March
5,
2004
MEMORANDUM
SUBJECT:
Transmittal
of
Meeting
Minutes
of
the
FIFRA
Scientific
Advisory
Panel
Meeting
Held
December
9,
2003
TO:
James
J.
Jones,
Director
Office
of
Pesticide
Programs
Charles
M.
Auer,
Director
Office
of
Pollution
Prevention
and
Toxics
FROM:
Steven
M.
Knott,
Designated
Federal
Official
FIFRA
Scientific
Advisory
Panel
Office
of
Science
Coordination
and
Policy
THRU:
Larry
C.
Dorsey,
Executive
Secretary
FIFRA
Scientific
Advisory
Panel
Office
of
Science
Coordination
and
Policy
Joseph
J.
Merenda,
Jr.,
Director
Office
of
Science
Coordination
and
Policy
Attached,
please
find
the
meeting
minutes
of
the
FIFRA
Scientific
Advisory
Panel
open
meeting
held
in
Arlington,
Virginia
on
December
9,
2003.
This
report
addresses
a
set
of
scientific
issues
being
considered
by
the
Environmental
Protection
Agency
regarding
the
proposed
science
policy:
PPAR­
 
agonist­
mediated
hepatocarcinogenesis
in
rodents
and
relevance
to
human
health
risk
assessment.

Attachment
cc:

Susan
Hazen
Adam
Sharp
Anne
Lindsay
Janet
Andersen
Steven
Bradbury
Page
2
of
2
cc:

William
Diamond
Debbie
Edwards
Arnold
Layne
Tina
Levine
Lois
Rossi
Frank
Sanders
Margaret
Stasikowski
Randolph
Perfetti
Karl
Baetcke
Vicki
Dellarco
Oscar
Hernandez
David
Lai
Elizabeth
Mendez
Esther
Rinde
Jennifer
Seed
William
Jordan
Douglas
Parsons
Daniel
Rosenblatt
David
Deegan
Vanessa
Vu
(
SAB)
OPP
Docket
FIFRA
Scientific
Advisory
Panel
Members
Gary
E.
Isom,
Ph.
D.
(
Session
Chair)
Stephen
M.
Roberts,
Ph.
D.
(
FIFRA
SAP
Chair)
Christopher
J.
Portier,
Ph.
D.

FQPA
Science
Review
Board
Members
George
B.
Corcoran,
Ph.
D.
Yvonne
P.
Dragan,
Ph.
D.
Ronald
N.
Hines,
Ph.
D.
Randy
L.
Jirtle,
Ph.
D.
Lisa
M.
Kamendulis,
Ph.
D.
James
P.
Kehrer,
Ph.
D.
Lois
D.
Lehman­
Mckeeman,
Ph.
D.
David
E.
Moody,
Ph.
D.
Daniel
J.
Noonan,
Ph.
D.
Carmen
E.
Perrone,
Ph.
D.
Martha
S.
Sandy,
Ph.
D.
Michael
D.
Wheeler,
Ph.
D.
SAP
Minutes
No.
2003­
05
A
Set
of
Scientific
Issues
Being
Considered
by
the
Environmental
Protection
Agency
Regarding:

Proposed
Science
Policy:
PPAR­
 
Agonist­
Mediated
Hepatocarcinogenesis
in
Rodents
and
Relevance
to
Human
Health
Risk
Assessment
December
9,
2003
FIFRA
Scientific
Advisory
Panel
Meeting,
Held
at
the
Holiday
Inn
National
Airport
Hotel,
Arlington,
Virginia
Page
2
of
35
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).
These
meeting
minutes
represent
the
views
and
recommendations
of
the
FIFRA
SAP,
not
the
United
States
Environmental
Protection
Agency
(
Agency).
The
content
of
these
meeting
minutes
does
not
represent
information
approved
or
disseminated
by
the
Agency.
These
meeting
minutes
have
not
been
reviewed
for
approval
by
the
Agency
and,
hence,
the
contents
of
this
report
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
was
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
Steven
Knott,
SAP
Designated
Federal
Official,
via
e­
mail
at
knott.
steven@.
epa.
gov.

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
within
the
structure
of
the
charge
by
the
Agency.
Page
3
of
35
CONTENTS
PARTICIPANTS...........................................................................................................
5
INTRODUCTION.........................................................................................................
6
PUBLIC
COMMENTERS............................................................................................
7
CHARGE.......................................................................................................................
7
SUMMARY
OF
PANEL
DISCUSSION
AND
RECOMMENDATIONS.................
10
PANEL
DELIBERATIONS
AND
RESPONSE
TO
CHARGE.................................
14
REFERENCES............................................................................................................
31
Page
4
of
35
SAP
Minutes
No.
2003­
05
A
Set
of
Scientific
Issues
Being
Considered
by
the
Environmental
Protection
Agency
Regarding:

Proposed
Science
Policy:
PPAR­
 
Agonist­
Mediated
Hepatocarcinogenesis
in
Rodents
and
Relevance
to
Human
Health
Risk
Assessment
December
9,
2003
FIFRA
Scientific
Advisory
Panel
Meeting,
Held
at
the
Holiday
Inn
National
Airport
Hotel,
Arlington,
Virginia
Steven
M.
Knott,
M.
S.
Gary
E.
Isom,
Ph.
D.
Designated
Federal
Official
FIFRA
SAP,
Session
Chair
FIFRA
Scientific
Advisory
Panel
FIFRA
Scientific
Advisory
Panel
Date:
March
5,
2004
Date:
March
5,
2004
Page
5
of
35
Federal
Insecticide,
Fungicide,
and
Rodenticide
Act
Scientific
Advisory
Panel
Meeting
December
9,
2003
Proposed
Science
Policy:
PPAR­
 
Agonist
Mediated
Hepatocarcinogenesis
in
Rodents
and
Relevance
to
Human
Health
Risk
Assessment
PARTICIPANTS
FIFRA
SAP,
Session
Chair
Gary
E.
Isom,
Ph.
D.,
Professor
of
Toxicology,
School
of
Pharmacy
and
Pharmacal
Sciences,
Purdue
University,
West
Lafayette,
IN
Designated
Federal
Official
Mr.
Steven
M.
Knott,
FIFRA
Scientific
Advisory
Panel
Staff,
Office
of
Science
Coordination
and
Policy,
EPA
FIFRA
Scientific
Advisory
Panel
Members
Stephen
M.
Roberts,
Ph.
D.
(
FIFRA
SAP
Chair),
Professor
&
Program
Director,
University
of
Florida,
Center
for
Environmental
&
Human
Toxicology,
Gainesville,
FL
Christopher
J.
Portier,
Ph.
D.,
Director,
Environmental
Toxicology
Program,
National
Institute
of
Environmental
Health
Sciences,
Research
Triangle
Park,
NC
FQPA
Science
Review
Board
Members
George
B.
Corcoran,
Ph.
D.,
Professor
&
Chairman,
Department
of
Pharmaceutical
Sciences,
Eugene
Applebaum
College
of
Pharmacy
&
Health
Sciences,
Wayne
State
University,
Detroit,
MI
Yvonne
P.
Dragan,
Ph.
D.,
Program
Director,
Hepatotoxicology
Center
for
Excellence,
National
Center
for
Toxicologic
Research,
U.
S.
Food
and
Drug
Administration,
Jefferson,
AR
Ronald
N.
Hines,
Ph.
D.,
Professor,
Departments
of
Pediatrics,
Pharmacology,
and
Toxicology,
Division
of
Clinical
Pharmacology,
Pharmacogenetics
and
Teratology,
Birth
Defects
Research
Center,
Medical
College
of
Wisconsin,
Milwaukee,
WI
Randy
L.
Jirtle,
Ph.
D.,
Professor,
Department
of
Radiology,
Duke
University
Medical
Center,
Durham,
NC
Page
6
of
35
Lisa
M.
Kamendulis,
Ph.
D.,
Assistant
Scientist,
Department
of
Pharmacology
and
Toxicology,
Division
of
Toxicology,
Indiana
University
School
of
Medicine,
Indianapolis,
IN
James
P.
Kehrer,
Ph.
D.,
Head,
Division
of
Pharmacology
and
Toxicology,
College
of
Pharmacy,
The
University
of
Texas
at
Austin,
Austin,
TX
Lois
D.
Lehman­
McKeeman,
Ph.
D.,
Research
Fellow,
Discovery
Toxicology,
Bristol­
Myers
Squibb
Company,
Princeton,
NJ
David
E.
Moody,
Ph.
D.,
Associate
Director,
Center
for
Human
Toxicology
and
Research
Professor,
Department
of
Pharmacology
and
Toxicology,
College
of
Pharmacy,
University
of
Utah,
Salt
Lake
City,
UT
Daniel
J.
Noonan,
Ph.
D.,
Associate
Professor,
Graduate
Center
for
Toxicology
and
Biochemistry
Department,
University
of
Kentucky,
Lexington,
KY
Carmen
E.
Perrone,
Ph.
D.,
Research
Assistant
Professor,
Department
of
Pathology,
New
York
Medical
College,
Valhalla,
NY
Martha
S.
Sandy,
Ph.
D.,
Chief,
Cancer
Toxicology
and
Epidemiology
Unit,
Reproductive
and
Cancer
Hazard
Assessment
Section,
Office
of
Environmental
Health
Hazard
Assessment,
California
Environmental
Protection
Agency,
Oakland,
CA
Michael
D.
Wheeler,
Ph.
D.,
Assistant
Professor
of
Pharmacology
and
Medicine,
University
of
North
Carolina,
Skipper
Bowles
Center
for
Alcohol
Studies,
Chapel
Hill,
NC
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
the
Proposed
Science
Policy:
Peroxisome
Proliferator
Activated
Receptor­
alpha
(
PPAR­
 )
Agonist­
Mediated
Hepatocarcinogenesis
in
Rodents
and
Relevance
to
Human
Health
Risk
Assessment.
Advance
notice
of
the
meeting
was
published
in
the
Federal
Register
on
October
24,
2003.
The
review
was
conducted
in
an
open
Panel
meeting
held
in
Arlington,
Virginia,
on
December
9,
2003.
Dr.
Gary
Isom
chaired
the
meeting.
Mr.
Steven
Knott
served
as
the
Designated
Federal
Official.

Dr.
Elizabeth
Mendez
(
Health
Effects
Division,
Office
of
Pesticide
Programs,
EPA)
provided
the
Agency
presentation
on
the
proposed
science
policy
regarding
PPAR­
 
Page
7
of
35
agonist­
mediated
hepatocarcinogenesis
in
rodents
and
relevance
to
human
health
risk
assessment.
Dr.
Jeff
Peters
(
Penn
State
University)
provided
a
presentation
on
the
paper
"
PPAR­
 
Agonist­
Induced
Rodent
Tumors:
Modes
of
Action
and
Human
Relevance"
(
Klaunig
et
al.,
2003).
The
paper
and
presentation
summarized
the
evaluation
of
a
working
group
convened
by
the
International
Life
Sciences
Institute,
Risk
Science
Institute.
This
evaluation,
along
with
the
pertinent
scientific
literature,
was
considered
by
EPA's
Office
of
Prevention,
Pesticides,
and
Toxic
Substances
in
developing
its
proposed
science
policy.
Ms.
Margaret
Stasikowski
(
Director,
Health
Effects
Division,
Office
of
Pesticide
Programs,
EPA)
provided
an
introduction
to
the
session
and
also
participated
in
the
meeting.
In
addition,
Dr.
Karl
Baetcke
(
Health
Effects
Division,
Office
of
Pesticide
Programs,
EPA),
Dr.
Jennifer
Seed
and
Dr.
David
Lai
(
both
from
the
Risk
Assessment
Division,
Office
of
Pollution
Prevention
and
Toxics,
EPA)
participated
in
the
session.

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.
These
meeting
minutes
address
the
information
provided
and
presented
at
the
meeting,
especially
the
response
to
the
charge
by
the
Agency.

PUBLIC
COMMENTERS
Oral
statements
were
presented
as
follows:

Jennifer
B.
Sass,
Ph.
D.,
Natural
Resources
Defense
Council
Written
statements
were
provided
as
follows:

Robert
A.
Bilott,
Taft,
Stettinius,
Hollister,
LLP
CHARGE
Developments
in
the
area
of
research
on
peroxisome
proliferating
chemicals
have
led
to
a
reevaluation
of
the
state
of
the
science
to
characterize
the
mode(
s)
of
action
(
i.
e.,
PPAR­
 
agonism)
and
the
human
relevance
of
rodent
tumors
induced
by
PPAR­
 
agonists.
Recently,
the
ILSI
Risk
Science
Institute
(
ILSI
RSI)
convened
a
large
expert
technical
group
to
evaluate
new
information
on
the
association
between
PPAR­
 
agonism
and
the
induction
of
tumors
by
peroxisome
proliferating
chemicals.
OPPTS
considered
the
2003
ILSI
report
as
well
as
the
pertinent
scientific
literature
in
developing
its
proposed
science
policy.

Please
provide
comment
and
advice
on
the
following
questions.
In
addressing
these
questions
consider
the
completeness
of
the
data
sets
evaluated.
Page
8
of
35
Issue
1:
Rodent
PPAR­
 
Mode
of
Action
(
MOA)
for
Hepatocarcinogenesis
OPPTS
has
concluded
that
there
is
sufficient
weight
of
evidence
to
establish
the
mode
of
action
(
MOA)
for
PPAR­
 
agonist­
induced
rodent
hepatocarcinogenesis.
It
is
proposed
in
the
OPPTS
document
that
PPAR­
 
agonists
activate
PPAR­
 
leading
to
an
increase
in
cell
proliferation
and
a
decrease
in
apoptosis,
and
eventually
further
clonal
expansion
of
preneoplastic
cells
and
formation
of
liver
tumors.
The
key
events
in
PPAR­
 
agonist­
induced
hepatocarcinogenesis
may
be
classified
as
either
causal
(
required
for
this
MOA)
or
associative
(
marker
of
PPAR­
 
agonism).

Question
1
­
Please
comment
on
the
weight
of
evidence
and
key
events
for
the
proposed
MOA
for
the
PPAR­
 
agonist­
induced
rodent
hepatocarcinogenesis.
Please
comment
on
the
adequacy
of
the
data
available
to
identify
the
key
events
in
the
PPAR­
 
MOA.
Discuss
whether
the
uncertainties
and
limitations
of
these
data
have
been
adequately
characterized.

