog
studies
detract
from
the
overall
conclusions
that
a
13­
week
study
is
sufficient
to
assess
the
toxicity
of
pesticides
in
dogs
for
determining
an
RfD;
in
addition,
these
cases
serve
to
identify
weaknesses
in
the
16
of
23
existing
study
designs.
However,
opinions
differed
regarding
the
practical
significance
of
a
small
difference
in
the
NOAEL
relative
to
the
100­
fold
default
uncertainty
factor
commonly
used
in
calculating
the
RfD.

The
more
detailed
comparison
and
analysis
presented
for
the
19
chemicals
identifies
several
common
issues
in
data
evaluation
for
hazard
identification
as
well
as
dose­
response
assessment.
Some
of
these
are:
data
interpretation
for
NOAEL
and
LOAEL
determination,
uneven
quality
of
studies
within
a
database
for
the
NOAEL
and
LOAEL
comparison,
different
levels
of
detail
in
toxicological
examinations
among
these
studies,
and
experimental
variability
that
is
further
compounded
by
the
relatively
small
sample
size
in
dog
studies.
New
and
more
severe
or
adverse
endpoints
also
can
arise
in
longer­
term
studies
as
critical
effects
at
the
LOAEL.
For
example,
in
the
case
of
cypermethrin,
death
is
one
of
the
observed
endpoints
at
the
1­
year
study
LOAEL.
This
death
occurred
at
a
lower
dose
than
the
NOAEL
from
the
13­
week
study
in
which
no
deaths
were
observed
(
page
20
of
the
Agency
position
paper
"
A
Comparison
of
Results
from
1­
or
2­
year
Dog
Studies
on
Pesticides
with
Dog
Studies
of
Shorter
Duration,"
dated
4/
7/
2005).
Even
in
cases
where
new
endpoints
from
the
1­
year
study
do
not
alter
the
NOAEL
determination,
such
studies
can
add
greater
weight
to
the
toxicological
evidence
that
may
otherwise
be
judged
as
of
equivocal
adversity
based
only
on
the
13­
week
study.
Another
issue
is
the
dose
spacing
and
selection.
Within
the
current
battery
of
toxicity
tests,
the
13­
week
studies
often
serve
to
guide
the
dose
spacing
and
selection
for
the
12­
month
studies.
Thus,
as
can
be
expected,
dose
selection
contributed
to
a
majority
of
the
NOAEL
and
LOAEL
differences
between
the
13­
week
and
1­
year
dog
studies
presented
in
the
Agency
analysis.
More
importantly,
within
the
NOAEL/
LOAEL
approach
to
doseresponse
assessment,
dose
selection
is
a
significant
factor
that
will
directly
impact
the
ultimate
determination
of
NOAEL
for
the
RfD
calculation.
These
issues
should
be
taken
into
consideration
in
drawing
conclusions
from
the
comparative
analysis
of
the
NOAEL
and
LOAEL.
Moreover,
these
considerations
will
greatly
facilitate
the
setting
of
the
study
protocol
for
13­
week
studies
and
deriving
guidance
for
their
evaluation
should
the
Agency
decide
to
eliminate
the
1­
year
study.

The
Panel
discussed
the
practical
impact
of
the
1­
year
dog
study
on
the
overall
RfD
determination.
One
debate
centered
on
the
quantitative
significance
of
a
different
NOAEL
that
could
result
from
not
having
the
1­
year
studies.
On
the
one
hand,
a
small
difference
in
the
NOAEL
can
be
relatively
minor
compared
to
the
100­
fold
uncertainty
factor
commonly
applied
to
the
RfD
derivation.
On
the
other
hand,
since
the
current
default
practice
is
to
consistently
apply
the
100­
fold
uncertainty
factor
for
inter­
species
and
inter­
individual
variations
in
calculating
the
RfD
from
all
critical
NOAELs,
a
given
magnitude
of
difference
in
the
NOAEL
would
proportionally
affect
the
final
regulatory
levels
and
can
be
significant
to
the
regulated
communities.
Another
Panel
debate
was
on
the
significance
of
missing
new
or
more
adverse
endpoints
that
are
elicited
beyond
13
weeks
of
exposure
but
at
the
same
dose
level
as
the
13­
week
LOAEL.
On
the
one
hand,
as
long
as
the
NOAEL
remains
the
same,
identifying
more
adverse
responses
in
longer
studies
would
not
have
practical
impact
on
the
RfD
determination.
However,
the
17
of
23
additional
information
could
provide
the
certainty
and
support
for
effects
of
equivocal
adversity
after
13­
weeks
of
dosing.
The
new
and
more
adverse
endpoints
from
a
chronic
study
would
also
provide
a
more
accurate
picture
of
the
potential
risk
of
long­
term
exposure
when
the
RfD
is
exceeded.
The
Panel
encouraged
the
Agency
to
include
these
issues
in
their
future
document.

