PRELIMINARY
RISK
ASSESSMENT
OF
THE
DEVELOPMENTAL
TOXICITY
ASSOCIATED
WITH
EXPOSURE
TO
PERFLUOROOCTANOIC
ACID
AND
ITS
SALTS
U.
S.
Environmental
Protection
Agency
Office
of
Pollution
Prevention
and
Toxics
Risk
Assessment
Division
April
10,
2003
RECEIVED
OPPT
NCIC
2003
APR
14
PM
4:
32
OPPT­
2003­
0012­
0002
Table
of
Contents
Executive
Summary
1
1.0
Scope
of
the
Assessment
6
2.0
Chemical
Identity
7
2.1
Physicochemical
Properties
7
3.0
Hazard
Characterization
9
3.1
Metabolism
and
Pharmacokinetics
in
Humans
9
3.1.1
Half­
life
Studies
in
Humans
9
3.2
Metabolism
and
Pharmacokinetic
Studies
in
Animals
10
3.2.1
Absorption
Studies
in
Animals
10
3.2.2
Distribution
Studies
in
Animals
11
3.2.3
Metabolism
Studies
in
Animals
14
3.2.4
Elimination
Studies
in
Animals
15
3.3
Epidemiology
Studies
17
3.3.1
Mortality
Studies
in
Humans
18
3.3.2
Hormone
Study
in
Humans
20
3.3.3
Study
on
Episodes
of
Care
(
Morbidity)
21
3.3.4
Medical
Surveillance
Studies
from
the
Antwerp
and
Decatur
Plants
23
3.3.5
Medical
Surveillance
Studies
from
the
Cottage
Grove
Plant
25
3.4
Prenatal
Developmental
Toxicity
Studies
in
Animals
27
3.5
Reproductive
Toxicity
Studies
in
Animals
31
4.0
Exposure
Characterization
40
4.1
Occupational
Exposures
40
4.2
Non­
occupational
Exposures
42
4.3
General
Population
Exposures
42
5.0
Preliminary
Risk
Characterization
45
5.1
Selection
of
Developmental
Endpoints
46
5.2
Serum
Levels
as
a
Measure
of
Internal
Dose
for
Humans
47
5.3
Serum
Levels
as
a
Measure
of
Internal
Dose
for
Animal
Studies
48
5.4
Calculation
of
MOEs
49
5.5
Uncertainties
in
the
Preliminary
Risk
Characterization
51
6.0
Overall
Conclusions
55
7.0
References
56
1
Executive
Summary
As
part
of
the
effort
by
the
Office
of
Pollution
Prevention
and
Toxics
(
OPPT)
to
understand
health
and
environmental
issues
presented
by
fluorochemicals
in
the
wake
of
unexpected
toxicological
and
bioaccumulation
discoveries
with
respect
to
perfluorooctane
sulfonates
(
PFOS),
OPPT
has
been
investigating
perfluorooctanoic
acid
(
PFOA)
and
its
salts.
PFOA
and
its
salts
are
fully
fluorinated
organic
compounds
that
can
be
produced
synthetically
or
through
the
degradation
or
metabolism
of
other
fluorochemical
products.
PFOA
is
primarily
used
as
a
reactive
intermediate,
while
its
salts
are
used
as
processing
aids
in
the
production
of
fluoropolymers
and
fluoroelastomers
and
in
other
surfactant
uses.
PFOA
and
its
salts
are
persistent
in
the
environment.

Human
Health
Effects
and
Biomonitoring
Little
information
is
available
concerning
the
pharmacokinetics
of
PFOA
and
its
salts
in
humans.
An
ongoing
5­
year,
half­
life
study
in
7
male
and
2
female
retired
workers
has
suggested
a
mean
serum
PFOA
half­
life
of
4.37
years
(
range,
1.50
 
13.49
years).
Animal
studies
have
shown
that
the
ammonium
salt
of
PFOA
(
APFO)
is
well
absorbed
following
oral
and
inhalation
exposure,
and
to
a
lesser
extent
following
dermal
exposure.
PFOA
distributes
primarily
to
the
liver
and
plasma.
It
does
not
partition
to
the
lipid
fraction
or
adipose
tissue.
PFOA
is
not
metabolized
and
there
is
evidence
of
enterohepatic
circulation
of
the
compound.
The
urine
is
the
major
route
of
excretion
of
PFOA
in
the
female
rat,
while
the
urine
and
feces
are
both
main
routes
of
excretion
in
male
rats.

There
are
gender
differences
in
the
elimination
of
PFOA
in
rats.
In
female
rats,
estimates
of
the
serum
half
life
range
from
1.9
to
24
hours,
while
in
male
rats
estimates
of
the
serum
half
life
range
from
4.4
to
9
days.
In
female
rats
elimination
of
PFOA
appears
to
be
biphasic;
a
fast
phase
occurs
with
a
half
life
of
approximately
2­
4
hours
while
a
slow
phase
occurs
with
a
half
life
of
approximately
24
hours.
The
rapid
excretion
of
PFOA
by
female
rats
is
due
to
active
renal
tubular
secretion
(
organic
acid
transport
system);
this
renal
tubular
secretion
is
believed
to
be
hormonally
controlled.
Hormonal
changes
during
pregnancy
do
not
appear
to
change
the
rate
of
elimination
in
rats.
This
gender
difference
has
not
been
observed
in
humans
based
on
the
limited
data
available
in
the
half
life
study
in
retired
workers.

While
the
environmental
concentrations
and
pathways
of
human
exposure
to
PFOA
and
its
salts
are
unknown,
there
are
limited
data
on
PFOA
serum
levels
in
workers
and
the
general
population.
Occupational
data
from
certain
plants
in
the
U.
S.
and
Belgium
that
manufacture
or
use
PFOA
indicate
that
mean
serum
levels
in
U.
S.
workers
in
2000
range
from
0.84
to
6.4
ppm.
At
another
U.
S.
plant
where
the
most
recently
reported
data
are
from
1997,
the
highest
level
reported
in
a
worker
was
81.3
ppm.
In
non­
occupational
populations,
serum
PFOA
levels
were
much
lower.
In
both
pooled
blood
bank
samples
and
in
individual
samples,
mean
serum
PFOA
levels
ranged
from
3
to
17
ppb.
The
highest
serum
PFOA
levels
of
the
general
public
were
reported
in
a
sample
of
children
from
different
geographic
regions
in
the
U.
S.
(
mean,
5.6
ppb;
range,
1.9
 
56.1
ppb).
2
Epidemiological
studies
on
the
effects
of
PFOA
in
humans
have
been
conducted
on
workers.
However,
these
studies
have
not
examined
developmental
outcomes.
The
majority
of
production
workers
at
facilities
that
produce
or
use
PFOA
are
male.
Two
mortality
studies,
a
morbidity
study,
and
studies
examining
effects
on
the
liver,
pancreas,
endocrine
system,
and
lipid
metabolism,
have
been
conducted
to
date.
In
addition,
a
longitudinal
study
of
the
worker
surveillance
data
recently
became
available.

A
retrospective
cohort
mortality
study
demonstrated
a
statistically
significant
association
between
prostate
cancer
mortality
and
employment
duration
in
the
chemical
facility
of
a
plant
that
manufactures
PFOA.
However,
in
a
recent
update
to
this
study
in
which
more
specific
exposure
measures
were
used,
a
significant
association
for
prostate
cancer
was
not
observed.
In
an
"
episodes
of
care"
study,
workers
with
the
highest
PFOA
exposures
for
the
longest
durations
sought
care
more
often
for
prostate
cancer
treatment
than
workers
with
lower
exposures.
However,
this
finding
was
not
statistically
significant
and
the
95%
confidence
interval
was
very
wide.

Another
study
reported
an
increase
in
estradiol
levels
in
workers
with
the
highest
PFOA
serum
levels;
however,
none
of
the
other
hormone
levels
analyzed
indicated
any
adverse
effects.
Some
of
the
same
employees
who
participated
in
the
hormone
study
also
were
included
in
a
study
of
cholecystokinin
(
CCK)
levels
in
employees.
No
positive
association
was
noted
between
CCK
values
and
PFOA.
The
other
available
study
examined
cholesterol
and
other
serum
components
in
workers.
There
did
not
appear
to
be
any
significant
differences
among
workers
of
different
exposure
levels.
At
plants
where
the
serum
PFOA
levels
were
lower,
cross­
sectional
and
longitudinal
studies
found
positive
significant
associations
between
PFOA
and
cholesterol
and
triglyceride
levels.
In
addition,
a
positive,
significant
association
was
reported
between
PFOA
and
T3
hormone
and
a
negative
association
with
HDL
in
the
cross­
sectional
study.
There
are
many
limitations
to
the
studies
conducted
to
date,
and
therefore,
all
of
these
results
must
be
interpreted
carefully.

Prenatal
developmental
toxicity
studies
in
rats
resulted
in
death
and
reduced
body
weight
in
dams
exposed
to
oral
doses
of
100
mg/
kg/
day
or
by
inhalation
to
25
mg/
m3
of
the
ammonium
salt
of
PFOA
(
APFO).
There
was
no
evidence
of
developmental
toxicity
after
oral
exposure
to
doses
as
high
as
150
mg/
kg/
day,
while
inhalation
exposure
to
25
mg/
m3
resulted
in
reduced
fetal
body
weights.
In
a
rabbit
oral
developmental
toxicity
study
there
was
a
significant
increase
in
skeletal
variations
after
exposure
to
50
mg/
kg/
day
APFO.
There
was
no
evidence
of
maternal
toxicity
at
50
mg/
kg/
day,
the
highest
dose
tested.

In
a
two­
generation
reproductive
toxicity
study
in
rats
exposed
to
0,
1,
3,
10,
and
30
mg/
kg/
day
APFO,
significant
increases
in
absolute
and
relative
liver
and
kidney
weights
were
observed
in
F0
males
at
1
mg/
kg/
day,
while
significant
reductions
in
absolute
and
relative
kidney
weights
were
observed
in
F0
females
at
30
mg/
kg/
day.
Reproductive
indices
were
not
affected
in
the
F0
animals.
Serum
levels
of
the
10
and
30
mg/
kg/
day
groups
were
measured
for
F0
males
after
mating
and
F0
females
at
weaning
of
the
F1
pups.
In
F0
males,
the
serum
levels
were
(
average
+
SD)
51.1+
9.30
and
45.3+
12.6
ug/
l,
respectively
for
the
10
and
30
mg/
kg/
day
groups,
and
in
F0
females,
the
serum
levels
were
0.37+
0.0805
and
1.02+
0.425
ug/
l,
respectively
for
the
10
and
30
3
mg/
kg/
day
groups.
In
F1
animals,
there
was
a
significant
reduction
in
mean
body
weight
(
sexes
combined)
during
lactation
in
the
30
mg/
kg/
day
group.
In
F1
females,
there
was
a
significant
increase
in
post
weaning
mortality,
a
significant
decrease
in
mean
body
weight,
and
a
significant
delay
in
sexual
maturation
at
30
mg/
kg/
day.
In
F1
males,
significant
decreases
in
body
weights
and
body
weight
gains,
and
significant
changes
in
absolute
liver
and
spleen
weights
and
in
the
ratios
of
liver,
kidney,
and
spleen
weights­
to­
brain
weights
were
observed
in
all
treated
groups.
The
increase
in
post
weaning
mortality
and
the
delay
in
sexual
maturation
were
also
noted
in
F1
males
at
30
mg/
kg/
day.
Reproductive
indices
were
not
affected
in
the
F1
animals.
The
LOAEL
for
the
F1
females
was
30
mg/
kg/
day,
and
the
NOAEL
was
10
mg/
kg/
day;
the
LOAEL
for
F1
males
was
1
mg/
kg/
day
and
a
NOAEL
was
not
determined.
It
should
be
noted
that
these
effect
levels
reflect
effects
seen
throughout
the
study
(
i.
e.
developmental
and
adult
exposures),
and
should
not
be
confused
with
the
effect
levels
that
are
used
in
the
preliminary
risk
assessment
for
strictly
developmental
exposures
and
effects.
The
difference
in
sensitivity
is
presumed
to
be
related
to
the
gender
difference
in
elimination
of
APFO.
No
treatment­
related
effects
were
observed
in
the
F2
generation.
However,
the
F2
pups
were
sacrificed
at
weaning,
and
thus
it
was
not
possible
to
ascertain
if
the
post­
weaning
effects
that
were
noted
in
the
F1
generation
occurred
in
the
F2
animals.

Preliminary
Risk
Assessment
This
preliminary
risk
assessment
focused
on
the
potential
risks
for
developmental
toxicity
associated
with
exposure
to
PFOA
and
its
salts.
A
margin
of
exposure
(
MOE)
approach
was
used;
the
MOE
is
calculated
as
the
ratio
of
the
NOAEL,
LOAEL,
or
BMDL
for
a
specific
endpoint
to
the
estimated
human
exposure
level.
The
MOE
does
not
provide
an
estimate
of
population
risk,
but
simply
describes
the
relative
"
distance"
between
the
exposure
level
and
the
NOAEL,
LOAEL,
or
BMDL.
For
many
risk
assessments,
the
MOE
is
calculated
as
the
ratio
of
the
administered
dose
from
the
animal
toxicology
study
to
the
estimated
human
exposure
level.
The
human
exposure
is
estimated
from
a
variety
of
potential
exposure
scenarios,
each
of
which
requires
a
variety
of
assumptions.
A
more
accurate
estimate
of
the
MOE
can
be
derived
if
measures
of
internal
dose
are
available
for
humans
and
the
animal
model.
In
this
preliminary
risk
assessment,
serum
levels
of
PFOA,
which
are
a
measure
of
internal
dose,
were
available
for
the
rat
two­
generation
reproductive
toxicology
study
and
from
human
biomonitoring
studies.
Thus,
internal
dose
was
used
for
the
calculation
of
MOEs
in
this
assessment.

For
this
preliminary
risk
assessment,
the
endpoints
from
the
two­
generation
reproductive
toxicity
study
that
were
considered
relevant
for
assessing
developmental
toxicity
included
the
significant
reduction
in
F1
mean
body
weight
during
lactation
(
sexes
combined).
In
addition,
for
F1
females,
postweaning
mortality
and
delayed
sexual
maturation
were
noted
at
30
mg/
kg/
day
APFO;
the
NOAEL
for
developmental
effects
for
F1
females
was
10
mg/
kg/
day.
Postweaning
mortality,
delayed
sexual
maturation
and
a
significant
reduction
in
postweaning
body
weights
were
noted
in
F1
males
at
30
mg/
kg/
day,
and
a
significant
reduction
in
postweaning
body
weight
was
noted
at
10
mg/
kg/
day.
For
F1
males,
the
LOAEL
for
developmental
effects
was
10
mg/
kg/
day
and
the
NOAEL
was
3
mg/
kg/
day.
Thus,
the
LOAEL
for
developmental
effects
from
the
study
was
10
mg/
kg/
day
and
the
NOAEL
was
3
mg/
kg/
day.
4
In
the
rat
two­
generation
reproductive
toxicity
study,
serum
levels
of
PFOA
were
only
measured
in
the
F0
animals.
In
order
to
use
these
serum
levels
as
surrogates
for
the
serum
levels
in
the
F1
animals,
several
areas
of
uncertainty
had
to
be
considered.
It
is
not
known
whether
the
effects
on
postweaning
mortality,
body
weight,
or
age
of
sexual
maturation
were
due
to
prenatal
exposures,
lactational
exposures,
postweaning
exposures,
or
a
combination
of
one
or
more
of
these
exposure
periods.
In
most
risk
assessments
of
developmental
toxicity,
no
attempt
is
made
to
determine
which
of
these
exposure
periods
is
important.
A
major
strength
of
this
preliminary
assessment
is
that
each
of
these
exposure
periods
was
considered
in
order
to
determine
the
appropriateness
and
uncertainties
associated
with
the
use
of
the
serum
levels
from
the
F0
animals.

It
was
reasoned
that
if
prenatal
and/
or
lactational
exposures
were
important
then
the
serum
levels
in
the
F0
females
would
be
the
most
appropriate
estimate
for
the
F1
animals.
If
postweaning
exposures
were
important
then
the
serum
levels
for
the
F0
males
would
be
the
most
appropriate
estimate
for
the
F1
males,
and
similarly
the
serum
levels
in
the
F0
females
would
be
the
most
appropriate
estimate
for
the
F1
females.
It
was
not
possible
to
make
a
"
direct"
estimate
of
F1
serum
levels
from
the
serum
levels
in
the
F0
females
for
several
reasons.
First,
there
is
a
gender
difference
in
the
elimination
of
PFOA
in
rats.
In
female
rats,
estimates
of
the
serum
half
life
range
from
1.9
to
24
hours,
while
in
male
rats
estimates
of
the
serum
half
life
range
from
4.4
to
9
days.
In
female
rats
elimination
of
PFOA
appears
to
be
biphasic;
a
fast
phase
occurs
with
a
half
life
of
approximately
2­
4
hours
while
a
slow
phase
occurs
with
a
half
life
of
approximately
24
hours.
In
the
two
generation
reproductive
toxicity
study,
the
animals
were
dosed
by
gavage
once
daily.
The
serum
levels
were
measured
24
hours
after
dosing.
Thus,
the
values
obtained
for
the
F0
females
represent
the
low
end
of
exposure.
With
no
knowledge
of
the
peak
exposures,
it
was
reasoned
that
it
was
unlikely
that
the
peak
exposure
would
be
higher
than
the
serum
level
in
the
F0
males
in
the
same
dose
group
since
they
would
tend
to
accumulate
PFOA
with
a
daily
dosing
regime.
Therefore,
the
strategy
that
was
employed
in
this
assessment
was
to
use
the
MOEs
that
were
calculated
from
the
serum
levels
in
the
F0
males
and
females
as
a
range
or
as
a
means
to
bracket
the
low
and
high
ends
of
exposure.

For
calculation
of
the
MOEs,
the
human
populations
that
were
considered
included
women
of
child
bearing
age
and
children.
Estimates
of
general
human
population
exposure
were
available
from
recent
analyses
of
individual
serum
samples
from
a
group
of
children
(
2­
12
years)
and
adults
(
20­
69
years).
For
the
populations
of
interest,
calculations
using
human
adult
serum
levels
and
children
serum
levels
in
combination
with
rat
serum
values
from
the
parental
(
F0)
females
and
males
produced
a
range
of
overlapping
MOE
values
that
extends
from
less
than
100
to
greater
than
9000.
There
are
a
number
of
important
uncertainties
discussed
in
this
document
that
provide
a
context
for
considering
these
MOEs
as
a
range
of
potential
values.

It
is
important
to
note
that
MOEs
that
were
calculated
from
the
serum
levels
in
the
F0
female
and
male
rats
provide
a
means
to
bracket
the
low
and
high
ends
of
experimental
animal
exposures.
This
is
an
unusual
situation
in
that
MOE
estimates,
which
typically
represent
point
estimates,
are
described
here
as
a
range
of
potential
values
due
to
uncertainties
in
the
rat
serum
data.
This
situation
arises
from
the
fact
that
the
available
data
do
not
allow
selection
of
a
particular
departure
point
for
the
MOE
calculations.
It
is
likely
that
MOEs
calculated
using
the
5
F0
female
rat
serum
level
are
lower
than
what
would
be
anticipated
in
the
human
population,
and
it
is
likely
that
MOEs
calculated
using
the
F0
male
rat
serum
level
are
higher
than
what
would
be
anticipated
in
the
human
population.
As
uncertainty
around
the
rat
serum
values
decreases
the
end
brackets
are
likely
to
shift
towards
the
middle
of
the
current
range.
Therefore,
MOE
values
presented
in
this
document
should
not
be
interpreted
as
representing
the
range
of
possible
MOEs
in
the
US
population.
It
is
likely
that
when
more
extensive
rat
kinetic
data
are
available,
the
resultant,
refined
estimated
range
of
MOEs
will
constitute
a
narrower
subset
of
the
range
presented
here.
Interpretation
of
the
significance
of
the
MOEs
for
ascertaining
potential
levels
of
concern
will
necessitate
a
better
understanding
of
the
appropriate
dose
metric
in
rats,
and
the
relationship
of
the
dose
metric
to
the
human
serum
levels.
6
1.0
Scope
of
the
Assessment
As
part
of
the
effort
by
the
Office
of
Pollution
Prevention
and
Toxics
(
OPPT)
to
understand
health
and
environmental
issues
presented
by
fluorochemicals
in
the
wake
of
unexpected
toxicological
and
bioaccumulation
discoveries
with
respect
to
perfluorooctane
sulfonates
(
PFOS),
OPPT
has
been
investigating
perfluorooctanoic
acid
and
its
salts
(
PFOA).
PFOA
and
its
salts
are
fully
fluorinated
organic
compounds
that
can
be
produced
synthetically
or
through
the
degradation
or
metabolism
of
other
fluorochemical
products.
PFOA
is
primarily
used
as
a
reactive
intermediate,
while
its
salts
are
used
as
processing
aids
in
the
production
of
fluoropolymers
and
fluoroelastomers
and
in
other
surfactant
uses.

OPPT
released
a
preliminary
Draft
Hazard
Assessment
of
Perfluorooctanoic
Acid
and
Its
Salts,
dated
February
20,
2002,
on
March
28,
2002,
and
issued
a
minor
correction
to
that
document
on
April
15,
2002.
That
draft
assessment
indicated
that
PFOA
and
its
salts
are
persistent
in
the
environment
and
in
humans
with
a
half
life
of
years.
The
assessment
noted
the
potential
systemic
toxicity
and
carcinogenicity
associated
with
the
ammonium
salt
of
PFOA
(
APFO),
which
has
been
the
focus
of
the
animal
toxicology
studies,
and
observed
that
blood
monitoring
data
suggested
widespread
exposure
to
the
general
population,
albeit
at
low
levels.
The
Agency
has
since
received
considerable
additional
animal
toxicology
data
on
APFO
that
suggest
a
potential
for
developmental/
reproductive
toxicity
and
immunotoxicity,
and
additional
human
biomonitoring
data
that
indicate
low
level
exposures
to
the
general
population
that
cannot
be
explained
at
this
time.

On
September
27,
2002,
the
Director
of
OPPT
issued
a
memorandum
announcing
that
OPPT
would
initiate
a
priority
review
to
determine
whether
PFOA
and
its
salts
meets
the
criteria
for
action
under
section
4(
f)
of
the
Toxic
Substances
Control
Act.
As
part
of
the
priority
review,
the
hazard
assessment
was
revised
and
released
on
September
30,
2002.
Another
revision
was
then
released
November
4,
2002.
In
addition,
OPPT
conducted
a
preliminary
risk
assessment
of
PFOA
and
its
salts.
OPPT
recognizes
that
there
is
a
wide
range
of
toxicological
endpoints
associated
with
exposure
to
APFO,
but
at
this
time
only
the
endpoints
that
are
included
in
section
4(
f)
were
considered;
these
include
cancer,
mutations,
and
birth
defects.
OPPT
did
not
include
gene
mutations
in
the
preliminary
risk
assessment
since
APFO
is
not
known
to
be
mutagenic.
In
addition,
APFO
is
a
peroxisome
proliferation
activating
receptor­"­
agonist
and
through
this
mode
of
action
could
lead
to
the
formation
of
liver
tumors
in
rodents.
The
relevance
of
this
mode
of
action
for
humans
is
currently
under
scientific
debate,
and
the
Agency
is
engaged
in
activities
to
resolve
this
issue.
Therefore,
at
this
time,
OPPT
has
narrowly
restricted
the
analysis
to
examine
only
the
potential
risks
of
developmental
toxicity.

The
relevant
information
pertaining
to
chemical
properties,
pharmacokinetics
and
metabolism,
epidemiology,
prenatal
developmental
toxicity,
reproductive
toxicity,
and
human
exposure
have
been
included
in
this
preliminary
risk
assessment.
Other
information
pertaining
to
systemic
toxicity,
carcinogenicity,
ecotoxicity,
production
and
uses,
fate
and
transport,
and
environmental
monitoring
can
be
found
in
the
Draft
Hazard
Assessment
of
Perfluorooctanoic
Acid
and
Its
Salts,
dated
November
4,
2002.
7
2.0
Chemical
Identity
Chemical
Name:
Perfluorooctanoic
Acid
Molecular
formula:
C8
H
F15
O2
Structural
formula:
F­
CF2­
CF2­
CF2­
CF2­
CF2­
CF2­
CF2­
C(=
O)­
X,

The
free
acid
and
some
common
derivatives
have
the
following
CAS
numbers:
The
perfluorooctanoate
anion
does
not
have
a
specific
CAS
number.

Free
Acid
(
X
=
OM+;
M
=
H)
[
335­
67­
1]
Ammonium
Salt
(
X
=
OM+;
M
=
NH4)
[
3825­
26­
1]
Sodium
Salt
(
X
=
OM+;
M
=
Na)
[
335­
95­
5]
Potassium
Salt
(
X
=
OM+;
M
=
K)
[
2395­
00­
8]
Silver
Salt
(
X
=
OM+;
M
=
Ag)
[
335­
93­
3]
Acid
Fluoride
(
X
=
F)
[
335­
66­
0]
Methyl
Ester
(
X
=
CH3)
[
376­
27­
2]
Ethyl
Ester
(
X
=
CH2­
CH3)
[
3108­
24­
5]

Synonyms:
1­
Octanoic
acid,
2,2,3,3,4,4,5,5,6,6,7,7,8,8,8­
pentadecafluoro­
PFOA
2.1
Physicochemical
Properties
PFOA
is
a
completely
fluorinated
organic
acid.
The
typical
structure
has
a
linear
chain
of
eight
carbon
atoms.
The
physical
chemical
properties
noted
below
are
for
the
free
acid,
unless
otherwise
stated.
The
data
for
the
free
acid,
pentadecafluorooctanoic
acid
[
335­
67­
1],
is
the
most
complete.
The
reported
vapor
pressure
of
10
mm
Hg
appears
high
for
a
low
melting
solid
when
compared
to
other
low
melting
solids
(
chloroacetic
acid:
solid;
MP
=
61
to
63
/
C;
BP
=
189
/
C;
VP
=
0.1
kPa
(
0.75
mm
Hg)
@
20
/
C;
NIOSH),
but
is
consistent
with
other
perfluorinated
compounds
with
similar
boiling
points
(
perfluorobutanoic
acid
BP
=
120
/
C,
VP
10
mm
Hg
@
20
/
C;
Beilstein,
1975).
Another
explanation
may
be
that
the
10
mm
vapor
pressure
was
measured
at
an
elevated
temperature
(
but
the
temperature
inadvertently
omitted),
as
perfluorooctanoic
acid
is
typically
handled
as
a
liquid
at
65
/
C
(
3M
data
sheet
for
FC­
26).
The
free
acid
is
expected
to
completely
dissociate
in
water,
leaving
the
anionic
carboxylate
in
the
water
and
the
perfluoroalkyl
chain
on
the
surface.
In
aqueous
solutions,
individual
molecules
of
PFOA
anion
loosely
associate
on
the
water
surface
and
partition
between
the
air
/
water
interface.
Several
reports
note
that
PFOA
salts
self­
associate
at
the
surface,
but
with
agitation
they
disperse
and
micelles
form
at
higher
concentrations.
(
Simister
et
al.,
1992;
Calfours,
1985;
Edwards,
1997).
Water
solubility
has
been
reported
for
PFOA,
but
it
is
unclear
whether
these
values
are
for
a
microdispersion
of
micelles,
rather
than
true
solubility.
Due
to
these
same
surface­
active
properties
of
PFOA,
and
the
test
protocol
for
the
OECD
shake
flask
method,
PFOA
is
anticipated
to
form
multiple
layers
in
octanol/
water,
much
like
those
observed
for
PFOS.
Therefore,
an
n­
octanol/
water
partition
coefficient
cannot
be
determined.
8
The
available
physicochemical
properties
for
the
PFOA
free
acid
are:

Molecular
weight:
414
(
Beilstein,
1975)
Melting
point:
45
 
50
/
C
(
Beilstein,
1975)
Boiling
point:
189
 
192
/
C
/
736
mm
Hg
(
Beilstein,
1975)
Vapor
pressure:
10
mm
Hg
@
25
/
C
(
approx.)
(
Exfluor
MSDS)
Water
solubility:
3.4
g/
L
(
telomeric
[
MP
=
34
/
C
ref.
0.01
­
0.02
mol/
L
~
4
­
8
g/
L)
(
MSDS
from
Merck,
Fischer,
and
Chinameilan
Internet
sites)
pKa:
2.5
(
USEPA
AR226­
0473)
pH
(
1g/
L):
2.6
(
MSDS
Merck)

The
PFOA
derivative
of
greatest
concern
and
most
wide
spread
use
is
the
ammonium
salt
(
APFO;
CAS
No.
3825­
26­
1).
The
water
solubility
of
APFO
has
been
inconsistently
reported.
One
3M
study
reported
the
water
solubility
of
APFO
to
be
>
10%.
It
was
noted
in
an
earlier
study
that
at
concentrations
of
20
g/
L,
the
solution
"
gelled"
(
3M,
1979).
These
numbers
seem
surprising
low
for
a
salt
in
light
of
Apollo
Scientific
selling
a
31%
aqueous
solution
of
APFO.
One
author
reported
the
APFO
partition
coefficient
log
Pow
=
5.
Another
author
reported
an
estimated
APFO
log
Pow
=
­
0.9.
This
value
might
not
be
accurate
due
to
the
estimation
method
used
(
Hansch
and
Leo
1979).
Again,
the
anticipated
formation
of
an
emulsified
layer
between
the
octanol
and
water
surface
interface
would
make
determination
of
log
Kow
impossible.