Issue
2:
Relative
Sensitivity
of
Fetal,
Neonatal,
and
Adult
Rodent
OPPTS
has
provided
a
review
of
the
ontogeny
of
PPAR­
 
expression
and
peroxisomal
assemblage
during
fetal
and
postnatal
development
in
rodents
as
well
as
an
analysis
of
the
available
data
evaluating
effects
on
peroxisomal
proliferation,
peroxisomal
enzyme
activity,
and
liver
weights
following
exposure
to
PPAR­
 
agonists
during
fetal
and
postnatal
development
in
rats
and
mice
(
see
Section
V
of
the
OPPTS
Document).
Based
on
this
analysis,
OPPTS
concluded
that
fetal
and
neonatal
rats
do
not
exhibit
an
increased
sensitivity
to
PPAR­
 
agonist­
induced
hepatocarcinogenicity
relative
to
the
adult
rodent.
Therefore,
any
conclusions
regarding
this
MOA
in
adult
rodents
would
also
apply
to
young
rodents,
and
similarly
any
conclusions
regarding
the
relevance
of
this
MOA
for
human
hepatocarcinogenesis
would
apply
to
the
young,
as
well
as
the
adults.

Question
2
­
Please
comment
on
the
weight
of
the
evidence
approach
and
mechanistic
data
used
to
support
this
conclusion.

Issue
3:
Human
Relevance
OPPTS
has
provided
an
analysis
of
a
variety
of
in
vitro
and
in
vivo
studies
on
the
key
events
pertaining
to
PPAR­
 
agonist­
induced
hepatocarcinogenesis
with
hamsters,
guinea
pigs,
non­
human
primates,
and
humans.
Based
on
the
weight
of
the
evidence,
OPPTS
concludes
that
although
PPAR­
 
agonists
can
induce
liver
tumors
in
rodents
and
while
PPAR­
 
is
functional
in
humans,
quantitatively,
humans
and
nonhuman
primates
are
refractory
to
the
hepatic
effects
of
PPAR­
 
agonists.

Therefore,
OPPTS
is
proposing
the
following
scientific
policy:
Page
9
of
35
When
liver
tumors
are
observed
in
long
term
studies
in
rats
and
mice,
and
1)
the
data
are
sufficient
to
establish
that
the
liver
tumors
are
a
result
of
a
PPAR­
 
agonist
MOA
and
2)
other
potential
MOAs
have
been
evaluated
and
found
not
operative,
the
evidence
of
liver
tumor
formation
in
rodents
should
not
be
used
to
characterize
potential
human
hazard.

Question
3
­
Please
comment
on
the
data
and
weight
of
evidence
regarding
the
hepatic
effects
of
PPAR­
 
agonists
in
humans,
and
please
comment
on
the
proposed
OPPTS's
science
policy
regarding
human
relevance.

Issue
4:
Data
Requirements
OPPTS
has
proposed
a
data
set
that
would
be
sufficient
to
demonstrate
that
PPAR­
 
agonism
is
the
MOA
for
the
induction
of
rodent
liver
tumors.
The
data
set
includes
evidence
of
PPAR­
 
agonism
(
i.
e.,
from
an
in
vitro
reporter
gene
assay),
in
vivo
evidence
of
an
increase
in
number
and
size
of
peroxisomes,
increases
in
the
activity
of
acyl
CoA
oxidase,
and
hepatic
cell
proliferation.
The
in
vivo
evidence
should
be
collected
from
studies
designed
to
provide
the
data
needed
to
show
dose­
response
and
temporal
concordance
between
precursor
events
and
liver
tumor
formation.

Question
4
­
Please
comment
in
general
on
the
proposed
data
set
and
particularly
on
its
adequacy
to
demonstrate
that
a
PPAR­
 
agonist­
mediated
MOA
is
operating
in
rodent
hepatocarcinogenesis.

Issue
5:
Other
Tumors
Induced
by
PPAR­
 
Agonists
Some
PPAR­
 
agonists
may
also
induce
pancreatic
acinar
cell
and
Leydig
cell
tumors
in
rats
and
modes
of
action
involving
agonism
of
PPAR­
 
have
been
proposed.
An
in
depth
analysis
of
these
tumors
is
provided
in
the
2003
ILSI
technical
panel
report.
Based
on
this
analysis,
OPPTS
agrees
that
the
data
available
to
date
are
insufficient
to
support
the
proposed
MOAs.

Thus,
OPPTS
is
proposing
the
following
science
policy:

Given
the
limited
evidence
available
to
support
that
a
chemical
may
induce
pancreatic
and
Leydig
cell
tumors
through
a
PPAR­
 
agonist
MOA,
the
evidence
is
inadequate
at
this
time
to
support
a
linkage
between
PPAR­
 
agonism
and
formation
of
these
tumor
types.
Thus,
it
is
presumed
that
chemicals
that
induce
pancreatic
or
Leydig
cell
tumors
may
pose
a
carcinogenic
hazard
for
humans.

Question
5
­
Please
comment
on
OPPTS's
conclusion
that
there
is
limited
evidence
that
a
chemical
may
induce
pancreatic
and
Leydig
cell
tumors
through
a
PPAR­
 
agonist
MOA,
Page
10
of
35
and
OPPTS's
proposed
science
policy
regarding
other
tumors
induced
by
PPAR­
 
agonists.

SUMMARY
OF
PANEL
DISCUSSION
AND
RECOMMENDATIONS
Rodent
PPAR­
 
Mode
of
Action
(
MOA)
for
Hepatocarcinogenesis
Overall,
the
majority
of
the
Panel
felt
the
evidence
in
support
of
the
proposed
MOA
for
PPAR­
 
agonist
induced
rodent
hepatocarcinogenesis
was
adequate,
though
the
opinions
of
individual
Panel
members
ranged
from
full
agreement
to
complete
disagreement.
The
key
event
in
the
MOA
is
PPAR­
 
activation.
PPAR­
 
activation
triggers
multiple
events
leading
to
tumorigenesis
but
the
PPAR­
 ­
altered
genes
in
the
causal
pathway
for
tumor
induction
have
not
been
identified.
While
some
of
the
key
events
that
occur
after
PPAR 
activation,
such
as
increased
cell
proliferation,
inhibition
of
apoptosis,
and
the
clonal
expansion
of
preneoplastic
lesions
are
known,
the
PPAR­
 
dependent
mechanism
for
the
perturbation
of
these
key
events
is
less
well
established.
Specifically,
mechanisms
and
steps
linking
key
events
downstream
of
PPAR­
 
activation
are
not
known.
The
data
are
sufficient
to
demonstrate
a
PPAR­
 
activation
dependence
to
the
MOA,
but
are
inadequate
to
provide
the
quantitative
linkages
associated
with
a
more
defined
mechanism
of
action.
The
Panel
members
agreed
that
additional
evidence
of
specific
alterations
associated
with
PPAR­
 
activation
would
greatly
strengthen
the
proposed
MOA.

There
was
agreement
among
most,
but
not
all,
of
the
Panel
that
data
from
the
PPAR­
 
­/­
(
null
or
knockout)
mouse
indicate
the
requirement
for
the
activation
of
PPAR­

 
in
the
MOA
of
the
hepatocarcinogenic
effect
of
these
agents.
That
the
PPAR­
 
null
mouse
fails
to
exhibit
the
key
and
associated
events
when
challenged
with
11
months
exposure
to
a
potent
PPAR­
 
agonist
at
a
dose
that
induces
100%
incidence
of
multiple
liver
adenomas
in
concurrently
exposed
control
(
wildtype)
mice
demonstrated
to
most,
but
not
all,
Panel
members
the
underlying
basis
of
the
MOA
statement.
A
few
Panel
members
expressed
concern
over
the
short
duration
of
the
studies
in
the
PPAR­
 
­/­
mouse
(
i.
e.,
11
months
vs.
24
months
in
standard
cancer
bioassays),
which
rendered
the
studies
incapable
of
assessing
the
lifetime
liver
cancer
risk
of
PPAR­
 
agonists
in
this
knockout
mouse
model,
and
thus,
inadequate
to
conclusively
demonstrate
that
PPAR­
 
activation
is
required
for
hepatocarcinogenesis.
One
Panel
member
did
not
find
the
weight
of
evidence
for
the
proposed
MOA
to
be
sufficient
based
on
the
current
absence
of
scientific
understanding
or
identification
of
any
of
the
intermediate
critical
events
on
the
causal
pathway
which
link
PPAR­
 
activation
with
increased
proliferation,
decreased
apoptosis,
clonal
expansion
of
preneoplastic
lesions,
or
liver
tumor
formation.
In
addition,
this
Panel
member
observed
that
there
is
a
large
body
of
data
demonstrating
that
PPAR­
 
agonists
activate
Kupffer
cells
through
a
PPAR­
 
independent
mechanism,
resulting
in
the
release
of
cytokines
capable
of
stimulating
parenchymal
cell
mitosis
and
suppressing
apoptosis.
Page
11
of
35
Relative
Sensitivity
of
Fetal,
Neonatal,
and
Adult
Rodent
The
Panel
does
not
support
the
OPPTS
conclusions
that
the
PPAR­
 
agonist
MOA
in
adult
rodents
would
also
apply
to
young
rodents,
and
similarly
any
conclusions
regarding
the
relevance
of
this
MOA
for
human
hepatocarcinogenesis
would
apply
to
the
young,
as
well
as
the
adults.
Differences
in
peroxisome
biogenesis
have
been
reported
during
the
ontogenic
development
of
rodents
and
humans.
While
the
assembly
of
peroxisomes
in
rats
and
mice,
including
the
insertion
of
 ­
oxidation
enzymes
into
the
peroxisomes,
occurs
near
birth,
the
assembly
of
human
peroxisomes
has
been
observed
as
early
as
8
weeks
of
gestation
(
Espeel,
et
al,
1997).
The
number
and
density
of
peroxisomes
plateau
by
17
weeks
of
gestation
in
humans.
Moreover,
acyl­
CoA
oxidase
and
3­
ketoacyl
CoA
thiolase
are
immunodetectable
in
the
peroxisomes
by
10
and
9
weeks
of
gestation,
respectively.
Thus,
this
suggests
differences
in
 ­
oxidation
capabilities
in
developing
rodents
and
humans.
It
was
also
considered
that
differences
in
cell
proliferation,
xenobiotic
metabolism,
and
other
factors
in
the
developing
rodent
(
or
human)
could
affect
sensitivity
to
PPAR­
 
hepatocarcinogenesis.
Therefore,
information
on
the
expression
of
the
PPAR­
 
during
ontogeny
as
well
as
responses
of
embryonic
and
fetal
human
hepatocytes
to
PPAR­
 
agonists
should
be
evaluated
before
concluding
that
the
developing
human
conceptus
is
unresponsive
to
PPAR­
 
agonist
exposures.

Human
Relevance
Overall,
the
majority
of
the
Panel
agreed
that
there
are
relevant
data
indicating
that
humans
are
less
sensitive
than
rodents
to
the
hepatic
effects
of
PPAR­
 
agonists.
However,
the
opinions
of
individual
Panel
members
ranged
from
full
agreement
with
the
proposed
OPPTS
policy
statement,
as
currently
written,
to
complete
disagreement.
The
majority
of
the
Panel
recognized
weaknesses
in
the
data
that
supported
the
policy
noting
in
particular
that
the
case
for
lack
of
human
relevance
was
deficient
in
the
human
data.
In
addition,
the
Panel
members
agreed
that
the
MOA
and
its
application
to
addressing
human
relevance
would
be
greatly
strengthened
by
additional
evidence
of
the
specific
alterations
associated
with
PPAR­
 
activation
that
lead
to
the
more
general
steps
of
hepatocellular
proliferation,
clonal
expansion
of
initiated
hepatocytes
and
tumor
development.
However,
the
Panel
was
divided
regarding
whether
such
additional
evidence
is
necessary
before
accepting
the
MOA
and
its
application
to
human
relevance.
Some
Panel
members
believed
that
the
data
failed
to
demonstrate
that
the
effect
could
only
occur
in
liver
and
that,
therefore,
the
policy
statement
should
be
limited
to
hepatocarcinogenic
effects
(
see
number
2
below).
Other
Panel
members
believed
that
the
overall
data
limitations
were
significant
enough
to
disagree
with
the
MOA
and
its
application
to
addressing
human
relevance.

As
noted
previously,
there
was
agreement
among
most,
but
not
all
of
the
Panel
that
data
from
PPAR­
 
null
mice
showing
that,
in
the
absence
of
the
receptor,
there
were
no
ensuing
changes
in
cell
proliferation
and
hepatic
tumor
formation,
was
strong
evidence
Page
12
of
35
that
activation
of
PPAR­
 
is
necessary
for
all
subsequent
steps
in
the
MOA.
It
also
was
noted
previously
that
a
few
Panel
members
expressed
concern
over
the
short
duration
of
the
studies
in
the
PPAR­
 
null
mice
(
i.
e.,
11
months
vs.
24
months
in
standard
cancer
bioassays),
which
rendered
the
studies
incapable
of
assessing
the
lifetime
liver
cancer
risk
of
PPAR­
 
agonists
in
this
knockout
mouse
model,
and
thus,
inadequate
to
conclusively
demonstrate
that
PPAR­
 
activation
is
required
for
hepatocarcinogenesis.
Considering
the
proposed
MOA,
there
was
agreement
that
PPAR­
 
is
present
in
humans
and
that
the
receptor
is
activated
in
human
liver
following
exposure
to
known
agonists.
Accordingly,
the
proposed
MOA
for
PPAR­
 
agonist­
induced
hepatocellular
carcinogenesis
in
rodents
is
plausible
for
humans.
There
was
also
agreement
that
the
nature
of
gene
expression
associated
with
hepatocellular
PPAR­
 
activation
is
qualitatively
different
between
humans
and
rodents.
This
difference
may
result
from
species
differences
in
peroxisome
proliferator
response
elements
(
PPREs),
but
there
are
few
data
available
that
identify
these
potentially
important
differences,
particularly
in
humans.
Humans
are
at
least
as
sensitive
to
activation
end­
points
that
lead
to
hypolipidemia
but
are
much
less
sensitive
to
other
end­
points
normally
associated
with
peroxisome
proliferation.
This
qualitative
difference
will
be
what
is
referred
to
in
subsequent
references
as
human
sensitivity.