Issue
3:
Recommended
durations
for
dog
studies
conducted
on
pharmaceutical
agents
versus
dog
studies
conducted
on
pesticides.

An
International
Conference
on
Harmonization
(
ICH)
workgroup
recommended
that,
in
the
case
of
pharmaceutical
agents,
the
animal
toxicity
database
should
include
a
dog
study
of
at
least
9
months
duration
(
DeGeorge
et
al.,
1999).
This
recommendation
appears
to
have
been
based
primarily
on
evidence
that
additional
toxicities
were
seen
from
9­
12
months
that
were
not
evident
at
6­
months.
EPA
understands
that
such
studies
are
often
conducted
with
dose
levels
in
the
range
of
dosages
anticipated
for
humans.
Therefore,
such
evidence
could
lead
to
an
adjustment
in
the
pharmacologically
active
doses
that
could
be
used
clinically
or
the
additional
toxicities
could
be
used
in
the
monitoring
of
clinical
parameters.
However,
it
appears
that
only
in
a
few
cases
did
the
additional
toxicities
seen
in
9­
12­
month
studies
lead
to
a
revision
in
the
margin
of
safety
(
i.
e.,
the
ratio
of
no
observed
adverse
effect
level
to
the
anticipated
human
exposure,
in
this
case
exposure
to
a
pharmacologically
active
dose);
it
is
unclear
to
what
extent
margin
of
safety
estimates
for
pharmaceuticals,
in
general,
would
be
affected.

EPA's
purpose
for
requiring
dog
studies
conducted
with
pesticides
is
to
identify
a
no
observed
adverse
effect
level
(
NOAEL)
and
the
lowest
observed
adverse
effect
level
(
LOAEL,
or
the
lowest
dose
that
produces
some
biologically
significant
evidence
of
toxicity).
The
effects
at
the
LOAEL
are
used
to
characterize
the
toxicity
that
may
be
expected
to
occur
if
exposure
to
a
pesticide
exceeds
an
RfD.
The
NOAEL
is
used
to
derive
an
RfD,
which
represents
a
dose
that
is
unlikely
to
produce
adverse
health
effects
in
humans
exposed
to
environmental
residues
of
a
pesticide
at
or
below
the
RfD.
In
contrast,
dog
studies
performed
with
pharmaceuticals
are
designed
to
ascertain
whether
adverse
effects
may
occur
in
humans
administered
a
pharmaceutical
chemical
at
pharmacologic
doses
and
to
determine
margins
of
safety.

The
analysis
of
13­
week
and
1­
or
2­
year
dog
studies
conducted
on
172
pesticides
by
Spielmann
and
Gerbracht
(
2001)
and
the
results
of
the
analysis
of
77
studies
on
pesticides
by
the
Office
of
Pesticide
Programs
has
provided
evidence
that
the
NOAELs
and
LOAELs
observed
in
13­
week
and
1­
or
2­
year
dog
studies
generally
do
not
differ.
Further,
Spielmann
and
Gerbracht
(
2001)
found
that
for
only
7
of
141
pesticides
were
significant
new
effects
found
in
1­
or
2­
year
dog
studies
that
were
not
seen
in
13­
week
dog
studies
or
chronic
studies
with
rats.
Based
on
the
results
of
these
analyses
of
dog
studies
conducted
with
pesticides,
EPA
has
concluded
that
a
dog
study
of
13­
weeks
duration
provides
sufficient
data
for
an
evaluation
of
potential
chronic
toxicity
in
this
species
along
with
other
routinely
required
toxicity
studies.
(
Note
that
a
chronic
dog
18
of
23
study
is
only
one
of
several
studies
that
may
be
selected
for
the
derivation
of
a
chronic
RfD;
other
possibilities
include
a
2­
year
rat
study,
a
2­
generation
rat
reproduction
study,
and
an
18­
month
mouse
study).