Determination
of
the
vapor
pressure
of
APFO
is
complicated.
A
vapor
pressure
of
7
x
10­
5
mm
Hg
at
20
/
C
has
been
reported
for
APFO;
however,
this
appears
to
be
too
low
for
a
material
that
sublimates
as
the
ammonium
salt
(
3M
Environmental
Laboratory,
1993).
The
ammonium
salt
begins
to
sublimate
at
130
/
C.
As
the
temperature
increases
from
when
APFO
begins
to
sublimate,
20%
of
the
sample
weight
is
lost
by
169
/
C.
Other
salts
(
Cs,
K,
Ag,
Pb,
Li)
do
not
demonstrate
similar
weight
loss
until
237
/
C
or
higher.
(
Lines,
1984).
Decomposition
of
different
salts
produces
perfluoroheptene
(
loss
of
metal
fluoride
and
carbon
dioxide).
This
occurs
at
320
/
C
for
the
sodium
salt
and
at
250­
290
/
C
for
the
silver
salt
(
Beilstein
1975).

The
physicochemical
properties
of
PFOA
and
its
common
derivatives
are
summarized
in
Table
1.
Table
1.
Reported
Physicochemical
Properties
Compound
CAS
REG
#
MP
BP
VP
Sol.­
H2O
Log
P*
R­
C(=
O)
Cl
335­
64­
8
131
/
C
R­
CO2H
335­
67­
1
55
/
C
189
/
C
10
mm
Hg
3.4
g/
L
R­
CO2­
NH4+
3825­
26­
1
130
/
C
(
sub)
sublimes
1
x
10E­
5
mm
Hg
20
g/
L
gels
R­
C(=
O)
OMe
376­
27­
2
159
/
C
pH
(
1
g
free
acid
/
L
Water)
=
1.5
 
2.5
Free
acid
pKa
is
approximately
0.6
Sodium
or
Silver
salts
of
PFOA
decompose
above
250
/
C
to
generate
perfluoroolefins.
 
Surfactants
traditionally
emulsify
octanol
and
water
9
3.0
Hazard
Characterization
3.1
Metabolism
and
Pharmacokinetics
in
Humans
3.1.1
Half­
life
Studies
in
Humans
There
are
very
limited
data
on
the
half­
life
of
PFOA.
With
the
exception
of
a
1980
study
in
which
total
organic
fluorine
in
blood
serum
was
measured
in
one
worker,
no
other
data
were
available
until
June
2000
(
Ubel
et
al.,
1980).
A
half­
life
study
on
27
retirees
from
the
Decatur
and
Cottage
Grove
3M
plants
was
undertaken,
in
which
serum
samples
were
drawn
every
6
months
over
a
5­
year
period.
Two
interim
reports
describing
the
results
thus
far
have
been
submitted
(
Burris
et
al.,
2000;
Burris
et
al.,
2002).
The
first
interim
report
suggested
a
median
serum
half­
life
of
PFOA
of
344
days,
with
a
range
of
109
to
1308
days.
The
two
highest
half­
life
calculations
were
for
the
2
female
retirees
who
participated
in
this
study
(
654
and
1308
days).

There
were
several
limitations
to
this
first
analysis
including:
1)
the
limited
data
available
and
the
range
of
serum
PFOA
levels
measured;
2)
serum
was
analyzed
after
each
collection
period
with
only
one
measurement
per
time
period
on
different
days
using
slightly
different
analytical
techniques;
and
3)
the
reference
material
purity
was
not
determined
until
after
the
first
3
samples
had
been
analyzed.

An
effort
was
made
to
minimize
experimental
error,
including
systematic
and
random
error
in
the
analytical
method.
Serum
samples
were
collected
from
9
of
the
original
27
subjects
over
4
time
periods
spanning
180
days,
measured
in
triplicate
with
all
time
points
from
each
subject
analyzed
in
the
same
analytical
run.
This
would
allow
for
statistical
evaluation
of
the
precision
of
the
measurement
and
assure
that
all
systematic
error
inherent
in
the
assay
equally
affected
each
sample
used
for
half­
life
determination.
Single
serum
measurements
were
made
on
samples
of
the
remaining
18
retirees,
but
were
not
included
because
triplicate
analyses
of
all
time
points
were
not
conducted.

Of
the
9
retirees
included
in
this
analysis,
there
were
7
males
and
2
females,
all
from
the
Decatur
plant.
The
average
age
of
the
retirees
was
61
years,
the
mean
number
of
years
worked
at
Decatur
was
27.7
years,
and
the
average
number
of
months
retired
from
the
plant
at
study
initiation
was
18.9.
The
average
body
mass
index
(
BMI)
of
this
group
was
27.9
(
range
22.5­
33,
SD
=
3.6).
The
mean
PFOA
value
at
study
initiation
was
0.72
ppm
(
range
0.06
 
1.84
ppm,
SD
=
0.64).

The
mean
serum
half­
life
for
PFOA
was
4.37
years
(
range
1.50
 
13.49
years,
SD
=
3.53).
Only
1
employee
had
a
half­
life
value
that
exceeded
4.3
years.
The
2
females
had
values
of
3.1
and
3.9
years.
Age,
BMI,
number
of
years
worked
or
years
since
retirement
were
not
significant
predictors
of
serum
half­
lives
in
multivariable
regression
analyses.

This
analysis
has
attempted
to
reduce
experimental
error
in
the
determination
of
a
half­
life
for
PFOA.
However,
two
issues
should
be
noted.
First,
the
effect
of
continual
non­
occupational,
low­
level
exposure
on
the
half­
life
is
unknown.
Second,
systematic
error
of
the
analytical
method
could
be
as
high
as
+/­
20%
and
still
satisfy
the
data
quality
criteria.
10
3.2
Metabolism
and
Pharmacokinetic
Studies
in
Animals
The
metabolism
and
pharmacokinetics
of
APFO
have
been
fairly
extensively
studied
in
animals.
Animal
studies
have
shown
that
APFO
is
well
absorbed
following
oral
and
inhalation
exposure,
and
to
a
lesser
extent
following
dermal
exposure.
The
compound
distributes
primarily
to
the
liver
and
plasma.
It
does
not
partition
to
the
lipid
fraction
or
adipose
tissue.
PFOA
is
not
metabolized
and
there
is
evidence
of
enterohepatic
circulation
of
the
compound.
The
urine
is
the
major
route
of
excretion
of
PFOA
in
the
female
rat,
while
the
urine
and
feces
are
both
major
routes
of
excretion
in
male
rats.
There
are
major
gender
differences
in
the
elimination
of
PFOA
in
rats.
In
female
rats,
estimates
of
the
serum
half
life
range
from
1.9
to
24
hours,
while
in
male
rats
estimates
of
the
serum
half
life
range
from
4.4
to
9
days.
In
female
rats
elimination
of
PFOA
appears
to
be
biphasic;
a
fast
phase
occurs
with
a
half
life
of
approximately
2­
4
hours
while
a
slow
phase
occurs
with
a
half
life
of
approximately
24
hours.
The
rapid
excretion
of
PFOA
by
female
rats
is
due
to
active
renal
tubular
secretion
(
organic
acid
transport
system);
this
renal
tubular
secretion
is
believed
to
be
hormonally
controlled.
Hormonal
changes
during
pregnancy
do
not
appear
to
change
the
rate
of
elimination
in
rats.
This
gender
difference
has
not
been
observed
in
primates
and
humans.
The
relevant
studies
are
summarized
below.

3.2.1
Absorption
Studies
in
Animals
PFOA
and
its
salts
are
well
absorbed
following
oral
and
inhalation
exposure,
and
to
a
lesser
extent
following
dermal
exposure.
The
serum
levels
of
PFOA
were
measured
in
a
twogeneration
reproductive
toxicity
study
in
rats.
These
results
are
presented
with
the
summary
of
that
study
in
section
3.5.
Other
studies
are
summarized
here.

Gibson
and
Johnson
(
1979)
administered
a
single
oral
dose
of
11.0
mg/
kg
14C­
PFOA
to
groups
of
3
male
CD
rats.
Twenty­
four
hr
after
administration,
at
least
93%
of
the
total
carbon­
14
was
absorbed;
the
elimination
half­
life
of
carbon­
14
from
the
plasma
was
4.8
days.

Ophaug
and
Singer
(
1980)
administered
2
ml
of
an
aqueous
solution
of
2
mg
PFOA
to
female
Holtzman
rats.
Seven
hundred
forty­
nine
ug
or
37%
of
the
fluorine
in
the
administered
dose
was
recovered
in
the
urine
within
4.5
hr
after
administration
of
PFOA.
The
quantity
of
nonionic
fluorine
recovered
in
the
urine
increased
to
61%,
76%
and
89%
at
8,
24
and
96
hr,
respectively,
after
administration.
Ionic
fluoride
and
total
fluorine
was
also
measured.
Four
and
half
hours
after
the
administration
of
PFOA,
serum
from
treated
rats
had
a
nonionic
fluorine
level
of
13.6
ppm,
virtually
all
of
which
was
bound
to
components
in
the
serum
and
was
not
ultrafilterable.
The
nonionic
fluorine
level
in
the
serum
decreased
to
11.2
ppm
at
8
hr,
0.35
ppm
at
24
hr,
and
0.08
ppm
at
96
hr.
Despite
the
large
increase
in
nonionic
fluorine
in
the
serum,
the
ionic
fluoride
level
was
only
0.03
ppm
and
remained
at
that
level
throughout
the
experiment.
Prior
to
administration
of
PFOA,
the
ionic
and
nonionic
fluorine
levels
in
serum
were
0.032
and
0.07
ppm,
respectively.
The
authors
concluded
that
PFOA
is
rapidly
absorbed
from
the
gastrointestinal
tract
and
then
rapidly
cleared
from
the
serum.

O'Malley
and
Ebbins
(
1981)
conducted
a
range
finding
study
that
indicates
significant
dermal
absorption
of
PFOA
in
male
and
female
New
Zealand
White
rabbits.
PFOA
at
concentrations
of
11
100,
1,000
and
2,000
mg/
kg
in
a
saline
slurry
was
applied
to
approximately
40%
of
the
shaved
trunk
of
the
animals
(
2/
sex/
group).
Animals
were
then
fitted
with
a
plastic
collar,
and
the
trunk
was
wrapped
with
impervious
plastic
sheeting.
The
exposure
period
was
24
hr/
day
5
days/
week
for
14
days.
Mortality
was
100%
(
4/
4)
in
the
2,000
mg/
kg
group,
75%
(
3/
4)
in
the
1,000
mg/
kg
group
and
0%
(
0/
4)
in
the
100
mg/
kg
group.

Kennedy
(
1985)
treated
rats
and
rabbits
dermally
with
a
total
of
10
applications
of
APFO
at
doses
of
0,
20,
200
or
2,000
mg/
kg.
Doses
were
applied
on
a
split
schedule
of
5
days
dosing,
2
days
of
rest,
and
5
days
of
dosing.
Treatment
resulted
in
elevated
blood
organofluorine
levels
that
increased
in
a
dose­
related
manner.

Kennedy
et
al.
(
1986)
exposed
male
rats
by
head­
only
inhalation
to
doses
of
0,
1,
8
or
84
mg/
m3
APFO
for
6/
hr/
day
5day/
wk
for
2
weeks.
Immediately
after
the
tenth
exposure,
mean
organofluoride
concentrations
in
the
blood
were
13,
47
and
108
ppm,
respectively
in
the
1,
8
and
84
mg/
m3
dose
groups.

3.2.2
Distribution
Studies
in
Animals
PFOA
distributes
primarily
to
the
liver,
plasma,
and
kidney,
and
to
a
lesser
extent,
other
tissues
of
the
body.
It
does
not
partition
to
the
lipid
fraction
or
adipose
tissue,
but
does
bind
to
macromolecules
in
the
tissues.
There
is
evidence
of
enterohepatic
circulation
of
the
compound.

Griffith
and
Long
(
1980)
determined
the
serum
and
liver
concentrations
of
PFOA
in
rhesus
monkeys
(
2
per
sex/
group)
in
a
90
day
oral
toxicity
study.
In
monkeys
that
received
a
dose
of
3
mg/
kg/
day,
mean
serum
PFOA
levels
were
50
ppm
in
males
and
58
ppm
in
females.
At
the
same
dose,
males
had
3
ppm
and
females
had
7
ppm
in
liver
samples.
At
a
dose
of
10
mg/
kg/
day,
male
monkeys
had
mean
serum
PFOA
levels
of
58
ppm
and
females
had
mean
levels
75
ppm.
Liver
levels
of
PFOA,
measured
as
organic
fluoride,
were
9
and
10
ppm
for
males
and
females,
respectively.

Johnson
et
al.
(
1984)
investigated
the
effect
of
feeding
cholestyramine
to
rats
on
the
fecal
elimination
of
APFO.
Since
APFO
exists
as
an
anion
at
physiologic
pH,
it
would
be
expected
to
complex
with
cholestyramine.
Ten
male
Charles
River
CD
rats,
12
weeks
of
age
and
weighing
300­
324
g,
were
given
a
single
iv
injection
of
13
mg/
kg
14C­
APFO.
Five
rats
were
given
4%
cholestyramine
in
feed.
Urine
and
feces
samples
were
collected
at
intervals
for
14
days,
at
which
time
the
animals
were
sacrificed
and
liver
samples
were
collected.
At
14
days
post
dose,
the
mean
percentage
of
PFOA
eliminated
in
the
feces
of
cholestyramine­
treated
rats
was
9.8­
fold
the
mean
percentage
eliminated
in
the
feces
of
rats
that
did
not
receive
cholestyramine..
Excretion
in
urine
was
41%
of
the
administered
dose
for
cholestyramine
treated
rats
and
67%
for
rats
that
did
not
receive
cholestyramine.
Carbon­
14
in
the
liver
equaled
4%
or
12.1
±
2.1
ug
eq/
g
in
cholestyramine
treated
rats
and
8
%
or
22.3
±
6.2
ug
eq/
g
rats
that
did
not
receive
cholestryamine.
In
plasma,
the
levels
were
5.1
±
1.7
ug
eq/
ml
in
cholestyramine
treated
rats
and
14.7
±
6.8
ug
eq/
ml
in
rats
that
did
not
receive
cholestyramine.
In
red
blood
cells,
the
levels
were
1.8
±
0.7
ug
eq/
ml
in
cholestryamine
treated
rats
and
4.2
±
2.4
ug
eq/
ml
in
rats
that
did
receive
cholestryamine.
The
high
concentration
of
14C­
APFO
in
the
liver
at
2
weeks
after
dosing
and
the
12
fact
that
cholestyramine
treatment
enhances
fecal
elimination
of
carbon­
14
nearly
10­
fold
suggests
that
there
is
enterohepatic
circulation
of
PFOA.

Hanhijarvi
et
al.
(
1987)
compared
the
disposition
of
PFOA
in
male
and
female
Wistar
rats
during
subchronic
administration.
PFOA
was
administered
by
gavage
to
48
newly­
weaned
animals
at
0,
3,
10,
and
30
mg/
kg/
day
for
28
consecutive
days.
Urine
was
collected
on
the
7th
and
28th
day
of
the
study.
At
the
end
of
the
study,
blood
was
collected
via
cardiac
puncture.
At
each
dose
level,
the
mean
PFOA
concentrations
in
the
plasma
of
the
male
rats
were
significantly
higher
than
those
of
the
female
rats.
The
mean
plasma
PFOA
concentrations
for
the
male
rats
were
48.6
±
26.5
ppm,
83.1
±
24.7
ppm,
and
53.4
±
11.2
ppm,
respectively
for
the
3,
10
and
30
mg/
kg/
day
dose
levels.
The
corresponding
figures
for
female
rats
were
2.43
±
5.96
ppm,
11.3
±
8.59
ppm,
and
9.06
±
8.80
ppm
respectively,
for
the
3,
10
and
30
mg/
kg/
day
dose
levels.
Although
the
plasma
PFOA
concentrations
were
significantly
higher
in
the
male
rats,
no
significant
histopathological
differences
between
the
sexes
were
observed
at
necropsy.

Ylinen
et
al.
(
1990)
studied
the
difference
between
male
and
female
Wistar
rats
in
the
distribution
and
accumulation
of
PFOA
after
single
and
subchronic
administration.
For
the
single
dose
study,
50
mg/
kg
of
PFOA
was
administered
by
ip
injection
to
groups
of
20
male
and
20
female
10
week
old
rats.
For
the
subchronic
study,
PFOA
was
administered
by
gavage
at
doses
of
3,
10,
and
30
mg/
kg/
day
to
groups
of
18
male
and
18
female
newly
weaned
rats.
For
both
studies,
samples
were
collected
for
determination
of
PFOA
levels
12
hr
after
treatment,
at
24­
168
hr
at
24
hr
intervals,
at
244
hr
and
at
336
hrs
after
treatment.
For
the
subchronic
study,
samples
were
also
taken
on
Day
28.
Serum
was
collected
by
cardiac
puncture;
the
brain
was
collected
afer
decapitation
and
at
necropsy
samples
from
the
liver,
kidney,
lung,
spleen,
ovary,
testis,
and
adipose
tissue
were
collected
and
frozen.
The
biological
half­
life
of
PFOA
in
the
serum
and
tissues
was
determined
from
the
linear
relationship
between
time
and
PFOA
concentration
in
the
semilogarithmic
plot.

In
the
single­
dose
study,
concentrations
of
PFOA
in
the
serum
and
tissues
were
higher
in
males
than
females
at
all
time
periods.
Twelve
hours
after
the
administration
of
PFOA
about
10%
of
the
administered
dose
was
found
in
the
serum
of
females,
whereas
about
40%
of
the
administered
dose
was
in
the
serum
of
males.
In
females,
the
concentration
of
PFOA
in
the
serum,
liver,
and
kidney
occurred
in
a
discontinuous
fashion,
indicating
distinct
phases.
The
half­
life
in
the
serum
was
24
hr
in
females
and
105
hr
in
the
males.
In
the
females,
a
half­
life
of
60
hr
was
estimated
in
the
liver
during
the
first
week.
In
the
males,
the
half­
life
in
liver
was
210
hr.
Although
PFOA
was
retained
by
the
liver,
it
was
not
found
in
the
lipid
fraction.
In
the
kidney,
the
half­
life
was
130
hr
and
145
hr
in
females
and
males,
respectively.
In
the
spleen,
the
half­
life
was
73
hr
in
the
females
and
170
hr
in
males.
PFOA
was
also
found
in
brain
tissue.
PFOA
was
not
detectable
in
adipose
tissue.

Samples
taken
on
the
28th
day
indicated
significantly
higher
PFOA
concentrations
in
the
serum
and
tissues
of
males
versus
females
at
all
three
dose
levels.
After
subchronic,
as
well
as
singledose
administration,
PFOA
was
mainly
distributed
in
the
serum
of
rats.
High
concentrations
of
PFOA
were
also
found
in
the
liver,
kidney,
and
lung
of
males
and
females.
At
the
30
mg/
kg/
day
dose
level,
females
and
males
exhibited,
respectively,
serum
concentrations
of
13.92
and
51.65
13
ppm,
liver
concentrations
of
6.64
ug/
g
and
49.77
ug/
g,
kidney
concentrations
of
12.54
ug/
g
and
39.81
ug/
g,
spleen
concentrations
of
1.59
ug/
g
and
4.10
ug/
g,
lung
concentrations
of
0.75
ug/
g
and
23.71
ug/
g,
and
brain
concentrations
of
0.044
ug/
g
and
0.710
ug/
g.
The
ovary
contained
1.16
ug/
g
and
the
testis
contained
7.22
ug/
g.
At
the
10
mg/
kg/
day
dose
level,
females
and
males
exhibited,
respectively,
serum
concentrations
of
12.47
and
87.27
ppm,
liver
concentrations
of
3.45
ug/
g
and
51.71
ug/
g,
kidney
concentrations
of
7.36
ug/
g
and
40.56
ug/
g,
spleen
concentrations
of
0.38
ug/
g
and
7.59
ug/
g,
lung
concentrations
of
0.22
ug/
g
and
22.58
ug/
g,
and
brain
concentrations
of
0.029
ug/
g
and
1.464
ug/
g.
The
ovary
contained
0.41
ug/
g
and
the
testis
contained
9.35
ug/
g.
At
the
3
mg/
kg/
day
dose
level,
females
and
males
exhibited,
respectively,
serum
concentrations
of
2.40
and
48.60
ppm,
liver
concentrations
of
1.81
ug/
g
and
39.90
ug/
g,
kidney
concentrations
of
0.06
ug/
g
and
1.55
ug/
g,
spleen
concentrations
of
0.15
ug/
g
and
4.75
ug/
g,
lung
concentrations
of
0.24
ug/
g
and
2.95
ug/
g,
and
brain
concentrations
of
<
limits
of
quantitation
and
0.398
ug/
g.
The
ovary
contained
less
than
the
limits
of
quantitation
and
the
testis
contained
6.24
ug/
g.
A
significant
positive
correlation
existed
between
the
administered
dose
and
the
concentration
of
PFOA
in
the
liver,
kidney,
spleen,
and
lung
of
females.
On
the
contrary,
no
significant
correlation
between
the
administered
dose
and
the
concentration
of
PFOA
was
observed
in
the
males,
as
10
mg/
kg/
day
produced
higher
PFOA
concentrations
in
the
serum
and
organs
than
30
mg/
kg/
day.
However,
in
males,
the
concentration
in
the
spleen,
testis,
and
brain
correlated
positively
with
the
concentration
in
the
serum.

Vanden
Heuvel
et
al.
(
1991b)
administered
9.4
umol/
kg
14C­
PFOA
by
ip
injection,
to
male
and
female
Harlan
Sprague­
Dawley
rats.
The
concentration
of
14C­
PFOA­
derived
radioactivity
in
the
blood
was
higher
and
eliminated
more
slowly
in
males
than
in
females.
In
males,
the
t
½
was
9
days
while
the
t
½
was
4
hr
in
females..
In
the
male
rats,
21%
of
the
administered
dose
was
present
in
the
liver
at
2
hr
after
treatment
followed
by
levels
in
the
plasma
and
kidney.
By
28
days
post­
treatment,
levels
in
the
liver
had
fallen
to
2%
of
the
administered
dose.
Far
lower
PFOA
concentrations
were
found
in
the
heart,
testis,
fat,
and
gastrocnemius
muscle.
In
females
at
2
hr
post
dose,
the
highest
concentrations
of
PFOA
were
found
in
the
plasma
followed
by
the
kidney,
liver
and
ovaries
in
that
order.
The
average
t
½
for
elimination
of
PFOA
from
the
liver
in
male
rats
was
11
days
compared
to
an
average
of
9
days
for
extrahepatic
tissues.
In
females,
the
average
t
½
for
tissue
elimination
was
approximately
3
hr.

Vanden
Heuvel
et
al.
(
1991a)
investigated
the
disposition
of
PFOA
in
perfused
male
rat
liver.
Liver
was
infused
with
0.08
umol
14C­
PFOA/
min
over
a
48
min
period
for
a
total
of
3.84
umol
14C.
Approximately
11%
of
the
cumulative
dose
of
14C­
PFOA
infused
was
extracted
by
the
liver
during
a
first
pass.
At
2
min,
the
cumulative
percent
of
PFOA
extracted
by
the
liver
was
33%;
that
was
substantially
greater
than
the
11%
cumulative
dose
of
14C
that
was
extracted
after
48
min
indicating
that
first­
pass
hepatic
uptake
of
PFOA
may
be
saturable.
Pooled
daily
urine
samples
taken
0­
4
days
post­
treatment
and
bile
extracts
analyzed
by
HPLC
contained
a
single
radioactive
peak
eluting
identically
to
the
parent
compound.
Tissues
were
taken
from
rats
treated
4,
14,
and
28
days
previously
with
14C­
PFOA
to
determine
the
presence
of
PFOA­
containing
lipid
conjugates.
Only
the
parent
compound
was
present
in
rat
tissues;
no
PFOA­
containing
hybrid
lipids
were
detected.
Fluoride
concentrations
in
plasma
and
urine
before
and
after
PFOA
treatment
were
unchanged,
indicating
that
PFOA
does
not
undergo
defluorination.
Female
rats
eliminated
PFOA­
derived
radioactivity
rapidly
in
the
urine
with
91%
of
the
dose
being
excreted
14
in
the
first
24
hr,
while
male
rats
excreted
only
6%
of
the
dose
in
the
same
time
period.
Negligible
radioactivity
was
recovered
in
the
feces
of
female
rats.
In
male
rats
during
the
28­
day
collection
period,
the
cumulative
excretion
of
PFOA­
derived
14C
in
urine
and
feces
was
36.4%
in
urine
and
35.1%
in
feces.
The
female
rat
retained
less
than
10%
of
the
administered
dose
after
24
hr,
while
the
male
rats
retained
30%
of
the
administered
dose
after
28
days.
The
whole­
body
elimination
half­
life
in
females
was
less
than
one
day,
and
in
males
it
was
15
days.
In
renalligated
rats
injected
ip
with
14C­
PFOA,
approximately
0.3%
of
the
PFOA­
derived
radioactivity
was
excreted
in
the
bile
after
6
hours.
No
sex­
related
difference
in
the
biliary
excretion
of
PFOA
was
observed
when
the
kidneys
were
ligated.

Vanden
Heuvel
et
al.
(
1992)
demonstrated
that
PFOA
covalently
binds
to
proteins
in
the
liver,
plasma,
and
testes
of
rats.
Carbon­
14­
labeled
PFOA
at
a
dose
of
9.4
umol/
kg
was
administered
by
ip
injection
to
six­
week
old
male
Harlan
Sprague­
Dawley
rats.
No
time­
dependent
changes
in
either
absolute
or
relative
concentrations
of
covalently
bound
PFOA­
derived
14C
were
found
at
2
hours,
1
and
4
days
post­
treatment.
Covalently
bound
PFOA
was
represented
by
0.1%
to
0.3%
of
the
tissue
14C
content.
The
absolute
concentration
of
covalently
bound
PFOA
was
significantly
higher
in
the
plasma
than
in
the
liver.
The
testes
had
the
highest
relative
concentration
of
covalently
bound
PFOA­
derived
radioactivity.

Johnson
(
1995a)
reported
on
the
disposition
of
the
tetrabutyl
ammonium
salt
of
perfluorooctanoic
acid
in
female
rabbits.
Individual
rabbits
were
given
intravenous
doses
of
0,
4,
16,
24
and
40
mg/
kg.
The
animal
give
40
mg/
kg
died
within
5
minutes
of
treatment.
All
other
animals
appeared
normal
throughout
the
study.
Serum
samples
were
analyzed
for
total
organic
fluorine
at
2,
4,
6,
8,
12,
24,
and
48
hours
post
dose.
At
2
hrs,
serum
organic
fluorine
levels
in
the
rabbits
that
received
0,
4,
16,
and
24
mg/
kg
were
1.25
ppm,
4.09
ppm,
14.9
ppm,
and
41.0
ppm,
respectively.
There
was
a
rapid
decrease
of
total
organic
fluorine
in
the
serum
with
time;
it
was
non­
detectable
at
48
hr.
The
biological
half­
life
was
on
the
order
of
4
hours.
The
total
organic
fluorine
levels
in
whole
liver
at
48
hr
post
dose
for
the
rabbits
that
received
0
mg/
kg,
4
mg/
kg,
16
mg/
kg,
and
24
mg/
kg
were
20
ug,
43
ug,
66
ug,
and
54
ug,
respectively.

3.2.3
Metabolism
Studies
in
Animals
Vanden
Heuvel
et
al.
(
1991b)
investigated
the
metabolism
of
PFOA
in
Harlan
Sprague­
Dawley
rats
administered
14C­
PFOA
(
9.4
umol/
kg,
ip).
Pooled
daily
urine
samples
(
0­
4
days
posttreatment
and
bile
extracts
analyzed
by
HPLC
contained
a
single
radioactive
peak
eluting
identically
to
the
parent
compound.
Tissues
were
taken
from
rats
treated
4,
14,
and
28
days
previously
with
14C­
PFOA
to
determine
the
presence
of
PFOA­
containing
lipid
conjugates.
Only
the
parent
compound
was
present
in
rat
tissues;
no
PFOA­
containing
hybrid
lipids
were
detected.
Fluoride
concentrations
in
plasma
and
urine
before
and
after
PFOA
treatment
were
unchanged,
indicating
that
PFOA
does
not
undergo
defluorination.