One
overall
concern
with
the
proposed
MOA
and
the
application
of
the
MOA
to
addressing
human
relevance
was
that,
whereas
PPAR­
 
activation
is
a
very
specific
component
of
the
MOA,
the
other
steps
deemed
to
be
causally­
related,
namely
increased
hepatocellular
proliferation
and
clonal
expansion
of
initiated
hepatocytes
leading
to
tumor
development,
were
very
general
and
non­
specific.
Overall,
the
Panel
members
agreed
that
additional
evidence
of
specific
alterations
associated
with
PPAR­
 
activation
in
primates
and
especially
humans
would
greatly
strengthen
the
proposed
MOA.

The
Panel
discussed
three
other
issues
relative
to
assessing
the
weight
of
evidence
regarding
the
hepatic
effects
of
PPAR­
 
agonists
in
humans,
and
the
proposed
science
policy
regarding
human
relevance.
These
included:

1.
The
use
of
the
word
"
refractory"
to
describe
the
human
response
to
PPAR­
 
activation
is
too
absolute.
The
Panel
agreed
that
"
less
sensitive"
is
a
more
appropriate
description
of
the
nature
of
the
human
response
relative
to
that
observed
in
rats
and
mice.

2.
The
policy
statement
drafted
by
OPPTS
concludes
with
the
phrase
"
evidence
of
liver
tumor
formation
in
rodents
should
not
be
used
to
characterize
potential
human
hazard."
After
some
discussion,
it
was
suggested
by
one
member
of
the
Panel,
and
supported
by
several
other
Panel
members,
that
this
phrase
should
be
modified
to
read,
"
evidence
of
liver
tumor
formation
in
rodents
should
not
be
used
to
characterize
potential
human
hepatocarcinogenic
hazard."

3.
One
member
of
the
Panel
expressed
a
concern,
which
was
shared
by
some
other
Panel
members,
that
the
MOA
and
evaluation
of
human
relevance
was
lacking
in
its
Page
13
of
35
assessment
of
altered
gene
expression
that
could
be
associated
with
altered
methylation
of
DNA.
There
is
evidence
that
DNA
methylation
is
modified
in
rodents
following
exposure
to
PPAR­
 
agonists
(
Ge
et
al.,
2001,
Ge
et
al.,
2002,
and
Pereira,
et
al.,
2004).
Given
the
accepted
role
for
DNA
methylation
in
gene
imprinting
and
the
loss
of
imprinting
in
cancer
etiology
(
see
for
example
McClachlan
et
al.,
2001),
such
a
role
for
PPAR­
 
agonists
in
causing
similar
alterations
in
humans
should
be
explored
before
human
relevance
can
be
appropriately
evaluated,
particularly
for
exposure
during
early
life
stages
and
for
questions
regarding
site
concordance.

Data
Requirements
There
was
general
consensus
among
the
Panel
that
the
proposed
data
set
was
adequate
and
provided
a
straight
forward
approach
to
classify
a
chemical
as
a
PPAR­
 
agonist.
The
Panel
also
concurred
that
the
use
of
PPAR­
 
knockout
mice
would
provide
definitive
evidence
to
classify
a
chemical
as
a
PPAR­
 
agonist,
but
that
the
proposed
data
set
would
be
sufficient
in
lieu
of
the
use
of
this
rather
costly
tool.

In
the
course
of
the
Panel's
discussion,
questions
for
clarification
were
posed
to
the
Agency
as
to
when
(
i.
e.,
before
or
after
a
positive
liver
tumor
finding
in
rodents)
this
set
of
assays
testing
for
PPAR­
 
agonist
activity
would
be
conducted.
The
Agency
indicated
that
data
demonstrating
PPAR­
 
agonist
activity
could
be
submitted
in
the
absence
of
testing
in
long­
term
carcinogenesis
studies.
In
response
to
this,
a
Panel
member
observed
that
in
the
absence
of
testing
in
standard
long­
term
rodent
carcinogenicity
studies,
it
is
not
possible
to
determine
whether
the
chemical
would
operate
through
a
PPAR­
 
agonist
MOA
producing
rodent
liver
tumors.
A
chemical
with
PPAR­
 
agonist
activity
may
either:
1)
not
cause
cancer
in
rodents,
2)
cause
liver
cancer
in
rodents
by
the
proposed
PPAR­
 
agonist
MOA,
3)
cause
liver
cancer
by
a
MOA
other
than
the
proposed
PPAR­
 
agonist
MOA
(
e.
g.,
cytotoxicity),
or
4)
cause
cancer
at
sites
other
than
the
liver
(
with
or
without
liver
cancer).
The
Panel
concurred
that
an
overriding
requirement
is
that
other
MOAs
have
been
excluded.
For
example,
rigorous
tests
must
be
performed
to
exclude
mutagenicity,
other
forms
of
DNA
damage
(
clastogenicity),
or
overt
cytotoxicity
directly
produced
by
the
test
compound,
or
its
metabolic
products.

Other
Tumors
Induced
by
PPAR­
 
Agonists
In
addition
to
the
hepatic
tumors
that
appear
to
be
a
general
occurrence
in
rats
and
mice,
nine
PPAR­
 
agonists
have
been
reported
to
induce
Leydig
cell
tumors
(
LCTs)
and
pancreatic
acinar
cell
tumors
(
PACTs)
in
rats.
Together
with
the
hepatic
tumors,
this
is
referred
to
as
the
tumor
triad.
The
Panel
was
in
agreement
with
the
OPPTS
conclusion
that
chemicals
that
induce
pancreatic
or
Leydig
cell
tumors
may
pose
a
carcinogenic
hazard
for
humans.

Given
the
limited
amount
of
data
available
on
the
true
MOA
for
LCTs
or
PACTs,
Page
14
of
35
including
the
possibility
raised
by
some
Panel
members
that
epigenetic
effects
of
the
PPAR­
 
agonists
may
occur,
it
is
not
possible
to
determine
whether
PPAR­
 
agonists
pose
a
carcinogenic
hazard
to
humans.
Thus,
the
conclusion
by
the
OPPTS
that
the
available
data
for
the
induction
of
rat
LCTs
and
PACTs
by
PPAR­
 
agonists
are
insufficient
to
conclude
that
the
sole
MOA
involves
the
PPAR­
 
receptor
is
considered
by
the
Panel
to
be
appropriate.
Further,
the
Panel
concurs
that
it
should
be
presumed
that
chemicals
that
induce
pancreatic
or
Leydig
cell
tumors
may
pose
a
carcinogenic
hazard
for
humans.

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

Response
to
Charge
Question
1
­
Rodent
PPAR­
 
Mode
of
Action
(
MOA)
for
Hepatocarcinogenesis
OPPTS
has
concluded
that
there
is
sufficient
weight
of
evidence
to
establish
the
MOA
for
PPAR­
 
agonist­
induced
rodent
hepatocarcinogenesis.
It
is
proposed
in
the
OPPTS
document
that
PPAR­
 
agonists
activate
PPAR­
 
leading
to
an
increase
in
cell
proliferation
and
a
decrease
in
apoptosis,
and
eventually
further
clonal
expansion
of
preneoplastic
cells
and
formation
of
liver
tumors.
The
key
events
in
PPAR­
 
agonistinduced
hepatocarcinogenesis
may
be
classified
as
either
causal
(
required
for
this
MOA)
or
associative
(
marker
of
PPAR­
 
agonism).

Please
comment
on
the
weight
of
evidence
and
key
events
for
the
proposed
MOA
for
the
PPAR­
 
agonist­
induced
rodent
hepatocarcinogenesis.
Please
comment
on
the
adequacy
of
the
data
available
to
identify
the
key
events
in
the
PPAR­
 
MOA.
Discuss
whether
the
uncertainties
and
limitations
of
these
data
have
been
adequately
characterized.

Response
Weight
of
the
Evidence
for
Proposed
MOA
Overall,
the
majority
of
the
Panel
felt
the
evidence
in
support
of
the
proposed
MOA
for
PPAR­
 
agonist
induced
rodent
hepatocarcinogenesis
was
adequate,
though
the
opinions
of
individual
Panel
members
ranged
from
full
agreement
to
complete
disagreement.
The
majority
of
the
Panel
felt
the
weight
of
evidence
in
support
of
the
proposed
MOA
in
rodents
is
adequate
for
PPAR­
 
agonists
in
which
hepatic
activation
of
PPAR­
 
results
in
the
key
downstream
events
of
increased
proliferation,
decreased
apoptosis,
and
clonal
expansion
of
preneoplastic
lesions
resulting
in
hepatocarcinogenesis.
Associated
events
(
indicators
of
PPAR­
 
activation)
include
induction
of
peroxisome
Page
15
of
35
proliferation
and
altered
expression
of
related
genes.
One
Panel
member
did
not
find
the
weight
of
evidence
for
the
proposed
MOA
to
be
sufficient,
based
on
the
current
absence
of
scientific
understanding
or
identification
of
any
of
the
intermediate
critical
events
on
the
causal
pathway
which
link
PPAR­
 
activation
with
increased
proliferation,
decreased
apoptosis,
clonal
expansion
of
preneoplastic
lesions,
or
liver
tumor
formation.
In
addition,
this
Panel
member
observed
that
there
is
a
large
body
of
data
demonstrating
that
PPAR­
 
agonists
activate
Kupffer
cells
through
a
PPAR­
 
independent
mechanism,
resulting
in
the
release
of
cytokines
capable
of
stimulating
parenchymal
cell
mitosis
and
suppressing
apoptosis
(
Rolfe
et
al.,
1997;
Rusyn
et
al.,
2001;
Parzefall
et
al.,
2001;
Hasmall
et
al.,
2001).

The
proposed
MOA
for
PPAR­
 
agonist
induced
rodent
hepatocarcinogenesis
is
based
on
a
considerable
body
of
evidence
that
has
accrued
over
the
past
3
decades,
and
particularly
on
the
more
recent
demonstration
of
a
lack
of
a
tumorigenic
response
in
the
PPAR­
 
­/­
mouse
after
11
months
of
PPAR­
 
agonist
administration
at
a
dose
that
induces
100%
incidence
of
liver
adenomas
in
concurrent
studies
in
the
PPAR­
 
+/+
mouse
with
the
same
genetic
background.
This
PPAR­
 
null
mouse
is
devoid
of
responses
indicative
of
PPAR­
 
agonism.
There
was
agreement
among
most,
but
not
all,
of
the
Panel
that
data
from
the
PPAR­
 
­/­
mouse
indicate
the
requirement
for
the
activation
of
PPAR­
 
in
the
MOA
of
the
hepatocarcinogenic
effect
of
these
agents.
A
few
Panel
members
expressed
concern
over
the
short
duration
of
the
studies
in
the
PPAR­
 
­/­
mouse
(
i.
e.,
11
months
vs.
24
months
in
standard
cancer
bioassays),
which
rendered
the
studies
incapable
of
assessing
the
lifetime
liver
cancer
risk
of
PPAR­
 
agonists
in
this
knockout
mouse
model,
and
thus,
inadequate
to
conclusively
demonstrate
that
PPAR­
 
activation
is
required
for
hepatocarcinogenesis.

Additional
supporting
evidence
for
the
MOA,
as
discussed
in
the
review
by
Klaunig
et
al.
(
2003)
comes
from
the
concordance
of
this
MOA
for
several
PPAR­
 
agonists,
dose
dependence
of
the
effect,
with
both
consistency
and
biological
plausibility
for
the
key
events.
One
Panel
member
noted
several
inconsistencies
in
the
supporting
data
however.
These
include
observations
from
long­
term
carcinogenicity
studies
of
the
PPAR­
 
agonist
gemfibrozil,
where
a
dose­
related
increase
in
liver
tumors
was
observed
in
male
rats,
while
in
females,
a
dose­
dependent
decrease
in
liver
tumors
was
seen
(
IARC,
1996).
In
another
example,
studies
in
rats
with
two
PPAR­
 
agonists,
WY­
14,463
and
DEHP,
demonstrated
that
doses
that
produced
equivalent
levels
of
hepatic
peroxisome
proliferation,
measured
as
peroxisome
number
and
peroxisomal
enzyme
activity,
produced
markedly
different
liver
tumor
incidences
(
Marsman
et
al.,
1988).
Another
Panel
member
noted
that
these
differences
may
be
due
to
sex,
species,
and
strain
differences
in
pharmacokinetics.

In
addition
to
the
above,
a
Panel
member
expressed
concern
with
the
lack
of
understanding
of
key
causal
events
in
the
proposed
MOA
intermediate
between
PPAR­
 
activation
and
cell
proliferation,
suppression
of
apoptosis
and
clonal
expansion,
given
that
Page
16
of
35
activation
of
PPAR­
 
results
in
regulation
of
a
multitude
of
genes
involved
in
a
variety
of
cellular
functions,
including
lipid
metabolism
and
transport,
amino
acid
metabolism,
signaling
molecules,
transcription
factors,
and
cell
cycle
and
growth
regulatory
proteins.

The
Panel
agreed
that
data
in
the
wild
type
and
the
PPAR­
 
knockout
mouse
would
be
strengthened
if
it
were
determined
that
the
null
mice
generated
on
a
129
genetic
background
are
not
resistant
to
liver
tumorigenesis
in
general,
as
opposed
to
specifically
resistant
to
PPAR­
 
agonists
(
see
Drinkwater
and
Bennett,
1991).
In
addition,
the
PPAR­

 
knockout
mouse
data
would
be
strengthened
by
a
demonstration
of
gene
dose
sensitivity.
The
Panel
members
also
agreed
that
additional
evidence
of
specific
alterations
associated
with
PPAR­
 
activation
would
greatly
strengthen
the
proposed
MOA.

Adequacy
of
the
Data
Though
the
opinions
of
individual
Panel
members
ranged
from
full
agreement
to
complete
disagreement,
overall,
the
majority
of
the
Panel
felt
the
data
supporting
the
key
events
associated
with
the
proposed
MOA
in
rats
and
mice
are
adequate,
but
recognized
areas
where
the
data
could
be
strengthened.
One
overall
concern
with
the
proposed
MOA
was
that,
whereas
PPAR­
 
activation
is
a
very
specific
component
of
the
MOA,
the
other
steps
deemed
to
be
causally
related,
namely
increased
hepatocellular
proliferation
and
clonal
expansion
of
the
initiated
hepatocytes
leading
to
tumor
development,
were
very
general
and
non­
specific.