Question
3A
Given
the
somewhat
different
objectives
of
dog
studies
conducted
on
pharmaceutical
agents
and
on
pesticides
please
comment
on
the
extent
to
which
the
recommendations
of
the
International
Conference
on
Harmonization
workgroup
may
be
relevant
to
the
use
of
dog
studies
in
risk
assessments
with
pesticides
and
whether
the
results
of
studies
performed
specifically
on
pesticides
support
EPA's
conclusion
that
a
dog
study
of
13
weeks
duration
along
with
rodent
chronic
data
is
adequate
for
providing
an
assessment
of
chronic
toxicity
for
purposes
of
deriving
chronic
RfDs.

Response
While
a
database
for
comparison
of
dose
duration
effects
could
contain
information
derived
from
testing
of
pesticides
and
pharmaceuticals,
the
objectives
of
such
testing
differ
for
the
two
classes
of
products,
and
this
can
be
considered
to
limit
the
utility
of
merging
the
data
or
conclusions.
Application
of
toxicity
findings
in
test
species
differs
between
pharmaceuticals
and
environmental
agents
on
several
counts,
and
as
such
the
conclusions
for
pharmaceuticals
are
not
directly
relevant
to
findings
influencing
RfDs.
Studies
of
pesticides
focus
on
testing
for
safety
related
to
low
level
environmental
exposures,
while
pharmaceutical
testing
focuses
on
effects
at
or
above
the
pharmacologically
active
range.
The
EPA's
application
of
the
toxicity
findings
is
more
focused
on
determination
of
"
virtually
safe"
exposures
(
doses)
as
a
derivative
of
toxicity
observations.
Many­
fold
margins
are
applied
to
findings
and
the
"
safe"
exposure
is
derived
from
the
most
sensitive
endpoint
from
any
study
in
any
species.
The
purpose
of
pharmaceutical
toxicology
studies
is
not
to
determine
virtually
safe
doses,
but
primarily
to
define
the
toxicity
profile
to
anticipate
what
toxicities
may
occur
under
conditions
of
human
testing,
employing
doses
that
are
high
relative
to
those
tested
for
pesticides
in
the
animals.
Clinical
trials
are
routinely
conducted
above
the
NOAEL
or
adverse
effect
level
in
either
or
both
test
species.
Thus,
findings
that
occur
at
exposures
higher
than
the
LOAEL
are
particularly
significant
in
evaluating
potential
risks
to
subjects
in
these
studies.
Newly
observed
findings
in
longer
term
studies,
even
when
not
establishing
new
safety
NOAELs,
can
contribute
to
significant,
relevant
safety
concerns
for
subjects
in
the
clinical
studies.
These
findings
above
the
LOAEL
are
less
of
a
consideration
in
defining
the
RfDs
for
pesticides.

A
further
difference
between
pharmaceuticals
and
pesticides
is
use
of
data
from
dog
studies
in
the
process
of
safety
evaluation.
Dog
studies
are
conducted
to
support
clinical
trials
in
drug
development.
Clinical
trial
safety
data
will
usually
supersede
the
preclinical
data
(
including
dog
toxicology
information)
prior
to
marketing
the
19
of
23
pharmaceutical.
For
pesticides,
however,
the
13­
week
and
1­
year
dog
studies
are
a
part
of
the
toxicological
program
that
will
be
used
directly
in
the
final
human
health
risk
assessment
and
to
set
regulatory
exposure
standards
to
prevent
any
adverse
public
health
effects
that
will
be
monitored
in
post­
marketing
surveillance
programs.