Ophaug
and
Singer
(
1980)
also
found
no
change
in
ionic
fluoride
level
in
the
serum
or
urine
following
oral
administration
of
PFOA
to
female
Holtzman
rats.
Ylinen
et
al.
(
1989)
found
no
evidence
of
phase
II
metabolism
of
PFOA
following
a
single
intraperitoneal
PFOA
dose
(
50
mg/
kg)
in
male
and
female
Wistar
rats.
15
3.2.4
Elimination
Studies
in
Animals
Gibson
and
Johnson
(
1980)
observed
a
sex
difference
in
extent
and
rate
of
excretion
of
total
carbon­
14
between
male
and
female
CD
rats
after
a
single
iv
dose
of
14C­
PFOA.
The
mean
dose
for
females
was
16.7
mg/
kg
while
that
for
males
was
13.1
mg/
kg.
Female
rats
excreted
essentially
all
of
the
administered
dose
via
the
urine
in
the
24
hours
after
treatment.
During
the
same
time
period,
male
rats
excreted
only
20%
of
the
total
dose.
Male
rats
excreted
83%
of
the
total
dose
via
the
urine
and
5.4%
via
the
feces
by
36
days
post
dose.
No
radioactivity
was
detected
in
tissues
of
female
rats
at
17
days
post
dose;
2.8%
of
the
total
dose
was
detected
in
the
liver
of
male
rats
and
1.1%
in
the
plasma
at
36
days
post
dose
with
lower
levels
equaling
<
0.5%
of
the
total
dose
in
other
organs.

The
urinary
excretion
of
APFO
in
rats
was
investigated
by
Hanhijarvi
et
al.
(
1982).
Four
male
and
six
female
Holtzman
rats
were
administered
2
mg
APFO
in
2
ml
aqueous
solution
by
stomach
intubation.
Seven
female
rats
were
administered
2
ml
distilled
water
as
controls.
The
animals
were
then
placed
in
metabolism
cages
with
rat
chow
and
tap
water.
Urine
was
collected
until
animals
were
sacrificed
at
24
hr
by
cardiac
puncture.
Serum
was
collected.
Ionic
fluoride
and
total
fluorine
content
of
serum
and
urine
were
determined,
and
nonionic
fluorine
was
calculated
as
the
difference.
For
clearance
studies
of
APFO
and
inulin,
the
rats
were
anesthetized
with
Inactin
and
the
femoral
artery
was
cannulated
for
continuous
infusion
of
5%
mannitol
in
isotonic
saline
while
the
femoral
artery
was
cannulated
for
drawing
blood
samples.
The
urinary
bladder
was
also
cannulated
for
serial
collections
of
urine.
When
the
urine
and
serum
collections
for
the
clearance
study
were
complete,
65­
68
mg/
kg
probenecid
was
administered
by
ip
injection
and
additional
clearance
tests
were
performed.
In
the
cumulative
excretion
study,
rats
were
dosed
iv
with
a
mixture
of
10%­
20%
radiolabeled­
APFO
and
80­
90%
unlabeled
APFO.
Five
percent
mannitol
was
infused
and
urine
specimens
were
collected
over
30­
min
intervals.
The
effect
of
probenecid
was
assessed
by
administering
65­
68
mg/
kg
by
ip
injection
at
least
30
min
prior
to
the
administration
of
APFO.
Twenty­
four
hours
after
oral
administration
of
APFO,
female
rats
had
excreted
76
±
2.7%
of
the
dose
in
the
urine
and
had
a
mean
serum
nonionic
fluorine
level
of
0.35
±
0.11
ppm,
while
male
rats
had
excreted
only
9.2
±
3.5%
of
the
dose
and
had
a
mean
serum
nonionic
fluorine
level
of
44.0
±
1.7
ppm.
97.5
±
0.25%
of
the
APFO
was
bound
in
the
plasma
of
both
male
and
female
rats.
The
clearance
studies
demonstrated
major
differences
between
the
sexes
in
rats.
The
APFO
clearance
in
female
rats
was
several
times
greater
than
the
inulin
clearance.
Administration
of
probenecid,
which
strongly
inhibits
active
renal
secretion
of
organic
acids,
reduced
the
APFO/
inulin
clearance
ratio
in
females
from
14.5
to
0.46.
APFO
clearance
was
reduced
from
5.8
to
0.11
ml/
min/
100g.
Net
APFO
excretion
was
reduced
from
4.6
ug/
min
to
0.13
ug/
min/
100g.
In
male
rats,
however,
the
APFO/
inulin
clearance
ratio
and
the
net
excretion
of
APFO
were
virtually
unaffected
by
probenecid.
In
the
males,
APFO
clearance
was
0.17
ml/
min/
100g,
the
APFO/
inulin
clearance
ratio
was
0.22,
and
net
APFO
excretion
was
0.17
ug/
min/
mg.
In
the
cumulative
excretion
studies,
female
rats
excreted
76%
of
the
administered
dose
of
APFO,
while
males
excreted
only
7.8%
of
the
administered
dose
over
a
7­
hr
period.
Probenecid
administration
modified
the
cumulative
excretion
curve
for
males
only
slightly.
However,
in
females
probenecid
markedly
reduced
APFO
elimination
to
11.8%.
The
authors
concluded
that
the
female
rat
possesses
an
active
secretory
mechanism
which
rapidly
eliminates
16
APFO
from
the
body.
This
secretory
mechanism
is
lacking
or
is
relatively
inactive
in
male
rats
and
accounts
for
the
greater
toxicity
of
APFO
in
males.

Hormonal
changes
during
pregnancy
do
not
appear
to
cause
a
change
in
the
rate
of
elimination
of
14C
after
oral
administration
of
a
single
dose
of
14C­
APFO
(
Gibson
and
Johnson,
1983).
At
8
or
9
days
after
conception,
four
pregnant
CD
rats
and
2
nonpregnant
female
CD
rats
were
given
a
mean
dose
of
15
mg/
kg
14C­
APFO.
Individual
urine
samples
were
collected
at
12,
24,
36,
and
48
hours
post
dose
and
analyzed
for
14C
content.
Essentially
all
of
the
14C
was
eliminated
via
the
urine
within
24
hours
for
both
groups
of
rats.

Hanhijarvi
et
al.
(
1988)
investigated
the
excretion
of
PFOA
in
the
beagle
dog.
Six
laboratory
bred
beagle
dogs,
3
males
and
3
females
were
given
an
iv
dose
of
30
mg/
kg
of
PFOA
followed
by
continuous
infusion
with
5%
mannitol.
Urine
was
collected
at
10
minute
intervals
for
60
min.
A
5
ml
blood
sample
was
collected
in
the
middle
of
each
urine
sampling
period.
Thirty
mg/
kg
probenecid
was
then
administered
by
iv
injection,
and
urine
and
blood
samples
were
collected
as
before.
Renal
clearance
of
PFOA
was
calculated
for
the
before
and
after
probenecid
injection
periods.
Four
additional
dogs,
2
males
and
2
females,
were
given
30
mg/
kg
of
PFOA
by
iv
injection.
These
dogs
were
kept
in
metabolism
cages,
and
blood
samples
were
collected
intermittently
for
30
days.
The
renal
clearance
rate
was
approximately
0.03
ml/
min/
kg.
Probenecid
significantly
reduced
the
PFOA
clearance
rate
in
both
sexes,
indicating
an
active
secretion
mechanism
for
PFOA.
The
plasma
half­
life
of
PFOA
was
473
hr
before
probenecid
administration
and
541
hr
after
in
male
dogs
and
202
hr
before
probenecid
and
305
hr
after
in
the
female
dogs.

Ylinen
et
al
(
1989)
studied
the
urinary
excretion
of
PFOA
in
male
Wistar
rats
after
castration
and
estradiol
administration.
They
also
studied
urinary
excretion
in
intact
males
and
females.
Twenty
male
rats
were
castrated
at
28
days
of
age
and
were
used
in
tests
of
PFOA
excretion
5
weeks
later.
Ten
castrated
and
10
intact
males
were
given
500
ug/
kg
estradiol
valerate
by
sc
injection
every
second
day
for
14
days
before
administration
of
PFOA.
PFOA
was
administered
as
a
single
ip
injection
at
50
mg/
kg.
Urine
was
collected
in
metabolism
cages
for
96
hr
after
PFOA
administration.
Blood
samples
were
collected
by
cardiac
puncture.
Six
female
rats
were
also
included
in
the
experiment.
Castration
and
administration
of
estradiol
to
the
male
rats
had
a
significant
stimulatory
effect
on
the
urinary
excretion
of
PFOA.
During
the
first
24
hours,
female
rats
excreted
72
±
5%
of
the
administered
dose
of
PFOA,
whereas
the
intact
males
excreted
only
9
±
4%.
After
the
estradiol
treatment,
both
the
intact
and
castrated
males
excreted
PFOA
in
amounts
similar
to
females,
61
±
19%
and
68
±
14%,
respectively.
The
castrated
males
without
estradiol
treatment
excreted
50
±
13%
of
the
administered
dose
of
PFOA
in
the
urine.
This
was
faster
than
the
intact
males
but
less
than
the
females
and
the
estrogen
treated
males.
At
the
end
of
the
test,
the
concentration
of
PFOA
in
the
serum
of
intact
males
was
17­
40
times
higher
than
the
concentration
PFOA
in
the
serum
of
other
groups.
There
was
no
statistically
significant
difference
in
the
serum
concentrations
between
the
other
groups.
PFOA
was
similarly
bound
by
the
proteins
in
the
serum
of
males
and
females.

Vanden
Heuvel
et
al.
(
1992a)
investigated
whether
androgens
or
estrogens
are
involved
in
the
marked
sex­
differences
in
the
urinary
excretion
of
PFOA.
Castrated
Harlan
Sprague­
Dawley
17
male
rats
were
given
9.4
umol/
kg,
14C­
PFOA
by
ip
injection.
Castration
increased
the
elimination
of
PFOA
in
the
urine
by
>
1­
fold
(
36%
of
the
dose
was
eliminated
in
4
days
versus
16%
in
controls),
demonstrating
that
a
factor
produced
by
the
testis
is
responsible
for
the
slow
elimination
of
PFOA
in
male
rats.
Castration
plus
17B­
estradiol
had
no
further
effect
on
PFOA
elimination
whereas
castration
plus
testosterone
replacement
at
the
physiological
level
reduced
PFOA
elimination
to
the
same
level
as
rats
with
intact
testis.
Thus,
in
male
rats,
testosterone
exerts
an
inhibitory
effect
on
renal
excretion
of
PFOA.
In
female
rats,
neither
ovariectomy
or
ovariectomy
plus
testosterone
affected
the
urinary
excretion
of
PFOA,
demonstrating
that
the
inhibitory
effect
of
testosterone
on
PFOA
renal
excretion
is
a
male­
specific
response.
Probenecid,
which
inhibits
the
renal
transport
system,
decreased
the
high
rate
of
PFOA
renal
excretion
in
castrated
males
but
had
no
effect
on
male
rats
with
intact
testis.

Kudo
et
al.
(
2002)
demonstrated
in
male
and
female
Wistar
rats
that
renal
clearance
(
CLR)
of
PFOA
and
the
renal
mRNA
levels
of
specific
organic
anion
transporters
are
markedly
affected
by
sex
hormones.
The
biological
half­
life
of
PFOA
in
male
rats
was
found
to
be
70
times
longer
(
5.7
days
versus
1.9
hours)
than
in
female
rats
and
this
difference
is
due
primarily
to
low
CLR
in
male
rats.
In
female
rats
there
appears
to
be
biphasic
elimination
of
PFOA;
the
fast
phase
occurs
with
a
half
life
of
approximately
1.9
hours
while
the
slow
phase
occurs
with
a
half
life
of
approximately
24
hours.
Castration
of
male
rats
caused
a
14­
fold
increase
in
CLR
of
PFOA.
The
elevated
PFOA
CLR
in
castrated
males
was
reduced
by
treating
them
with
testosterone.
Treatment
of
male
rats
with
estradiol
increased
the
CLR
of
PFOA.
In
female
rats,
ovariectomy
caused
a
significant
increase
in
CLR
of
PFOA,
which
was
reduced
by
estradiol
treatment.
Treatments
of
female
rats
with
testosterone
reduced
the
CLR
of
PFOA.
Treatment
with
probenecid,
a
known
inhibitor
of
organic
anion
transporters,
markedly
reduced
the
CLR
of
PFOA
in
male
rats,
castrated
male
rats,
and
female
rats.
To
identify
the
transporter
molecules
that
are
responsible
for
PFOA
transport
in
the
rat
kidney,
renal
mRNA
levels
of
specific
organic
anion
transporters
were
determined
in
male
and
female
rats
under
various
hormonal
states
and
compared
with
the
CLR
of
PFOA.
The
level
of
OAT2
mRNA
in
male
rats
was
only
13%
that
in
female
rats.
Castration
or
estradiol
treatment
increased
the
level
of
OAT2
mRNA
whereas
treatment
of
castrated
male
rats
with
testosterone
reduced
it.
Ovariectomy
of
female
rats
significantly
increased
the
level
of
OAT3
mRNA.
Multiple
regression
analysis
of
the
data
suggested
that
organic
anion
transporter
2
(
OAT2)
and
OAT3
are
responsible
for
urinary
elimination
of
PFOA
in
the
rat.

3.3
Epidemiology
Studies
3M
has
conducted
several
epidemiology
and
medical
surveillance
studies
of
the
workers
at
its
Decatur,
Antwerp,
and
Cottage
Grove
plants.
However,
these
studies
have
not
provided
information
regarding
the
potential
for
developmental
toxicity
since
the
majority
of
production
workers
at
facilities
that
produce
or
use
PFOA
are
male,
and
reproductive
outcomes
have
not
been
examined.
The
results
of
these
studies
are
summarized
below
for
the
readers'
information.
18
3.3.1
Mortality
Studies
in
Humans
A
retrospective
cohort
mortality
study
was
performed
on
employees
at
the
Cottage
Grove,
MN
plant
which
produces
APFO
(
Gilliland
and
Mandel,
1993).
At
this
plant,
APFO
production
was
limited
to
the
Chemical
Division.
The
cohort
consisted
of
workers
who
had
been
employed
at
the
plant
for
at
least
6
months
between
January
1947
and
December
1983.
Death
certificates
of
all
of
the
workers
were
obtained
to
determine
cause
of
death.
There
was
almost
complete
follow­
up
(
99.5%)
of
all
of
the
study
participants.
The
exposure
status
of
the
workers
was
categorized
based
on
their
job
histories.
If
they
had
been
employed
for
at
least
1
month
in
the
Chemical
Division,
they
were
considered
exposed.
All
others
were
considered
to
be
not
exposed
to
PFOA.
The
number
of
months
employed
in
the
Chemical
Division
provided
the
cumulative
exposure
measurements.
Of
the
3537
(
2788
men
and
749
women)
employees
who
participated
in
this
study,
398
(
348
men
and
50
women)
were
deceased.
Eleven
of
the
50
women
and
148
of
the
348
men
worked
in
the
Chemical
Division,
and
therefore,
were
considered
exposed
to
PFOA.

Standardized
Mortality
Ratios
(
SMRs),
adjusted
for
age,
sex,
and
race
were
calculated
and
compared
to
U.
S.
and
Minnesota
white
death
rates
for
men.
For
women,
only
state
rates
were
available.
The
SMRs
for
males
were
stratified
for
3
latency
periods
(
10,
15,
and
20
years)
and
3
periods
of
duration
of
employment
(
5,
10,
and
20
years).

For
all
female
employees,
the
SMRs
for
all
causes
and
for
all
cancers
were
less
than
1.
The
only
elevated
(
although
not
significant)
SMR
was
for
lymphopoietic
cancer,
and
was
based
on
only
3
deaths.
When
exposure
status
was
considered,
SMRs
for
all
causes
of
death
and
for
all
cancers
were
significantly
lower
than
expected,
based
on
the
U.
S.
rates,
for
both
the
Chemical
Division
workers
and
the
other
employees
of
the
plant.

In
all
male
workers
at
the
plant,
the
SMRs
were
close
to
1
for
most
of
the
causes
of
death
when
compared
to
both
the
U.
S.
and
the
Minnesota
death
rates.
When
latency
and
duration
of
employment
were
considered,
there
were
no
elevated
SMRs.
When
employee
deaths
in
the
Chemical
Division
were
compared
to
Minnesota
death
rates,
the
SMR
for
prostate
cancer
for
workers
in
the
Chemical
Division
was
2.03
(
95%
CI
.55
­
4.59).
This
was
based
on
4
deaths
(
1.97
expected).
There
was
also
a
statistically
significant
association
with
length
of
employment
in
the
Chemical
Division
and
prostate
cancer
mortality.
Based
on
the
results
of
proportional
hazard
models,
the
relative
risk
for
a
1­
year
increase
in
employment
in
the
Chemical
Division
was
1.13
(
95%
CI
1.01
to
1.27).
It
rose
to
3.3
(
95%
CI
1.02
­
10.6)
for
workers
employed
in
the
Chemical
Division
for
10
years
when
compared
to
the
other
employees
in
the
plant.
The
SMR
for
workers
not
employed
in
the
Chemical
Division
was
less
than
expected
for
prostate
cancer
(.
58).

An
update
of
this
study
was
conducted
to
include
the
death
experience
of
employees
through
1997
(
Alexander,
2001a).
The
cohort
consisted
of
3992
workers.
The
eligibility
requirement
was
increased
to
1
year
of
employment
at
the
Cottage
Grove
plant,
and
the
exposure
categories
were
changed
to
be
more
specific.
Workers
were
placed
into
3
exposure
groups
based
on
job
history
information:
definite
PFOA
exposure
(
n
=
492,
jobs
where
cell
generation,
drying,
shipping
and
packaging
of
PFOA
occurred
throughout
the
history
of
the
plant);
probable
PFOA
exposure
(
n
=
19
1685,
other
chemical
division
jobs
where
exposure
to
PFOA
was
possible
but
with
lower
or
transient
exposures);
and
not
exposed
to
fluorochemicals
(
n
=
1815,
primarily
non­
chemical
division
jobs).

In
this
new
cohort,
607
deaths
were
identified:
46
of
these
deaths
were
in
the
PFOA
exposure
group,
267
in
the
probable
exposure
group,
and
294
in
the
non­
exposed
group.
When
all
employees
were
compared
to
the
state
mortality
rates,
SMRs
were
less
than
1
or
only
slightly
higher
for
all
of
the
causes
of
death
analyzed.
None
of
the
SMRs
were
statistically
significant
at
p
=
.05.
The
highest
SMR
reported
was
for
bladder
cancer
(
SMR
=
1.31,
95%
CI
=
0.42
 
3.05).
Five
deaths
were
observed
(
3.83
expected).

A
few
SMRs
were
elevated
for
employees
in
the
definite
PFOA
exposure
group:
2
deaths
from
cancer
of
the
large
intestine
(
SMR
=
1.67,
95%
CI
=
0.02
 
6.02),
1
from
pancreatic
cancer
(
SMR
=
1.34,
95%
CI
=
0.03
 
7.42),
and
1
from
prostate
cancer
(
SMR
=
1.30,
95%
CI
=
0.03
 
7.20).
In
addition,
employees
in
the
definite
PFOA
exposure
group
were
2.5
times
more
likely
to
die
from
cerebrovascular
disease
(
5
deaths
observed,
1.94
expected;
95%
CI
=
0.84
 
6.03).

In
the
probable
exposure
group,
3
SMRs
should
be
noted:
cancer
of
the
testis
and
other
male
genital
organs
(
SMR
=
2.75,
95%
CI
=
0.07
 
15.3);
pancreatic
cancer
(
SMR
=
1.24,
95%
CI
=
0.45
 
2.70);
and
malignant
melanoma
of
the
skin
(
SMR
=
1.42,
95%
CI
=
0.17
 
5.11).
Only
1,
6,
and
2
cases
were
observed,
respectively.
The
SMR
for
prostate
cancer
in
this
group
was
0.86
(
95%
CI
=
0.28
 
2.02)
(
n
=
5).

There
were
no
notable
excesses
in
SMRs
in
the
non­
exposed
group,
except
for
cancer
of
the
bladder
and
other
urinary
organs.
Four
cases
were
observed
and
only
1.89
were
expected
(
95%
CI
=
0.58
 
5.40).

It
is
difficult
to
interpret
the
results
of
the
prostate
cancer
deaths
between
the
first
study
and
the
update
because
the
exposure
categories
were
modified
in
the
update.
Only
1
death
was
reported
in
the
definite
exposure
group
and
5
were
observed
in
the
probable
exposure
group.
All
of
these
deaths
would
have
been
placed
in
the
chemical
plant
employees
exposure
group
in
the
first
study.
The
number
of
years
that
these
employees
worked
at
the
plant
and/
or
were
exposed
to
PFOA
was
not
reported.
This
is
important
because
even
1
prostate
cancer
death
in
the
definite
PFOA
exposure
group
resulted
in
an
elevated
SMR
for
the
group.
Therefore,
if
any
of
the
employees'
exposures
were
misclassified,
the
results
of
the
analysis
could
be
altered
significantly.

The
excess
mortality
in
cerebrovascular
disease
noted
in
employees
in
the
definite
exposure
group
was
further
analyzed
based
on
number
of
years
of
employment
at
the
plant.
Three
of
the
5
deaths
occurred
in
workers
who
were
employed
in
jobs
with
definite
PFOA
exposure
for
more
than
5
years
but
less
than
10
years
(
SMR
=
15.03,
95%
CI
=
3.02
 
43.91).
The
other
2
occurred
in
employees
with
less
than
1
year
of
definite
exposure.
The
SMR
was
6.9
(
95%
CI
=
1.39
 
20.24)
for
employees
with
greater
than
5
years
of
definite
PFOA
exposure.
In
order
to
confirm
that
the
results
regarding
cerebrovascular
disease
were
not
an
artifact
of
death
certificate
coding,
regional
mortality
rates
were
used
for
the
reference
population.
The
results
did
not
change.
20
When
these
deaths
were
further
analyzed
by
cumulative
exposure
(
time­
weighted
according
to
exposure
category),
workers
with
27
years
of
exposure
in
probable
PFOA
exposed
jobs
or
those
with
9
years
of
definite
PFOA
exposure
were
3.3
times
more
likely
to
die
of
cerebrovascular
disease
than
the
general
population.
A
dose­
response
relationship
was
not
observed
with
years
of
exposure.

It
is
difficult
to
compare
the
results
of
the
first
and
second
mortality
studies
at
the
Cottage
Grove
plant
since
the
exposure
categories
were
modified.
Although
the
potential
for
exposure
misclassification
was
certainly
more
likely
in
the
first
study,
it
may
still
have
occurred
in
the
update
as
well.
It
is
difficult
to
judge
the
reliability
of
the
exposure
categories
that
were
defined
without
measured
exposures.
Although
serum
PFOA
measurements
were
considered
in
the
exposure
matrix
developed
for
the
update,
they
were
not
directly
used.
In
the
second
study,
the
chemical
plant
employees
were
sub­
divided
into
PFOA­
exposed
groups,
and
the
film
plant
employees
essentially
remained
in
the
"
non­
exposed"
group.
This
was
an
effort
to
more
accurately
classify
exposures;
however,
these
new
categories
do
not
take
into
account
duration
of
exposure
or
length
of
employment.
Another
limitation
to
this
study
is
that
17
death
certificates
were
not
located
for
deceased
employees
and
therefore
were
not
included
in
the
study.
The
inclusion
or
exclusion
of
these
deaths
could
change
the
analyses
for
the
causes
of
death
that
had
a
small
number
of
cases.
Follow
up
of
worker
mortality
at
Cottage
Grove
(
and
Decatur)
needs
to
continue.
Although
there
were
more
than
200
additional
deaths
included
in
this
analysis,
it
is
a
small
number
and
the
cohort
is
still
relatively
young.
Given
the
results
of
studies
on
fluorochemicals
in
both
animals
and
humans,
further
analysis
is
warranted.

3.3.2
Hormone
Study
in
Humans
Endocrine
effects
have
been
associated
with
PFOA
exposure
in
animals;
therefore,
medical
surveillance
data,
including
hormone
testing,
from
male
employees
only
of
the
Cottage
Grove,
Minnesota
plant
were
analyzed
(
Olsen,
et
al.,
1998a).
PFOA
serum
levels
were
obtained
for
volunteer
workers
in
1993
(
n
=
111)
and
1995
(
n
=
80).
Sixty­
eight
employees
were
common
to
both
sampling
periods.
In
1993,
the
range
of
PFOA
was
0­
80
ppm
(
although
80
ppm
was
the
limit
of
detection
that
year,
so
it
could
have
been
higher)
and
0­
115
ppm
in
1995
using
thermospray
mass
spectrophotometry
assay.
Eleven
hormones
were
assayed
from
the
serum
samples.
They
were:
cortisol,
dehydroepiandrosterone
sulfate
(
DHEAS),
estradiol,
FSH,
17
gamma­
hydroxyprogesterone
(
17­
HP),
free
testosterone,
total
testosterone,
LH,
prolactin,
thyroid­
stimulating
hormone
(
TSH)
and
sex
hormone­
binding
globulin
(
SHBG).
Employees
were
placed
into
4
exposure
categories
based
on
their
serum
PFOA
levels:
0­
1
ppm,
1­
<
10
ppm,
10­
<
30
ppm,
and
>
30
ppm.
Statistical
methods
used
to
compare
PFOA
levels
and
hormone
values
included:
multivariable
regression
analysis,
ANOVA,
and
Pearson
correlation
coefficients.

PFOA
was
not
highly
correlated
with
any
of
the
hormones
or
with
the
following
covariates:
age,
alcohol
consumption,
BMI,
or
cigarettes.
Most
of
the
employees
had
PFOA
serum
levels
less
than
10
ppm.
In
1993,
only
12
employees
had
serum
levels
>
10
ppm,
and
15
in
1995.
However,
these
levels
ranged
from
approximately
10
ppm
to
over
114
ppm.
There
were
only
4
employees
in
the
>
30
ppm
PFOA
group
in
1993
and
only
5
in
1995.
Therefore,
it
is
likely
that
there
was
not
21
enough
power
to
detect
differences
in
either
of
the
highest
categories.
The
mean
age
of
the
employees
in
the
highest
exposure
category
was
the
lowest
in
both
1993
and
1995
(
33.3
years
and
38.2
years,
respectively).
Although
not
significantly
different
from
the
other
categories,
BMI
was
slightly
higher
in
the
highest
PFOA
category.

Estradiol
was
highly
correlated
with
BMI
(
r
=
.41,
p
<
.001
in
1993,
and
r
=
.30,
p
<
.01
in
1995).
In
1995,
all
5
employees
with
PFOA
levels
>
30
ppm
had
BMIs
>
28,
although
this
effect
was
not
observed
in
1993.
Estradiol
levels
in
the
>
30
ppm
group
in
both
years
were
10%
higher
than
the
other
PFOA
groups;
however,
the
difference
was
not
statistically
significant.
The
authors
postulate
that
the
study
may
not
have
been
sensitive
enough
to
detect
an
association
between
PFOA
and
estradiol
because
measured
serum
PFOA
levels
were
likely
below
the
observable
effect
levels
suggested
in
animal
studies
(
55
ppm
PFOA
in
the
CD
rat).
Only
3
employees
in
this
study
had
PFOA
serum
levels
this
high.
They
also
suggest
that
the
higher
estradiol
levels
in
the
highest
exposure
category
could
suggest
a
threshold
relationship
between
PFOA
and
estradiol.

Free
testosterone
was
highly
correlated
with
age
in
both
1993
and
1995.
The
authors
did
not
report
a
negative
association
between
PFOA
serum
levels
and
testosterone.
There
were
no
statistically
significant
trends
noted
for
PFOA
and
either
bound
or
free
testosterone.
However,
17­
HP,
a
precursor
of
testosterone,
was
highest
in
the
>
30
ppm
PFOA
group
in
both
1993
and
1995.
In
1995,
PFOA
was
significantly
associated
with
17­
HP
in
regression
models
adjusted
for
possible
confounders.
However,
the
authors
state
that
this
association
was
based
on
the
results
of
one
employee
(
data
were
not
provided
in
the
report).
There
were
no
significant
associations
between
PFOA
and
cortisol,
DHEAS,
FSH,
LH,
and
SHBG.

There
are
several
design
issues
that
should
be
noted
when
evaluating
the
results
of
this
study.
First,
although
there
were
2
study
years
(
1993
and
1995),
the
populations
were
not
independent.
Sixty­
eight
employees
participated
in
both
years.
Second,
there
were
31
fewer
employees
who
participated
in
the
study
in
1995,
thus
reducing
the
power
of
the
study.
There
were
also
very
few
employees
in
either
year
with
serum
PFOA
levels
greater
than
10
ppm.
Third,
the
cross­
sectional
design
of
the
study
does
not
allow
for
analysis
of
temporality
of
an
association.
Since
the
halflife
of
PFOA
is
at
least
1
year,
the
authors
suggest
that
it
is
possible
that
there
may
be
some
biological
accommodation
to
the
effects
of
PFOA.
Fourth,
only
one
sample
was
taken
for
each
hormone
for
each
of
the
study
years.
In
order
to
get
more
accurate
measurements
for
some
of
the
hormones,
pooled
blood
taken
in
a
short
time
period
should
have
been
used
for
each
participant.
Fifth,
some
of
the
associations
that
were
measured
in
this
study
were
done
based
on
the
results
of
an
earlier
paper
that
linked
PFOA
with
increased
estradiol
and
decreased
testosterone
levels.
However,
total
serum
organic
fluorine
was
measured
in
that
study
instead
of
PFOA,
making
it
difficult
to
compare
the
results.
Finally,
there
may
have
been
some
measurement
error
of
some
of
the
confounding
variables.