In
support
of
the
adequacy
of
the
data,
the
key
events
and
the
associated
events
have
been
demonstrated
to
occur
following
administration
of
PPAR­
 
agonists.
These
data
have
been
derived
from
many
laboratories
over
the
course
of
the
last
30
years.
Many
of
the
associative
events
are
highly
correlated
markers
of
PPAR­
 
agonist
exposure
and
potential
contributors
to
the
causal
events
in
the
proposed
MOA.
The
mechanistic
linkage
between
the
required
step
of
PPAR­
 
activation
and
the
key
events
(
increased
cell
proliferation,
decreased
apoptosis,
and
clonal
expansion
of
preneoplastic
hepatic
lesions)
has
not
been
determined.
Although
having
these
steps
in
the
mechanism
of
PPAR­
 
induced
rat
and
mouse
hepatocarcinogenesis
would
strengthen
the
MOA,
the
majority
of
the
Panel
agreed
that
the
current
dataset
is
adequate
to
support
the
MOA.
That
the
PPAR­
 
null
mouse
fails
to
exhibit
the
key
and
associated
events
when
challenged
with
11
months
exposure
to
a
potent
PPAR­
 
agonist
at
a
dose
that
induces
100%
incidence
of
multiple
liver
adenomas
in
concurrently
exposed
control
(
wildtype)
mice
demonstrated
to
most,
but
not
all,
Panel
members
the
underlying
basis
of
the
MOA
statement.

Several
concerns
regarding
the
adequacy
of
the
data
also
were
discussed.
As
previously
noted,
a
few
Panel
members
expressed
concern
over
the
short
duration
of
the
studies
in
PPAR­
 
null
mice
which
rendered
the
studies
inadequate
to
conclusively
demonstrate
that
PPAR­
 
activation
is
required
for
hepatocarcinogenesis.
One
member
of
the
Panel
was
concerned
that
the
data
were
not
adequate
to
identify
the
key
events
in
the
Page
17
of
35
MOA
for
PPAR­
 
agonist
induced
rodent
hepatocarcinogenesis,
stating
that
although
PPAR­
 
activation
is
believed
to
be
the
earliest
key
event,
none
of
the
many
genes
whose
expression
is
regulated
by
PPAR­
 
has
been
identified
as
being
in
the
causal
pathway
for
liver
tumorigenesis.
More
data
are
needed
to
establish
and
link
the
events
that
have
been
proposed
as
key
causal
events
in
the
proposed
MOA.
In
addition,
a
number
of
studies
provide
compelling
data
that
suggest
that
a
PPAR­
 
independent
event,
namely
Kupffer
cell
activation,
is
required
for
increased
hepatocyte
proliferation
by
PPAR­
 
agonists.
The
Panel
member
felt
that
more
data
characterizing
the
relationship
between
Kupffer
cell
activation,
and
the
cytokines
that
are
released
upon
activation
in
hepatocarcinogenesis,
and
PPAR­
 
activation
were
needed
before
the
identification
of
key
events
in
the
MOA
could
be
properly
evaluated.
Another
member
of
the
Panel
expressed
concern,
which
was
shared
by
some
other
Panel
members,
that
data
were
lacking
on
the
potential
roles
alterations
in
DNA
methylation
and
chromatin
structure
play
in
the
hepatocarcinogenic
MOA
of
PPAR­
 
agonists.

Uncertainties
and
Inadequacies
of
the
Data
Limitations
of
the
available
data
have
been
detailed
in
the
Klaunig
et
al.
(
2003)
review.
As
noted
above,
the
mechanism
for
the
induction
of
cell
proliferation
and
apoptosis
suppression
induced
by
PPAR­
 
agonists
is
not
known.
One
significant
factor
to
consider
is
the
role
of
nonparenchymal
hepatic
cells
in
these
processes.
For
example,
Kupffer
cells
release
cytokines,
some
of
which
are
mitogenic
to
parenchymal
cells
and
some
that
affect
parenchymal
cell
apoptosis.
In
addition,
many
of
the
enzymes
used
as
indicators
of
PPAR­
 
activation
are
regulated
through
a
well
defined
mechanism
of
action
that
involves
altered
transcription
of
PPRE
containing
genes.
Because
this
pathway
of
PPAR­
  
dependent
alteration
of
gene
regulation
is
only
associated
with
PPAR­
 
activation
and
not
with
the
regulation
of
key
events
in
the
MOA,
other
mechanisms
for
induction
of
the
key
events
need
to
be
considered.
Specific
uncertainties
may
include
whether
agents
must
be
metabolized
from
a
pro­
form
to
an
active­
form
to
be
able
to
modulate
the
PPAR­
 
pathway,
the
induction
of
PPREs,
or
other
indirect
events.

Many,
but
not
all,
agents
that
demonstrate
an
ability
to
induce
peroxisomes
in
rats
and
mice
also
induce
a
neoplastic
response
in
the
liver
of
rats
and
mice.
Morphologic
and
biochemical
evidence
of
peroxisome
proliferation
in
rat
and
mouse
liver
is
supportive
evidence
of
the
proposed
MOA.
It
should
be
noted
that
these
remain
associated
key
events
that
are
not
proposed
at
this
time
to
be
causally
related
to
tumor
formation.
The
Panel
agreed
that
there
were
considerable
uncertainties
as
to
the
significance
of
associated
key
events,
such
as
hepatic
acyl
CoA
oxidase
induction,
with
regard
to
the
tumor
forming
potential
of
PPAR­
 
agonists
in
rats
and
mice.
PPAR­
 
agonists
can
bind
directly
to
PPAR­
 ,
but
may
also
perturb
interactions
with
the
RXR
binding
partner,
the
binding
of
co­
activators
and
co­
repressors
to
the
receptor,
or
the
availability
and
action
of
endogenous
ligands
or
inhibitors.
Page
18
of
35
Question
2
­
Relative
Sensitivity
of
Fetal,
Neonatal,
and
Adult
Rodents
OPPTS
has
provided
a
review
of
the
ontogeny
of
PPAR­
 
expression
and
peroxisomal
assemblage
during
fetal
and
postnatal
development
in
rodents
as
well
as
an
analysis
of
the
available
data
evaluating
effects
on
peroxisomal
proliferation,
peroxisomal
enzyme
activity,
and
liver
weights
following
exposure
to
PPAR­
 
agonists
during
fetal
and
postnatal
development
in
rats
and
mice
(
see
Section
V
of
the
OPPTS
Document).
Based
on
this
analysis,
OPPTS
concluded
that
fetal
and
neonatal
rats
do
not
exhibit
an
increased
sensitivity
to
PPAR­
 
agonist­
induced
hepatocarcinogenicity
relative
to
the
adult
rodent.
Therefore,
any
conclusions
regarding
this
MOA
in
adult
rodents
would
also
apply
to
young
rodents,
and
similarly
any
conclusions
regarding
the
relevance
of
this
MOA
for
human
hepatocarcinogenesis
would
apply
to
the
young,
as
well
as
the
adults.

Please
comment
on
the
weight
of
the
evidence
approach
and
mechanistic
data
used
to
support
this
conclusion.

Response
The
Panel
does
not
support
the
OPPTS
conclusions.
Although
fetal
and
embryonic
rats
and
mice
respond
to
PPAR­
 
agonists
as
demonstrated
by
changes
in
peroxisomal
enzyme
activities,
strong
evidence
demonstrating
that
fetal
and
neonatal
rats
do
not
exhibit
an
increased
sensitivity
to
PPAR­
 
agonist­
induced
hepatocarcinogenesis
is
lacking.
Moreover,
conclusions
regarding
this
MOA
for
human
hepatocarcinogenesis
should
not
be
applied
to
developing
humans.

As
discussed
in
the
response
to
question
1,
the
proposed
MOA
involves
activation
of
PPAR­
 ,
which
regulates
the
expression
of
numerous
genes,
including
several
that
encode
for
peroxisomal
enzymes,
and
identifies
as
key
causal
events
increases
in
cell
proliferation,
inhibition
of
apoptosis,
and
clonal
expansion
of
preneoplastic
lesions,
which
result
in
the
formation
of
liver
tumors.
Published
reports
have
shown
that
both
the
expression
of
PPAR­
 
and
the
assembly
of
peroxisomes
occur
late
in
the
development
of
rats
and
mice.
Furthermore,
it
has
been
shown
that,
as
in
adult
livers,
embryonic,
fetal
and
neonatal
livers
of
rats
and
mice
respond
to
PPAR­
 
agonists
by
increasing
peroxisome
number,
peroxisome
volume
density,
liver
weight,
and
the
expression
of
the
peroxisomal
enzyme
palmitoyl
CoA
oxidase.
This
suggests
that
at
least
some
of
the
cellular
macromolecules
involved
in
the
proposed
PPAR­
 
agonist
MOA
are
functional
and
responsive
to
PPAR­
 
agonists
in
rat
and
mouse
embryonic,
fetal,
and
neonatal
livers.
However,
data
on
the
hepatocarcinogenic
response
of
rat
and
mouse
embryonic,
fetal,
and
neonatal
livers
to
PPAR­
 
agonists
are
lacking
and,
therefore,
no
conclusions
can
be
made
at
this
time
as
to
the
relative
sensitivity
of
these
early
life
stages
to
PPAR­
 
agonist
induced
hepatocarcinogenicity.

Although
the
exposure
of
pregnant
rats
and
mice
led
to
increases
in
peroxisomal
Page
19
of
35
enzyme
activities
and
increases
in
liver
weight
in
embryonic,
fetal,
and
neonatal
liver
tissues,
other
parameters
involved
in
the
proposed
MOA,
such
as
cell
proliferation,
inhibition
of
apoptosis
and
clonal
expansion
of
preneoplastic
cells,
were
not
examined
in
these
studies.
In
addition,
responses
to
PPAR­
 
agonists
in
the
fetal
and
neonatal
rat
and
mouse,
as
measured
by
the
peroxisomal
enzyme
expression
levels,
suggest
that
there
are
differences
in
young
animals
relative
to
adults.
It
is
unclear
how
these
differences
in
enzyme
expression
levels
might
translate
into
differences
in
sensitivity
to
hepatocarcinogenesis.
Regarding
the
comparison
of
changes
in
liver
weights
across
early
and
later
life
stages,
it
is
inappropriate
to
assume
that
a
given
proliferative
response
seen
at
one
stage
of
life
is
equivalent
to
a
similar
proliferative
response
at
another
stage
of
life.
For
example,
an
increase
in
liver
weight
during
the
neonatal
period
might
result
in
a
much
greater
lifetime
risk
of
cancer
than
an
equivalent
increase
occurring
during
adulthood,
because
a
larger
number
of
cells
in
the
neonatal
liver
will
undergo
multiple
cell
divisions
than
in
the
adult.
Finally,
none
of
the
studies
examining
the
response
of
the
rodent
in
utero
or
during
early
life
stages
were
carried
out
with
the
late
onset
of
tumors
as
a
specific
endpoint.
A
two­
generation
study
conducted
in
mice
was
designed
as
a
reproductive
study
and
not
as
a
cancer
study.
Thus,
no
liver
pathology
was
documented
from
F1
male
and
female
mice
after
approximately
4
and
6
months
of
exposure,
respectively
(
one
Panel
member
noted
that
complete
pathology
was
not
evaluated
in
this
study).
The
available
data
pertain
to
effects
that
have
not
been
demonstrated
as
causally
linked
to
the
carcinogenic
MOA
of
these
agents.
The
relevance
of
the
induction
of
peroxisomes
or
peroxisomal
enzymes
to
the
carcinogenic
process
has
not
been
established.
As
stated
above,
there
is
the
possibility
that
developing
organs
and
tissues
may
respond
differently
to
peroxisome
proliferators
compared
to
adult
organs
and
tissues.
There
may
also
be
PPAR­
 
independent
effects
occurring
in
the
young
animal
that
result
in
an
increased
cancer
risk.
In
the
absence
of
this
information,
conclusions
regarding
the
sensitivity
of
developing
rodents
to
PPAR­
 
agonists
cannot
be
formulated.
Chemical
exposures
early
in
development
could
increase
the
sensitivity
to
cancer
risk.
It
is
known
that
PPAR­
 
modulates
metabolic
pathways
other
than
 ­
oxidation
of
fatty
acids,
such
as
glucose
and
amino
acid
metabolism.
Moreover,
PPAR­
 
is
a
transcription
factor
involved
in
the
modulation
of
gene
expression.
PPAR­
 
agonists
not
only
modulate
the
expression
of
genes
with
PPREs
but
they
may
also
regulate
gene
expression
by
altering
levels
of
gene
methylation
(
Ge,
et
al.,
2001).
Such
DNA
methylation
is
known
to
be
involved
in
imprinting
and
alterations
or
loss
of
imprinting
can
directly
or
indirectly
impact
disease
risk
at
later
life
stages
(
Cui,
H.
et
al.,
2003).

Conclusions
regarding
the
relevance
of
the
PPAR­
 
agonist
MOA
for
human
hepatocarcinogenesis
applied
to
adults
may
not
apply
to
the
young.
In
contrast
to
adult
human
liver,
there
are
no
data
establishing
PPAR­
 
expression
levels
in
embryonic,
fetal
and
neonatal
human
liver.
To
date,
there
is
only
one
publication
reporting
the
effects
of
one
PPAR­
 
agonist
in
lactating
non­
human
primates
(
Cappon
et
al.
2002).
In
this
report,
the
exposure
of
four
Rhesus
monkey
females
to
HCFC­
123
for
short
periods
of
time
decreased
the
activities
of
cytochrome
P450
enzymes
and
acyl
CoA
oxidase
in
maternal
Page
20
of
35
monkey
liver,
as
well
as
induced
centrilobular
hepatocyte
vacuolation,
necrosis
and
mild
to
moderate
inflammation;
however,
no
histological
or
biochemical
data
were
reported
from
the
infant
monkeys.
Non­
human
primate
studies
investigating
preneoplastic
and
neoplastic
effects
of
fetal
or
neonatal
exposure
to
PPAR­
 
agonists
would
be
desirable.