Despite
these
differences
in
application
of
the
dog
data,
the
Panel
agreed
that
the
data
from
pharmaceutical
experience
were
informative.
Some
of
the
Panel
members
noted
that
the
retrospective
analysis
of
the
pharmaceutical
data
supported
the
reduction
of
the
length
of
dog
studies
for
pesticides.
However,
other
Panel
members,
given
the
totality
of
the
data
available,
did
not
revise
their
view
that
the
duration
of
the
dog
study
could
not
be
reduced
in
accordance
with
the
EPA
proposal.

In
the
view
of
several
Panel
members,
the
safety
derived
from
an
RfD
calculation,
because
of
the
100
plus
fold
uncertainty
factors
applied
was
robust
enough
that
it
mitigated
the
small
changes
in
NOAEL/
LOAEL
observed
in
the
overall
dataset,
rendering
them
less
meaningful.
Given
these
observations
in
conjunction
with
the
totality
of
the
data
presented,
the
Panel
discussed
two
potential
paths
forward.
One
path
forward
would
be
to
retain
the
one­
year
dog
study
requirement
until
limitations
in
the
Agency
analysis
have
been
addressed.
If
the
results
of
the
analysis
continue
to
indicate
little
added
value
from
the
one­
year
dog
studies,
the
Agency
could
move
toward
eliminating
them
on
a
stronger
basis.
Another
path
forward
would
be
to
move
immediately
toward
eliminating
the
oneyear
dog
study
requirement,
but
employ
an
uncertainty
factor
until
such
time
as
the
limitations
in
the
analysis
have
been
addressed.
At
that
point,
if
there
is
confidence
that
data
from
chronic
rodent
and
13­
week
dog
studies
are
sufficient
for
deriving
RfDs,
the
uncertainty
factor
could
be
removed.

One
conclusion
regarding
the
pharmaceutical
dataset
is
that
its
prominence
and
discussion
in
the
report
were
excessive
in
relation
to
discussions
of
the
pesticide
dataset.
Regardless
of
these
divergent
views
on
the
application
and
extent
of
discussion
of
the
pharmaceutical
data
and
its
relevance
to
pesticides,
an
important
observation
is
that
the
pharmaceutical
data
underwent
a
review
process
under
the
auspices
of
The
International
Conference
on
Harmonization
of
Technical
Requirements
for
the
Registration
of
Pharmaceuticals
for
Human
Use
(
ICH).
This
process
was
critical
in
standardizing
testing
recommendations
for
pharmaceuticals
and
it
would
seem
valuable
that
a
similar
standardized
approach
be
sought
for
pesticides
through
international
discussions.

Question
3B
EPA's
analysis
showed
that
there
was
no
additional
qualitative
hazard
information
provided
by
a
1­
year
dog
study
which
would
raise
significant
concerns
that
would
lead
to
the
application
of
uncertainty
factors
in
addition
to
the
standard
factors
when
deriving
an
RfD.
Please
comment
on
the
soundness
of
this
conclusion.
20
of
23
Response
Based
on
the
data
provided,
several
of
the
Panel
members
concluded
that
the
EPA
analysis
could
be
viewed
as
defensible.
However,
most
of
the
Panel,
including
several
who
considered
the
statement
defensible
on
the
face
value
of
the
data
analyzed,
voiced
concern
that
the
data
used
were
not
adequately
inclusive.
A
substantial
portion
of
the
dog
data
was
excluded
because
the
RfD
was
based
on
rodent
data,
rather
than
dog
findings.
Most
of
the
Panel
agreed
these
additional
data
need
to
be
analyzed
before
the
conclusion
to
eliminate
the
longer­
term
dog
study
could
be
fully
supported.
Some
Panel
members
indicated
that
they
could
not
comment
on
the
soundness
of
the
conclusion
without
more
details
on
the
type(
s)
of
additional
qualitative
hazard
information
provided
by
the
1­
year
dog
study
that
might
lead
to
use
of
an
additional
uncertainty
factor
in
deriving
the
RfD.
As
this
can
be
a
subjective
decision,
some
additional
discussion
of
the
basis
for
this
determination
is
needed
to
assess
the
soundness
of
the
conclusion.