3.3.3
Study
on
Episodes
of
Care
(
Morbidity)

In
order
to
gain
additional
insight
into
the
effects
of
fluorochemical
exposure
on
workers'
health,
an
"
episode
of
care"
analysis
was
undertaken
at
the
Decatur
plant
to
screen
for
morbidity
22
outcomes
that
may
be
associated
with
long­
term,
high
exposure
to
fluorochemicals
(
Olsen
et
al.,
2001g).
An
"
episode
of
care"
is
a
series
of
health
care
services
provided
from
the
start
of
a
particular
disease
or
condition
until
solution
or
resolution
of
that
problem.
Episodes
of
care
were
identified
in
employees'
health
claims
records
using
Clinical
Care
Groups
(
CCG)
software.
All
inpatient
and
outpatient
visits
to
health
care
providers,
procedures,
ancillary
services
and
prescription
drugs
used
in
the
diagnosis,
treatment,
and
management
of
over
400
diseases
or
conditions
were
tracked.

Episodes
of
care
were
analyzed
for
652
chemical
employees
and
659
film
plant
employees
who
worked
at
the
Decatur
plant
for
at
least
1
year
between
January
1,
1993
and
December
31,
1998.
Based
on
work
history
records,
employees
were
placed
into
different
comparison
groups:
Group
A
consisted
of
all
film
and
chemical
plant
workers;
Group
B
had
employees
who
only
worked
in
either
the
film
or
chemical
plant;
Group
C
consisted
of
employees
who
worked
in
jobs
with
high
POSF
exposures;
and
Group
D
had
employees
who
worked
in
high
exposures
in
the
chemical
plant
for
10
years
or
more
prior
to
the
onset
of
the
study.
Film
plant
employees
were
considered
to
have
little
or
no
fluorochemical
exposure,
while
chemical
plant
employees
were
assumed
to
have
the
highest
exposures.

Ratios
of
observed
to
expected
episodes
of
care
were
calculated
for
each
plant.
Expected
numbers
were
based
on
3M's
employee
population
experience
using
indirect
standardization
techniques.
A
ratio
of
the
chemical
plant's
observed
to
expected
experience
divided
by
the
film
plant's
observed
to
expected
experience
was
calculated
to
provide
a
relative
risk
ratio
for
each
episode
of
care
(
RREpC).
For
each
RREpC,
95%
confidence
intervals
were
calculated.
Episodes
of
care
that
were
of
greatest
interest
were
those
which
had
been
reported
in
animal
or
epidemiologic
literature
on
PFOS
and
PFOA:
liver
and
bladder
cancer,
endocrine
disorders
involving
the
thyroid
gland
and
lipid
metabolism,
disorders
of
the
liver
and
biliary
tract,
and
reproductive
disorders.

The
only
increased
risk
of
episodes
for
these
conditions
of
a
priori
interest
were
for
neoplasms
of
the
male
reproductive
system
and
for
the
overall
category
of
cancers
and
benign
growths
(
which
included
cancer
of
the
male
reproductive
system).
There
was
an
increased
risk
of
episodes
for
the
overall
cancer
category
for
all
4
comparison
groups.
The
risk
ratio
was
greatest
in
the
group
of
employees
with
the
highest
and
longest
exposures
to
fluorochemicals
(
RREpC
=
1.6,
95%
CI
=
1.2
 
2.1).
Increased
risk
of
episodes
in
long­
time,
high­
exposure
employees
also
was
reported
for
male
reproductive
cancers
(
RREpC
=
9.7,
95%
CI
=
1.1
­
458).
It
should
be
noted
that
the
confidence
interval
is
very
wide
for
male
reproductive
cancers
and
the
sub­
category
of
prostate
cancer.
Five
episodes
of
care
were
observed
for
reproductive
cancers
in
chemical
plant
employees
(
1.8
expected),
of
which
4
were
prostate
cancers
(
RREpC
=
8.2,
95%
CI
=
0.8
­
399).
One
episode
of
prostate
cancer
was
observed
in
film
plant
employees
(
3.4
expected).
This
finding
should
be
noted
because
an
excess
in
prostate
cancer
mortality
was
observed
in
the
Cottage
Grove
plant
mortality
study
when
there
were
only
2
exposure
categories
(
chemical
division
employees
and
non­
chemical
division
employees).
The
update
of
the
study
sub­
divided
the
chemical
plant
employees
and
did
not
corroborate
this
finding
when
exposures
were
divided
into
definitely
exposed
and
probably
exposed
employees.
23
There
was
an
increased
risk
of
episodes
for
neoplasms
of
the
gastrointestinal
tract
in
the
high
exposure
group
(
RREpC
=
1.8,
95%
CI
=
1.2­
3.0)
and
the
long­
term
employment,
high
exposure
group
(
RREpC
=
2.9,
95%
CI
=
1.7
 
5.2).
Most
of
the
episodes
were
attributable
to
benign
colonic
polyps.
Similar
numbers
of
episodes
were
reported
in
film
and
chemical
plant
employees.

In
the
entire
cohort,
only
1
episode
of
care
was
reported
for
liver
cancer
(
0.6
expected)
and
1
for
bladder
cancer
(
1.5
expected).
Both
occurred
in
film
plant
employees.
Only
2
cases
of
cirrhosis
of
the
liver
were
observed
(
0.9
expected),
both
in
the
chemical
plant.
There
was
a
greater
risk
of
lower
urinary
tract
infections
in
chemical
plant
employees,
but
they
were
mostly
due
to
recurring
episodes
of
care
by
the
same
employees.
It
is
difficult
to
draw
any
conclusions
about
these
observations,
given
the
small
number
of
episodes
reported.

Chemical
plant
employees
in
the
high
exposure,
long­
term
employment
group
were
2
½
times
more
likely
to
seek
care
for
disorders
of
the
biliary
tract
than
their
counterparts
in
the
film
plant
(
RREpC
=
2.6,
95%
CI
=
1.2
­
5.5).
Eighteen
episodes
of
care
were
observed
in
chemical
plant
employees
and
14
in
film
plant
workers.
The
sub­
categories
that
influenced
this
observation
were
episodes
of
cholelithiasis
with
acute
cholecystitis
and
cholelithiasis
with
chronic
or
unspecified
cholecystitis.
Most
of
the
observed
cases
occurred
in
chemical
plant
employees.

Risk
ratios
of
episodes
of
care
for
endocrine
disorders,
which
included
sub­
categories
of
thyroid
disease,
diabetes,
hyperlipidemia,
and
other
endocrine
or
nutritional
disorders,
were
not
elevated
in
the
comparison
groups.
Conditions
which
were
not
identified
a
priori
but
which
excluded
the
null
hypothesis
in
the
95%
confidence
interval
for
the
high
exposure,
long­
term
employment
group
included:
disorders
of
the
pancreas,
cystitis,
and
lower
urinary
tract
infections.

The
results
of
this
study
should
only
be
used
for
hypothesis
generation.
Although
the
episode
of
care
design
allowed
for
a
direct
comparison
of
workers
with
similar
demographics
but
different
exposures,
there
are
many
limitations
to
this
design.
The
limitations
include:
1)
episodes
of
care
are
reported,
not
disease
incidence,
2)
the
data
are
difficult
to
interpret
because
a
large
RREpC
may
not
necessarily
indicate
high
risk
of
incidence
of
disease,
3)
many
of
the
risk
ratios
for
episodes
of
care
had
very
wide
confidence
intervals,
thereby
providing
unstable
results,
4)
the
analysis
was
limited
to
6
years,
5)
the
utilization
of
health
care
services
may
reflect
local
medical
practice
patterns,
6)
individuals
may
be
counted
more
than
once
in
the
database
because
they
can
be
categorized
under
larger
or
smaller
disease
classifications,
7)
episodes
of
care
may
include
the
same
individual
several
times,
8)
not
all
employees
were
included
in
the
database,
such
as
those
on
long­
term
disability,
9)
the
analysis
may
be
limited
by
the
software
used,
which
may
misclassify
episodes
of
care,
10)
the
software
may
assign
2
different
diagnoses
to
the
same
episode,
and
11)
certain
services,
such
as
lab
procedures
may
not
have
been
reported
in
the
database.

3.3.4
Medical
Surveillance
Studies
from
the
Antwerp
and
Decatur
Plants
A
cross­
sectional
analysis
of
the
data
from
the
2000
medical
surveillance
program
at
the
Decatur
and
Antwerp
plants
was
undertaken
to
determine
if
there
were
any
associations
between
PFOA
24
and
hematology,
clinical
chemistries,
and
hormonal
parameters
of
volunteer
employees
(
Olsen,
et
al.,
2001e).
The
data
were
analyzed
for
all
employees
from
both
plant
locations.
Mean
PFOA
serum
levels
were
1.03
ppm
for
all
male
employees
at
the
Antwerp
plant
and
1.90
ppm
for
all
male
employees
at
the
Decatur
plant.
Male
production
employees
at
the
Decatur
plant
had
significantly
higher
(
p
<
.05)
mean
serum
levels
(
2.34
ppm)
than
those
at
the
Antwerp
plant
(
1.28
ppm).
Non­
production
employees
at
both
plants
had
mean
levels
below
1
ppm.
PFOA
serum
levels
were
higher
than
the
PFOS
serum
values
at
both
plants,
especially
the
Decatur
plant
where
serum
levels
are
higher
overall.
In
addition,
values
for
total
organic
fluorine
were
even
higher
than
the
PFOA
levels.

Multivariable
regression
analyses
were
conducted
to
adjust
for
possible
confounders
that
may
affect
the
results
of
the
clinical
chemistry
tests.
The
following
variables
were
included:
production
job
(
yes
or
no),
plant,
age,
body
mass
index
(
BMI),
cigarettes/
day,
drinks/
day
and
years
worked
at
the
plant.
A
positive
significant
association
was
reported
between
PFOA
and
cholesterol
(
p
=
.05)
and
PFOA
and
triglycerides
(
p
=
.002).
Age
was
also
significant
in
both
analyses.
Alcohol
consumed
per
day
was
significant
in
the
cholesterol
model,
while
BMI
and
cigarettes
smoked
per
day
was
significant
for
triglycerides.
When
both
PFOA
and
PFOS
were
included
in
the
analyses,
neither
reached
statistical
significance
in
the
cholesterol
model,
while
PFOA
remained
significant
(
p
=
.02)
in
the
triglycerides
model.
HDL
was
negatively
associated
with
PFOA
(
p
=
.04)
and
remained
significant
(
p
=
.04)
when
both
PFOA
and
PFOS
were
included
in
the
model.
A
positive
association
(
p
=
.01)
between
T3
and
PFOA
was
also
observed
and
remained
statistically
significant
(
p
=
.05)
when
PFOS
was
included
in
the
model.
BMI,
cigarettes/
day,
alcohol/
day
were
also
significant
in
the
model.
None
of
the
other
clinical
chemistry,
thyroid
or
hematology
measures
were
significantly
associated
with
PFOA
in
the
regression
model.

A
longitudinal
analysis
of
the
above
data
and
previous
medical
surveillance
results
was
performed
to
determine
whether
occupational
exposure
to
fluorochemicals
over
time
is
related
to
changes
in
clinical
chemistry
and
lipid
results
in
employees
of
the
Antwerp
and
Decatur
facilities
(
Olsen,
et
al.,
2001f).
The
clinical
chemistries
included:
cholesterol,
HDL,
triglycerides,
alkaline
phosphatase,
gamma
glutamyl
transferase
(
GGT),
aspartate
aminotransferase
(
AST),
alanine
aminotransferase
(
ALT),
total
and
direct
bilirubin.
Medical
surveillance
data
from
1995,
1997,
and
2000
were
analyzed
using
multivariable
regression.
The
plants
were
analyzed
using
3
subcohorts
that
included
those
who
participated
in
2
or
more
medical
exams
between
1995
and
2000.
A
total
of
175
male
employees
voluntarily
participated
in
the
2000
surveillance
and
at
least
one
other.
Only
41
employees
were
participants
in
all
3
surveillance
periods.

When
mean
serum
PFOA
levels
were
compared
by
surveillance
year,
PFOA
levels
in
the
employees
participating
in
medical
surveillance
at
the
Antwerp
plant
increased
between
1994/
95
and
1997
and
then
decreased
slightly
between
1997
and
2000.
At
the
Decatur
plant,
PFOA
serum
levels
decreased
between
1994/
95
and
1997
and
then
increased
between
1997
and
2000.
When
analyzed
using
mixed
model
multivariable
regression
and
combining
Antwerp
and
Decatur
employees,
there
was
a
statistically
significant
positive
association
between
PFOA
and
serum
cholesterol
(
p
=
.0008)
and
triglycerides
(
p
=
.0002)
over
time.
When
analyzed
by
plant
25
and
also
by
subcohort,
these
associations
were
limited
to
the
Antwerp
employees
(
p
=
.005)
and,
in
particular,
the
21
Antwerp
employees
who
participated
in
all
3
surveillance
years
(
p
=
.001).
However,
the
association
between
PFOA
and
triglycerides
was
also
statistically
significant
(
p
=
.02)
for
the
subgroup
in
which
employees
participated
in
biomonitoring
in
1994/
95
and
2000.
There
was
not
a
significant
association
between
PFOA
and
triglycerides
among
Decatur
workers.
There
were
no
significant
associations
between
PFOA
and
changes
over
time
in
HDL,
alkaline
phosphatase,
GGT,
AST,
ALT,
total
bilirubin,
and
direct
bilirubin.

There
are
several
limitations
to
the
2000
cross­
sectional
and
longitudinal
studies
including:
1)
serum
PFOA
levels
were
significantly
higher
at
the
Decatur
plant
than
at
the
Antwerp
plant,
2)
all
participants
were
volunteers,
3)
there
were
several
consistent
differences
in
clinical
chemistry
profiles
and
demographics
between
employees
of
the
Decatur
and
Antwerp
plants
(
Antwerp
employees
as
compared
to
Decatur
employees
had
lower
PFOA
serum
levels,
were
younger,
had
lower
BMIs,
worked
fewer
years,
had
higher
alcohol
consumption,
higher
mean
HDL
and
bilirubin
values,
lower
mean
triglyceride,
alkaline
phosphatase,
GGT,
AST,
and
ALT
values,
and
mean
thyroid
hormone
values
tended
to
be
higher),
4)
PFOS
and
other
perfluorinated
chemicals
are
also
present
in
these
plants,
5)
in
the
cross­
sectional
study,
plant
populations
cannot
be
compared
because
they
were
placed
into
quartiles
based
on
PFOS
serum
distributions
only
which
were
different
for
each
subgroup
and
not
applicable
to
PFOA,
6)
only
one
measurement
at
a
certain
point
in
time
was
collected
for
each
clinical
chemistry
test,
and
7)
PFOA
serum
levels
overall
have
been
increasing
over
time
in
these
employees.
In
addition,
in
the
longitudinal
study
only
a
small
number
of
employees
participated
in
all
3
sampling
periods
(
24%),
different
labs
and
analytical
techniques
for
PFOA
were
used
each
year,
and
female
employees
could
not
be
analyzed
because
of
the
small
number
of
participants.

3.3.5
Medical
Surveillance
Studies
from
the
Cottage
Grove
Plant
A
voluntary
medical
surveillance
program
was
offered
to
employees
of
the
Cottage
Grove,
Minnesota
plant
in
1993,
1995,
and
1997
(
n
=
111,
80
and
74
employees,
respectively)
(
Olsen,
et
al.,
1998b,
Olsen
et
al.,
2000).
The
clinical
chemistry
parameters
(
cholesterol,
hepatic
enzymes,
and
lipoprotein
levels)
used
in
the
longitudinal
and
cross­
sectional
studies
of
the
Antwerp
and
Decatur
plants
were
also
used
in
this
study.
In
addition,
in
1997
only,
cholecystokinin­
33
(
CCK)
was
also
measured
at
the
Cottage
Grove
plant.
CCK
levels
were
observed
because
certain
research
has
suggested
that
pancreas
acinar
cell
adenomas
seen
in
rats
exposed
to
PFOA
may
be
the
result
of
increased
CCK
levels
(
Obourn,
et
al.,
1997).

Only
male
employees
involved
in
PFOA
production
were
included
in
the
study.
Sixty­
eight
employees
were
common
to
the
1993
and
1995
sampling
periods,
21
were
common
between
1995
and
1997,
and
17
participated
in
all
three
surveillance
years.
Mean
serum
PFOA
levels
and
ranges
are
provided
in
Table
2
of
the
Biomonitoring
Section
of
this
report.
It
should
be
noted
that
Cottage
Grove
has
the
highest
serum
PFOA
levels
of
the
3
plants
studied.

Employees'
serum
PFOA
levels
were
stratified
into
3
categories
(<
1,
1­
<
10,
and
$
10
ppm),
chosen
to
provide
a
greater
number
of
employees
in
the
$
10
ppm
category.
As
employees'
mean
serum
PFOA
levels
increased,
no
statistically
significant
abnormal
liver
function
tests,
26
hypolipidemia,
or
cholestasis
were
observed
in
any
of
the
sampling
years.
Multivariable
regression
analyses
controlling
for
potential
confounders
(
age,
alcohol
consumption,
BMI,
and
cigarettes
smoked)
yielded
similar
results.
The
authors
also
reported
that
renal
function,
blood
glucose,
and
hematology
measures
were
not
associated
with
serum
PFOA
levels;
however,
these
data
were
not
provided
in
the
paper.

The
mean
CCK
value
reported
for
the
1997
sample
was
28.5
pg/
ml
(
range
8.8
­
86.7
pg/
ml).
The
means
in
the
2
serum
categories
<
10
ppm
were
at
least
50%
higher
than
in
the
$
10
ppm
category.
A
statistically
significant
(
p
=
.03)
negative
association
between
mean
CCK
levels
and
the
3
PFOA
serum
categories
was
observed.
A
scatter
plot
of
the
natural
log
of
CCK
and
PFOA
shows
that
all
but
2
CCK
values
are
within
the
assay's
reference
range
of
0
­
80
pg/
ml.
Both
of
these
employees
(
CCK
values
of
80.5
and
86.7
pg/
ml)
had
serum
PFOA
levels
less
than
10
ppm
(
0.6
and
5.6
ppm,
respectively).
A
multiple
regression
model
of
the
natural
log
of
CCK
and
serum
PFOA
levels
continued
to
display
a
negative
association
after
adjusting
for
potential
confounders.

The
cross­
sectional
design
is
a
limitation
of
this
study.
Only
17
subjects
were
common
to
all
3
sampling
years.
In
addition,
the
medical
surveillance
program
is
a
voluntary
one.
The
participation
rate
of
eligible
production
employees
decreased
from
approximately
70%
in
1993
to
50%
in
1997.
Also,
the
laboratory
reference
range
changed
substantially
for
ALT
in
1997.
Finally,
different
analytical
methods
were
used
to
measure
serum
PFOA.
Serum
PFOA
was
determined
by
electrospray
high­
performance
liquid
chromatography/
mass
spectrometry
in
1997,
but
by
thermospray
in
1993
and
1995.

An
earlier
medical
surveillance
study
on
workers
who
were
employed
in
the
1980'
s
was
conducted
at
the
Cottage
Grove
plant;
however,
total
serum
fluorine
was
measured
instead
of
PFOA
(
Gilliland
and
Mandel,
1996).
Based
on
animal
studies
that
reported
that
animals
exposed
to
PFOA
develop
hepatomegaly
and
alterations
in
lipid
metabolism,
a
cross­
sectional,
occupational
study
was
performed
to
determine
if
similar
effects
are
present
in
workers
exposed
to
PFOA.
In
a
PFOA
production
facility,
115
workers
were
studied
to
determine
whether
serum
PFOA
affected
their
cholesterol,
lipoproteins,
and
hepatic
enzymes.
Forty­
eight
workers
who
were
exposed
to
PFOA
from
1985­
1989
were
included
in
the
study
(
96%
participation
rate).
Sixty­
five
employees
who
either
volunteered
or
were
asked
to
participate,
were
included
in
the
unexposed
group.
These
employees
were
assumed
to
have
little
or
no
PFOA
exposure
based
on
their
job
description.
However,
when
serum
levels
were
analyzed,
it
was
noted
that
this
group
of
workers
had
PFOA
levels
much
greater
than
the
general
population.
Therefore,
instead
of
job
categories,
total
serum
fluorine
was
used
to
classify
workers
into
exposure
groups.

Total
serum
fluorine
was
used
as
a
surrogate
measure
for
PFOA.
Serum
PFOA
was
not
measured,
due
to
the
cost
of
analyzing
the
samples.
Blood
samples
were
analyzed
for
total
serum
fluorine,
serum
glutamyl
oxaloacetic
transaminase
(
SGOT
or
AST),
serum
glutamyl
pyruvic
transaminase
(
SGPT
or
ALT),
gamma
glutamyl
transferase
(
GGT),
cholesterol,
low­
density
lipoproteins
(
LDL),
and
high­
density
lipoproteins
(
HDL).
All
of
the
participants
were
placed
into
five
categories
of
total
serum
fluorine
levels:
<
1
ppm,
1­
3
ppm,
>
3
­
10
ppm,
>
10
­
15
ppm,
and
>
15
ppm.
The
range
of
the
serum
fluorine
values
was
0
to
26
ppm
(
mean
3.3
ppm).
27
Approximately
half
of
the
workers
fell
into
the
>
1
­
3
ppm
category,
while
23
had
serum
levels
<
1
ppm
and
11
had
levels
>
10
ppm.

There
were
no
significant
differences
between
exposure
categories
when
analyzed
using
univariate
analyses
for
cholesterol,
LDL,
and
HDL.
In
the
multivariate
analysis,
there
was
not
a
significant
association
between
total
serum
fluorine
and
cholesterol
or
LDL
after
adjusting
for
alcohol
consumption,
age,
BMI,
and
cigarette
smoking.
There
were
no
statistically
significant
differences
among
the
exposure
categories
of
total
serum
fluorine
for
AST,
ALT
and
GGT.
However,
increases
in
AST
and
ALT
occurred
with
increasing
total
serum
fluorine
levels
in
obese
workers
(
BMI
=
35
kg/
m2).
This
result
was
not
observed
when
PFOA
was
measured
directly
in
serum
of
workers
in
1993,
1995,
or
1997
surveillance
data
of
employees
of
the
Cottage
Grove
plant
(
Olsen,
et
al.,
2000).

Since
PFOA
was
not
measured
directly
and
there
is
no
exposure
information
provided
on
the
employees
(
e.
g.
length
of
employment/
exposure),
the
results
of
the
study
provide
limited
information.
The
authors
state
that
no
adverse
clinical
outcomes
related
to
PFOA
exposure
have
been
observed
in
these
employees;
however,
it
is
not
clear
that
there
has
been
follow­
up
of
former
employees.
In
addition,
the
range
of
results
reported
for
the
liver
enzymes
were
fairly
wide
for
many
of
the
exposure
categories,
indicating
variability
in
the
results.
Given
that
only
one
sample
was
taken
from
each
employee,
this
is
not
surprising.
It
would
be
much
more
helpful
to
have
several
samples
taken
over
time
to
ensure
their
reliability.
It
also
would
have
been
interesting
to
compare
the
results
of
the
workers
who
were
known
to
be
exposed
to
PFOA
to
those
who
were
originally
thought
not
to
be
exposed
to
see
if
there
were
any
differences
among
the
employees
in
these
groups.
There
were
more
of
the
"
unexposed"
employees
(
n
=
65)
participating
in
the
study
than
those
who
worked
in
PFOA
production
(
n
=
48).

3.4
Prenatal
Developmental
Toxicity
Studies
in
Animals
Several
prenatal
developmental
toxicity
studies
of
APFO
have
been
conducted.
These
include
two
oral
studies
in
rats,
one
oral
study
in
rabbits,
and
one
inhalation
study
in
rats.

Gortner
(
1981)
administered
time­
mated
Sprague­
Dawley
rats
(
22
per
group)
doses
of
0,
0.05,
1.5,
5,
and
150
mg/
kg/
day
APFO
in
distilled
water
by
gavage
on
gestation
days
(
GD)
6­
15.
Doses
were
adjusted
according
to
body
weight.
Dams
were
monitored
on
GD
3­
20
for
clinical
signs
of
toxicity.
Individual
body
weights
were
recorded
on
GD
3,
6,
9,
12,
15,
and
20.
Animals
were
sacrificed
on
GD
20
by
cervical
dislocation
and
the
ovaries,
uteri,
and
contents
were
examined
for
the
number
of
corpora
lutea,
number
of
viable
and
non­
viable
fetuses,
number
of
resorption
sites,
and
number
of
implantation
sites.
Fetuses
were
weighed
and
sexed
and
subjected
to
external
gross
necropsy.
Approximately
one­
third
of
the
fetuses
were
fixed
in
Bouin's
solution
and
examined
for
visceral
abnormalities
by
free­
hand
sectioning.
The
remaining
fetuses
were
subjected
to
skeletal
examination
using
alizarin
red.

Signs
of
maternal
toxicity
consisted
of
statistically
significant
reductions
in
mean
maternal
body
weights
on
GD
9,
12,
and
15
at
the
high­
dose
group
of
150
mg/
kg/
day.
Mean
maternal
body
weight
on
GD
20
continued
to
remain
lower
than
controls,
although
the
difference
was
not
28
statistically
significant.
Other
signs
of
maternal
toxicity
that
occurred
only
at
the
high­
dose
group
included
ataxia
and
death
in
three
rat
dams.
No
other
effects
were
reported.
Administration
of
APFO
during
gestation
did
not
appear
to
affect
the
ovaries
or
reproductive
tract
of
the
dams.
Under
the
conditions
of
the
study,
a
NOAEL
of
5
mg/
kg/
day
and
a
LOAEL
of
150
mg/
kg/
day
for
maternal
toxicity
were
indicated.

A
significantly
higher
incidence
in
fetuses
with
one
missing
sternebrae
was
observed
at
the
highdose
group
of
150
mg/
kg/
day;
however
this
skeletal
variation
also
occurred
in
the
controls
and
the
other
three
dose
groups
(
at
similar
incidence
but
lower
than
the
high­
dose
group)
and
therefore
was
not
considered
to
be
treatment­
related.
No
significant
differences
between
treated
and
control
groups
were
noted
for
other
developmental
parameters
that
included
the
mean
number
of
males
and
females,
total
and
dead
fetuses,
the
mean
number
of
resorption
sites,
implantation
sites,
corpora
lutea
and
mean
fetus
weights.
Likewise,
a
fetal
lens
finding
initially
described
as
a
variety
of
abnormal
morphological
changes
localized
to
the
area
of
the
embryonal
nucleus,
was
later
determined
to
be
an
artifact
of
the
free­
hand
sectioning
technique
and
therefore
not
considered
to
be
treatment­
related.
Under
the
conditions
of
the
study,
a
NOAEL
for
developmental
toxicity
of
150
mg/
kg/
day
(
highest
dose
group)
was
indicated.

A
second
oral
prenatal
developmental
toxicity
study
was
conducted
in
rabbits
(
Gortner,
1982).
Based
on
the
results
of
a
range­
finding
study,
an
upper
dose
level
of
50
mg/
kg/
day
was
set
for
the
definitive
study
in
which
four
groups
of
18
pregnant
New
Zealand
White
rabbits
were
administered
0,
1.5,
5,
and
50
mg/
kg/
day
APFO
in
distilled
water
by
gavage
on
gestation
days
6­
18.
Pregnancy
was
established
in
each
sexually
mature
female
by
iv
injection
of
pituitary
lutenizing
hormone
in
order
to
induce
ovulation,
followed
by
artificial
insemination
with
0.5
ml
of
pooled
semen
collected
from
male
rabbits;
the
day
of
insemination
was
designated
as
day
0
of
gestation.
A
constant
dose
volume
of
1
ml/
kg
was
administered.
Individual
body
weights
were
measured
on
GD
3,
6,
9,
12,
15,
18,
and
29.
The
does
were
observed
daily
on
GD
3­
29
for
abnormal
clinical
signs.
On
GD
29,
the
does
were
euthanized
and
the
ovaries,
uterus
and
contents
examined
for
the
number
of
corpora
lutea,
live
and
dead
fetuses,
resorptions
and
implantation
sites.
Fetuses
were
examined
for
gross
abnormalities
and
placed
in
a
370
C
incubator
for
a
24­
hour
survival
check.
Pups
were
subsequently
euthanized
and
examined
for
visceral
and
skeletal
abnormalities.

Signs
of
maternal
toxicity
consisted
of
statistically
significant
transient
reductions
in
body
weight
gain
on
GD
6­
9
when
compared
to
controls;
body
weight
gains
returned
to
control
levels
on
GD
12­
29.
Administration
of
APFO
during
gestation
did
not
appear
to
affect
the
ovaries
or
reproductive
tract
contents
of
the
does.
No
clinical
or
other
treatment­
related
signs
were
reported.
Under
the
conditions
of
the
study,
a
NOAEL
of
50
mg/
kg/
day,
the
highest
dose
tested,
for
maternal
toxicity
was
indicated.