In
contrast
to
embryonic
and
fetal
rodent
liver
in
which
cytochrome
P450
enzymes
are
expressed
near,
during
and
after
birth
(
Ring
et
al.
1999),
embryonic
and
fetal
human
livers
possess
metabolic
activation
capabilities
resulting
from
the
early
developmental
expression
of
cytochrome
P450
enzymes.
Moreover,
the
expression
profiles
of
xenobiotic
metabolizing
enzymes
and
isozymes
are
different
in
embryonic,
fetal,
neonatal
and
adult
human
livers.
Like
the
gene
expression
profile
of
xenobiotic
metabolizing
enzymes,
it
is
difficult
to
disregard
the
possibility
that
there
could
be
differences
between
the
expression
of
PPAR­
 
and
its
transcriptional
co­
factors
in
the
human
conceptus
and
adult
human
liver.
In
addition,
metabolic
differences
in
rats
and
mice
play
an
important
role
in
determining
the
degree
of
response
to
some
PPAR­
 
agonists
(
Lake,
1995)
and
that
could
also
apply
to
the
human
conceptus.

Differences
in
peroxisome
biogenesis
have
been
reported
during
the
ontogenic
development
of
rodents
and
humans.
While
the
assembly
of
peroxisomes
in
rats
and
mice,
including
the
insertion
of
 ­
oxidation
enzymes
into
the
peroxisomes,
occurs
near
birth,
the
assembly
of
human
peroxisomes
has
been
observed
as
early
as
8
weeks
of
gestation
(
Espeel,
et
al,
1997).
The
number
and
density
of
peroxisomes
plateau
by
17
weeks
of
gestation
in
humans.
Moreover,
acyl­
CoA
oxidase
and
3­
ketoacyl
CoA
thiolase
are
immunodetectable
in
the
peroxisomes
by
10
and
9
weeks
of
gestation,
respectively.
These
observations
suggest
differences
in
 ­
oxidation
capabilities
in
developing
rodents
and
humans
and
therefore
information
on
the
expression
of
the
PPAR­
 
during
ontogeny,
as
well
as
responses
to
PPAR­
 
agonists
in
embryonic
and
fetal
human
hepatocytes
should
be
evaluated
before
concluding
that
the
developing
human
conceptus
is
unresponsive
to
PPAR­
 
agonist
exposures.

There
are
numerous
uncertainties
concerning
the
relevance
of
the
PPAR­
 
agonist
MOA
for
human
hepatocarcinogenesis
in
the
young.
These
uncertainties
stem
largely
from
our
incomplete
understanding
of
the
species­
specific
differences
in
sensitivity.
Although
numerous
mechanisms
have
been
posited
(
see
Klauning
et
al.,
2003),
none
have
adequate
data
supporting
their
validity.
Some
of
these
include
differences
in
the
PPREs
in
specific
critical
genes,
species­
specific
co­
factors
that
suppress
transactivation
ability
of
the
ligand
activated
PPAR­
 ,
sequence
differences
that
result
in
the
prevalence
of
inactive,
splice
variants
and/
or
dominant
negative
PPAR­
 
gene
products,
perturbation
of
RXR
binding
partner
interactions
with
other
nuclear
receptors,
and
polymorphisms
that
result
in
a
less
efficient
transcription
factor.
Most
importantly,
there
is
no
reason
to
eliminate
the
possibility
that
one
or
more
of
these
scenarios
would
function
differently
in
the
human
fetus,
neonate
or
infant
relative
to
the
adult,
impacting
both
MOA
and
sensitivity
at
these
different
life
stages.
Page
21
of
35
Question
3
 
Human
Relevance
OPPTS
has
provided
an
analysis
of
a
variety
of
in
vitro
and
in
vivo
studies
on
the
key
events
pertaining
to
PPAR­
 
agonist­
induced
hepatocarcinogenesis
with
hamsters,
guinea
pigs,
non­
human
primates,
and
humans.
Based
on
the
weight
of
the
evidence,
OPPTS
concludes
that
although
PPAR­
 
agonists
can
induce
liver
tumors
in
rodents
and
while
PPAR­
 
is
functional
in
humans,
quantitatively,
humans
and
nonhuman
primates
are
refractory
to
the
hepatic
effects
of
PPAR­
 
agonists.

Therefore,
OPPTS
is
proposing
the
following
scientific
policy:

When
liver
tumors
are
observed
in
long
term
studies
in
rats
and
mice,
and
1)
the
data
are
sufficient
to
establish
that
the
liver
tumors
are
a
result
of
a
PPAR­
 
agonist
MOA
and
2)
other
potential
MOAs
have
been
evaluated
and
found
not
operative,
the
evidence
of
liver
tumor
formation
in
rodents
should
not
be
used
to
characterize
potential
human
hazard.

Please
comment
on
the
data
and
weight
of
evidence
regarding
the
hepatic
effects
of
PPAR­
 
agonists
in
humans,
and
please
comment
on
the
proposed
OPPTS's
science
policy
regarding
human
relevance.

Response
Overall,
the
majority
of
the
Panel
agreed
that
there
are
relevant
data
indicating
that
humans
are
less
sensitive
than
rodents
to
the
hepatic
effects
of
PPAR­
 
agonists.
However,
the
opinions
of
individual
Panel
members
ranged
from
full
agreement
with
the
proposed
OPPTS
policy
statement,
as
currently
written,
to
complete
disagreement.
The
majority
of
the
Panel
recognized
weaknesses
in
the
data
that
supported
the
policy
noting
in
particular
that
the
case
for
lack
of
human
relevance
was
deficient
in
the
human
data.
In
addition,
the
Panel
members
agreed
that
the
MOA
and
its
application
to
addressing
human
relevance
would
be
greatly
strengthened
by
additional
evidence
of
the
specific
alterations
associated
with
PPAR­
 
activation
that
lead
to
the
more
general
steps
of
hepatocellular
proliferation,
clonal
expansion
of
initiated
hepatocytes
and
tumor
development.
However,
the
Panel
was
divided
regarding
whether
such
additional
evidence
is
necessary
before
accepting
the
MOA
and
its
application
to
human
relevance.
Some
Panel
members
believed
that
the
data
failed
to
demonstrate
that
the
effect
could
only
occur
in
liver
and
that,
therefore,
the
policy
statement
should
be
limited
to
hepatocarcinogenic
effects
(
see
number
2
below).
Other
Panel
members
believed
that
the
overall
data
limitations
were
significant
enough
to
disagree
with
the
MOA
and
its
application
to
addressing
human
relevance.

Over
the
past
30
years,
a
variety
of
data
have
been
accumulated
that
demonstrate
Page
22
of
35
species­
specific
sensitivities
to
agonist
activation
of
PPAR­
 ,
PPAR­
 
agonist­
induced
liver
peroxisome
proliferation
and
PPAR­
 
agonist­
induced
hepatocarcinogenesis.
As
noted
in
the
response
to
question
1,
there
was
agreement
among
most,
but
not
all
of
the
Panel
that
data
from
PPAR­
 
null
mice,
showing
that
in
the
absence
of
the
receptor,
there
were
no
ensuing
changes
in
cell
proliferation
and
hepatic
tumor
formation,
was
strong
evidence
that
activation
of
PPAR­
 
is
necessary
for
all
subsequent
steps
in
the
MOA.
It
also
was
noted
in
the
response
to
question
1
that
a
few
Panel
members
expressed
concern
over
the
short
duration
of
the
studies
in
the
PPAR­
 
null
mice
(
i.
e.,
11
months
vs.
24
months
in
standard
cancer
bioassays),
which
rendered
the
studies
incapable
of
assessing
the
lifetime
liver
cancer
risk
of
PPAR­
 
agonists
in
this
knockout
mouse
model,
and
thus,
inadequate
to
conclusively
demonstrate
that
PPAR­
 
activation
is
required
for
hepatocarcinogenesis.
Considering
the
proposed
MOA,
there
was
agreement
that
PPAR­

 
is
present
in
humans
and
that
the
receptor
is
activated
in
human
liver
following
exposure
to
known
agonists.
Accordingly,
the
proposed
MOA
for
PPAR­
 
agonist­
induced
hepatocellular
carcinogenesis
in
rodents
is
plausible
for
humans.
There
was
also
agreement
that
the
nature
of
gene
expression
associated
with
hepatocellular
PPAR­
 
activation
is
qualitatively
different
between
humans
and
rodents.
This
difference
may
result
from
species
differences
in
PPREs,
but
there
are
few
data
available
that
identify
these
potentially
important
differences,
particularly
in
humans.
Humans
are
at
least
as
sensitive
to
activation
end­
points
that
lead
to
hypolipidemia
but
are
much
less
sensitive
to
other
end­
points
normally
associated
with
peroxisome
proliferation.
This
qualitative
difference
will
be
what
is
referred
to
in
subsequent
references
as
human
sensitivity.

One
overall
concern
with
the
proposed
MOA
was
noted
in
the
response
to
question
1
and
is
also
a
concern
regarding
the
application
of
the
MOA
to
addressing
human
relevance.
Whereas
PPAR­
 
activation
is
a
very
specific
component
of
the
MOA,
the
other
steps
deemed
to
be
causally­
related,
namely
increased
hepatocellular
proliferation
and
clonal
expansion
of
initiated
hepatocytes
leading
to
tumor
development
were
very
general
and
non­
specific.
Overall,
the
Panel
members
agreed
that
additional
evidence
of
specific
alterations
associated
with
PPAR­
 
activation
in
primates
and
especially
humans
would
greatly
strengthen
the
proposed
MOA.

Although
much
of
the
data
cumulatively
support
the
hypothesis
that
agonistinduced
human
PPAR­
 
(
hPPAR­
 )
activation
fails
to
follow
the
MOA
seen
in
rodent
livers,
namely,
increased
liver
cell
proliferation,
decreased
apoptosis,
formation
of
preneoplastic
foci
and
clonal
expansion
of
these
foci
into
liver
tumors,
the
weight
of
evidence
for
this
MOA
and
consequences
of
agonist­
induced
PPAR­
 
activation
events
in
humans
is
less
well
defined
than
in
rodents.
Human
liver
biopsy
data,
while
limited,
indicate
that
clinical
administration
of
PPAR­
 
agonists
results
in
increases
in
the
number
and
volume
density
of
hepatic
peroxisomes.
The
Panel
agreed
that
the
available
cancer
epidemiological
data
on
pharmacologic
PPAR­
 
agonists
are
too
limited
in
study
size
and
duration
to
provide
any
relevant
information
to
evaluate
human
relevance.
As
such,
data
from
other
animals,
including
non­
human
primates,
along
with
in
vitro
studies
in
human
Page
23
of
35
hepatocytes,
or
cell
lines,
provide
the
basis
for
evaluating
the
relevance
of
the
proposed
MOA
in
humans.

The
available
data
from
other
animals
includes
guinea
pigs,
hamsters,
dogs
and
non­
human
primates.
In
all
cases,
these
animals
demonstrate
reduced
liver
sensitivities
to
PPAR­
 
agonists.
Hamsters
have
a
functional
PPAR­
 
receptor
but
are
intermediate
in
response
between
rats
(
and
mice)
and
humans,
and
no
increased
cell
proliferation
or
liver
tumors
have
been
observed
in
hamsters
(
Lake
et
al.,
1993).
Similarly,
PPAR­
 
is
constitutively
present
in
guinea
pigs,
albeit
at
lower
levels
than
rats
or
mice,
and
guinea
pigs
are
also
less
sensitive
than
rats
and
mice
to
PPAR­
 
activation
(
Roberts
et
al.,
2000).
Data
from
non­
human
primates
are
limited,
but
generally
indicate
that
PPAR­
 
agonists
do
not
elicit
the
typical
pattern
of
histopathological
and
biochemical
changes
associated
with
peroxisome
proliferation
in
rats
and
mice,
as
the
non­
human
primate
responses
to
PPAR­
 
agonists
have
involved
changes
of
lesser
magnitude
in
fewer
of
the
histopathological
and
biochemical
markers
(
Reddy
et
al.,
1984;
Lalwani
et
al.,
1985;
Lake
et
al.,
1989;
Graham
et
al.,
1994;
and
Kurata
et
al.,
1998).
Collectively,
the
Panel
was
split
on
the
applicability
of
data
from
other
animals
to
contribute
to
a
weight
of
evidence
regarding
the
hepatocarcinogenic
effects
of
PPAR­
 
agonists
in
humans.
All
Panel
members
recognized
that
the
data
on
non­
rodent,
non­
human
species
provided
relevant
information
on
the
reduced
activity
of
PPAR­
 
agonists
and
contributed
to
the
MOA.
Also,
while
all
Panel
members
recognized
the
limitations
of
these
data
(
number
of
compounds
studied,
study
sizes,
and
study
durations),
some
believed
that
the
data
were
sufficient
to
conclude
the
MOA
was
working,
whereas
others
were
concerned
that
the
limitations
were
significant
enough
to
disagree
with
the
MOA.

There
was
a
general
consensus
that
the
data
linking
PPAR­
 
activation
to
increased
cell
proliferation
in
all
species
was
relatively
weak.
The
strongest
evidence
in
support
of
the
importance
of
this
step
in
subsequent
tumor
development
is
derived
from
the
PPAR­
 
knockout
mouse
studies
in
which
no
increase
in
hepatic
cell
proliferation
and
no
tumors
are
observed
after
11
months
of
treatment
(
Peters
et
al.,
1997).
The
Panel
was
again
divided
on
the
conclusions
that
can
be
reached
from
studies
in
the
knockout
mouse,
as
some
were
convinced
by
such
data,
whereas
others
felt
that
the
overall
susceptibility
of
this
mouse
model
to
hepatocarcinogenesis
in
11
months
had
not
been
defined.