Some
Panel
members
stated
that
given
the
inadequacy
of
the
analysis
in
relation
to
this
issue,
they
could
not
make
a
recommendation
on
whether
or
how
large
an
uncertainty
factor
would
be
appropriate
to
apply
to
a
shorter
duration
study.
In
the
event
that
the
Agency
eliminates
the
1­
year
study
from
data
requirements,
some
recommended
that
the
use
of
an
additional
uncertainty
factor
be
retained
as
one
possible
tool
for
establishing
the
RfD
based
on
a
13­
week
dog
study
if
there
are
indications
that
the
dog
may
be
the
more
sensitive
test
species
and
that
the
13­
week
study
cannot
sufficiently
address
the
toxicity
evaluation
that
could
be
anticipated
from
the
1­
year
study.

As
an
illustration
of
the
concern
the
following
is
cited.
The
determination
of
NOAEL
and
LOAEL
are
important
for
dose­
response
assessment.
Beyond
this,
the
toxicity
endpoints
and
their
severity
or
adversity
also
should/
do
contribute
to
both
the
hazard
identification
and
the
dose­
response.
In
the
case
of
cypermethrin,
the
endpoint
from
the
13­
week
study
was
tremors,
but
the
endpoint
from
the
1­
year
study
included
death
not
observed
in
the
shorter
term
study.
Although
the
NOAEL
and
LOAEL
are
the
same
for
both
endpoints,
the
adversity
of
death
as
an
endpoint
should
inform
and
impact
the
risk
assessment.
In
this
case,
without
the
1­
year
study,
the
severity
of
the
endpoint
at
the
LOAEL
would
be
missed.

Issue
4:
Refinement
of
the
13­
week
dog
study
The
dog,
as
a
second
non­
rodent
species,
is
well
accepted
and
established
as
an
important
regulatory
data
requirement.

Question
4
If
the
13­
week
dog
study
is
considered
adequate
for
hazard
identification,
how
could
the
best
use
be
made
of
this
13­
week
study
to
characterize
potential
human
health
effects
(
e.
g.,
increase
the
number
of
animals
21
of
23
evaluated,
measuring
additional
parameters
such
as
blood
pressure
and
ECG,
obtaining
toxicokinetic
data)?

Response
The
Panel
discussed
whether
the
intent
of
the
question
was
to
focus
specifically
on
hazard
identification,
or
if
suggestions
to
improve
the
dose­
response
assessment
value
of
13­
week
studies
were
also
relevant.
After
querying
Agency
scientists,
the
Panel
directed
their
comments
to
both
hazard
identification
and
dose­
response
assessment.

The
existing
standardized
protocol
for
the
13­
week
dog
study
is
not
likely
to
be
optimal
for
determining
NOAEL
and
LOAEL
for
a
given
pesticide
under
consideration.
The
optimal
design
to
detect
the
NOAEL
and
LOAEL
for
any
single
pesticide
will
depend
on
the
underlying
(
and
unknown)
form
of
the
true
dose/
response
curve
and
the
interindividual
variability
in
the
effects
of
interest.
The
designer
of
the
study
can
control
the
statistical
power
to
detect
effects
(
or
absence)
by
increasing
the
number
of
doses,
optimally
spacing
the
dose
levels
and
increasing
the
number
of
subjects
tested
at
each
dose
level.
To
investigate
how
potential
changes
to
the
existing
13­
week
protocol
for
dog
studies
might
generally
improve
the
power
to
detect
the
true
NOAEL
and
LOAEL,
existing
data
from
archived
studies
can
be
used
to
estimate
the
inter­
individual
variability
of
responses
of
interest.
Using
these
estimates
from
the
historical
data
and
the
established
NOAEL/
LOAEL
values,
standard
programs
for
power
analysis
could
be
used
to
simulate
the
effect
of
increasing
sample
size
per
dose,
numbers
of
doses
and
spacing
of
doses
on
the
power
to
detect
true
NOAEL
and
LOAEL
values.