No
significant
differences
were
noted
between
controls
and
treated
groups
for
the
number
of
males
and
females,
dead
or
live
fetuses,
and
fetal
weights.
Likewise,
there
were
no
significant
differences
reported
for
the
number
of
resorption
and
implantation
sites,
corpora
lutea,
the
conception
incidence,
abortion
rate,
or
the
24­
hour
mortality
incidence
of
the
fetuses.
Gross
necropsy
and
skeletal/
visceral
examinations
were
unremarkable.
The
only
sign
of
developmental
29
toxicity
consisted
of
a
dose­
related
increase
in
a
skeletal
variation,
extra
ribs
or
13th
rib,
with
statistical
significance
at
the
high­
dose
group
(
38%
at
50
mg/
kg/
day,
30%
at
5
mg/
kg/
day,
20%
at
1.5
mg/
kg/
day,
and
16
%
at
0
mg/
kg/
day).
A
statistically
significant
increase
in
13th
ribsspurred
occurred
in
the
mid­
dose
group
of
5
mg/
kg/
day;
however,
the
biological
significance
of
this
effect
is
uncertain
since
in
both
the
high­
and
low­
dose
groups,
this
effect
occurred
at
the
same
rate
and
was
not
statistically
significantly
different
from
controls.
Therefore,
under
the
conditions
of
the
study,
a
LOAEL
for
developmental
toxicity
of
50
mg/
kg/
day
(
highest
dose
group)
was
indicated.

Staples
et
al.
(
1984)
also
conducted
a
developmental
toxicity
study
of
APFO.
The
study
design
consisted
of
an
inhalation
and
an
oral
portion,
each
with
two
trials
or
experiments.
In
the
first
trial
the
dams
were
sacrificed
on
GD
21;
while
in
the
second
trial,
the
dams
were
allowed
to
litter
and
the
pups
were
sacrificed
on
day
35­
post
partum.
For
the
inhalation
portion
of
the
study,
the
two
trials
consisted
of
12
pregnant
Sprague­
Dawley
rats
per
group
exposed
to
0,
0.1,
1,
10,
and
25
mg/
m3
APFO
for
6
hours/
day,
on
GD
6­
15.
In
the
oral
portion
of
the
study,
25
and
12
Sprague­
Dawley
rats
for
the
first
and
second
trials,
respectively,
were
administered
0
and
100
mg/
kg/
day
APFO
in
corn
oil
by
gavage
on
GD
6­
15.
For
both
routes
of
administration,
females
were
mated
on
an
as­
needed
basis
and
when
the
number
of
mated
females
was
bred,
they
were
ranked
within
breeding
days
by
body
weight
and
assigned
to
groups
by
rotation
in
order
of
rank.
Finally,
two
additional
groups
(
six
dams
per
group)
were
added
to
each
trial
that
was
pair­
fed
to
the
10
and
25
mg/
m3
groups.

For
trial
one,
the
dams
were
weighed
on
GD
1,
6,
9,
13,
16,
and
21
and
observed
daily
for
abnormal
clinical
signs.
On
GD
21,
the
dams
were
sacrificed
by
cervical
dislocation
and
examined
for
any
gross
abnormalities,
liver
weights
were
recorded
and
the
reproductive
status
of
each
animal
was
evaluated.
The
ovaries,
uterus
and
contents
were
examined
for
the
number
of
corpora
lutea,
live
and
dead
fetuses,
resorptions
and
implantation
sites.
Pups
(
live
and
dead)
were
counted,
weighed
and
sexed
and
examined
for
external,
visceral,
and
skeletal
alterations.
The
heads
of
all
control
and
high­
dosed
group
fetuses
were
examined
for
visceral
alterations
as
well
as
macro­
and
microscopic
evaluation
of
the
eyes.

For
trial
two,
in
which
the
dams
were
allowed
to
litter,
the
procedure
was
the
same
as
that
for
trial
one
up
to
GD
21.
Two
days
before
the
expected
day
of
parturition,
each
dam
was
housed
in
an
individual
cage.
The
date
of
parturition
was
noted
and
designated
Day
1
PP.
Dams
were
weighed
and
examined
for
clinical
signs
on
Days
1,
7,
14,
and
22
PP.
On
Day
23
PP
all
dams
were
sacrificed.
Pups
were
counted,
weighed,
and
examined
for
external
alterations.
Each
pup
was
subsequently
weighed
and
inspected
for
adverse
clinical
signs
on
Days
4,
7,
14,
and
22
PP.
The
eyes
of
the
pups
were
also
examined
on
Days
15
and
17
PP
for
the
inhalation
portion
and
on
Days
27
and
31
PP
for
the
gavage
portion
of
the
study.
Pups
were
sacrificed
on
Day
35
PP
and
examined
for
visceral
and
skeletal
alterations.

In
trial
one
of
the
inhalation
study,
treatment­
related
clinical
signs
of
maternal
toxicity
occurred
at
10
and
25
mg/
m3
and
consisted
of
wet
abdomens,
chromodacryorrhea,
chromorhinorrhea,
a
general
unkempt
appearance,
and
lethargy
in
four
dams
at
the
end
of
the
exposure
period
(
highconcentration
group
only).
Three
out
of
12
dams
died
during
treatment
at
25
mg/
m3
(
on
GD
12,
30
13,
and
17).
Food
consumption
was
significantly
reduced
at
both
10
and
25
mg/
m3;
however,
no
significant
differences
were
noted
between
treated
and
pair­
fed
groups.
Significant
reductions
in
body
weight
were
also
observed
at
these
concentrations,
with
statistical
significance
at
the
highconcentration
only.
Likewise,
statistically
significant
increases
in
mean
liver
weights
were
seen
at
the
high­
concentration
group.
Under
the
conditions
of
the
study,
a
NOAEL
and
LOAEL
for
maternal
toxicity
of
1
and
10
mg/
m3,
respectively,
were
indicated.

No
effects
were
observed
on
the
maintenance
of
pregnancy
or
the
incidence
of
resorptions.
Mean
fetal
body
weights
were
significantly
decreased
in
the
25­
mg/
m3
groups
and
in
the
control
group
pair­
fed
25
mg/
m3.
However,
interpretation
of
the
decreased
fetal
body
weight
is
difficult
given
the
high
incidence
of
mortality
in
the
dams.
Under
EPA
guidance,
data
at
doses
exceeding
10%
mortality
are
generally
discounted.
Under
the
conditions
of
the
study,
a
NOAEL
and
LOAEL
for
developmental
toxicity
of
10
and
25
mg/
m3,
respectively,
were
indicated.

In
trial
two
of
the
inhalation
study,
clinical
signs
of
maternal
toxicity
seen
at
10
and
25
mg/
m3
were
similar
in
type
and
incidence
to
those
described
for
trial
one.
Maternal
body
weight
gain
during
treatment
at
25
mg/
m3
was
less
than
controls,
although
the
difference
was
not
statistically
significant.
In
addition,
2
out
of
12
dams
died
during
treatment
at
25
mg/
m3.
No
other
treatmentrelated
effects
were
reported,
nor
were
any
adverse
effects
noted
for
any
of
the
measurements
of
reproductive
performance.
Under
the
conditions
of
the
study,
a
NOAEL
and
LOAEL
for
maternal
toxicity
of
1
and
10
mg/
m3,
respectively,
were
indicated.

Signs
of
developmental
toxicity
in
this
group
consisted
of
statistically
significant
reductions
in
pup
body
weight
on
Day
1
PP
(
6.1
g
at
25
mg/
m3
vs.
6.8
g
in
controls).
On
Days
4
and
22
PP,
pup
body
weights
continued
to
remain
lower
than
controls,
although
the
difference
was
not
statistically
significant
(
Day
4
PP:
9.7
g
at
25
mg/
m3
vs.
10.3
in
controls;
Day
22
PP:
49.0
g
at
25
mg/
m3
vs.
50.1
in
controls).
No
significant
effects
were
reported
following
external
examination
of
the
pups
or
with
ophthalmoscopic
examination
of
the
eyes.
Again,
interpretation
of
these
effects
is
problematic
given
the
high
incidence
of
maternal
mortality.
Under
the
conditions
of
the
study,
a
NOAEL
and
LOAEL
for
developmental
toxicity
of
10
and
25
mg/
m3,
respectively,
were
indicated.

In
trial
one
of
the
oral
study,
three
out
of
25
dams
died
during
treatment
of
100
mg/
kg
APFO
during
gestation
(
one
death
on
GD
11;
two
on
GD
12).
Clinical
signs
of
maternal
toxicity
in
the
dams
that
died
were
similar
to
those
seen
with
inhalation
exposure.
Food
consumption
and
body
weights
were
reduced
in
treated
animals
compared
to
controls.
No
adverse
signs
of
toxicity
were
noted
for
any
of
the
reproductive
parameters
such
as
maintenance
of
pregnancy
or
incidence
of
resorptions.
Likewise,
no
significant
differences
between
treated
and
control
groups
were
noted
for
fetal
weights,
or
in
the
incidences
of
malformations
and
variations;
nor
were
there
any
effects
noted
following
microscopic
examination
of
the
eyes.

In
trial
two
of
the
oral
study,
similar
observations
for
clinical
signs
were
noted
for
the
dams
as
in
trial
one.
Likewise,
no
adverse
effects
on
reproductive
performance
or
in
any
of
the
fetal
observations
were
noted.
31
3.5
Reproductive
Toxicity
Studies
in
Animals
York
(
2002)
conducted
an
oral
two­
generation
reproductive
toxicity
study
of
APFO,
which
is
summarized
below.
Although
this
preliminary
risk
assessment
focuses
on
developmental
toxicity,
the
summary
below
of
the
two
generation
reproductive
toxicity
study
includes
all
endpoints.

Five
groups
of
30
Sprague­
Dawley
rats
per
sex
per
dose
group
were
administered
APFO
by
gavage
at
doses
of
0,
1,
3,
10,
and
30
mg/
kg/
day
six
weeks
prior
to
and
during
mating.
Treatment
of
the
F0
male
rats
continued
until
mating
was
confirmed,
and
treatment
of
the
F0
female
rats
continued
throughout
gestation,
parturition,
and
lactation.

The
F0
animals
were
examined
twice
daily
for
clinical
signs,
abortions,
premature
deliveries,
and
deaths.
Body
weights
of
F0
male
rats
were
recorded
weekly
during
the
dosage
period
and
then
on
the
day
of
sacrifice.
Body
weights
of
F0
female
rats
were
recorded
weekly
during
the
pre­
and
cohabitation
periods
and
then
on
gestation
days
(
GD)
0,
7,
10,
14,
18,
21,
and
25
(
if
necessary)
and
on
lactation
days
(
LD)
1,
5,
8,
11,
15,
and
22
(
terminal
body
weight).
Food
consumption
values
in
F0
male
rats
were
recorded
weekly
during
the
treatment
period,
while
in
F0
female
rats,
values
were
recorded
weekly
during
the
precohabitation
period,
on
GDs
0,
7,
10,
14,
18,
21,
and
25
and
on
LDs
1,
5,
8,
11,
and
15.

Estrous
cycling
was
evaluated
daily
by
examination
of
vaginal
cytology
beginning
21
days
before
the
scheduled
cohabitation
period
and
continuing
until
confirmation
of
mating
by
the
presence
of
sperm
in
a
vaginal
smear
or
confirmation
of
a
copulatory
plug.
On
the
day
of
scheduled
sacrifice,
the
stage
of
the
estrous
cycle
was
assessed.

For
mating,
one
male
rat
and
one
female
rat
per
group
were
cohabitated
for
a
maximum
of
14
days.
Female
rats
with
evidence
of
sperm
in
a
vaginal
smear
or
copulatory
plug
were
designated
as
GD
0
and
assigned
to
individual
housing.
Parental
females
were
evaluated
for
length
of
gestation,
fertility
index,
gestation
index,
number
and
sex
of
offspring
per
litter,
number
of
implantation
sites,
general
condition
of
the
dam
and
litter
during
the
postpartum
period,
litter
size
and
viability,
viability
index,
lactation
index,
percent
survival,
and
sex
ratio.
Maternal
behavior
of
the
dams
was
recorded
on
LDs
1,
5,
8,
15,
and
22.

F0
generation
animals
were
sacrificed
by
carbon
dioxide
asphyxiation
(
day
106
to
110
of
the
study
for
male
rats,
i.
e.,
after
completion
of
the
cohabitation
period;
and
LD
22
for
female
rats),
necropsied,
and
examined
for
gross
lesions.
Gross
necropsy
included
examination
of
external
surfaces
and
orifices,
as
well
as
internal
examination
of
tissues
and
organs.
Individual
organs
were
weighed
and
organ­
to­
body
weight
and
organ­
to­
brain
weight
ratios
were
calculated
for
the
brain,
kidneys,
spleen,
ovaries,
testes,
thymus,
liver,
adrenal
glands,
pituitary,
uterus
with
oviducts
and
cervix,
left
epididymis
(
whole
and
cauda),
right
epididymis,
prostate
and
seminal
vesicles,
(
with
coagulating
glands
and
with
and
without
fluid).
Tissues
retained
in
neutral
buffered
10%
formalin
for
possible
histological
evaluation
included
the
pituitary,
adrenal
glands,
vagina,
uterus,
with
oviducts,
cervix
and
ovaries,
right
testis,
seminal
vesicles,
right
epididymis,
and
prostate.
Histological
examination
was
performed
on
tissues
from
10
randomly
selected
rats
32
per
sex
from
the
control
and
high
dosage
groups.
All
gross
lesions
were
examined
histologically.
All
F0
generation
rats
that
died
or
appeared
moribund
were
also
examined.

Histological
examination
of
the
reproductive
organs
in
the
low­
and
mid­
dose
groups
was
conducted
in
rats
that
exhibited
reduced
fertility
by
either
failing
to
mate,
conceive,
sire,
or
deliver
healthy
offspring;
or
for
which
estrous
cyclicity
or
sperm
number,
motility,
or
morphology
were
altered.
Sperm
number,
motility,
and
morphology
were
evaluated
in
the
left
cauda
epididymis
of
F0
generation
male
rats;
testicular
spermatid
concentrations
were
evaluated
in
the
left
testis.
The
number
and
distribution
of
implantation
sites
were
recorded
in
F0
generation
female
rats.
Rats
that
did
not
deliver
a
litter
were
sacrificed
on
GD
25
and
examined
for
pregnancy
status.
Uteri
of
apparently
nonpregnant
rats
were
examined
to
confirm
the
absence
of
implantation
sites.
A
gross
necropsy
of
the
thoracic,
abdominal
and
pelvic
viscera
was
performed.
Female
rats
without
a
confirmed
mating
date
that
did
not
deliver
a
litter
were
sacrificed
on
an
estimated
day
25
of
gestation.

At
scheduled
sacrifice,
after
completion
of
the
cohabitation
period
in
F0
male
rats
and
on
LD
22
in
F0
female
rats,
blood
samples
(
10
males
and
10
females
each
for
the
10
and
30
mg/
kg/
day
dose
groups;
3
males
and
3
females
for
the
control
group)
were
collected
and
frozen
for
future
analysis.
The
methods
section
cites
that
liver
samples
were
also
collected,
but
no
other
details
were
provided
and
the
results
did
not
appear
to
be
available
at
the
time
of
the
report.

The
F1
generation
pups
in
each
litter
were
counted
once
daily.
The
litter
sizes
were
not
standardized
on
day
4.
Physical
signs
(
including
variations
from
expected
lactation
behavior
and
gross
external
physical
anomalies)
were
recorded
for
the
pups
each
day.
Pup
body
weights
were
recorded
on
LDs
1,
5,
8,
15
and
22.
On
LD
12,
all
F1
generation
male
pups
were
examined
for
the
presence
of
nipples.
Pups
that
died
before
examination
of
the
litter
for
pup
viability
on
LD
1
were
evaluated
for
vital
status
at
birth.
Pups
found
dead
on
LDs
2
to
22
were
examined
for
gross
lesions
and
for
the
cause
of
death.
All
F1
generation
rats
were
weaned
on
LD
22
based
on
observed
growth
and
viability
of
these
pups.

At
weaning
(
LD
22),
two
F1
generation
pups
per
sex
per
litter
per
group
(
60
male
and
60
female
pups
per
group)
were
selected
for
continued
evaluation,
resulting
in
600
total
rats
(
300
rats
per
sex)
assigned
to
the
five
dosage
groups.
At
least
two
male
pups
and
two
female
pups
per
litter,
when
possible,
were
selected.
F1
generation
pups
not
selected
for
continued
observation
for
sexual
maturation
were
sacrificed.
Three
pups
per
sex
per
litter
were
examined
for
gross
lesions.
Necropsy
included
a
single
cross­
section
of
the
head
at
the
level
of
the
frontal­
parietal
suture
and
examination
of
the
cross­
sectioned
brain
for
apparent
hydrocephaly.
The
brain,
spleen
and
thymus
from
one
of
the
three
selected
pups
per
sex
per
litter
were
weighed
and
the
brain,
spleen,
and
thymus
from
the
three
selected
pups
per
sex
per
litter
were
retained
for
possible
histological
evaluation.
All
remaining
pups
were
discarded
without
further
examination.

The
F1
generation
rats
were
given
the
same
dosage
level
of
the
test
substance
and
in
the
same
manner
as
their
respective
F0
generation
sires
and
dams.
Dosages
were
given
once
daily,
beginning
at
weaning
and
continuing
until
the
day
before
sacrifice.
F1
generation
female
rats
were
examined
for
age
of
vaginal
patency,
beginning
on
day
28
postpartum
(
LD
28).
F1
33
generation
male
rats
were
evaluated
for
age
of
preputial
separation,
beginning
on
day
39
postpartum
(
LD
39).
Body
weights
were
recorded
when
rats
reached
sexual
maturation.

Following
sexual
maturation,
a
table
of
random
units
was
used
to
select
one
male
and
one
female
per
litter
per
group
for
continuation
through
mating
to
produce
the
F2
generation.
The
remaining
F1
animals
were
sacrificed.

Estrous
cycling
was
evaluated
daily
by
examination
of
vaginal
cytology
beginning
21
days
before
the
scheduled
cohabitation
period
and
continuing
until
confirmation
of
mating
by
the
presence
of
sperm
in
a
vaginal
smear
or
confirmation
of
a
copulatory
plug.
On
the
day
of
scheduled
sacrifice,
the
stage
of
the
estrous
cycle
was
assessed.

A
table
of
random
units
was
used
to
assign
F1
generation
rats
to
cohabitation,
one
male
rat
per
female
rat.
If
random
assignment
to
cohabitation
resulted
in
the
pairing
of
F1
generation
siblings,
an
alternate
assignment
was
made.
The
cohabitation
period
consisted
of
a
maximum
of
14
days.

Body
weights
of
the
F1
generation
male
rats
were
recorded
weekly
during
the
postweaning
period
and
on
the
day
of
sacrifice.
Body
weights
of
the
F1
generation
female
rats
were
recorded
weekly
during
the
postweaning
period
to
cohabitation,
and
on
GDs
0,
7,
10,
14,
18,
21
and
25
(
if
necessary)
and
on
LDs
1,
5,
8,
11,
15
and
22.
Food
consumption
values
for
the
F1
generation
male
rats
were
recorded
weekly
during
the
dosage
period.
Food
consumption
values
for
the
F1
generation
female
rats
were
recorded
weekly
during
the
postweaning
period
to
cohabitation,
on
GDs
0,
7,
10,
14,
18,
21
and
25
and
on
LDs
1,
5,
8,
11
and
15.
Because
pups
begin
to
consume
maternal
food
on
or
about
LD
15,
food
consumption
values
were
not
tabulated
after
LD
15.

At
scheduled
sacrifice,
the
F1
animals
were
subjected
to
gross
necropsy,
and
selected
organs
were
weighed
and
examined
histologically
as
described
above
for
the
F0
animals.
Sperm
analyses
were
also
conducted
as
described
for
the
F0
animals.

F2
generation
litters
were
examined
after
delivery
to
identify
the
number
and
sex
of
pups,
stillbirths,
live
births
and
gross
alterations.
Each
litter
was
evaluated
for
viability
at
least
twice
each
day
of
the
22­
day
postpartum
period.
Dead
pups
observed
at
these
times
were
removed
from
the
nesting
box.
Anogenital
distance
was
measured
for
all
live
F2
generation
pups
on
LDs
1
and
22.

Parental
Males
(
F0)

One
F0
male
rat
in
the
30
mg/
kg/
day
dose
group
was
sacrificed
on
day
45
of
the
study
due
to
adverse
clinical
signs
(
emaciation,
cold­
to­
touch,
and
decreased
motor
activity).
Necroscopic
examination
in
that
animal
revealed
a
pale
and
tan
liver,
and
red
testes.
All
other
F0
generation
male
rats
survived
to
scheduled
sacrifice.
Statistically
significant
increases
in
clinical
signs
were
also
observed
in
male
rats
in
the
high­
dose
group
that
included
dehydration,
urine­
stained
abdominal
fur,
and
ungroomed
coat.
34
Significant
reductions
in
body
weight
and
body
weight
gain
were
reported
for
most
of
the
dosage
period
and
continuing
until
termination
of
the
study
in
the
3,
10,
and
30
mg/
kg/
day
dose
groups.
Absolute
food
consumption
values
were
also
significantly
reduced
during
these
periods
at
the
30
mg/
kg/
day
dose
group,
while
significant
increases
in
relative
food
consumption
values
were
observed
in
the
3,
10,
and
30
mg/
kg/
day
within
those
same
periods.

No
treatment­
related
effects
were
reported
at
any
dose
level
for
any
of
the
mating
and
fertility
parameters
assessed,
including
numbers
of
days
to
inseminate,
numbers
of
rats
that
mated,
fertility
index,
numbers
of
rats
with
confirmed
mating
dates
during
the
first
and
second
week
of
cohabitation,
and
numbers
of
pregnant
rats
per
rats
in
cohabitation.
At
necropsy,
none
of
the
sperm
parameters
evaluated
(
sperm
number,
motility,
or
morphology)
were
affected
by
treatment
at
any
dose
level.

At
necropsy,
statistically
significant
reductions
in
terminal
body
weights
were
seen
at
3,
10,
and
30
mg/
kg/
day.
Absolute
weights
of
the
left
and
right
epididymides,
left
cauda
epididymis,
seminal
vesicles
(
with
and
without
fluid),
prostate,
pituitary,
left
and
right
adrenals,
spleen,
and
thymus
were
also
significantly
reduced
at
30
mg/
kg/
day.
The
absolute
weight
of
the
seminal
vesicles
without
fluid
was
significantly
reduced
in
the
10
mg/
kg/
day
dose
group.
The
absolute
weight
of
the
liver
was
significantly
increased
in
all
dose­
groups.
Kidney
weights
were
significantly
increased
in
the
1,
3,
and
10
mg/
kg/
day
dose
groups,
but
significantly
decreased
in
the
30
mg/
kg/
day
group.
All
organ
weight­
to­
terminal
body
weight
and
ratios
were
significantly
increased
in
all
treated
groups.
Organ
weight­
to­
brain
weight
ratios
were
significantly
reduced
for
some
organs
at
the
high
dose
group,
and
significantly
increased
for
other
organs
among
all
treated
groups.

No
treatment­
related
effects
were
seen
at
necropsy
or
upon
microscopic
examination
of
the
reproductive
organs,
with
the
exception
of
increased
thickness
and
prominence
of
the
zona
glomerulosa
and
vacuolation
of
the
cells
of
the
adrenal
cortex
in
the
10
and
30
mg/
kg/
day
dose
groups.

Serum
analysis
for
the
F0
generation
males
sampled
at
the
end
of
cohabitation
showed
that
PFOA
was
present
in
all
samples
tested,
including
controls.
Control
males
had
an
average
concentration
of
0.0344+
0.0148
ppm
PFOA.
Levels
of
PFOA
found
in
male
sera
remained
the
same
between
the
two
dose
groups;
treated
males
had
51.1+
9.30
and
45.3+
12.6
ppm,
respectively
for
the
10
and
30
mg/
kg/
day
dose
groups.

Parental
Females
(
F0)

No
treatment­
related
deaths
or
adverse
clinical
signs
were
reported
in
parental
females
at
any
dose
level.
No
treatment­
related
effects
were
reported
for
body
weights,
body
weight
gains,
and
absolute
and
relative
food
consumption
values.

There
were
no
treatment­
related
effects
on
estrous
cyclicity,
mating
or
fertility
parameters.
None
of
the
natural
delivery
and
litter
observations
were
affected
by
treatment,
that
is,
the
numbers
of
dams
delivering
litters,
the
duration
of
gestation,
the
averages
for
implantation
sites
per
35
delivered
litter,
the
gestation
index
(
number
of
dams
with
one
or
more
liveborn
pups/
number
of
pregnant
rats),
the
numbers
of
dams
with
stillborn
pups,
dams
with
all
pups
dying,
liveborn
and
stillborn
pups
viability
index,
pup
sex
ratios,
and
mean
birth
weights
were
comparable
to
controls
among
all
treated
groups.

Necropsy
and
histopathological
evaluation
were
also
unremarkable.
Terminal
body
weights,
organ
weights,
and
organ­
to­
terminal
body
weight
ratios
were
comparable
to
control
values
for
all
treated
groups,
except
for
kidney
and
liver
weights.
The
weights
of
the
left
and
right
kidney,
and
the
ratios
of
these
organ
weights­
to­
terminal
body
weight
and
of
the
left
kidney
weight­
tobrain
weight
were
significantly
reduced
at
the
highest
dose
of
30
mg/
kg/
day.
The
ratio
of
liver
weights­
to­
terminal
body
weight
was
also
significantly
reduced
at
3
and
10
mg/
kg/
day.

Results
of
the
serum
analysis
in
F0
generation
females
sampled
on
LD
22
showed
that
PFOA
was
present
in
all
samples
tested,
except
in
controls
where
the
level
was
below
the
limits
of
quantitation
(
0.00528
ppm).
Levels
of
PFOA
found
in
female
sera
increased
between
the
two
dose
groups;
treated
females
had
an
average
concentration
of
0.37+
0.0805
and
1.02+
0.425
ppm,
respectively
for
the
10
and
30
mg/
kg/
day
dose
groups.

F1
Generation
No
effects
were
reported
at
any
dose
level
for
the
viability
and
lactation
indices.
No
differences
between
treated
and
control
groups
were
noted
for
the
numbers
of
pups
surviving
per
litter,
the
percentage
of
male
pups,
litter
size
and
average
pup
body
weight
per
litter
at
birth.
At
30
mg/
kg/
day,
one
pup
from
one
dam
died
prior
to
weaning
on
lactation
day
1
(
LD1).
Additionally,
on
lactation
days
6
and
8,
statistically
significant
increases
in
the
numbers
of
pups
found
dead
were
observed
at
3
and
30
mg/
kg/
day.
According
to
the
study
authors,
this
was
not
considered
to
be
treatment
related
because
they
did
not
occur
in
a
dose­
related
manner
and
did
not
appear
to
affect
any
other
measures
of
pup
viability
including
numbers
of
surviving
pups
per
litter
and
live
litter
size
at
weighing.
An
independent
statistical
analysis
was
conducted
by
US
EPA
(
2002b).
No
significant
differences
were
observed
between
dose
groups
and
the
response
did
not
have
any
trend
in
dose.

Pup
body
weight
on
a
per
litter
basis
(
sexes
combined)
was
reduced
throughout
lactation
in
the
30
mg/
kg/
day
group,
and
statistical
significance
was
achieved
on
days
1,
5,
and
8.
Of
the
pups
necropsied
at
weaning,
no
statistically
significant,
treatment­
related
differences
were
observed
for
the
weights
of
the
brain,
spleen
and
thymus
and
the
ratios
of
these
organ
weights
to
the
terminal
body
weight
and
brain
weight.

F1
Males
Significant
increases
in
treatment­
related
deaths
(
5/
60
animals)
were
reported
in
F1
males
in
the
high
dose
group
of
30
mg/
kg/
day
between
days
2­
4
postweaning.
One
rat
was
moribund
sacrificed
on
day
39
postweaning
and
another
was
found
dead
on
day
107
postweaning.
36
Statistically
significant
increases
in
clinical
signs
of
toxicity
were
also
observed
in
F1
males
during
most
of
entire
postweaning
period.
These
signs
included
an
increased
incidence
of
annular
constriction
of
the
tail
at
all
doses,
with
statistical
significance
at
the
1,
10,
and
30
mg/
kg/
day;
a
significant
increase
at
10
and
30
mg/
kg/
day
in
the
number
of
male
rats
that
were
emaciated;
and
a
significant
increase
in
the
incidence
of
urine­
stained
abdominal
fur,
decreased
motor
activity,
and
abdominal
distention
at
30
mg/
kg/
day.

Body
weights
and
body
weight
gains
were
statistically
significantly
reduced
prior
to
and
during
cohabitation
and
during
the
entire
dosing
period
in
all
treated
groups.
Statistically
significant
reductions
in
body
weights
were
observed
at
10
and
30
mg/
kg/
day
during
days
8­
15,
22­
29,
29­
36,
43­
50,
and
50­
57
postweaning.
Body
weight
gains
were
also
significantly
reduced
in
the
30
mg/
kg/
day
group
on
days
1­
8,
15­
22,
36­
43,
57­
64,
and
64­
70
postweaning.
Statistically
significant,
dose­
related
reductions
in
body
weight
gains
were
observed
for
the
entire
dosage
period
(
days
1­
113
postweaning).
Absolute
food
consumption
values
were
significantly
reduced
at
10
and
30
mg/
kg/
day
during
the
entire
precohabitation
period
(
days
1­
70
postweaning),
while
relative
food
consumption
values
were
significantly
increased.