The
strength
of
the
hypothesis
that
humans
are
less
sensitive
to
agonist­
induced
PPAR­
 ­
mediated
hepatocarcinogenesis
lies
in
the
human
primary
hepatocyte
data.
The
Panel
was
again
divided
on
the
interpretation
and
utility
of
these
data.
First,
there
was
a
difference
of
opinion
on
the
applicability
of
the
in
vitro
studies
used
to
assess
the
ability
of
human
hepatocytes
to
proliferate
in
response
to
treatment
with
a
PPAR­
 
agonist.
Although
limited
in
total
sample
size,
these
studies
have
shown
that
in
vitro
cultured
human
hepatocytes
respond
differently
to
PPAR­
 
agonists
when
compared
to
in
vitro
cultured
rodent
hepatocytes.
As
discussed
in
more
detail
below,
whether
these
differences
are
attributable
to
true
interspecies
differences
or
reflect
differences
in
human
and
rodent
Page
24
of
35
hepatocyte
culture
preparations
remains
an
open
question.
In
parallel
experiments
with
in
vitro
cultured
rodent
hepatocytes,
in
vitro
cultured
human
hepatocytes
fail
to
display
several
of
the
key
responses
deemed
essential
for
the
MOA
in
agonist­
induced
PPAR­
 ­
mediated
rodent
hepatocarcinogenesis,
those
being
increased
cell
proliferation
and
decreased
apoptosis.
Furthermore,
in
vitro
cultured
human
hepatocytes
appear
to
be
less
responsive
to
upregulation
of
peroxisomal
genes
and
proliferation
of
peroxisomes,
two
key
associative
events
of
agonist­
induced
PPAR­
 ­
mediated
rodent
hepatocarcinogenesis.
Several
Panel
members
suggested
that
further
experiments
in
human
primary
hepatocytes
(
co­
cultured
with
and
without
Kupffer
cells;
see
comments
below)
would
be
useful
if
they
provide
additional
biochemical
data
that
demonstrate
reduced
levels
of
PPAR­
 
expression
in
human
liver
and
an
inability
for
agonist­
induced
PPAR­
 
to
modulate
the
gene
expression
for
several
key
peroxisomal
enzymes.
Such
experiments
would
strongly
support
the
hypothesis
that
human
liver
cells
are
less
sensitive
to
agonist­
induced
PPAR­
 ­
mediated
hepatocarcinogenesis.
Positive
controls
for
known
hPPAR­
 
responsive
gene
products
should
be
included
in
such
experiments
(
see,
for
example,
Lawrence
et
al.
2001).

Those
who
disagreed
with
the
conclusions
noted
above
based
their
opinion
largely
on
data
that
suggest
that
Kupffer
cells
are
required
to
elicit
a
proliferative
response
in
cultured
hepatocytes.
Specifically,
evidence
is
emerging
that
supports
a
role
for
Kupffer
cell
activation
on
the
induction
of
DNA
synthesis,
and
subsequent
neoplastic
development
following
PPAR­
 
agonist
treatment.
In
vivo
studies
have
shown
that
depletion
of
Kupffer
cells
or
inhibition
of
Kupffer
cell
activation
prevents
the
induction
of
DNA
synthesis
by
several
PPAR­
 
agonists.
These
findings
suggest
that
the
lack
of
response
from
PPAR­
 
agonist
exposure
in
human
hepatocytes
in
vitro,
may
be
due
to
the
lack
of
nonparenchymal
cells
in
the
hepatocyte
preparations.
For
example,
the
growth
permissive
factors
released
from
activated
Kupffer
cells
following
PPAR­
 
agonist
exposure
are
absent
and
may
explain
the
lack
of
induction
of
DNA
synthesis
seen
in
cultured
human
hepatocytes.
Support
for
this
possibility
has
been
demonstrated
in
rodent
cultures
in
vitro
(
Rose,
et
al.,
1999).
In
these
studies,
PPAR­
 
agonists
were
unable
to
induce
DNA
synthesis
in
purified
preparations
of
rodent
hepatocytes
(
devoid
of
nonparenchymal
cells),
while
PPAR­
 
agonist­
induced
DNA
synthesis
was
restored
upon
the
addition
of
nonparenchymal
cells,
or
medium
derived
from
activated
Kupffer
cells,
to
the
purified
hepatocyte
cultures.

It
was
noted
that
arguments
against
the
involvement
of
the
Kupffer
cells
comes
from
studies
in
the
PPAR­
 
null
mice.
In
these
mice,
agonists
failed
to
elicit
a
DNA
synthetic
response.
Since
this
model
is
replete
with
Kupffer
cells,
the
lack
of
DNA
synthesis
has
been
interpreted
as
indicating
that
the
Kupffer
cell
is
not
required.
On
the
other
hand,
some
members
of
the
Panel
felt
that
the
communication
and/
or
interplay
between
PPAR­
 
agonism
and
Kupffer
cells
has
not
been
fully
characterized
and
as
such,
the
null
mouse,
lacking
PPAR­
 ,
is
not
directly
applicable
to
the
human
situation
in
which
PPAR­
 
is
present
and
can
be
activated.
Page
25
of
35
With
regard
to
the
human
data,
the
Panel
noted
deficiencies
arising
from
studies
in
which
the
duration
of
exposures
to
PPAR­
 
agonists
were
significantly
less
than
lifetime,
the
exposure
levels
were
at
therapeutic
doses,
and
the
populations
of
exposed
individuals
were
fairly
small.
As
stated
previously,
the
Panel
agreed
that
the
available
cancer
epidemiological
data
on
pharmacologic
PPAR­
 
agonists
are
too
limited
in
study
size
and
duration
to
provide
sufficient
information
to
evaluate
human
risk
potential.
Although
the
human
data
are
limited,
the
existing
data
do
provide
some
important
information
for
consideration.
Human
liver
contains
functional
PPAR­
 
receptors
and
the
fibrate
class
of
drugs
is
able
to
activate
this
receptor
to
alter
the
expression
of
genes
involved
in
lipid
metabolism
that
induce
hypolipidemia.
Chronic
exposure
data
reported
in
humans
for
two
different
PPAR­
 
agonists
suggest
that
humans
do
not
respond
to
PPAR­
 
agonists
by
an
increase
of
the
associated
key
events
(
such
as
cell
proliferation,
suppressed
apoptosis,
and
clonal
expansion
of
preneoplastic
hepatic
lesions)
observed
during
PPAR­
 
activation
in
rats
and
mice
exposed
to
these
agonists.
In
addition
to
the
short
duration
of
exposure
and
the
use
of
therapeutic
doses
(
lower
than
the
doses
used
in
studies
with
rats
and
mice),
the
limitations
of
these
studies
include
the
use
of
weak
agonists.
The
human
epidemiology
data
from
short
duration
follow
up
(
5
year
time
period)
indicated
an
early
increase
in
GI
tract
tumors,
although
liver
cancer
was
not
reported
independently.
However,
no
differences
were
noted
after
8
years
of
follow
up.
Evidence
for
peroxisome
proliferation
and
increased
cell
proliferation
was
lacking
in
human
liver
biopsies.
Problems
with
these
observations
include
the
high
variability
in
assessing
peroxisome
increases
in
biopsy
material
that
are
not
representative
of
all
zones
of
the
liver,
and
whether
the
timing
of
biopsy
sample
acquisition
was
appropriate
for
detecting
an
increase
in
cell
proliferation.
A
slight
increase
in
the
number
and
density
of
peroxisomes
is
observed
in
humans
with
chronic
exposure
to
therapeutic
levels
of
a
PPAR­
 
agonist.
This
level
is
indicative
of
normal
physiologic
or
metabolic
changes
and
is
lower
than
the
approximately
three
fold
level
defined
by
Ashby
et
al.
(
1994)
as
the
threshold
level
of
peroxisome
induction
associated
with
liver
cancer
risk
in
rats
and
mice.
These
observations
in
humans
are
strengthened
by
the
studies
of
chronic
exposure
of
non­
human
primates
to
PPAR­
 
agonists
for
5
or
more
years.
Again,
the
number
of
non­
human
primates
exposed
was
limited
and
the
duration
of
exposure
was
less
than
lifetime.
Assessment
of
the
presence
or
absence
of
PPAR­
 
regulated
gene
expression
and
of
preneoplastic
lesions
needs
to
be
detailed
in
primates
compared
to
rats
and
mice
following
exposure
to
PPAR­
 
agonists.
The
non­
human
primate
appears
to
have
a
markedly
attenuated
response
to
fairly
potent
PPAR­
 
agonists
(
e.
g.,
ciprofibrate)
compared
with
rats
and
mice,
although,
as
with
the
human
data,
the
PPAR­
 
agonist
challenge
has
been
at
lower
doses
of
shorter
duration.
Studies
by
Pugh
et
al.,
(
2000)
wherein
numerous
PPAR 
agonists
were
administered
to
nonhuman
primates
support
this
contention
in
that
a
lack
of
increase
in
liver
weights
indicates
a
lack
of
cell
proliferation
as
verified
by
replicative
DNA
synthesis.

The
Panel
discussed
three
other
issues
relative
to
assessing
weight
of
evidence
regarding
the
hepatic
effects
of
PPAR­
 
agonists
in
humans,
and
the
proposed
science
policy
regarding
human
relevance.
These
included:
Page
26
of
35
1.
The
use
of
the
word
"
refractory"
to
describe
the
human
response
to
PPAR­
 
activation
is
too
absolute.
The
Panel
agreed
that
"
less
sensitive"
is
a
more
appropriate
description
of
the
nature
of
the
human
response
relative
to
that
observed
in
rats
and
mice.

2.
The
policy
statement
drafted
by
OPPTS
concludes
with
the
phrase
"
evidence
of
liver
tumor
formation
in
rodents
should
not
be
used
to
characterize
potential
human
hazard."
After
some
discussion,
it
was
suggested
by
one
member
of
the
Panel,
and
supported
by
several
other
Panel
members,
that
this
phrase
should
be
modified
to
read,
"
evidence
of
liver
tumor
formation
in
rodents
should
not
be
used
to
characterize
potential
human
hepatocarcinogenic
hazard."

3.
One
member
of
the
Panel
expressed
concern,
which
was
shared
by
some
other
Panel
members,
that
the
MOA
and
evaluation
of
human
relevance
was
lacking
in
its
assessment
of
altered
gene
expression
that
could
be
associated
with
altered
methylation
of
DNA.
There
is
evidence
that
DNA
methylation
is
modified
in
rodents
following
exposure
to
PPAR­
 
agonists
(
Ge
et
al.,
2001,
Ge
et
al.,
2002,
and
Pereira,
et
al.,
2004).
Given
the
accepted
role
for
DNA
methylation
in
gene
imprinting
and
the
loss
of
imprinting
in
cancer
etiology
(
see
for
example
McClachlan
et
al.,
2001),
such
a
role
for
PPAR­
 
agonists
in
causing
similar
alterations
in
humans
should
be
explored
before
human
relevance
can
be
appropriately
evaluated,
particularly
for
exposure
during
early
life
stages
and
for
questions
regarding
site
concordance.

Question
4
 
Data
Requirements
OPPTS
has
proposed
a
data
set
that
would
be
sufficient
to
demonstrate
that
PPAR­
 
agonism
is
the
MOA
for
the
induction
of
rodent
liver
tumors.
The
data
set
includes
evidence
of
PPAR­
 
agonism
(
i.
e.,
from
an
in
vitro
reporter
gene
assay),
in
vivo
evidence
of
an
increase
in
number
and
size
of
peroxisomes,
increases
in
the
activity
of
acyl
CoA
oxidase,
and
hepatic
cell
proliferation.
The
in
vivo
evidence
should
be
collected
from
studies
designed
to
provide
the
data
needed
to
show
dose­
response
and
temporal
concordance
between
precursor
events
and
liver
tumor
formation.

Please
comment
in
general
on
the
proposed
data
set
and
particularly
on
its
adequacy
to
demonstrate
that
a
PPAR­
 
agonist­
mediated
MOA
is
operating
in
rodent
hepatocarcinogenesis.

Response
Data
requirements
refer
to
the
experimental
data
needed
to
demonstrate
that
a
compound
acts
through
a
PPAR­
 
agonist
MOA.
These
data
may
be
used
subsequent
to
a
bioassay
that
finds
induction
of
hepatic
tumors
to
demonstrate
such
tumors
arose
from
a
PPAR­
 
agonist
MOA,
or
subsequent
to
initial
(
sub)
acute
experiments
to
assist
in
the
Page
27
of
35
subsequent
experimental
design
of
long­
term
experiments
for
submission
to
the
Agency.
This
use
of
the
data
may
dictate
some
differences
in
the
data
requirements
needed.
The
following
discussion
focuses
on
requirements
after
a
positive
bioassay,
with
suggestions
provided
for
the
converse
situation.

There
was
general
consensus
among
the
Panel
that
the
proposed
data
set
was
adequate
and
provided
a
straight
forward
approach
to
classifying
a
chemical
as
a
PPAR­
 
agonist.
The
Panel
also
concurred
that
the
use
of
PPAR­
 
knockout
mice
would
be
definitive
evidence
to
ascribe
a
chemical
as
a
PPAR­
 
agonist,
but
that
the
proposed
data
set
would
be
sufficient
in
lieu
of
the
use
of
this
rather
costly
tool.
While
the
Panel
agreed
with
these
data
needs,
they
suggested
some
clarifications
and
additional
supportive
approaches.

The
clarifications
indicated
were
as
follows:
the
term
`
direct
DNA
reactivity'
may
need
to
be
clarified
as
`
direct'
may
be
interpreted
by
some
to
mean
"
without
metabolism";
in
keeping
with
the
ILSI
document
(
Klaunig
et
al.,
2003),
rather
than
using
the
term
`
mutagenicity'
alone,
the
terms
`
mutagenicity
and/
or
clastogenicity'
may
be
more
appropriate;
palmitoyl
CoA
activity
is
simply
a
substrate­
specific
name
for
acyl
CoA
oxidase
activity;
and
microsomal
fatty
acid
oxidation
(
as
opposed
to
microsomal
fatty
acid
omega­
oxidation)
is
not
specific
enough
to
designate
CYP4A
activity.

In
the
course
of
the
Panel's
discussion,
questions
for
clarification
were
posed
to
the
Agency
as
to
when
(
i.
e.,
before
or
after
a
positive
liver
tumor
finding
in
rodents)
this
set
of
assays
testing
for
PPAR­
 
agonist
activity
would
be
conducted.
The
Agency
indicated
that
data
demonstrating
PPAR­
 
agonist
activity
could
be
submitted
in
the
absence
of
testing
in
long­
term
carcinogenesis
studies.
In
response
to
this,
a
Panel
member
observed
that
in
the
absence
of
testing
in
standard
long­
term
rodent
carcinogenicity
studies,
it
is
not
possible
to
determine
whether
the
chemical
would
operate
through
a
PPAR­
 
agonist
MOA
producing
rodent
liver
tumors.
A
chemical
with
PPAR­
 
agonist
activity
may
either:
1)
not
cause
cancer
in
rodents,
2)
cause
liver
cancer
in
rodents
by
the
proposed
PPAR­
 
agonist
MOA,
3)
cause
liver
cancer
by
a
MOA
other
than
the
proposed
PPAR­
 
agonist
MOA
(
e.
g.,
cytotoxicity),
or
4)
cause
cancer
at
sites
other
than
the
liver
(
with
or
without
liver
cancer).
The
Panel
concurred
that
an
overriding
requirement
is
that
other
MOAs
have
been
excluded.
For
example,
rigorous
tests
must
be
performed
to
exclude
mutagenicity,
other
forms
of
DNA
damage
(
clastogenicity),
or
overt
cytotoxicity
directly
produced
by
the
test
compound,
or
its
metabolic
products.