It
was
generally
agreed
that
refinements
of
the
13­
week
dog
study
would
add
confidence
and
decrease
uncertainty
in
the
calculation
of
RfDs
compared
to
the
present
study
protocols.
Several
refinements
were
recommended
by
the
Panel.

1)
Study
design
could
be
optimized
by
the
use
of
all
available
data
prior
to
the
conduct
of
the
dog
studies.
These
data
may
include,
but
are
not
limited
to,
structure­
activity
relationships,
physiologically­
based
pharmacokinetic
modeling,
etc.

2)
Increasing
the
number
of
dose
groups
treated
would
result
in
more
accurate
calculation
of
RfDs,
NOAELs,
and
LOAELs.
It
was
pointed
out
that
NOAELs
must
be
based
on
actual
doses
used
in
a
study,
and
having
a
small
number
of
dose
groups
limits
the
datasets
that
can
be
used
to
set
NOAELs.
Present
guidelines
use
three
treatment
groups,
and
the
recommendation
was
to
consider
increasing
this,
perhaps
to
four
treatment
groups.
While
there
was
discussion
of
increasing
the
number
of
animals
per
dose
group,
it
was
pointed
out
that
the
dog
studies
are
generally
not
analyzed
statistically
due
to
the
limited
number
of
animals
per
group.
One
Panel
member
commented
that
increasing
the
number
per
group
from
four
to
six
or
eight
would
not
increase
the
statistical
power
to
a
convincing
degree,
and
22
of
23
increasing
the
number
to
do
so
is
impractical.
It
was
noted
that
the
likelihood
of
observing
an
adverse
outcome
was
a
function
of
the
number
of
dose
groups,
the
number
of
animals
per
dose
group,
and
the
spacing
of
the
dose
groups.
Increasing
the
number
of
dose
groups
would
contribute
to
the
other
two
variables.
In
Table
A3
of
the
Agency's
document
"
A
Comparison
of
Results
from
1­
or
2­
year
Dog
Studies
on
Pesticides
with
Dog
Studies
of
Shorter
Duration,"
dated
4/
7/
2005,
11
pesticides
were
judged
to
be
associated
with
differences
in
dose
selection
and
dose
spacing.
Increasing
these
two
variables
could
increase
the
accuracy
of
calculations
of
NOAEL/
LOAEL
for
these
chemicals.
This
was
also
favored
as
a
way
to
define
more
accurately
the
dose­
response
curve
if
that
would
help
reduce
uncertainty.
It
was
also
recommended
that
a
summary
of
the
pertinent
current
study
guidelines
be
included
in
the
Agency
position
document
as
an
appendix.

3)
Comparative
toxicokinetics
was
widely
accepted
as
providing
additional
data
that
could
be
used
to
compare
the
responses
between
rats
and
dogs.
While
extensive
studies
on
additional
animals
using
radiolabeled
compounds
was
not
favored,
following
unlabeled
compound
in
blood,
urine
and
feces
during
the
course
of
the
13­
week
study
would
be
practical
and
add
to
the
strength
of
the
study,
especially
when
the
rat
and
dog
results
varied
widely.
Kinetics
of
absorption
and
elimination,
and
bioavailability
estimates
would
be
particularly
valuable
information
to
acquire.
However,
this
most
probably
will
require
an
independent
study.

4)
In
vitro
metabolism
studies
could
provide
important
information
about
similarities
and
differences
in
the
biotransformation
pathways
between
rats,
dogs,
and
humans.
The
relevant
value
of
the
rodent
and
dog
studies
to
human
health
assessment
could
be
better
assessed
with
such
metabolism
data.

5)
There
was
a
suggestion
that,
even
if
the
1
year
study
requirement
for
pesticide
toxicity
testing
was
eliminated,
a
chronic
study
could
still
be
required
if
the
shorter
(
13
week
study)
resulted
in
certain
"
triggers."
These
triggers
may
be
events
such
as
data
from
the
dog
study
indicating
that
the
dog
is
a
much
more
sensitive
species
than
the
rat,
data
from
outside
sources
that
indicate
that
early
changes
to
chemical
exposure
are
not
representative
of
toxicity
following
chronic
exposure,
etc.