Statistically
significant
(
p<
0.01)
delays
in
sexual
maturation
(
the
average
day
of
preputial
separation)
were
observed
in
high­
dose
animals
versus
concurrent
controls
(
52.2
days
of
age
versus
48.5
days
of
age,
respectively).

No
apparent
effects
were
observed
on
any
of
the
mating
or
fertility
parameters
including
fertility
and
pregnancy
indices
(
number
of
pregnancies
per
number
of
rats
that
mated
and
rats
in
cohabitation,
respectively),
the
number
of
days
to
inseminate,
the
number
of
rats
that
mated,
and
the
number
of
rats
with
confirmed
mating
dates
during
the
first
week.
No
statistically
significant,
treatment­
related
effects
were
observed
on
any
of
the
sperm
parameters
(
motility,
concentration,
or
morphology).

Necroscopic
examination
revealed
statistically
significant
treatment­
related
effects
at
3,
10,
and
30
mg/
kg/
day
ranging
from
tan
areas
in
the
lateral
and
median
lobes
of
the
liver
to
moderate
to
slight
dilation
of
the
pelvis
of
one
or
both
kidneys.

Statistically
significant,
dose­
related
decreases
in
terminal
body
weights
of
parental
F1
males
were
observed
in
the
1,3,10,
and
30
mg/
kg/
day
dose
groups.
The
absolute
weights
of
the
liver
were
significantly
increased
and
the
absolute
weights
of
the
spleen
were
significantly
decreased
at
all
treated
groups.
The
absolute
weights
of
the
left
and/
or
right
kidneys
were
significantly
increased
in
the
1
and
3
mg/
kg/
day
dose
groups
and
significantly
decreased
in
the
30
mg/
kg/
day
dose
group.
The
absolute
weight
of
the
thymus
was
also
significantly
decreased
in
the
10
and
30
mg/
kg/
day
dose
groups.
The
absolute
weight
of
the
prostate,
brain
and
left
adrenal
gland
were
significantly
decreased
in
the
30
mg/
kg/
day
dosage
group.
The
ratios
of
the
weights
of
the
seminal
vesicles,
with
and
without
fluid,
liver
and
left
and
right
kidneys
to
the
terminal
body
weights
were
significantly
increased
in
all
treated
groups.
The
ratios
of
the
weights
of
the
left
testis,
with
and
without
the
tunica
albuginea
and
the
right
testis
to
the
terminal
body
weight,
were
significantly
increased
at
3
mg/
kg/
day
and
higher.
The
ratios
of
the
weights
of
the
left
epididymis,
left
cauda
epididymis,
right
epididymis
and
brain
to
the
terminal
body
weight
were
37
significantly
increased
at
10
mg/
kg/
day
and
higher.
The
ratios
of
the
weight
of
the
seminal
vesicles
with
fluid
to
the
brain
weight
were
increased
at
1
mg/
kg/
day
and
higher,
with
statistical
significance
at
1
and
10
mg/
kg/
day.
The
ratios
of
the
liver
weight­
to­
brain
weight
were
significantly
increased
in
the
1
mg/
kg/
day
and
higher
dosage
groups,
and
the
ratios
of
the
left
and
right
kidney
weights­
to­
brain
weight
were
significantly
increased
in
all
treated
groups.
The
ratios
of
the
spleen
weight­
to­
brain
weight
were
significantly
decreased
at
1
mg/
kg/
day
and
higher,
and
the
ratios
of
the
thymus
weight­
to­
brain
weight
were
significantly
decreased
at
10
and
30
mg/
kg/
day.
The
ratios
of
the
left
and
right
testes
weight­
to­
brain
weight
were
increased
in
the
3
mg/
kg/
day
and
higher
dosage
groups.
These
ratios
were
significantly
increased
at
10
mg/
kg/
day
(
right
testis
only)
and
30
mg/
kg/
day.

Histopathologic
examination
of
the
reproductive
organs
was
unremarkable;
however,
treatmentrelated
microscopic
changes
were
observed
in
the
adrenal
glands
of
high­
dose
animals
(
cytoplasmic
hypertrophy
and
vacuolation
of
the
cells
of
the
adrenal
cortex)
and
in
the
liver
of
animals
treated
with
3,
10,
and
30
mg/
kg/
day
(
hepatocellular
hypertrophy).
No
other
treatmentrelated
effects
were
reported.

F1
Females
A
statistically
significant
increase
in
treatment­
related
mortality
(
6/
60
animals)
was
observed
in
F1
females
on
postweaning
days
2­
8
at
the
highest
dose
of
30
mg/
kg/
day.
No
adverse
clinical
signs
of
treatment­
related
toxicity
were
reported
for
any
dose
level
during
any
time
of
the
study
period.

Statistically
significant
decreases
in
body
weights
and
body
weight
gains
were
observed
in
highdose
animals
on
days
8,
15,
22,
29,
50,
and
57
postweaning,
during
precohabitation
(
recorded
on
the
day
cohabitation
began,
when
F1
generation
rats
were
92­
106
days
of
age),
and
during
gestation
and
lactation.
Statistically
significant
decreases
in
absolute
food
consumption
were
observed
during
days
1­
8,
8­
15
postweaning,
during
precohabitation
and
during
gestation
and
lactation
in
animals
treated
with
30
mg/
kg/
day.
Relative
food
consumption
values
were
comparable
across
all
treated
groups.

Statistically
significant
(
p<
0.01)
delays
in
sexual
maturation
(
the
average
day
of
vaginal
patency)
were
observed
in
high­
dose
animals
versus
concurrent
controls
(
36.6
days
of
age
versus
34.9
days
of
age,
respectively).

Prior
to
mating,
the
study
authors
noted
a
statistically
significant
increase
in
the
average
numbers
of
estrous
stages
per
21
days
in
high­
dose
animals
(
5.4
versus
4.7
in
controls).
For
this
calculation,
the
number
of
independent
occurrences
of
estrus
in
the
21
days
of
observation
was
determined.
This
type
of
calculation
can
be
used
as
a
screen
for
effects
on
the
estrous
cycle,
but
a
more
detailed
analysis
should
then
be
conducted
to
determine
whether
there
is
truly
an
effect.
3M
Company
(
2002)
recently
completed
an
analysis
that
showed
there
were
no
effects
on
the
estrous
cycle;
there
were
no
differences
in
the
number
of
females
with
>
3
days
of
estrus
or
with
>
4
days
of
diestrus
in
the
control
and
high
dose
groups.
Analyses
conducted
by
the
US
EPA
(
2002a)
also
demonstrated
that
there
were
no
differences
in
the
estrous
cycle
among
the
control
38
and
high
dose
groups.
The
cycles
were
evaluated
as
having
either
regular
4­
5
day
cycles,
uneven
cycling
(
defined
as
brief
periods
with
irregular
pattern)
or
periods
of
prolonged
diestrus
(
defined
as
4­
6
day
diestrus
periods)
extended
estrus
(
defined
as
3
or
4
days
of
cornified
smears),
possibly
pseudopregnant,
(
defined
as
6­
greater
days
of
leukocytes)
or
persistent
estrus
(
defined
as
5­
or
greater
days
of
cornified
smears).
The
two
groups
were
not
different
in
any
of
the
parameters
measured.
Thus,
the
increase
in
the
number
of
estrous
stages
per
21
days
that
was
noted
by
the
study
authors
is
due
to
the
way
in
which
the
calculation
was
done,
and
is
not
biologically
meaningful.

No
effects
on
any
of
the
mating
and
fertility
parameters
(
numbers
of
days
in
cohabitation,
numbers
of
rats
that
mated,
fertility
index,
rats
with
confirmed
mating
dates
during
the
first
week
of
cohabitation
and
number
of
rats
pregnant
per
rats
in
cohabitation).

All
natural
delivery
observations
were
unaffected
by
treatment
at
any
dose
level.
Numbers
of
dams
delivering
litters,
the
duration
of
gestation,
averages
for
implantation
sites
per
delivered
litter,
the
gestation
index
(
number
of
dams
with
one
or
more
liveborn
pups/
number
of
pregnant
rats),
the
numbers
of
dams
with
stillborn
pups,
dams
with
all
pups
dying
and
liveborn
and
stillborn
pups
were
comparable
among
treated
and
control
groups.

No
treatment­
related
effects
were
observed
on
terminal
body
weights.
The
absolute
weight
of
the
pituitary
and
the
ratios
of
the
pituitary
weight­
to­
terminal
body
weight
and
to
the
brain
weight
were
significantly
decreased
at
3
mg/
kg/
day
and
higher,
but
did
not
show
a
dose­
response.
No
other
differences
were
reported
for
the
absolute
weights
or
ratios
for
other
organs
evaluated.
No
treatment­
related
effects
were
reported
following
necroscopic
and
histopathologic
examinations.

F2
Generation
No
treatment­
related
adverse
clinical
signs
were
observed
at
any
dose
level.
Likewise,
no
treatment­
related
effects
were
reported
following
necroscopic
examination,
with
the
exception
of
no
milk
in
the
stomach
of
the
pups
that
were
found
dead.
The
total
number
of
pups
found
either
dead
or
stillborn
did
not
show
a
dose­
response
(
3/
28,
6/
28,
10/
28,
10/
28,
and
6/
28
in
0,
1,
3,
10,
and
30
mg/
kg/
day
dose
groups,
respectively;
ratios
refer
to
total
pups/
total
number
of
litters)
and
therefore
were
unlikely
related
to
treatment.

No
effects
were
reported
at
any
dose
level
for
the
viability
and
lactation
indices.
No
differences
between
treated
and
control
groups
were
noted
for
the
numbers
of
pups
surviving
per
litter,
the
percentage
of
male
pups,
litter
size
and
average
pup
body
weight
per
litter
when
measured
on
LDs
1,
5,
8,
15,
or
22.
Anogenital
distances
measured
for
F2
male
and
female
pups
on
LDs
1
and
22
were
also
comparable
among
the
five
dosage
groups
and
did
not
differ
significantly.

Statistically
significant
increases
(
p<
0.01)
in
the
number
of
pups
found
dead
were
observed
on
lactation
day
1
in
the
3
and
10
mg/
kg/
day
groups.
According
to
the
study
authors,
this
was
not
considered
to
be
treatment
related
because
they
did
not
occur
in
a
dose­
related
manner
and
did
not
appear
to
affect
any
other
measures
of
pup
viability
including
numbers
of
surviving
pups
per
litter
and
live
litter
size
at
weighing.
An
independent
statistical
analysis
was
conducted
by
US
39
EPA
(
2002b).
No
significant
differences
were
observed
between
dose
groups
and
the
response
did
not
have
any
trend
in
dose.

Terminal
body
weights
in
F2
pups
were
not
significantly
different
from
controls.
Absolute
weights
of
the
brain,
spleen
and
thymus
and
the
ratios
of
these
organ
weights­
to­
terminal
body
weight
and
to
brain
weight
were
also
comparable
among
treated
and
control
groups.

Conclusions
Dosing
with
APFO
at
30
mg/
kg/
day
appeared
to
delay
the
onset
of
sexual
maturation
in
both
male
and
female
F1
offspring.
The
authors
of
the
study
contend
that
the
delays
in
sexual
maturation
(
preputial
separation
or
vaginal
patency)
observed
in
high­
dose
animals
are
due
to
the
fact
that
these
animals
have
a
decreased
gestational
age,
a
variable
which
they
have
defined
as
the
time
in
days
from
evidence
of
mating
in
the
F0
generation
until
evidence
of
sexual
maturation
in
the
F1
generation.
The
authors
state
that
gestational
age
appeared
to
be
decreased
in
high­
dose
animals
at
the
time
of
acquisition
(
the
time
when
sexual
maturation
was
reached),
which
they
believe
meant
the
animals
in
that
group
were
younger
and
more
immature
than
the
control
group,
in
which
there
was
no
significant
difference
in
sexual
maturation.

In
order
to
test
this
hypothesis,
the
authors
covaried
separately
the
decreases
in
body
weight
and
in
gestational
age
with
the
delays
in
sexual
maturation
in
order
to
determine
whether
or
not
body
weights
and
gestational
age
were
a
contributing
factor.
When
the
body
weight
was
covaried
with
the
time
to
sexual
maturation,
the
time
to
sexual
maturation
showed
a
dose
related
delay
that
was
statistically
significant
at
the
p<
0.05.
This
suggests
that
the
delay
in
sexual
maturation
was
partly
related
to
body
weight,
but
not
entirely.
When
gestational
age
was
covaried
with
the
time
to
sexual
maturation,
there
was
no
significant
difference
in
the
time
of
onset
of
sexual
maturation
between
controls
and
high­
dose
animals.
This
indicates
that
the
effect
of
delayed
sexual
maturation
could
possibly
be
attributed
to
decreased
gestational
age.

While
it
is
known
and
commonly
accepted
that
changes
in
the
body
weights
of
offspring
can
affect
the
time
to
sexual
maturation,
whether
or
not
gestational
age,
as
defined
by
the
authors,
also
affects
the
time
of
sexual
maturation
is
purely
speculative,
especially
since
there
were
no
data
provided
by
the
authors
to
support
this
relationship.
Additionally,
covaring
gestational
age
with
time
to
sexual
maturation
is
problematic
from
a
statistical
standpoint.
Since
there
was
no
significant
change
in
the
length
of
gestation
at
30
mg/
kg/
day,
based
on
the
authors'
definition
of
`
gestational
age',
the
decreases
in
gestational
age
would
have
to
be
due
mostly
to
changes
in
time
to
sexual
maturation.
Therefore,
sexual
maturation
is
essentially
being
covaried
with
itself.
Still,
even
if
a
relationship
between
gestational
age
and
time
to
sexual
maturation
were
shown,
it
merely
offers
an
explanation
for
the
observed
delays
in
sexual
maturation
in
high­
dose
animals,
but
does
not
diminish
its
significance.

Therefore,
under
the
conditions
of
the
study,
the
LOAEL
for
F0
parental
males
is
considered
to
be
1
mg/
kg/
day,
the
lowest
dose
tested,
based
on
significant
increases
in
the
liver
and
kidney
weights­
to­
terminal
body
weight
and
to
brain
weight
ratios.
A
NOAEL
for
the
F0
parental
males
could
not
be
determined
since
treatment­
related
effects
were
seen
at
all
doses
tested.
40
The
NOAEL
and
LOAEL
for
F0
parental
females
are
considered
to
be
10
and
30
mg/
kg/
day,
respectively,
based
on
significant
reductions
in
kidney
weight
and
kidney
weight­
to­
terminal
body
weight
and
to
brain
weight
ratios
observed
at
the
highest
dose.

The
LOAEL
for
F1
generation
males
is
considered
to
be
1
mg/
kg/
day,
based
on
significant,
dose­
related
decreases
in
body
weights
and
body
weight
gains
(
observed
prior
to
and
during
cohabitation
and
during
the
entire
dosing
period),
and
in
terminal
body
weights;
and
significant
changes
in
absolute
liver
and
spleen
weights
and
in
the
ratios
of
liver,
kidney,
and
spleen
weights­
to­
brain
weights.
A
NOAEL
for
the
F1
males
could
not
be
determined
since
treatmentrelated
effects
were
seen
at
all
doses
tested.

The
NOAEL
and
LOAEL
for
F1
generation
females
are
considered
to
be
10
and
30
mg/
kg/
day,
respectively,
based
on
statistically
significant
increases
in
postweaning
mortality,
delays
in
sexual
maturation
(
time
to
vaginal
patency),
decreases
in
body
weight
and
body
weight
gains,
and
decreases
in
absolute
food
consumption,
all
observed
at
the
highest
dose
tested.

The
NOAEL
for
the
F2
generation
offspring
was
considered
to
be
30
mg/
kg/
day.
No
treatmentrelated
effects
were
observed
at
any
doses
tested
in
the
study.
However,
it
should
be
noted
that
the
F2
pups
were
sacrificed
at
weaning,
and
thus
it
was
not
possible
to
ascertain
the
potential
post­
weaning
effects
that
were
noted
in
the
F1
generation.

It
is
important
to
note
that
the
LOAELs
and
NOAELs
summarized
above
are
for
the
entire
duration
of
the
study,
and
therefore
represent
effects
resulting
from
developmental
and/
or
adult
exposures.
These
effect
levels
differ
from
those
described
in
section
5.1.
The
LOAELs
and
NOAELs
described
in
section
5.1
only
refer
to
effects
resulting
from
developmental
exposures.

4.0
Exposure
Characterization
PFOA
has
been
detected
in
human
serum
of
workers
occupationally
exposed
to
APFO,
and
it
has
also
been
measured
in
the
general
population.
In
general
the
levels
in
the
general
population
are
much
lower
than
in
the
workers.
However,
it
should
be
noted
that
the
highest
levels
reported
to
date
in
the
general
population
are
similar
to
some
of
the
lowest
levels
in
workers
exposed
to
PFOA
occupationally.
It
is
not
known
what
the
environmental
concentrations
of
APFO
are
or
the
pathways
of
exposure
to
the
general
population.

4.1
Occupational
exposures
3M
has
been
offering
voluntary
medical
surveillance
to
workers
at
plants
that
produce
or
use
perfluorinated
compounds
since
1976.
Serum
PFOA
levels
have
been
measured
and
reported
since
1993.
Prior
to
this
time,
only
total
organic
fluorine
was
measured.
The
results
of
biomonitoring
for
PFOA
have
been
reported
for
3
plants:
Cottage
Grove,
MN;
Decatur,
AL
and
Antwerp,
Belgium.
Surveillance
years
include
1993,
1995,
1997,
1998,
and
2000,
although
not
all
of
the
plants
offered
surveillance
in
all
of
these
years.
The
1998
data
reported
for
the
Decatur
plant
consist
of
a
random
sample
of
employees;
however,
volunteers
participated
in
all
of
the
41
other
sampling
periods
for
all
of
the
plants.
The
results
of
these
studies
are
summarized
in
Table
2.

Table
2.
Summary
of
Occupational
Exposures
(
ppm)

Plant
Arithmetic
Mean
Range
Geometric
Mean
95%
CI
Cottage
Grove
1997
(
n
=
7
4)
1995
(
n
=
80)
1993
(
n
=
111)
6.4
6.8
5.0
0.1
­
81.3
0
­
114.1
0
­
80.0
NA
NA
NA
NA
NA
NA
Decatur
2000
(
n
=
263)
1998
(
n
=
126)
1997
(
n
=
126)
1995
(
n
=
90)
1.78
1.54
1.57
1.46
0.04
­
12.7
0.02
­
6.76
NA
NA
1.13
0.9
NA
NA
0.99
­
1.3
0.72
­
1.12
NA
NA
Antwerp
2000
(
n
=
258)
1995
(
n
=
93)
0.84
1.13
0.01
­
7.04
0
­
13.2
0.33
NA
0.27
­
0.4
NA
Building
236
2000
(
n
=
45)
0.106
0.008
­
0.668
0.053
0.037
­
0.076
Mean
serum
PFOA
levels
have
increased
slightly
at
both
the
Cottage
Grove
and
Decatur
plants
since
1993.
Workers
at
the
Cottage
Grove
plant,
where
PFOA
exposures
are
highest,
have
the
highest
PFOA
serum
levels.
The
latest
sample
was
in
1997
(
Olsen,
et
al.,
1998b).
The
mean
serum
PFOA
level
was
6.4
ppm
(
range
=
0.1
 
81.3
ppm).
Only
74
employees
participated
in
the
1997
surveillance.
The
eligible
voluntary
participation
rates
ranged
from
approximately
50%
in
1997
to
70%
in
1993.

At
the
Decatur
plant,
263
of
500
employees
participated
in
2000
(
Olsen,
et
al.,
2001a).
The
mean
serum
PFOA
level
was
1.78
ppm.
It
was
higher
in
males
(
n
=
215)
than
females
(
n
=
48),
1.90
and
1.23
ppm,
respectively.
In
addition,
male
production
employees
had
higher
mean
serum
levels
(
2.34
ppm).
Five
employees
had
serum
levels
greater
than
5
ppm,
the
Biological
Limit
Value
established
by
the
3M
Exposure
Guideline
Committee.
Cell
operators
had
the
largest
increase
in
serum
PFOA
between
1998
and
2000.
The
highest
level
was
in
a
chemical
operator
on
the
Scotchgard
team
(
12.70
ppm).
The
mean
level
for
the
rest
of
the
members
of
the
team
was
5.06
ppm
(
range
5
­
9
ppm).
Other
job
categories
did
not
exhibit
such
a
large
increase.
3M
reports
that
this
is
due
to
increased
PFOA
production
at
the
Decatur
plant
beginning
in
1999.
Serum
PFOA
levels
for
the
Antwerp
plant
are
lower
than
at
Decatur
or
Cottage
Grove,
and
have
decreased
slightly
since
1995
(
Olsen,
et
al.,
2001b).
Participation
in
medical
surveillance
at
the
42
Antwerp
plant
was
the
highest
it
had
ever
been
in
2000
(
258
volunteers
out
of
340
workers).
The
mean
serum
PFOA
level
was
0.84
ppm,
and
the
highest
serum
level
reported
was
7.04
ppm.
As
in
the
Decatur
plant,
males
(
n
=
209)
had
higher
mean
serum
PFOA
levels
(
1.03
ppm)
than
females
(
n
=
49,
0.07
ppm).
Three
employees
had
levels
greater
than
5
ppm.

3M's
Specialty
Materials
Manufacturing
Division
Laboratories,
where
employees
perform
fluorochemical
research
(
Building
236),
conducted
voluntary
biomonitoring
of
45
employees
in
2000
(
Olsen,
et
al.,
2001c).
The
mean
PFOA
serum
level
was
0.106
ppm
(
range
0.008
 
0.668
ppm).

4.2
Non­
occupational
Exposures
Serum
PFOA
levels
in
corporate
staff
and
managers
at
a
3M
plant
in
St.
Paul,
MN,
where
occupational
exposure
to
PFOA
should
not
have
occurred,
were
reported
(
3M
Report,
1999).
Four
of
31
employees
had
serum
PFOA
levels
greater
than
the
detection
limit
of
10
ppb.
The
mean
for
these
employees
was
12.5
ppb.

4.3
General
Population
Exposures
Data
on
PFOA
levels
in
the
general
population
are
very
limited.
They
are
very
recent
so
that
trends
over
time
cannot
be
established.
The
mean
serum
PFOA
levels
are
lower
in
the
general
population
than
in
workers
exposed
to
PFOA.
The
available
data
are
summarized
in
Table
3.
43
Table
3.
Summary
of
General
Population
Exposures
(
ppb)

Sample
Arithmetic
Mean
Range
Geometric
Mean
95%
CI
Pooled
Samples
Commerical
sources
of
blood,
1999
(
n
=
35
lots)
3
1
­
13
NA
NA
Blood
banks,
1998
(
n
=
18
lots,
340­
680
donors)
17*
12
­
22
NA
NA
Individual
Samples
American
Red
Cross
blood
banks,
2000
(
n
=
645)
5.6
1.9
­
52.3
4.6
4.3
­
4.8
Elderly
(
65
­
96
years),
2000
(
n
=
238)
NA
1.4
­
16.7
4.2
3.9
­
4.5
Children
(
2
­
12
years),
1995
(
n
=
598)
5.6
1.9
­
56.1
4.9
4.7
­
5.1
*
PFOA
detected
in
about
1/
3
of
the
pooled
samples
but
quantifiable
in
only
2.

Pooled
blood
samples
from
U.
S.
blood
banks
indicate
mean
PFOA
levels
of
3
to
17
ppb
(
3M
Company,
Feb.
5,
1999;
3M
Company,
May
26,
1999).
The
highest
pooled
sample
reported
was
22
ppb.
Samples
were
collected
in
1998
and
1999.
However,
it
cannot
be
assumed
that
these
levels
are
generalizable
to
the
U.
S.
population
for
several
reasons:
1)
blood
donors
are
a
unique
group
that
does
not
necessarily
reflect
the
U.
S.
population
as
a
whole,
2)
many
of
the
blood
banks
originally
contacted
for
possible
inclusion
in
the
study
declined
to
participate,
3)
only
a
small
number
of
samples
have
actually
been
analyzed
for
PFOA,
and
4)
no
other
data
such
as
age,
sex,
or
other
demographic
information
are
available
on
the
donors.

Individual
blood
samples
from
3
different
age
populations
were
recently
analyzed
for
PFOA
and
other
fluorochemicals
using
high­
pressure
liquid
chromatography/
electrospray
tandem
mass
spectrometry
(
HPLC/
ESMSMS)
(
Olsen,
et
al.,
2002a,
2002b,
2002c).
The
studies'
participants
included
adult
blood
donors,
an
elderly
population
participating
in
a
prospective
study
in
Seattle,
WA,
and
children
from
23
states
participating
in
a
clinical
trial.
Overall,
the
PFOA
geometric
means
were
similar
across
all
3
populations
(
4.6
ppb,
4.2
ppb,
and
4.9
ppb,
respectively).
The
geometric
means
and
95%
tolerance
limits
(
the
exposure
below
which
95%
of
the
population
is
expected
to
be
found)
and
their
upper
bounds
were
comparable
across
all
3
studies.
However,
the
upper
ranges
for
the
children
and
adults
were
much
higher
than
for
the
elderly
population.
It
is
not
clear
whether
this
is
the
result
of
geographic
differences
in
PFOA
levels
or
some
other
factor.
It
should
be
noted
that
PFOS
and
PFOA
were
highly
correlated
in
all
three
studies
(
r
=
.63,
r
=
.70,
and
r
=
.75)
and
that
PFOA
did
not
meet
the
criteria
for
a
log
normal
distribution
44
based
on
the
Shapiro­
Wilk
test
in
any
of
the
studies.
However,
the
data
appeared
to
have
a
log
normal
distribution
and
therefore
geometric
means
were
calculated.
The
authors
suggest
that
it
may
be
due
to
the
greater
proportion
of
subjects
with
values
less
than
the
lower
limit
of
quantitation
(
LLOQ);
however,
only
12
of
the
1481
total
samples
were
below
the
LLOQ.
In
those
instances
where
a
sample
was
measured
below
the
LLOQ,
the
midpoint
between
zero
and
the
LLOQ
was
used
for
calculation
of
the
geometric
mean.
The
details
of
each
study
are
provided
below.

Blood
samples
from
645
U.
S.
adult
blood
donors
(
332
males,
313
females),
ages
20­
69,
were
obtained
from
six
American
Red
Cross
blood
banks
located
in:
Los
Angeles,
CA;
Minneapolis/
St.
Paul,
MN;
Charlotte,
NC;
Boston,
MA;
Portland,
OR,
and
Hagerstown,
MD
(
Olsen,
et
al.,
2002a).
Each
blood
bank
was
requested
to
provide
approximately
10
samples
per
10­
year
age
intervals
(
20­
29,
30­
39,
etc.)
for
each
sex.
The
only
demographic
factors
known
for
each
donor
were
age,
gender,
and
location.

The
geometric
mean
serum
PFOA
level
was
4.6
ppb.
The
range
was
<
lower
limit
of
quantitation
(
1.9
ppb)
to
52.3
ppb.
Only
2
samples
were
less
than
the
LLOQ.
Males
had
significantly
higher
(
p
<
.05)
geometric
mean
PFOA
levels
than
females
(
4.9
ppb
vs.
4.2
ppb).
Age
was
not
an
important
predictor
of
adult
serum
fluorochemical
concentrations.
When
stratified
by
geographic
location,
the
highest
geometric
mean
for
PFOA
was
in
the
samples
from
Charlotte,
NC
(
6.3
ppb,
range:
2.1
 
29.0)
and
the
lowest
from
Portland
(
3.6
ppb,
range:
2.1
 
16.7).
The
highest
individual
value
was
reported
in
Hagerstown
(
52.3
ppb).

Serum
PFOA
levels
were
reported
for
238
(
118
males
and
120
females)
elderly
volunteers
in
Seattle
participating
in
a
study
designed
to
examine
cognitive
function
in
adults
aged
65­
96
(
Olsen,
et
al.,
2002b).
Age,
gender
and
number
of
years'
residence
in
Seattle
were
the
only
data
available
on
the
participants.
Most
of
the
participants
were
under
the
age
of
85
and
had
lived
in
the
Seattle
area
for
over
50
years.

The
geometric
mean
of
PFOA
for
all
samples
was
4.2
ppb
(
95%
CI,
3.9
 
4.5).
The
range
was
1.4
 
16.7
ppb.
Only
5
samples
were
less
than
the
LLOQ
of
1.4
ppb.
There
was
no
significant
(
p
<
.05)
difference
in
geometric
means
for
males
and
females.
In
simple
linear
regression
analyses,
age
was
negatively
(
p
<
.05)
associated
with
PFOA
in
elderly
men
and
women.
In
bootstrap
analyses,
the
mean
of
the
95%
tolerance
limit
for
PFOA
was
9.7
ppb
with
an
upper
95%
confidence
limit
of
11.3
ppb.
PFOS
and
PFOA
were
highly
correlated
(
r
=
.75)
in
this
study.