The
Panel
also
concurred
that
direct
evidence
of
the
activation
of
PPAR­
 
is
required
to
show
that
complementary
in
vivo
results
do
not
result
from
activation
of
other
PPARs
or
from
an
unknown
mechanism
as
exemplified
by
dehydroepiandrosterone
(
DHEA)
(
Isseman
and
Green,
1990,
Peters,
et
al.,
1996
and
Waxman,
1996).
The
activation
of
PPAR­
 
is
often
demonstrated
using
chimeric
systems
that
include
an
expression
system
for
the
PPAR­
 
receptor
and
a
reporting
system
that
includes
the
PPRE
Page
28
of
35
in
the
promoting
region.
It
was
recommended
by
one
Panel
member
that
this
study
could
be
supplemented
by
gene­
dosage
experiments
in
knockout
mice
or
transgenic
mice
overexpressing
the
receptor
with
respective
loss
or
exacerbation
of
responsiveness.
These
experiments
would
demonstrate
a
direct
effect
of
the
receptor
on
a
true
genomic
PPRE,
rather
than
a
construct.
It
was
also
recommended
that
it
be
acknowledged
that
in
some
cases
a
metabolite
of
the
test
compound
may
be
a
more
suitable
substrate
to
use
in
these
experiments.
Direct
involvement
of
PPAR­
 
can
alternatively
be
assessed
using
in
vivo
experiments
with
wild
type
(
PPAR 
+/+)
and
knockout
mice
(
PPAR 
­/­);
endpoints
for
these
in
vivo
experiments
are
discussed
below.
Compounds
with
positive
bioassays
in
rats
but
not
mice
would
not
be
suitable
for
this
alternative
approach.

In
vivo
experiments
should
be
conducted
using
doses
that
produce
positive
bioassays;
as
they
are
normally
(
sub)
acute
they
will
meet
temporal
requirements
that
they
occur
prior
to
tumor
formation.
Of
highest
priority,
they
must
demonstrate
an
increase
in
hepatocyte
cell
replication/
reduced
apoptosis,
induction
of
peroxisomal
acylCoA
oxidase
and
an
increase
in
number
and
volume
percent
of
peroxisomes.
Demonstration
of
induction
of
other
enzymes
with
PPRE
sequences
in
the
promoter
region
(
CYP4A,
carnitine
acetyl
transferase,
fatty
acid
binding
protein,
etc.)
or
catalase
provides
supportive
evidence.
It
was
recommended
that
at
least
one
`
supportive
example
of
enzyme
induction'
be
included.
Induction
of
enzymes
can
be
demonstrated
from
increased
enzyme
activity
and/
or
increased
expression
of
mRNA.
It
was
also
noted
that
the
need
to
show
both
increases
in
peroxisome
volume
percent
and
density
would
require
morphometric
analysis
of
liver
sections
examined
by
electron
microscopy
(
demonstration
of
increased
density,
but
not
volume
percent,
could
be
approached
using
light
microscopic
methods).

One
Panel
member
suggested
that
when
acute
evidence
of
a
PPAR­
 
agonist
MOA
has
been
found
prior
to
long­
term
dosing
studies,
the
evidence
of
the
MOA
can
be
further
enhanced
by
inclusion
of
an
initiation/
promotion
test
system
where
the
test
compound
is
administered
as
the
promoter
after
suitable
initiation.
These
experiments
demonstrate
the
key
event
of
clonal
expansion.
In
addition,
there
are
some
(
immuno)
histochemical
stains
that
can
be
used
to
show
a
greater
degree
of
specificity
for
this
MOA.
It
was
acknowledged
that
while
such
experiments
would
further
support
the
MOA,
they
were
fairly
time­
and
cost­
inefficient
with
regard
to
the
main
objective
of
demonstrating
that
the
compound
is
a
PPAR­
 
agonist.

Question
5
 
Other
Tumors
Induced
by
PPAR­
 
Agonists
Some
PPAR­
 
agonists
may
also
induce
pancreatic
acinar
cell
and
Leydig
cell
tumors
in
rats
and
modes
of
action
involving
agonism
of
PPAR­
 
have
been
proposed.
An
in
depth
analysis
of
these
tumors
is
provided
in
the
2003
ILSI
technical
panel
report.
Based
on
this
analysis,
OPPTS
agrees
that
the
data
available
to
date
are
insufficient
to
support
the
proposed
MOAs.
Page
29
of
35
Thus,
OPPTS
is
proposing
the
following
science
policy:

Given
the
limited
evidence
available
to
support
that
a
chemical
may
induce
pancreatic
and
Leydig
cell
tumors
through
a
PPAR­
 
agonist
MOA,
the
evidence
is
inadequate
at
this
time
to
support
a
linkage
between
PPAR­
 
agonism
and
formation
of
these
tumor
types.
Thus,
it
is
presumed
that
chemicals
that
induce
pancreatic
or
Leydig
cell
tumors
may
pose
a
carcinogenic
hazard
for
humans.

Please
comment
on
OPPTS's
conclusion
that
there
is
limited
evidence
that
a
chemical
may
induce
pancreatic
and
Leydig
cell
tumors
through
a
PPAR­
 
agonist
MOA,
and
OPPTS's
proposed
science
policy
regarding
other
tumors
induced
by
PPAR­
 
agonists.

Response
In
addition
to
the
hepatic
tumors
that
appear
to
be
a
general
occurrence
in
rats
and
mice,
nine
PPAR­
 
agonists
have
been
reported
to
induce
Leydig
cell
tumors
(
LCTs)
and
pancreatic
acinar
cell
tumors
(
PACTs)
in
rats.
Together
with
the
hepatic
tumors,
this
is
referred
to
as
the
tumor
triad.
The
Panel
was
in
agreement
with
the
OPPTS
conclusion
that
chemicals
that
induce
pancreatic
or
Leydig
cell
tumors
may
pose
a
carcinogenic
hazard
for
humans.

LCTs
were
most
apparent
when
PPAR­
 
agonists
were
tested
in
non­
F344
male
rats,
likely
because
by
2
years
of
age,
the
F344
rat
has
virtually
a
100%
incidence
of
spontaneously
occurring
LCTs.
This
will
obscure
any
ability
to
detect
a
xenobioticinduced
testicular
tumor
in
this
strain.
The
finding
that
a
relationship
appears
to
exist
between
PPAR­
 
agonists
and
LCT
formation
has
led
to
speculation
that
many,
if
not
all,
such
agonists
would
induce
this
tumor
if
tested
adequately
in
a
rat
strain
other
than
F344.
This
speculation
has
been
supported
by
limited
studies
in
other
strains
(
Biegel,
et
al.,
2001,
Maltoni,
et
al.,
1988
and
Mennear,
1988).

It
was
originally
hypothesized
that
PPAR­
 
agonists
cause
LCTs
by
a
PPAR­
 ­
dependent
mechanism
similar
to
that
of
the
liver.
However,
evidence
exists
using
PPAR­
 
null
mice
(
Ward
et
al.,
1998)
to
suggest
that
the
PPAR­
 
agonist
DEHP
can
induce
toxic
lesions
in
kidney
and
testis
independently
of
this
receptor.
In
addition,
although
Leydig
and
pancreatic
acinar
cells
contain
PPAR­
 ,
agonists
do
not
appear
to
induce
peroxisome
proliferation
in
these
cells.
This
suggests
that
tumors
developing
in
these
tissues
in
rats
do
so
via
different
mechanisms
than
in
the
liver
where
peroxisome
proliferation
is
always
observed.
A
prevailing
hypothesis
is
that
PPAR­
 
agonists
cause
an
increase
in
estradiol
that
promotes
the
secretion
of
transforming
growth
factor
(
TGF­
 ).
Evidence
in
support
of
this
hypothesis
is
that
PPAR­
 
agonists
increase
the
expression
of
aromatase,
an
enzyme
that
under
normal
conditions
maintains
serum
estradiol
concentrations
by
Page
30
of
35
converting
testosterone
to
estradiol
(
Biegel,
et
al.,
1995).
Estradiol
stimulates
TGF­
 
production
which
induces
Leydig
cell
proliferation
(
Khan,
Teerds,
and
Dorrington,
1992).

Another
proposed
MOA
of
PPAR­
 
agonist­
induced
LCTs
is
that
they
cause
testicular
hypertrophy
by
decreasing
testosterone
biosynthesis,
leading
to
an
imbalance
of
androgen/
estrogen
levels.
This
leads
to
an
increase
in
leutinizing
hormone
(
LH)
which
promotes
LCTs.
However,
it
is
not
known
whether
steroid
synthesis
pathways
in
testis
are
regulated
by
PPAR­
 ,
and
in
Cook
et
al.
(
2001)
no
changes
in
LH
were
observed.

The
Panel
agreed
that
although
some
data
suggest
LCTs
may
involve
PPAR­
 ,
additional
research
will
be
required
to
confirm
this
role.
In
addition,
the
link
to
PPAR­
 
activation
is
considered
tenuous
because
limited
studies
of
PPAR­
 

agonists
in
other
animal
species,
such
as
the
mouse,
hamster
and
nonhuman
primates,
did
not
show
extrahepatic
carcinogenic
responses,
including
PACTs
and
LCTs.
As
noted
previously,
the
Panel
agreed
that
the
available
cancer
epidemiological
data
for
pharmaceutical
PPAR­
 
agonists
are
too
limited
in
study
size
and
duration
to
be
informative
as
to
cancer
risk
at
any
site.
While
LCT
data
in
mice
remain
limited,
this
species
difference
from
rats
is
certainly
indicative
of
some
unique
feature
either
in
rats
which
causes
the
tumors,
or
in
mice
which
are
resistant.
Further
data
are
needed
to
determine
which
is
the
case.
It
is
also
noteworthy
that
the
spontaneous
rate
of
LCTs
is
much
lower
in
humans
than
in
rats
suggesting
innate
resistance
to
this
type
of
cancer,
and
that
rat
and
human
Leydig
cells
respond
differently
to
human
chorionic
gonadotropin
(
human
cells
undergo
hypertrophy
while
rat
cells
proliferate).
Finally,
a
human
condition
with
constant
LH
receptor
activation
does
not
lead
to
LCTs,
even
though
this
is
one
of
the
major
proposed
MOAs
in
rats.

Key
events
in
the
postulated
MOA
for
PACTs
in
rats
are
considered
to
begin
with
PPAR­
 
activation
in
the
liver,
followed
by
changes
in
bile
composition
and
a
decrease
in
its
synthesis.
This
results
in
cholestasis
and
a
sustained
increase
in
cholecystokinin.
This
stimulates
acinar
cell
proliferation
and
promotes
the
development
of
PACTs.
If
this
is
true,
then
the
rat
PACTs
are
secondary
to
the
liver
effects
of
PPAR­
 
agonists.
Some
data
indicate
that
many
of
the
non­
hepatocarcinogenic
parameters
and
symptoms
manifested
in
rodents
upon
long­
term
administration
of
PPAR­
 
agonists
are
also
manifested
in
humans.
This
is
particularly
true
since
it
has
been
observed
in
rodents
that
long­
term
administration
of
PPAR­
 
agonists
results
in
marked
changes
in
bile
acid
secretion
and
composition.
In
human
studies
it
is
also
established,
by
multiple
investigators,
that
fibric
acid
drug
treatment
increases
biliary
cholesterol
and
induces
supersaturation
of
bile.
Studies
demonstrating
that
hPPAR­
 
is
functional
in
the
regulation
of
a
variety
of
enzymes
associated
with
bile
acid
metabolism
in
human
liver
cells
would
suggest
that
the
risks
of
PACTs
in
humans
exposed
to
PPAR­
 
agonists
could
involve
a
PPAR­
 
mechanism.
However,
the
data
are
not
sufficient
to
firmly
conclude
that
this
MOA
is
operative.
Furthermore,
the
difference
between
rodents
and
humans
in
the
cellular
origin
of
pancreatic
tumors
(
acinar
in
rat,
ductal
in
humans)
suggests
that
these
animal
data
are
of
Page
31
of
35
limited
relevance
to
humans.
Again,
although
data
in
other
species
are
limited,
only
rats
have
shown
these
tumors.

Finally,
in
addition
to
PPAR­
 
agonism
as
a
potential
MOA
of
extrahepatic
tumors,
as
noted
previously,
one
member
of
the
Panel
expressed
concern,
which
was
shared
by
some
other
Panel
members,
that
consideration
needs
to
be
given
to
epigenetic
phenomena
that
may
be
activated
by
these
chemicals.
DNA
methylation
and
chromatin
structure
alterations
are
significant
nongenotoxic
mechanisms
involved
in
deregulating
gene
function.
Furthermore,
PPAR­
 
agonists
inhibit
methylation
during
DNA
replication
(
Ge
et
al.,
2001),
thereby
altering
the
cellular
epigenome.
This
is
important
since
the
earliest
change
identified
in
tumor
cells
compared
to
their
normal
counterpart
is
genomewide
hypomethylation
(
Feinberg
and
Vogelstein,
1983).
These
changes
can
be
particularly
critical
during
development,
including
puberty,
but
even
adults
vary
dramatically
in
their
susceptibility
to
cancer
because
of
marked
differences
in
the
epigenome.
For
example,
there
is
now
evidence
that
approximately
10%
of
the
human
population
is
at
high
risk,
at
least
for
colon
cancer,
because
of
either
an
inability
to
maintain
imprinting
at
the
IGF2
locus
or
exposure
early
in
development
to
some
environmental
factor
resulted
in
IGF2
loss
of
imprinting
(
Cui
et
al.,
2003).
It
is
conceivable
that
these
"
preneoplastic"
individuals
are
more
susceptible
to
PPAR­
 
agonists
than
the
general
population.

In
summary,
given
the
limited
amount
of
data
available
on
the
true
MOA
for
LCTs
or
PACTs,
including
the
possibility
raised
by
some
Panel
members
that
epigenetic
effects
of
the
PPAR­
 
agonists
may
occur,
it
is
not
possible
to
determine
whether
PPAR­
 
agonists
pose
a
carcinogenic
hazard
to
humans.
Thus,
the
conclusion
by
the
OPPTS
that
the
available
data
for
the
induction
of
rat
LCTs
and
PACTs
by
PPAR­
 
agonists
are
insufficient
to
conclude
that
the
sole
MOA
involves
the
PPAR­
 
receptor
is
considered
by
the
Panel
to
be
appropriate.
Further,
the
Panel
concurs
that
it
should
be
presumed
that
chemicals
that
induce
pancreatic
or
Leydig
cell
tumors
may
pose
a
carcinogenic
hazard
for
humans.