Additional
General
Comments
from
the
Panel
In
addition
to
providing
comments
to
the
four
questions
posed
by
the
Agency,
the
Panel
proposed
that
the
Agency
consider
the
following:

The
Panel
cautioned
that
conclusions
based
on
analysis
of
the
existing
database
may
not
be
valid
for
new
chemicals
undergoing
review
or
in
development,
especially
if
they
have
different
modes
of
action
from
those
previously
evaluated.
23
of
23
The
Panel
suggested
two
interrelated
approaches
to
assess
the
impact
and
role
of
dog
studies
in
the
RfD
determination.
First,
instead
of
the
1.5­
fold
impact
on
the
RfD
used
by
the
Agency
as
the
criterion
for
data
comparison
between
the
13­
week
and
1­
year
dog
studies,
the
Agency
may
consider
assessing
the
impact
of
requiring
only
a
13­
week
study
against
a
pre­
defined
magnitude
of
deviation
from
the
current
RfD
as
the
worst
case
scenario.
Second,
assess
whether
any
dog
study
is
needed
to
establish
the
RfD.
To
answer
this
question,
the
Agency
may
want
to
re­
evaluate
their
dataset,
comparing
the
RfD
that
would
be
calculated
from
the
appropriate
rodent
study
to
the
RfD
derived
from
the
dog
study
and
determine
the
magnitude
of
their
differences.
If
the
difference
is
never
greater
than
"
x",
then
it
may
be
possible
to
use
only
rat
data
with
an
additional
uncertainty
factor
of
"
x".
If
the
RfD
with
this
additional
uncertainty
factor
is
acceptable
(
assures
safety)
for
a
given
pesticide
there
may
not
be
a
need
for
dog
data,
unless
there
is
a
question
about
target
organ
toxicity
(
qualitative).

In
assessing
the
usefulness
of
the
1­
year
dog
study,
the
Agency
should
also
consider
the
potential
consequence
of
seeking
to
rely
on
a
single
test
with
a
study
length
that
is
short
in
comparison
with
the
normal
lifespan
of
the
animal.
The
specific
concern
is
the
possibility
of
not
attaining
the
purported
advantage
of
using
fewer
animals.
One
scenario
is
the
potential
for
poorly
defined
NOAEL
due
to
the
sub­
optimal
spacing
and
range
of
doses
that
can
result
from
the
absence
of
a
range
finding
study.
In
this
case,
instead
of
using
a
single
study
for
both
hazard
identification
and
dose­
response
assessment,
a
repeated
study
may
become
desirable
merely
for
re­
defining
the
NOAEL.
The
shorter
duration
of
the
study
is
likely
to
make
the
decision
to
repeat
a
study
more
desirable
and
attainable
than
would
a
long­
term
study.
However,
a
repeated
study
practically
nullifies
the
purported
advantage
of
reducing
the
use
of
animals
in
toxicity
testing.
Another
scenario
is
the
need
to
more
carefully
consider
the
design
and
conduct
of
the
13­
week
study
to
assure
the
detection
of
adverse
effects
when
a
1­
year
study
is
no
longer
available
to
confirm
and
support
the
findings
from
a
single
13­
week
study.
One
way
to
strengthen
the
13­
week
study
is
through
increasing
the
number
of
dose
groups
over
the
current
protocol
of
employing
3
treatment
groups.
However,
increasing
dose
groups
would
also
mean
less
reduction
of
use
of
animals
from
the
current
testing
requirements.
In
order
to
retain
the
purported
desire
to
reduce
the
use
of
animals,
a
third
scenario
may
be
to
require
the
chemical
sponsors
to
abide
by
the
NOAEL
established
from
the
single
dog
study
that
meets
the
prescribed
study
protocol
without
the
recourse
of
repeating
the
study,
even
if
the
resulting
NOAEL
is
lower
than
desirable.