A
sample
of
598
children,
ages
2­
12
years
old,
participating
in
a
study
of
group
A
streptococcal
infections,
was
analyzed
for
serum
PFOA
levels
(
Olsen,
et
al.,
2002c).
The
samples
were
collected
in
1994­
1995
from
children
residing
in
23
states
and
the
District
of
Columbia.
PFOA
did
not
meet
the
criteria
for
a
log
normal
distribution
based
on
the
Shapiro­
Wilk
test.
The
authors
suggest
that
it
may
be
due
to
the
greater
proportion
of
subjects
with
values
<
LLOQ
for
PFOA;
however,
only
5
samples
were
less
than
the
LLOQ
of
1.9
ppb.
The
geometric
mean
of
PFOA
for
all
of
the
participants
was
4.9
ppb
(
95%
CI,
4.7
 
5.1).
The
range
was
1.9
to
56.1
ppb.
Male
children
had
significantly
(
p<.
01)
higher
geometric
mean
serum
PFOA
levels
than
females:
5.2
ppb
and
4.7
ppb,
respectively.
In
simple
linear
regression
analyses,
age
was
significantly
(
p
<
45
.05)
negatively
associated
with
PFOA
in
both
males
and
females.
When
stratified
by
age,
the
geometric
mean
of
PFOA
was
highest
at
age
4
(
5.7
ppb)
and
lowest
at
age
12
(
3.5
ppb).
Although
the
data
were
not
reported,
a
graphical
presentation
of
log
PFOA
levels
for
each
state
by
gender
looked
similar
across
the
states;
however,
it
is
difficult
to
interpret
these
data
without
the
data
and
given
the
limited
sample
size
for
each
gender/
location
subgroup.
In
bootstrap
analyses,
the
mean
of
the
95%
tolerance
limit
for
PFOA
was
10.1
ppb
with
an
upper
95%
confidence
limit
of
11.0
ppb.
PFOS
and
PFOA
were
highly
correlated
(
r
=
.70)
in
this
study.
PFOA
and
PFHS
(
perfluorohexanesulfonate)
were
also
correlated
(
r
=
.48).

The
above
3
studies
indicate
similar
geometric
means
and
ranges
of
PFOA
among
sampled
adults,
children,
and
an
elderly
population.
However,
an
unexpected
finding
was
the
level
of
PFHS
and
M570
(
N­
methyl
perfluorooctanesulfonamidoacetate)
in
children.
These
serum
levels
were
much
higher
in
the
sampled
children
than
in
the
sampled
adults
or
elderly.
It
is
not
clear
why
this
occurred,
but
it
is
probably
due
to
a
different
exposure
pattern
in
children.

In
another
study,
the
PFOA
concentration
was
analyzed
in
human
sera
and
liver
samples
(
Olsen
et
al.,
2001d).
Thirty­
one
donor
samples
were
obtained
from
16
males
and
15
females
over
an
18­
month
period
from
the
International
Institute
for
the
Advancement
of
Medicine
(
IIAM).
The
average
age
of
the
male
donors
was
50
years
(
SD
15.6,
range
5­
69)
and
the
average
age
of
the
female
donors
was
45
years
(
SD
18.5,
range
13­
74).
The
causes
of
death
were
intracranial
hemorrhage
(
n
=
16
or
52%),
motor
vehicle
accident
(
n
=
7
or
23%),
head
trauma
(
n
=
4
or
13%),
brain
tumor
(
n
=
2
or
6%),
drug
overdose
(
n
=
1
or
3%)
and
respiratory
arrest
(
n
=
1
or
3%).
Both
serum
and
liver
tissue
were
obtained
from
23
donors;
7
donors
contributed
liver
tissue
only
and
1
donor
contributed
serum
only.
Serum
samples
were
obtained
from
5
ml
of
blood;
liver
samples
consisted
of
10
g
of
tissue.
Samples
were
frozen
at
IIAM
and
shipped
frozen
to
3M
for
analysis.
Samples
were
extracted
using
an
ion­
pairing
extraction
procedure
and
were
quantitatively
assayed
using
HPLC­
ESMSMS
and
evaluated
versus
an
unextracted
curve.
Extensive
matrix
spike
studies
were
performed
to
evaluate
the
precision
and
accuracy
of
the
extraction
procedure.
Serum
values
for
PFOA
ranged
from
<
LOQ
(<
3.0)
 
7.0
ng/
mL.
Assuming
the
midpoint
value
between
zero
and
LOQ
serum
value
for
samples
<
LOQ,
the
mean
serum
PFOA
level
was
3.1
ng/
mL
with
a
geometric
mean
of
2.5
ng/
mL.
No
liver
to
serum
ratios
were
provided
because
more
than
90%
of
the
individual
liver
samples
were
<
LOQ.

5.0
Preliminary
Risk
Assessment
For
this
preliminary
risk
assessment,
a
margin
of
exposure
(
MOE)
approach
was
used
to
describe
the
potential
for
developmental
toxicity
associated
with
exposure
to
PFOA
and
its
salts.
The
MOE
is
calculated
as
the
ratio
of
the
NOAEL,
LOAEL,
or
BMDL
for
a
specific
endpoint
to
the
estimated
human
exposure
level.
The
MOE
does
not
provide
an
estimate
of
population
risk,
but
simply
describes
the
relative
"
distance"
between
the
exposure
level
and
the
NOAEL,
LOAEL,
or
BMDL.

For
many
risk
assessments,
the
MOE
is
calculated
as
the
ratio
of
the
administered
dose
from
the
animal
toxicology
study
to
the
estimated
human
exposure
level.
The
human
exposure
is
estimated
from
a
variety
of
potential
exposure
scenarios,
each
of
which
requires
a
variety
of
46
assumptions.
A
more
accurate
estimate
of
the
MOE
can
be
derived
if
measures
of
internal
dose
are
available
for
humans
and
the
animal
model.
In
this
preliminary
risk
assessment,
serum
levels
of
PFOA,
which
is
a
measure
of
internal
dose,
were
available
for
the
animal
toxicology
studies
and
from
human
biomonitoring
studies.
Thus,
internal
dose
was
used
for
the
calculation
of
MOEs
in
this
assessment.

5.1
Selection
of
Developmental
Endpoints
As
stated
in
the
section
entitled
"
Scope
of
the
Assessment",
the
purpose
of
this
preliminary
assessment
was
to
determine
the
potential
of
developmental
toxicity
associated
with
exposure
to
PFOA
and
its
salts.
It
was
therefore
necessary
to
determine
which
endpoints
from
the
animal
toxicology
studies
would
be
relevant
for
this
assessment.
The
selection
of
developmental
endpoints
for
this
assessment
was
based
on
the
Agency's
Developmental
Toxicity
Risk
Assessment
Guidelines
(
EPA,
1991).
According
to
the
guidelines,
the
period
of
exposure
for
developmental
toxicity
is
prior
to
conception
to
either
parent,
through
prenatal
development
and
continuing
until
sexual
maturation.
In
contrast,
the
period
during
which
a
developmental
effect
may
be
manifested
includes
the
entire
lifespan
of
the
organism.

Several
oral
prenatal
developmental
toxicity
studies
of
APFO
have
been
conducted
in
rats
and
rabbits.
A
summary
of
the
exposure
duration
and
the
LOAELs
and
NOAELs
are
presented
in
Table
4.

Table
4.
Summary
of
Oral
Prenatal
Developmental
Toxicity
Studies
Species
Exposure
Duration
Time
Endpoints
Assessed
LOAEL
(
mg/
kg/
day)
NOAEL
(
mg/
kg/
day)
Reference
Sprague­
Dawley
rat
(
n=
22/
group)
GD
6­
15
GD
20
none
150
Gortner,
1981
Sprague­
Dawley
rat
(
n=
25/
group)
GD
6­
15
GD
20
none
100
Staples,
1984
Sprague­
Dawley
rat
(
n=
12/
group)
GD
6­
15
PND
35
none
100
Staples,
1984
New
Zealand
white
rabbit
(
n=
18/
group)
GD
6­
18
GD
29
50*
5
Gortner,
1982
*
­
there
was
a
dose­
related
increase
in
a
skeletal
variation,
extra
ribs
or
13th
rib,
with
statistical
significance
at
the
high­
dose
group
(
38%
at
50
mg/
kg/
day,
30%
at
5
mg/
kg/
day,
20%
at
1.5
mg/
kg/
day,
and
16
%
at
0
mg/
kg/
day).
47
In
addition,
developmental
effects
were
observed
in
the
oral
two
generation
reproductive
toxicity
study
that
was
conducted
in
Sprague­
Dawley
rats
(
York,
2002).
For
selection
of
the
developmental
endpoints
from
this
study,
attention
was
focused
on
effects
that
were
noted
during
the
period
of
developmental
exposure.
Thus,
only
effects
that
occurred
up
to
sexual
maturation
were
considered
relevant
for
this
preliminary
risk
characterization.
In
the
high
dose
group
administered
30
mg/
kg/
day
APFO,
there
was
a
reduction
in
F1
mean
body
weight
on
a
litter
basis
during
lactation
(
sexes
combined).
For
F1
females,
there
was
a
significant
increase
in
mortality
mainly
during
the
first
few
days
after
weaning,
and
a
significant
delay
in
the
timing
of
sexual
maturation
in
the
30
mg/
kg/
day
group.
For
F1
females,
the
LOAEL
for
developmental
toxicity
was
considered
to
be
30
mg/
kg/
day,
and
the
NOAEL
was
10
mg/
kg/
day.
For
F1
males
in
the
30
mg/
kg/
day
group,
there
was
an
increase
in
mortality
mainly
during
the
first
few
days
after
weaning,
a
significant
delay
in
the
timing
of
sexual
maturation,
and
a
significant
reduction
in
mean
postweaning
body
weight
that
began
on
day
8
and
continued
through
the
duration
of
the
study.
At
10
mg/
kg/
day,
mean
body
weights
were
significantly
reduced
beginning
on
day
36
postweaning
and
continuing
through
the
duration
of
the
study.
For
F1
males,
the
LOAEL
for
developmental
toxicity
was
considered
to
be
10
mg//
kg/
day,
and
the
NOAEL
was
3
mg/
kg/
day.
The
NOAEL
for
the
F2
generation
offspring
was
considered
to
be
30
mg/
kg/
day.
No
treatmentrelated
effects
were
observed
at
any
doses
tested
in
the
study.
However,
it
should
be
noted
that
the
F2
pups
were
sacrificed
at
weaning,
and
thus
it
was
not
possible
to
ascertain
the
potential
post­
weaning
effects
that
were
noted
in
the
F1
generation.

The
database
that
is
available
to
examine
the
potential
developmental
toxicity
associated
with
oral
exposure
to
APFO
thus
consists
of
two
prenatal
studies
in
Sprague­
Dawley
rats,
a
prenatal
study
in
New
Zealand
white
rabbits,
and
a
two
generation
reproductive
toxicity
study
in
Sprague­
Dawley
rats.
Since
no
developmental
toxicity
was
noted
in
the
prenatal
studies
in
Sprague­
Dawley
rats
these
studies
were
not
considered
for
the
calculation
of
the
MOEs.
The
effects
noted
in
the
prenatal
study
in
New
Zealand
white
rabbits
and
in
the
two
generation
reproductive
toxicity
study
in
Sprague­
Dawley
rats
were
considered
important
for
the
calculation
of
the
MOEs.

5.2
Use
of
Serum
Levels
as
a
Measure
of
Internal
Dose
for
Humans
Serum
levels
of
PFOA
were
available
from
human
biomonitoring
studies.
These
provide
a
measure
of
total
human
exposure
and
a
measure
of
internal
dose.
The
populations
that
were
considered
relevant
for
assessing
the
potential
for
developmental
toxicity
included
children
and
women
of
child
bearing
age.
Estimates
of
general
human
population
exposure
were
available
from
recent
analyses
of
individual
serum
samples
from
a
group
of
children
(
2­
12
years)
and
adults
(
20­
69
years).
Individual
serum
data
were
also
available
from
a
recent
analysis
of
a
group
of
elderly
adults
(
65­
96
years).
However,
these
data
were
not
included
in
this
analysis
given
that
the
concerns
are
for
women
of
child
bearing
age.
The
data
obtained
from
the
pooled
blood
samples
from
blood
banks
and
commercial
sources
were
not
used
in
the
calculation
of
the
MOEs
given
the
limitations
of
these
data.

A
summary
of
the
human
serum
levels
of
PFOA
that
were
considered
in
the
calculation
of
MOEs
is
provided
in
Table
5.
The
arithmetic
means
and
ranges
are
presented
in
order
to
display
48
the
higher
end
of
the
range
of
PFOA
serum
levels
in
a
small
segment
of
these
populations.
The
geometric
means
and
95%
confidence
intervals
are
presented
because
the
serum
levels
represent
what
appears
to
be
a
log
normal
distribution.
Gender
specific
data
were
available
for
the
geometric
mean
and
range,
but
not
for
the
arithmetic
mean.
Since
the
geometric
means
for
the
males
and
females
were
very
similar
(
4.2
ppb
for
females
and
4.9
ppb
for
males),
the
value
used
in
the
MOE
calculation
was
the
mean
for
sexes
combined
which
was
4.6
ppb.
The
value
for
sexes
combined
was
used
so
that
MOEs
could
be
calculated
using
both
the
arithmetic
and
geometric
means.
The
use
of
the
value
of
4.6
ppb
versus
the
gender
specific
levels
of
4.2
and
4.9
ppb
has
minimal
impact
on
the
resulting
MOE
in
this
preliminary
risk
assessment.

Table
5.
Summary
of
Levels
of
PFOA
in
the
Serum
of
Human
Populations
Population
Arithmetic
Mean
Range
Geometric
Mean
95%
CI
Adults
(
20
­
69
years,
American
Red
Cross
blood
banks,
2000,
n=
645)
5.6
ppb
1.9
­
52.3
ppb
4.6
ppb
4.3
­
4.8
ppb
Children
(
2­
12
years,
1995,
n=
598)
5.6
ppb
1.9
­
56.1
ppb
4.9
ppb
4.7
­
5.1
ppb
5.3
Use
of
Serum
Levels
as
a
Measure
of
Internal
Dose
for
Animal
Studies
Serum
levels
of
PFOA
were
available
as
a
measure
of
the
internal
dose
of
humans.
Thus,
only
those
animal
studies
with
serum
levels
of
PFOA
were
considered
for
the
calculation
of
the
MOEs.
Since
no
information
on
serum
levels
was
available
for
the
prenatal
developmental
toxicity
study
in
New
Zealand
white
rabbits,
MOEs
were
not
calculated
for
the
endpoints
from
this
study.
Serum
levels
were
available
for
the
two
generation
reproductive
study
in
Sprague­
Dawley
rats.
Therefore,
this
study
was
used
for
the
calculation
of
the
MOEs.

In
the
two
generation
reproductive
toxicity
study
in
Sprague­
Dawley
rats,
the
LOAEL
for
developmental
toxicity
was
considered
to
be
30
mg/
kg/
day,
and
the
NOAEL
was
10
mg/
kg/
day
for
the
F1
females.
For
the
F1
males,
the
LOAEL
for
developmental
toxicity
was
considered
to
be
10
mg/
kg/
day,
and
the
NOAEL
was
3
mg/
kg/
day.
Serum
levels
of
PFOA
were
not
measured
in
the
F1
animals.
Serum
levels
were
available
for
F0
animals
in
the
control,
10
and
30
mg/
kg/
day
groups.
Serum
levels
were
not
measured
in
the
1
and
3
mg/
kg/
day
groups.
The
serum
levels
were
measured
in
the
F0
males
at
the
end
of
the
cohabitation
period,
while
serum
levels
were
measured
on
lactation
day
22
in
the
F0
females.
For
both
sexes,
the
serum
levels
were
measured
24
hours
after
the
administration
of
the
last
dose.

For
this
preliminary
risk
assessment,
the
serum
levels
for
the
F0
animals
were
used
to
provide
an
estimated
range
of
potential
serum
levels
in
the
F1
animals.
The
following
approach
was
employed.
It
was
reasoned
that
if
prenatal
and/
or
lactational
exposures
were
important
in
49
eliciting
the
developmental
effects,
then
the
serum
levels
in
the
dam
(
i.
e.
F0
females
who
were
being
administered
APFO)
would
be
the
most
representative
of
the
serum
levels
in
the
F1
pups.
The
serum
levels
in
the
F0
males
would
not
be
informative.
It
was
further
reasoned
that
if
postweaning
exposures
were
important
then
the
serum
levels
for
the
F0
males
would
be
the
most
appropriate
estimate
for
the
F1
males,
and
similarly
the
serum
levels
in
the
F0
females
would
be
the
most
appropriate
estimate
for
the
F1
females.

If
prenatal
and/
or
lactational
exposures
were
important,
several
other
factors
had
to
be
considered.
First,
serum
levels
were
measured
in
the
F0
females
24
hours
after
dosing.
Given
that
the
serum
half­
life
in
female
rats
is
less
than
24
hours,
this
value
would
represent
the
low
end
of
exposure.
Since
the
peak
exposure
of
the
F0
females
is
not
known,
it
was
reasoned
that
it
was
unlikely
that
the
peak
exposure
to
the
F0
females
was
higher
than
the
serum
level
in
the
F0
males
in
the
same
dose
group
since
the
serum
half
life
of
PFOA
in
male
rats
is
4.4
­
9
days,
and
therefore
with
a
daily
dosing
regime
they
would
tend
to
accumulate
PFOA.
Therefore,
the
serum
levels
in
the
F0
males
and
females
in
the
10
mg/
kg/
day
were
used
to
provide
a
range
of
values
for
the
calculation
of
the
MOEs;
the
serum
levels
were
51.1
and
0.37
ppm
in
the
F0
males
and
females,
respectively.
The
serum
levels
of
the
F0
females
probably
represent
a
low
internal
dose
for
the
F1
animals
and
the
serum
levels
of
the
F0
males
probably
represent
a
high
internal
dose
for
the
F1
animals.

For
the
scenario
where
postweaning
exposures
are
important,
the
serum
levels
in
F0
females
would
be
the
most
appropriate
estimate
of
the
serum
levels
in
F1
females.
For
F1
females,
the
same
rationale
was
applied
for
this
scenario
as
was
applied
to
the
previous
scenario
for
prenatal
and/
or
lactational
exposures.
Therefore,
the
serum
levels
in
the
F0
males
and
females
in
the
10
mg/
kg/
day
were
used
to
provide
a
range
of
values
for
the
calculation
of
the
MOEs.

Similarly,
for
the
scenario
where
postweaning
exposures
are
important,
the
serum
levels
in
the
F0
males
would
be
the
most
appropriate
estimate
of
the
serum
levels
in
F1
males.
For
this
assessment,
the
LOAEL
for
developmental
effects
in
the
F1
males
was
considered
to
be10
mg/
kg/
day
and
the
NOAEL
was
considered
to
be
3
mg/
kg/
day.
However,
serum
levels
for
the
F0
males
were
only
measured
in
the
10
and
30
mg/
kg/
day
groups;
there
was
no
information
available
for
the
3
mg/
kg/
day
group.
In
addition,
it
was
not
possible
to
extrapolate
the
serum
levels
to
the
lower
administered
doses
as
the
values
appear
to
have
reached
a
plateau
at
10
and
30
mg/
kg/
day
(
51.1
and
45.3
ppm,
respectively)
and
are
not
linear.
Therefore,
for
this
preliminary
assessment
the
serum
levels
from
the
F0
males
in
the
10
mg/
kg/
day
group
were
used
in
the
calculation
of
the
MOEs.

5.4
Calculation
of
MOEs
The
human
populations
of
concern
for
this
preliminary
assessment
are
women
of
child
bearing
age
and
children.
As
stated
in
section
5.2,
the
serum
data
from
the
American
Red
Cross
study
included
both
men
and
women,
ages
20­
69.
As
explained
in
section
5.2,
since
the
data
were
not
consistently
reported
separately
for
each
gender
and
since
the
geometric
means
were
very
similar
for
males
and
females,
the
means
for
the
sexes
combined
are
used
as
a
surrogate
for
women
of
child
bearing
age.
The
MOEs
were
calculated
by
dividing
the
serum
values
for
the
F0
female
50
and
male
rats
in
the
10
mg/
kg/
day
group
(
0.37
ppm
and
51.1
ppm,
respectively)
in
the
two
generation
reproductive
toxicity
study
by
the
American
Red
Cross
blood
samples
and
children's
samples
presented
in
Table
5.
The
MOEs
for
potentially
exposed
populations
are
presented
in
Table
6.

It
is
important
to
note
that
MOEs
that
were
calculated
from
the
serum
levels
in
the
F0
female
and
male
rats
provide
a
means
to
bracket
the
low
and
high
ends
of
reported
experimental
exposures.
This
is
an
unusual
situation
in
that
MOE
estimates,
which
typically
represent
point
estimates,
are
described
here
as
a
range
of
potential
values
due
to
uncertainties
in
the
rat
serum
data.
This
situation
arises
from
the
fact
that
the
available
data
do
not
allow
selection
of
a
particular
departure
point
for
the
MOE
calculations.
It
is
likely
that
the
MOEs
calculated
using
the
F0
female
rat
serum
level
are
lower
than
what
would
be
anticipated
in
the
human
population,
and
it
is
likely
that
MOEs
calculated
using
the
F0
male
rat
serum
level
are
higher
than
what
would
be
anticipated
in
the
human
population.
As
uncertainty
around
the
rat
serum
values
decreases
the
end
brackets
are
likely
to
shift
towards
the
middle
of
the
current
range.
Therefore,
MOE
values
presented
in
Table
6
should
not
be
interpreted
as
representing
the
range
of
possible
MOEs
in
the
US
population.
It
is
likely
that
when
more
extensive
rat
kinetic
data
are
available,
the
resultant,
refined
estimated
range
of
MOEs
will
constitute
a
narrower
subset
of
the
range
presented
here.

Table
6.
MOE
Calculations
for
Potentially
Exposed
Populations
Using
F0
Rat
Serum
Values
and
Human
Serum
Values
Human
Population
MOE
values
calculated
using
rat
serum
values
from
the
2­
generation
reproductive
study1
Women
of
Childbearing
Age2
Arithmetic
mean
Geometric
mean
66
­­­­­
9125
80
­­­­­
11,109
Children3
Arithmetic
mean
Geometric
mean
66
­­­­
9125
75
­­­­
10,429
1Estimated
MOE
values
are
bracketed
by
the
serum
level
concentrations
in
the
F0
females
(
0.31
ppm)
and
the
F0
males
(
51.1
ppm).
2The
American
Red
Cross
serum
samples
were
used
as
an
estimate
of
the
arithmetic
and
geometric
means
in
women
of
child
bearing
age.
3The
samples
from
the
Children's
study
were
used
as
an
estimate
of
the
arithmetic
and
geometric
means.
51
The
MOE
that
was
calculated
at
the
low
end
of
the
range
using
the
rat
female
F0
serum
levels
and
the
arithmetic
mean
for
the
adults
is
66,
while
the
MOE
that
was
calculated
using
the
geometric
mean
for
the
adults
is
80.
The
MOEs
that
were
calculated
at
the
high
end
of
the
range
using
the
serum
levels
of
the
rat
F0
males
and
the
arithmetic
or
geometric
means
for
the
adults
range
from
approximately
9,000­
11,000.
If
the
upper
and
lower
values
of
the
human
serum
levels
are
used
in
the
MOE
calculation
with
the
rat
F0
females,
the
resulting
MOEs
are
195
and
7.
If
the
upper
and
lower
values
of
the
human
serum
levels
are
used
in
the
MOE
calculation
with
the
rat
F0
males,
the
resulting
MOEs
are
26,895
and
911.

The
MOEs
for
children
were
calculated
by
dividing
the
serum
values
for
the
rat
F0
females
and
males
in
the
10
mg/
kg/
day
group
(
0.37
ppm
and
51.1
ppm,
respectively)
by
the
biomonitoring
values
presented
in
Table
5.
The
MOE
that
was
calculated
at
the
low
end
of
the
range
using
the
rat
female
F0
serum
levels
and
the
arithmetic
mean
for
the
children
is
66,
while
the
MOE
that
was
calculated
using
the
geometric
mean
for
the
children
is
75.
The
MOEs
that
were
calculated
at
the
high
end
of
the
range
using
the
serum
levels
of
the
rat
F0
males
and
the
arithmetic
or
geometric
means
for
the
children
range
from
approximately
9,000­
10,400.
If
the
upper
and
lower
values
of
the
childrens'
serum
levels
are
used
in
the
MOE
calculation
with
the
F0
females,
the
resulting
MOEs
are
195
and
6.6.
If
the
upper
and
lower
values
of
the
childrens'
serum
levels
are
used
in
the
MOE
calculation
with
the
F0
males,
the
MOEs
are
26,895
and
911.

5.5
Uncertainties
in
the
Preliminary
Risk
Characterization
Some
of
the
uncertainties
encountered
in
this
preliminary
risk
assessment
are
common
for
many
risk
assessments.
One
such
example
pertains
to
the
choice
of
the
animal
model.
In
this
preliminary
assessment,
serum
levels
were
not
available
for
the
oral
rabbit
prenatal
developmental
toxicity
study
and
therefore
this
study
was
not
used
in
the
calculation
of
the
MOEs.
It
is
not
known
whether
rabbits
are
a
more
appropriate
animal
model
than
rats.

Other
uncertainties
that
are
common
for
many
assessments
have
been
avoided
in
this
preliminary
assessment
through
the
use
of
a
measure
of
internal
dose
of
PFOA
for
both
humans
and
the
rat
animal
model.
This
approach
has
avoided
many
of
the
pitfalls
encountered
in
trying
to
estimate
human
exposure
levels
through
the
application
of
various
models
and
assumptions.
Unlike
many
environmental
chemicals
where
it
is
only
hypothesized
that
humans
are
exposed,
serum
data
from
humans
gives
direct
evidence
that
exposure
to
PFOA
has
occurred
in
the
general
public,
and
provides
a
measure
of
internal
dose.
Although,
it
is
not
known
when
or
how
exposure
has
occurred,
this
is
less
of
an
issue
for
PFOA
given
its
long
half
life
in
humans.
Therefore,
the
MOEs
calculated
in
this
preliminary
risk
assessment
can
be
considered
to
be
a
more
accurate
comparison
of
the
relative
"
distance"
between
the
exposure
level
and
the
NOAEL
than
if
administered
dose
had
been
used.

Although
the
use
of
serum
levels
as
a
measure
of
internal
dose
introduces
uncertainties
that
are
unique
to
this
assessment,
many
of
these
have
been
accounted
for
through
the
use
of
a
range
of
MOEs.
For
example,
one
area
of
uncertainty
pertains
to
the
use
of
serum
data
for
the
F0
animals
as
estimates
of
serum
levels
in
the
F1
animals.
As
noted
in
the
previous
section,
it
is
not
known
whether
the
effects
on
postweaning
mortality,
body
weight,
or
age
of
sexual
maturation
were
due
52
to
prenatal
exposures,
lactational
exposures,
postweaning
exposures,
or
a
combination
of
one
or
more
of
these
exposure
periods.
In
most
risk
assessments,
no
attempt
is
made
to
determine
which
of
these
exposure
periods
is
important.
A
major
strength
of
this
preliminary
assessment
is
that
each
of
these
exposure
periods
was
considered
in
order
to
determine
the
appropriateness
and
uncertainties
associated
with
the
use
of
the
serum
levels
from
the
F0
animals.

It
was
reasoned
that
if
prenatal
and/
or
lactational
exposures
were
important
then
the
serum
levels
in
the
F0
females
would
be
the
most
appropriate
estimate
for
the
F1
animals.
If
postweaning
exposures
were
important
then
the
serum
levels
for
the
F0
males
would
be
the
most
appropriate
estimate
for
the
F1
males,
and
similarly
the
serum
levels
in
the
F0
females
would
be
the
most
appropriate
estimate
for
the
F1
females.
It
was
not
possible
to
make
a
"
direct"
estimate
of
F1
serum
levels
from
the
serum
levels
in
the
F0
females
for
the
prenatal
and/
or
lactational
exposure
scenario
for
several
reasons.
First,
as
noted
in
the
previous
sections,
there
is
a
gender
difference
in
the
elimination
of
PFOA
in
rats.
In
female
rats,
estimates
of
the
serum
half
life
range
from
1.9
to
24
hours,
while
in
male
rats
estimates
of
the
serum
half
life
range
from
4.4
to
9
days.
In
female
rats
elimination
of
PFOA
appears
to
be
biphasic;
a
fast
phase
occurs
with
a
half
life
of
approximately
2­
4
hours
while
a
slow
phase
occurs
with
a
half
life
of
approximately
24
hours.
In
the
two
generation
reproductive
toxicity
study,
the
animals
were
dosed
by
gavage
once
daily.
The
serum
levels
were
measured
24
hours
after
dosing.
Thus,
the
values
obtained
for
the
F0
females
represent
the
low
end
of
exposure.
With
no
knowledge
of
the
peak
exposures,
it
was
reasoned
that
it
was
unlikely
that
the
peak
exposure
would
be
higher
than
the
serum
level
in
the
F0
males
in
the
same
dose
group
since
they
would
tend
to
accumulate
PFOA
with
a
daily
dosing
regime.
Therefore,
the
strategy
that
was
employed
in
this
assessment
was
to
use
the
MOEs
that
were
calculated
from
the
serum
levels
in
the
F0
females
and
males
as
a
range
or
as
a
means
to
bracket
the
low
and
high
ends
of
exposure.
Thus,
it
is
likely
that
the
MOEs
that
were
calculated
using
the
F0
female
serum
levels
are
probably
too
low,
while
the
values
calculated
from
the
F0
male
serum
levels
are
probably
too
high.