REFERENCES
Ashby,
J.,
Brady,
A.,
Elcombe,
C.,
Elliott,
B.,
Ishmael,
J.,
Odum,
J.,
Tugwood,
J.,
Kettle,
S.,
and
Purchase,
I.,
(
1994),
Mechanistically­
based
human
hazard
assessment
of
peroxisome
proliferator­
induced
hepatocarcinogenesis,
Human
and
Experimental
Toxicology,
Suppl.
2,
S1­
117.

Biegel,
L.,
Liu,
R.,
Hurtt,
M.,
and
Cook,
J.,
(
1995),
Effects
of
ammonium
perfluorooctanate
on
Leydig
cell
function:
in
vitro,
in
vivo,
and
ex
vivo
studies,
Toxicology
and
Applied
Pharmacology,
134(
1),
pp.
18­
25.
Page
32
of
35
Cappon,
G.,
Keller,
D.,
Brock,
W.,
Slauter,
R.,
and
Hurtt,
M.,
(
2002),
Effects
of
HCFC­
123
exposure
to
maternal
and
infant
Rhesus
monkeys
on
hepatic
biochemistry,
lactational
parameters,
and
postnatal
growth,
Drug
and
Chemical
Toxicology,
25(
4),
pp.
481­
496.

Cui,
H.,
Cruz­
Correa,
M.,
Giardiello,
F.,
Hutcheon,
D.,
Kafonek,
D.,
Brandenburg,
S.,
Wu,
Y.,
He,
X.,
Powe,
N.,
and
Feinberg,
A.,
(
2003),
Loss
of
IGF2
imprinting:
a
potential
marker
of
colorectal
cancer
risk,
Science,
299(
5613),
pp.
1753­
1755.

Drinkwater,
N.
and
Bennett,
L.,
(
1991),
Genetic
control
of
carcinogenesis
in
experimental
animals,
Progress
in
Experimental
Tumor
Research,
33,
pp.
1­
20.

Espeel,
M.,
Depreter,
M.,
Nardacci,
R.,
D'Herde,
K.,
Kerckaert,
I.,
Stefanini,
S.,
and
Roels,
F.,
(
1997),
Biogenesis
of
peroxisomes
in
fetal
liver,
Microscopy
Research
and
Technique,
39(
5),
pp.
453­
466.

Feinberg,
A.
and
Vogelstein,
B.,
(
1983),
Hypomethylation
distinguishes
genes
of
some
human
cancers
from
their
normal
counterparts,
Nature,
301(
5895),
pp.
89­
92.

Ge,
R.,
Wang,
W.,
Kramer,
P.,
Yang,
S.,
Tao,
L.,
and
Pereira,
M.,
(
2001),
Wy­
14,643­
induced
hypomethylation
of
the
c­
myc
gene
in
mouse
liver,
Toxicological
Sciences,
62,
pp.
28­
35.

Ge,
R.,
Tao,
L.,
Kramer,
P.,
Cunningham,
M.,
and
Pereira,
M.,
(
2002),
Effect
of
peroxisome
proliferators
on
the
methylation
and
protein
level
of
the
c­
myc
protooncogene
in
B6C3F1
mice
liver,
Journal
of
Biochemical
and
Molecular
Toxicology,
16,
pp.
41­
47.

Graham,
M.,
Wilson,
S.,
Winham,
M.,
Spencer,
A.,
Rees,
J.,
Old,
S.,
and
Bonner,
F.,
(
1994),
Lack
of
peroxisome
proliferation
in
marmoset
liver
following
treatment
with
ciprofibrate
for
3
years,
Fundamental
and
Applied
Toxicology,
22(
1),
pp.
58­
64.

Hasmall,
S.,
James,
N.,
Hedley,
K.,
Olsen,
K.,
Roberts,
R.,
(
2001),
Mouse
hepatocyte
response
to
peroxisome
proliferators:
dependency
on
hepatic
nonparenchymal
cells
and
peroxisome
proliferator
activated
receptor
alpha
(
PPARalpha),
Archives
of
Toxicology,
75(
6),
pp.
357­
361.

IARC,
(
1996),
IARC
Monographs
on
the
Evaluation
of
Carcinogenic
Risks
to
Humans,
Volume
66
 
Some
Pharmaceutical
Drugs.

Isseman,
I.,
and
Green,
S.,
(
1990),
Activation
of
a
member
of
the
steroid
hormone
receptor
superfamily
by
peroxisome
proliferators,
Nature,
347(
6294),
pp.
645­
650.

Khan,
S.,
Teerds,
K.,
and
Dorrington,
J.,
(
1992),
Growth
factor
requirements
for
DNA
Page
33
of
35
synthesis
by
Leydig
cells
from
the
immature
rat,
Biology
of
Reproduction,
46(
3),
pp.
335­
341.

Klaunig,
J.,
Babich,
M.,
Baetcke,
K.,
Cook,
J.,
Corton,
J.,
David,
R.,
Deluca,
J.,
Lai,
D.,
Mckee,
R.,
Peters,
J.,
Roberts,
R.,
and
Fenner­
Crisp,
P.,
(
2003),
PPAR­
 
agonist­
induced
rodent
tumors:
modes
of
action
and
human
relevance,
Critical
Reviews
in
Toxicology,
33(
6),
pp.
655­
780.

Kurata,
Y.,
Kidachi,
F.,
Yokoyama,
M.,
Toyota,
N.,
Tsuchitani,
M.,
and
Katoh,
M.,
(
1998),
Subchronic
toxicity
of
di(
2­
ethylhexyl)
phthalate
in
common
marmosets:
lack
of
hepatic
peroxisome
proliferation,
testicular
atrophy,
or
pancreatic
acinar
cell
hyperplasia,
Toxicological
Sciences,
42(
1),
pp.
49­
56.

Lake,
B.,
Evans,
J.,
Gray,
T.,
Korosi,
S.,
and
North,
C.,
(
1989),
Comparative
studies
on
nafenopin­
induced
hepatic
peroxisome
proliferation
in
the
rat,
Syrian
hamster,
guinea
pig,
and
marmoset,
Toxicology
and
Applied
Pharmacology,
99(
1),
pp.
148­
160.

Lake,
B.,
Evans,
J.,
Cunningham,
M.,
and
Price,
R.,
(
1993),
Comparison
of
the
hepatic
effects
of
nafenopin
and
WY
14­
463
on
peroxisome
proliferation
and
cell
replication
in
the
rat
and
Syrian
hamster,
Environmental
Health
Perspectives,
101,
Suppl.
5,
pp.
241­
247.

Lake,
B.,
(
1995),
Peroxisome
Proliferation:
Current
mechanisms
relating
to
nongenotoxic
carcinogenesis,
Toxicology
Letters,
82­
83,
pp.
673­
681.

Lalwani,
N.,
Reddy,
M.,
Ghosh,
S.,
Barnard,
S.,
Molello,
J.,
and
Reddy,
J.,
(
1985),
Induction
of
fatty
acid
beta­
oxidation
and
peroxisome
proliferation
in
the
liver
of
Rhesus
monkeys
by
DL­
040,
a
new
hypolipidemic
agent,
Biochemical
Pharmacology,
34(
19),
pp.
3473­
3482.

Lawrence,
J.,
Li,
Y.,
Chen,
S.,
Deluca,
J.,
Berger,
J.,
Umbenhauer,
D.,
Moller,
D.,
and
Zhou,
G.,
(
2001),
Differential
gene
regulation
in
human
versus
rodent
hepatocytes,
by
PPAR
alpha:
PPAR
alpha
fails
to
induce
peroxisome
proliferation
associated
genes
in
human
cells
independently
of
receptor
levels,
Journal
of
Biological
Chemistry,
276,
pp.
31521­
31527.

Maltoni,
C.,
Lefemine,
G.,
Cotti,
G.,
and
Perino,
G.,
(
1988),
Long­
term
carcinogenicity
bioassays
on
trichloroethylene
administered
by
inhalation
to
Sprague­
Dawley
rats
and
Swill
and
B6C3F1
mice,
Annals
of
the
New
York
Academy
of
Sciences,
534,
pp.
316­
342.

Marsman,
D.,
Cattley,
R.,
Conway,
J.,
and
Popp,
J.
(
1988),
Relationship
of
hepatic
peroxisome
proliferation
and
replicative
DNA
synthesis
to
the
hepatocarcinogenicity
of
the
peroxisome
proliferators
di(
2­
ethylhexyl)
phthalate
and
[
4­
chloro­
6­(
2,3­
xylidino)­
2­
pyrimidinylthio]
acetic
acid
(
Wy­
14,643)
in
rats,
Cancer
Research,
48(
23),
pp.
6739­
6744.
Page
34
of
35
McClachlan,
J.,
Burow,
M.,
Chiang,
T.,
and
Li,
S.,
(
2001),
Gene
imprinting
in
developmental
toxicology:
a
possible
interface
between
physiology
and
pathology,
Toxicology
Letters,
120,
pp.
161­
164.

Mennear,
J.,
(
1988),
Toxicology
and
carcinogenesis
studies
of
trichloroethylene
in
four
strains
of
rats,
NTP
Technical
Report
273
(
NIH
Publication
No.
88­
2529),
Research
Triangle
Park,
NC.

Parzefall,
W.,
Berger,
W.,
Kainzbauer,
E.,
Teufelhofer,
O.,
Schulte
Herman,
R.,
Thurman,
R.,
(
2001),
Peroxisome
proliferators
do
not
increase
DNA
synthesis
in
purified
rat
hepatocytes,
Carcinogenesis,
22(
3),
pp.
519­
523.

Pereira,
M.,
Wang,
W.,
Kramer,
P.,
and
Tao,
L.,
(
2004),
Prevention
by
methionine
of
dichloroacetic
acid­
induced
liver
cancer
and
DNA
hypomethylation
in
mice,
Toxicological
Sciences,
77,
pp.
243­
248.

Peters,
J.,
Zhou,
Y.,
Ram,
P.,
Lee,
S.,
Gonzalez,
F.,
and
Waxman,
D.,
(
1996),
Peroxisome
proliferator­
activated
receptor
alpha
required
for
gene
induction
by
dehydroepiandrosterone­
3
beta­
sulfate,
Molecular
Pharmacology,
50(
1),
pp.
67­
74.

Peters,
J.,
Cattley,
R.,
and
Gonzalez,
F.,
(
1997),
Role
of
PPAR­
 
in
the
mechanism
of
action
of
the
nongenotoxic
carcinogen
and
peroxisome
proliferator
WY
14,643,
Carcinogenesis,
18,
pp.
2029­
2033.

Pugh,
G.
Jr.,
Isenberg,
J.,
Kamendulis,
L.,
Ackley,
D.,
Clare,
L.,
Brown,
R.,
Lington,
A.,
Smith,
J.,
and
Klaunig,
J.,
(
2000),
Effects
of
di­
isononyl
phthalate,
di­
2­
ethylhexyl
phthalate,
and
clofibrate
in
Cynomolgus
monkeys,
Toxicological
Sciences,
56(
1),
pp.
181­
188.

Reddy,
J.,
Lalwani,
N.,
Qureshi,
S.,
Reddy,
M.,
and
Moehle,
C.,
(
1984),
Induction
of
hepatic
peroxisome
proliferation
in
nonrodent
species,
including
primates,
American
Journal
of
Pathology,
114(
1),
pp.
171­
183.

Ring,
J.,
Ghabrial,
H.,
Ching,
M.,
Smallwood,
R.,
and
Morgan,
D.,
(
1999),
Fetal
hepatic
drug
elimination,
Pharmacology
and
Therapeutics,
84(
3),
pp.
429­
445.

Roberts,
R.,
James,
N.,
Hasmall,
S.,
Holden,
P.,
Lambe,
K.,
Macdonald,
N.,
West,
D.,
Woodyatt,
N.,
and
Whitcome,
D.,
(
2000),
Apoptosis
and
proliferation
in
nongenotoxic
carcinogenesis:
species
differences
and
role
of
PPARalpha,
Toxicology
Letters,
112­
113,
pp.
49­
57.

Rolfe,
M.,
James,
N.,
and
Roberts,
R.,
(
1997),
Tumor
necrosis
factor
alpha
(
TNF
alpha)
Page
35
of
35
suppresses
apoptosis
and
induces
DNA
synthesis
in
rodent
hepatocytes:
a
mediator
of
the
hepatocarcinogenicity
of
peroxisome
proliferators?,
Carcinogenesis,
18(
11),
pp.
2277­
2280.

Rose,
M.,
Rivera,
C.,
Bradford,
B.,
Graves,
L.,
Cattley,
R.,
Schoonhoven,
R.,
Swenberg,
J.,
and
Thurman,
R.,
(
1999),
Kupffer
cell
oxidant
production
is
central
to
the
mechanism
of
peroxisome
proliferators,
Carcinogenesis,
20(
1),
pp.
27­
33.

Rusyn,
I.,
Kadiiski,
M.,
Dikalova,
A.,
Kono,
H.,
Yin,
M.,
Tsuchiya,
K.,
Mason,
R.,
Peters,
J.,
Gonzalez,
F.,
Segal,
B.,
Holland,
S.,
and
Thurman,
R.,
(
2001),
Phthalates
rapidly
increase
production
of
reactive
oxygen
species
in
vivo:
role
of
Kupffer
cells,
Molecular
Pharmacology,
59(
4),
pp.
744­
750.

Ward,
J.,
Peters,
J.,
Perella,
C.,
and
Gonzalez,
F.,
(
1998),
Receptor
and
nonreceptormediated
organ­
specific
toxicity
of
di(
2­
ethylhexyl)
phthalate
(
DEHP)
in
peroxisome
proliferator­
activated
receptor
alpha­
null
mice,
Toxicologic
Pathology,
26(
2),
pp.
240­
246.

Waxman,
D.,
(
1996),
Role
of
metabolism
in
the
activation
of
dehydroepiandrosterone
as
a
peroxisome
proliferator,
Journal
of
Endocrinology,
150
Suppl:
S129­
147.