As
noted
above,
the
serum
levels
from
the
F0
females
would
be
the
most
appropriate
estimate
for
the
F1
females
if
postweaning
exposures
were
important.
Given
the
issues
associated
with
the
gender
difference
in
elimination
of
PFOA
in
rats
and
the
timing
of
serum
collection
in
the
F0
females,
the
same
logic
which
was
applied
for
the
prenatal
and
lactational
exposures
was
also
used
for
this
scenario.
Again,
the
MOEs
that
were
calculated
from
the
serum
levels
in
the
F0
females
and
males
were
viewed
as
a
range
or
as
a
means
to
bracket
the
low
and
high
ends
of
potential
exposure.

Similarly,
the
serum
levels
from
the
F0
males
would
be
the
most
appropriate
estimate
for
the
F1
males
if
postweaning
exposures
were
important.
As
stated
above,
the
LOAEL
for
developmental
effects
in
the
F1
males
was
considered
to
be10
mg/
kg/
day
and
the
NOAEL
was
considered
to
be
3
mg/
kg/
day.
However,
serum
levels
for
the
F0
males
were
only
measured
in
the
10
and
30
mg/
kg/
day
groups;
there
was
no
information
available
for
the
3
mg/
kg/
day
group.
In
addition,
it
was
not
possible
to
extrapolate
the
serum
levels
to
the
lower
administered
doses
as
the
values
appear
to
have
reached
a
plateau
at
10
and
30
mg/
kg/
day
(
51.1
and
45.3
ppm,
respectively)
and
are
not
linear.
Thus,
if
postweaning
exposures
were
important,
the
MOEs
based
on
the
serum
levels
in
the
F0
males
would
be
too
high
and
underestimate
potential
risks.
Again,
this
53
uncertainty
has
been
accounted
for
in
this
preliminary
assessment
through
the
calculation
of
a
range
of
MOEs.

In
the
scenarios
presented
above,
the
serum
levels
of
the
F0
males
were
viewed
as
an
overestimate
given
that
male
rats
tend
to
accumulate
PFOA
given
the
long
half
life
and
the
fact
that
saturation
had
apparently
occurred.
One
line
of
reasoning
that
would
support
the
statement
that
the
serum
levels
from
the
F0
males
are
probably
too
high
an
estimate
of
peak
exposure
in
the
F0
females
is
provided
by
consideration
of
the
serum
half
life
of
PFOA
in
female
rats.
Four
studies
have
examined
serum
half
life
in
a
variety
of
rat
strains;
these
studies
have
employed
different
routes
of
exposure
and
different
methods
for
quantitating
serum
PFOA
levels.
Two
of
these
studies
were
most
relevant
to
the
rat
strain
and
route
employed
in
the
two
generation
reproductive
toxicity
study.
Ophaug
and
Singer
(
1980)
estimated
a
serum
half
life
of
10
hours
after
administering
female
Holtzman
rats
a
gavage
dose
of
8
mg/
kg,
while
Vanden
Heuvel
et
al.
(
1991b)
estimated
a
serum
half
life
of
4
hours
after
administering
female
Harlan
Sprague­
Dawley
rats
an
intraperitoneal
dose
of
4
mg/
kg.
Using
a
serum
half
life
value
of
10
hours
would
yield
a
peak
internal
dose
of
approximately
1.85
ppm,
while
a
serum
half
life
of
4
hours
would
yield
a
peak
internal
dose
of
approximately
24.7
ppm.
While
neither
of
these
values
may
be
absolutely
correct
given
that
serum
half
life
has
not
been
reported
for
the
rat
strain
and
dosing
conditions
used
in
the
two
generation
reproductive
toxicity
study,
both
values
do
support
the
statement
that
the
serum
levels
in
the
F0
males
are
probably
too
high
(
1.85
and
24.7
ppm
versus
51.1
ppm).
Thus
the
resultant
MOEs
using
the
F0
male
serum
values
probably
underestimate
the
potential
risks.

It
is
possible
that
lactational
and
postweaning
exposures
to
the
F1
pups
may
have
been
higher
than
the
exposures
to
the
F0
females
due
to
the
time
required
for
maturation
of
the
clearance
mechanism.
First,
the
clearance
is
under
hormonal
control
and
these
hormones
do
not
reach
adult
levels
until
puberty.
Second,
recent
studies
have
examined
the
role
of
three
organic
anion
transporters
(
OAT),
OAT1,
OAT2,
and
OAT3,
in
the
urinary
elimination
of
PFOA
in
the
rat.
Kudo
et
al.
(
2002)
has
provided
evidence
that
OAT2
and
OAT3
may
be
involved.
A
study
of
developmental
and
gender­
specific
influences
on
the
expression
of
rat
OATs
in
the
kidney
has
shown
that
at
birth
all
OAT
mRNA
levels
are
low
(
Buist
et
al.,
2002).
Renal
OAT1
expression
approaches
adult
level
at
30
days,
where
at
day
40
and
45
OAT1
levels
were
greater
in
males
than
females.
OAT2
expression
was
minimal
through
day
30
but
increased
dramatically
only
in
females
at
day
35.
OAT3
expression
matured
the
earliest
and
reached
adult
levels
at
10
days.
If
these
OATs
are
important,
then
this
developmental
profile
would
suggest
that
the
PFOA
clearance
in
juvenile
female
rats
is
less
than
in
adult
female
rats.
Therefore,
the
serum
levels
in
the
females
prior
to
sexual
maturation
may
be
higher
than
in
the
adult,
and
the
resultant
MOEs
would
also
be
higher.
However,
it
is
important
to
note
that
the
MOEs
in
this
situation
would
still
fall
within
the
range
of
MOEs
that
were
calculated
in
this
preliminary
assessment.

There
are
several
other
uncertainties
unique
to
this
assessment.
The
first
pertains
to
the
choice
of
developmental
endpoints.
This
preliminary
assessment
utilized
endpoints
from
the
two
generation
reproductive
toxicity
study
that
could
be
directly
attributed
to
developmental
exposures.
Another
way
to
ascertain
potential
developmental
effects
would
be
to
compare
the
magnitude
and
dose
levels
of
the
systemic
toxicity
that
was
observed
in
the
F1
animals
54
(
developmental
and
adult
exposures)
at
the
end
of
the
study
period
with
those
systemic
effects
noted
in
the
F0
animals
(
adult
exposures
only).
For
example,
if
organ
weight
changes
were
greater
in
F1
than
F0
animals
or
if
changes
occurred
at
lower
doses
in
the
F1
animals,
this
may
be
indicative
of
the
importance
of
developmental
exposures.
In
the
F0
males,
there
were
significant
reductions
in
the
absolute
weights
of
the
left
and
right
epididymides,
left
cauda
epididymis,
seminal
vesicles
(
with
and
without
fluid),
prostate,
pituitary,
left
and
right
adrenals,
spleen,
and
thymus
at
30
mg/
kg/
day.
In
the
F1
males,
the
effects
on
many
of
these
organ
weights
occurred
at
lower
doses.
This
may
indicate
that
the
developmental
exposures
were
important,
and
that
the
LOAEL
and
NOAEL
are
lower
than
the
values
used
in
this
preliminary
assessment.
If
true,
this
preliminary
assessment
would
have
underestimated
potential
risks.

Another
area
of
uncertainty
pertains
to
the
use
of
the
serum
levels
as
a
measure
of
internal
dose.
PFOA
behaves
differently
from
many
other
persistent
environmental
contaminants
in
that
it
is
not
stored
in
adipose
tissue.
It
is
not
clear
whether
it
binds
to
proteins
or
other
macromolecules,
but
it
is
clear,
based
on
animal
studies,
that
PFOA
enters
enterohepatic
circulation.
Animal
studies
indicate
that
PFOA
mainly
partitions
to
blood
serum
and
the
liver.
In
this
preliminary
assessment,
serum
levels
were
used
primarily
because
this
was
the
only
information
available
for
humans.
However,
it
is
possible
that
area
under
the
curve,
liver
levels,
the
ratio
of
serum
to
liver
levels,
or
some
other
measure
may
be
more
appropriate
dose
metrics.

In
addition,
the
apparent
biphasic
elimination
of
PFOA
in
female
rats
raises
an
important
issue,
whether
the
observed
effects
are
due
to
the
kinetics
associated
with
the
fast
or
slow
elimination
phase
in
the
females.
The
data
currently
available
do
not
allow
development
of
this
potentially
important
issue.
In
this
preliminary
assessment,
the
serum
values
for
the
F0
males
were
used
to
provide
an
estimate
of
the
peak
exposure
of
the
F0
females.
The
MOEs
that
were
then
calculated
from
the
serum
levels
in
the
F0
females
and
males
provide
a
means
to
bracket
the
high
and
low
ends
of
exposure.
If
the
effects
are
associated
with
the
slow
elimination
phase,
then
the
serum
levels
in
the
F0
females
and
the
resultant
MOEs
would
be
more
realistic.

Another
area
of
uncertainty
pertains
to
the
differences
in
the
serum
half
life
in
humans
and
rats.
In
humans,
the
serum
half
life
is
years,
while
in
the
rat
it
is
less
than
24
hours
in
females
and
105
hours
in
males.
To
date,
there
is
no
evidence
of
a
gender
difference
in
the
elimination
of
PFOA
in
humans.
Thus,
humans
appear
to
be
more
similar
to
male
rats
in
that
they
will
tend
to
accumulate
PFOA
and
will
have
a
more
continuous
internal
exposure.
It
is
not
known
how
this
would
impact
potential
risks
of
developmental
toxicity
in
humans.

Finally,
there
is
some
uncertainty
regarding
the
use
of
the
human
biomonitoring
data.
Although
the
available
data
include
a
range
of
populations
with
various
demographics,
there
may
be
some
populations
that
may
not
be
represented.
Since
it
is
unknown
how
the
human
exposures
are
occurring,
proximity
to
a
manufacturing
plant
may
be
a
factor
in
exposure.
However,
populations
living
near
the
plants
were
not
sampled.
Therefore,
it
is
possible
that
PFOA
serum
levels
may
be
underestimated
for
certain
portions
of
the
U.
S.
population.
The
children's
sample
was
derived
from
blood
collected
in
1994/
1995;
therefore,
it
may
not
reflect
the
current
status
of
PFOA
in
children's
blood.
It
is
not
clear
how
PFOA
may
affect
more
sensitive
subpopulations
or
if
their
exposures
would
vary.
55
6.0
Overall
Conclusions
This
preliminary
risk
assessment
focused
on
the
potential
risks
for
developmental
toxicity
associated
with
exposure
to
PFOA
and
its
salts.
Concerns
for
developmental
toxicity
were
raised
from
the
results
of
a
rat
two­
generation
reproductive
toxicity
study
of
APFO
.
In
this
study,
there
was
a
reduction
in
F1
mean
body
weight
on
a
litter
basis
during
lactation
(
sexes
combined).
Postweaning
mortality
and
delayed
sexual
maturation
were
noted
in
F1
females
administered
30
mg/
kg/
day
APFO;
the
NOAEL
for
developmental
effects
for
F1
females
was
10
mg/
kg/
day.
Postweaning
mortality,
delayed
sexual
maturation
and
a
significant
reduction
in
postweaning
body
weights
were
noted
in
F1
males
at
30
mg/
kg/
day,
and
a
significant
reduction
in
postweaning
body
weight
was
noted
at
10
mg/
kg/
day.
For
F1
males,
the
LOAEL
for
developmental
effects
was
10
mg/
kg/
day
and
the
NOAEL
was
3
mg/
kg/
day.

For
calculation
of
the
MOEs,
the
human
populations
that
were
considered
included
women
of
child
bearing
age
and
children.
Estimates
of
general
human
population
exposure
were
available
from
recent
analyses
of
individual
serum
samples
from
a
group
of
children
(
2­
12
years)
and
adults
(
20­
69
years).
For
the
populations
of
interest,
calculations
using
human
adult
serum
levels
and
children
serum
levels
in
combination
with
rat
serum
values
from
the
parental
(
F0)
females
and
males
produced
a
range
of
overlapping
MOE
values
that
extends
from
less
than
100
to
greater
than
9000.
There
are
a
number
of
important
uncertainties
discussed
in
this
document
that
provide
a
context
for
considering
these
MOEs
as
a
range
of
potential
values.
Interpretation
of
the
significance
of
the
MOEs
for
ascertaining
potential
levels
of
concern
in
exposed
populations
will
necessitate
a
better
understanding
of
the
appropriate
measure
of
exposure
in
rats,
and
the
relationship
of
the
latter
to
human
serum
levels.
56
7.0
References
3M
Company.
(
1979)
Technical
report
summary
­
final
comprehensive
report:
FC­
143.
(
U.
S.
EPA
AR­
226
528).

3M
Environmental
Laboratory.
(
1993)
Impinger
studies
of
volatility
of
FC­
95
and
FC­
143.
St.
Paul,
MN.
3M
Laboratories.
3M
Lab
Request
Number
L3306.

Alexander,
BH.
(
2001a)
Mortality
study
of
workers
employed
at
the
3M
Cottage
Grove
facility.
Final
Report.
Division
of
Environmental
and
Occupational
Health,
School
of
Public
Health,
University
of
Minnesota,
April
26,
2001.

Alexander,
BH.
(
2001b)
Mortality
study
of
workers
employed
at
the
3M
Decatur
facility.
Final
Report.
Division
of
Environmental
and
Occupational
Health,
School
of
Public
Health,
University
of
Minnesota,
April
26,
2001.

Beilstein
(
1975)
Beilstein
Handbook
4th
Work
Volume
2
Part
2
page
994.

Bluist,
S;
Cherrington,
NJ;
Choudhuri,
J;
Hartley,
DP;
Klaassen,
CD.
(
2002).
Gender­
specific
and
developmental
influences
on
the
expression
of
rat
organic
anion
transporters.
J.
Pharmacol.
Exper.
Ther.
301:
145­
151.

Burris,
JM;
Olsen,
G;
Simpson,
C;
et
al.
(
2000)
Determination
of
serum
half­
lives
of
several
fluorochemicals.
Interim
Report
#
1,
Corporate
Occupational
Medicine
Department,
3M
Company.

Burris,
JM;
Lundberg,
JK;
Olsen,
GW;
et
al.
(
2002)
Determination
of
serum
half­
lives
of
several
fluorochemicals.
Interim
Report
#
2.
3M
Medical
Department.

Calfours,
J;
Stilbs,
P.
(
1985)
Solubilization
in
sodium
perfluorooctanoate
micelles:
a
multicomponent
self­
diffusion
study.
Colloid
Interface
Sci
103:
332­
336.

Edwards,
PJB;
Jolley,
KW;
Smith,
MH;
et
al.
(
1997)
Solvent
isotope
effect
on
the
self­
assembly
liquid
crystalline
phase
behavior
in
aqueous
solutions
of
ammonium
pentadecafluorooctanoate.
Langmuir
13(
10):
2665­
2669.

Gibson,
SJ;
Johnson,
JD.
(
1979)
Absorption
of
FC­
143­
14C
In
Rats
After
a
Single
Oral
Dose.
Riker
Laboratories,
Inc.,
Subsidiary
of
3M,
St.
Paul,
Minnesota.

Gibson,
SJ;
Johnson,
JD.
(
1980)
Extent
and
route
of
excretion
and
tissue
distribution
of
total
carbon­
14
in
male
and
female
rats
after
a
single
IV
dose
of
FC­
143­
14C.
Riker
Laboratories,
Inc.,
Subsidiary
of
3M,
St.
Paul,
MN.
57
Gibson,
SJ;
Johnson,
JD.
(
1983)
Extent
and
route
of
excretion
of
total
carbon­
14
in
pregnant
rats
after
a
single
oral
dose
of
ammonium
14
C­
perfluorooctanoate.
Riker
Laboratories,
Inc.,
Subsidiary
of
3M,
St.
Paul,
MN.

Gilliland,
F.
(
1992)
Fluorochemicals
and
human
health:
Studies
in
an
occupational
cohort.
Doctoral
thesis,
Division
of
Environmental
and
Occupational
Health,
University
of
Minnesota.

Gilliland,
FD;
Mandel,
JS.
(
1993)
Mortality
among
employees
of
a
perfluorooctanoic
acid
production
plant.
JOM
35(
9):
950­
954.

Gilliland,
FD;
Mandel,
JS.
(
1996)
Serum
perfluorooctanoic
acid
and
hepatic
enzymes,
lipoproteins,
and
cholesterol:
A
study
of
occupationally
exposed
men.
Am
J
Ind
Med
29:
560­
568.

Gortner,
EG.
(
1981)
Oral
teratology
study
of
T­
2998CoC
in
rats.
Safety
Evaluation
Laboratory
and
Riker
Laboratories,
Inc.
Experiment
Number:
0681TR0110,
December
1981.

Gortner,
EG.
(
1982)
Oral
teratology
study
of
T­
3141CoC
in
rabbits.
Safety
Evaluation
Laboratory
and
Riker
Laboratories,
Inc.
Experiment
Number:
0681TB0398,
February
1982.

Griffith,
FD;
Long,
JE.
(
1980)
Animal
toxicity
studies
with
ammonium
perfluorooctanoate.
Am
Ind
Hyg
Assoc
J
41(
8):
576­
583.

Hanhijarvi,
H;
Ophaug,
RH;
Singer,
L.
(
1982)
The
sex­
related
difference
in
perfluorooctanoate
excretion
in
the
rat.
Proc
Soc
Exp
Biol
Med
171:
50­
55.

Hanhijarvi,
H;
Ylinen,
M;
Kojo,
A;
et
al.
(
1987)
Elimination
and
toxicity
of
perfluorooctanoic
acid
during
subchronic
administration
in
the
Wistar
rat.
Pharmacol.
Toxicol.
61:
66­
68.

Hanhijarvi,
H;
Ylinen,
M;
Haaranen,
T;
et
al.
(
1988)
A
proposed
species
difference
in
the
renal
excretion
of
perfluorooctanoic
acid
in
the
beagle
dog
and
rat.
In:
New
developments
in
biosciences:
their
implications
for
laboratory
animal
science.
Beynen,
AC;
Solleveld,
HA,
Eds.
Martinus
Nijhoff
Publishers.
Dordrecht,
Netherlands.

Hansch,
C;
Leo,
A;
Eds.
(
1979)
The
fragment
method
of
calculated
partition
coefficients.
In:
Substituent
Constants
for
Correlation
Analysis
and
Chemistry
and
Biology,
Chapter
IV.
John
Wiley
and
Sons,
Inc.
58
Johnson,
JD.
(
1995a)
Final
report,
analytical
study,
single­
dose
intravenous
pharmacokinetic
study
of
T­
6067
in
rabbits.
Study
Number:
AMDT­
120694.1.
3M
Environmental
Technology
&
Services,
St.
Paul,
MN.

Johnson,
JD.
(
1995b)
Final
report,
analytical
study,
single­
dose
absorption/
toxicity
study
of
T­
6067,
T­
6068,
and
T­
6069
in
rabbits.
Study
Number:
AMDT­
011095.1.
3M
Environmental
Technology
&
Services,
St.
Paul,
MN.

Johnson,
JD;
Gibson,
SJ;
Ober,
RE.
(
1984)
Cholestyramine­
enhanced
fecal
elimination
of
carbon­
14
in
rats
after
administration
of
ammonium
[
14C]
perfluorooctanoate
or
potassium
[
14C]
perfluorooctanesulfonate.
Fundam
Appl
Toxicol
4:
972­
976.

Kennedy,
G.
L.
(
1985).
Dermal
toxicity
of
ammonium
perfluorooctanoate.
Toxicol.
Appl.
Pharmacol.
81(
2):
348­
355.

Kennedy,
G.
L;
Hall,
GT;
Brittelli,
MR;
Barnes,
JR;
Chen,
HC.
(
1986).
Inhalation
toxicity
of
ammonium
perfluorooctanoate.
Food
Chem.
Toxicol.
24(
12):
1325­
1329.

Kudo,
N;
Katakura,
M;
Sato,
Y;
Kawashima,
Y.
(
2002).
Sex
hormone­
regulated
renal
transport
of
perfluorooctanoic
acid.
Chem.
Biol.
Interact.
139:
301­
316.

Lines,
D;
Sutcliffe,
H.
(
1984)
Preparation
and
properties
of
some
salts
of
perfluorooctanoic
acid.
Journal
of
Fluorine
Chemistry
25:
505­
512.

Obourn,
J.
D.,
Frame.
S.
R.,
Bell,
R.
H.
Jr.,
Longnecker,
D.
S.,
Elliott,
G.
S.
and
Cook,
J.
C.
1997.
Mechanisms
for
the
pancreatic
oncogenic
effects
of
the
peroxisome
proliferator
Wyeth­
14,643.
Toxicol.
Appl.
Pharmacol.
145:
425­
436.

Olsen,
GW;
Gilliland,
FD;
Burlew,
MM;
et
al.
(
1998a)
An
epidemiologic
investigation
of
reproductive
hormones
in
men
with
occupational
exposure
to
perfluorooctanoic
acid.
JOEM
40(
7):
614­
622.

Olsen,
GW;
Burris,
JM;
Burlew,
MM;
Mandel,
JH.
(
1998b)
3M
Final
report:
an
epidemiologic
investigation
of
plasma
cholecystokinin,
hepatic
function
and
serum
perfluorooctanoic
acid
levels
in
production
workers.
3M
Company,
St.
Paul,
MN.
September
4.

Olsen,
G.
W.,
Burris,
J.
M.,
Burlew,
M.
M.,
Mandel,
J.
H.
2000.
Plasma
cholecystokinin
and
hepatic
enzymes,
cholesterol
and
lipoproteins
in
ammonium
perfluorooctanoate
production
workers.
Drug
Chem
Tox.
23(
4):
603­
620.
59
Olsen,
GW,
Logan,
PW,
Simpson,
CA,
Burris,
JM,
Burlew,
MM,
Lundberg,
JK,
Mandel,
JH.
2001a.
Descriptive
summary
of
serum
fluorochemical
levels
among
employee
participants
of
the
year
2000
Decatur
fluorochemical
medical
surveillance
program.
Final
Report.
March
19,
2001.

Olsen,
GW,
Schmickler,
M,
Tierens,
JM,
Logan,
PW,
Burris,
JM,
Burlew,
MM,
Lundberg,
JK,
Mandel,
JH.
2001b.
Descriptive
summary
of
serum
fluorochemical
levels
among
employee
participants
of
the
year
2000
Antwerp
fluorochemical
medical
surveillance
program.
Final
Report.
March
19,
2001.

Olsen,
GW,
Madsen,
DC,
Burris,
JM,
Mandel,
JH.
2001c.
Descriptive
summary
of
serum
fluorochemical
levels
among
236
building
employees.
Final
Report.
March
19,
2001.

Olsen,
GW,
Hansen,
Clemen,
LA,
Burris,
JM,
Mandel,
JH.
2001d.
Identification
of
Fluorochemicals
in
Human
Tissue.
Final
Report.
Epidemiology,
220­
3W­
05,
Medical
Department,
3M
Company,
St.
Paul,
MN
55144.

Olsen
GW,
Burlew
MM,
Burris
JM,
Mandel
JH.
2001e.
A
cross­
sectional
analysis
of
serum
perfluorooctanesulfonate
(
PFOS)
and
perfluorooctanoate
(
PFOA)
in
relation
to
clinical
chemistry,
thyroid
hormone,
hematology
and
urinalysis
results
from
male
and
female
employee
participants
of
the
2000
Antwerp
and
Decatur
fluorochemical
medical
surveillance
program.
Final
report.
3M
Medical
Department.

Olsen,
G.
W.,
Burlew,
M.
M,
Burris,
J.
M.,
Mandel,
J.
H.
2001f.
A
Longitudinal
Analysis
of
Serum
Perfluorooctanesulfonate
(
PFOS)
and
Perfluorooctanoate
(
PFOA)
Levels
in
Relation
to
Lipid
and
Hepatic
Clinical
Chemistry
Test
Results
from
Male
Employee
Participants
of
the
1994/
95,
1997,
and
2000
Fluorochemical
Medical
Surveillance
Program.
3M
Final
Report.

Olsen,
GW,
Burlew,
MM,
Hocking,
BB,
Skratt,
JC,
Burris,
JM,
Mandel,
JH.
2001g.
An
epidemiologic
analysis
of
episodes
of
care
of
3M
Decatur
chemical
and
film
plant
employees,
1993­
1998.
Final
Report.
May
18,
2001.

Olsen,
GW,
Burris,
JM,
Lundberg,
JK,
Hansen,
KJ,
Mandel,
JH,
Zobel,
LR.
2002a.
Identification
of
fluorochemicals
in
human
sera.
I.
American
Red
Cross
adult
blood
donors.
Final
report.
3M
Medical
Department.

Olsen,
GW,
Burris,
JM,
Lundberg,
JK,
Hansen,
KJ,
Mandel,
JH,
Zobel,
LR.
2002b.
Identification
of
fluorochemicals
in
human
sera.
II.
Elderly
participants
of
the
Adult
Changes
in
Thought
Study,
Seattle,
Washington.
Final
Report.
3M
Medical
Department.
60
Olsen,
GW,
Burris,
JM,
Lundberg,
JK,
Hansen,
KJ,
Mandel,
JH,
Zobel,
LR.
2002c.
Identification
of
fluorochemicals
in
human
sera.
III.
Pediatric
participants
in
a
Group
A
Streptococci
clinical
trial
investigation.
Final
Report.
3M
Medical
Department.

O'Malley,
KD;
Ebbins,
KL.
(
1981)
Repeat
application
28
day
percutaneous
absorption
study
with
T­
2618CoC
in
albino
rabbits.
Riker
Laboratories,
St.
Paul,
MN.

Ophaug,
RH;
Singer,
L.
(
1980)
Metabolic
handling
of
perfluorooctanoic
acid
in
rats.
Proc
Soc
Exp
Biol
Med
163:
19­
23.

Simister,
EA;
Lee,
EM;
Lu,
JR;
Thomas,
RK;
Ottewill,
RH;
Rennie,
AR;
Penfold,
J.
(
1992).
Adsorption
of
ammonium
perfluoroocanoate
and
ammonium
decanoate
at
the
air­
solution
interface.
J
Chem
Soc,
Faraday
Trans
88(
20):
3033­
41.

Staples,
RE;
Burgess,
BA;
Kerns,
WD.
(
1984)
The
embryo­
fetal
toxicity
and
teratogenic
potential
of
ammonium
perfluorooctanoate
(
APFO)
in
the
rat.
Fundam
Appl
Toxicol
4:
429­
440.

Ubel
FA;
Sorenson,
SD;
Roach,
DE.
(
1980)
Health
status
of
plant
workers
exposed
to
fluorochemicals­­
a
preliminary
report.
Am
Ind
Hyg
Assoc
J
41:
584­
589.

U.
S.
EPA.
(
1991)
Guidelines
for
developmental
toxicity
risk
assessment.
Federal
Register
56(
234):
63798­
63826.

USEPA
2002a.
Memorandum
from
Dr.
Ralph
Cooper,
NHEERL,
to
Dr.
Jennifer
Seed,
dated
October
2,
2002.

USEPA
2002b.
Memorandum
from
Dr.
Elizabeth
Margosches
to
Dr.
Katherine
Anitole,
dated
October
21,
2002.

Vanden
Heuvel,
JP;
Kuslikis,
BI;
Peterson,
RE.
(
1991a)
Covalent
binding
of
perfluorinated
fatty
acids
to
proteins
in
the
plasma,
liver
and
testes
of
rats.
Chem­
Biol
Interact
82:
317­
328.

Vanden
Heuvel,
JP;
Kuslikis,
BI;
Van
Rafelghem,
ML;
et
al.
(
1991b)
Tissue
distribution,
metabolism,
and
elimination
of
perfluorooctanoic
acid
in
male
and
female
rats.
J
Biochem
Toxicol
6(
2):
83­
92.

Vanden
Heuvel,
JP;
Davis,
JW;
Sommers,
R;
et
al.
(
1992)
Renal
excretion
of
perfluorooctanoic
acid
in
male
rats:
Inhibitory
effect
of
testosterone.
J
Biochem
Toxicol
7(
1):
31­
36.
61
Ylinen,
M;
Hanhijarvi,
H;
Jaakonaho,
I;
et
al.
(
1989)
Stimulation
by
estradiol
of
the
urinary
excretion
of
perfluorooctanoic
acid
in
the
male
rat.
Pharmacol
Toxicol
65:
274­
277.

Ylinen,
M;
Kojo,
A;
Hanhijdrvi,
H;
et
al.
(
1990)
Disposition
of
perfluorooctanoic
acid
in
the
rat
after
single
and
subchronic
administration.
Bull
Environ
Contam
Toxicol
44:
46­
53.

York,
RG.
(
2002)
Oral
(
gavage)
two­
generation
(
one
litter
per
generation)
reproduction
study
of
ammonium
perfluorooctanoic
(
APFO)
in
rats.
Argus
Research
Laboratories,
Inc.
Protocol
Number:
418­
020,
Sponsor
Study
Number:
T­
6889.6,
March
26,
2002.
