1
FIFRA
Scientific
Advisory
Panel
Background
Document
May
4­
6,
2004
meeting
Proposed
Hazard
Identification
Methodology
for
Assessment
of
Dermal
Sensitization
Risk
Timothy
F.
McMahon,
Ph.
D.
Jonathan
Chen,
Ph.
D.
Antimicrobials
Division
Office
of
Pesticide
Programs
U.
S.
Environmental
Protection
Agency
Washington,
D.
C.
20460
2
Introduction
Among
the
many
types
of
risk
assessments
conducted
in
the
EPA's
Office
of
Pesticide
Programs
(
OPP)
is
an
assessment
of
the
dermal
sensitization
potential
of
pesticide
chemicals.
As
noted
in
40
CFR
798.4100,
"
Information
derived
from
tests
for
skin
sensitization
serves
to
identify
the
possible
hazard
to
a
population
repeatedly
exposed
to
a
test
substance."
Information
from
this
test
is
qualitatively
assessed
and,
if
appropriate,
precautionary
language
is
included
on
the
pesticide
label
as
well
as
the
Occupational
Safety
and
Health
Administration's
Material
Data
Safety
Sheets
(
MSDS).
Occupational
dermal
exposures
to
known
or
suspected
dermal
sensitizing
pesticide
chemicals
can
be
then
dealt
with
appropriately,
either
through
engineering
controls
or
use
of
personal
protective
equipment.
Non­
occupational
exposures
can
normally
be
dealt
with
through
precautionary
label
statements.

Data
available
through
the
National
Institute
for
Occupational
Safety
and
Health
(
NIOSH)

indicate
that
occupational
exposure
to
dermal
irritants
and
contact
allergens
accounts
for
a
significant
number
of
occupational
illness
and
that
chemical
agents
are
the
most
frequent
cause
of
such
illness.
Specific
national
occupational
disease
and
illness
data
are
available
from
the
U.
S.

Bureau
of
Labor
Statistics
(
BLS).
The
BLS
conducts
annual
surveys
of
approximately
174,000
employers,
selected
to
represent
all
private
industries
in
the
U.
S.
The
goal
is
to
ascertain
the
total
numbers
and
incidence
rates
of
occupational
injuries
and
illnesses.
The
survey
results
are
then
projected
to
estimate
the
numbers
and
incidence
rates
of
injuries
and
illnesses
in
the
American
working
population.
All
occupational
skin
diseases
or
disorders,
including
occupational
contact
dermatitis
(
OCD),
are
tabulated
in
this
survey.
In
1999,
of
over
372,000
occupational
illnesses
reported,
12%
were
reported
as
skin
diseases/
disorders,
making
this
the
most
common
non
3
trauma
related
occupational
illness
(
NIOSH,
2001).
The
economic
impact
of
occupational
skin
diseases/
disorders
is
also
significant.
As
measured
by
direct
medical
cost
and
worker
compensation,
the
total
annual
cost
of
occupational
skin
disease
may
range
from
$
222
million
to
$
1
billion
(
NIOSH,
2001).

While
pesticide
chemicals
can
usually
be
labeled
to
warn
of
potential
dermal
sensitization
effects,

there
also
exists
the
manufacture
of
"
treated
articles
or
substances"
(
40
CFR
152.25),
in
which
a
registered
pesticide
is
incorporated
into
the
article
to
protect
the
integrity
of
the
article
or
substance
itself
(
such
as
paint
treated
with
a
pesticide
to
protect
the
paint
coating
or
wood
products
treated
to
protect
the
wood
against
fungal
or
insect
decay).
Under
such
circumstances
of
use,
the
general
public
may
unknowingly
be
exposed
to
pesticide
chemical
residue
in
the
treated
article.
Therefore,
prior
to
such
use,
the
pesticide
chemical
must
first
be
registered
under
FIFRA,
which
requires
that
the
manufacturer
of
the
pesticide
demonstrate
that
it
can
be
used
without
unreasonable
risks
to
humans
or
the
environment.
Treated
articles
such
as
preserved
wood,
however,
do
not
bear
a
pesticide
label
or
effectively
use
other
communication
methods
to
inform
and
protect
people
against
potential
hazards,
including
the
potential
for
dermal
sensitization
Purpose
The
focus
of
this
paper
is
on
the
Agency's
interest
in
developing
the
foundation
of
a
scientifically
sound
approach
to
quantitative
assessment
of
dermal
sensitization
risk
to
pesticide
chemicals,

including
pesticide
chemicals
that
are
incorporated
into
other
materials
(
i.
e.
treated
articles).
The
4
Agency
is
interested
in
obtaining
expert
advice
on
methods
published
in
the
scientific
literature
that
have
been
proposed
for
use
in
determining
induction
thresholds
and
elicitation
thresholds
for
chemicals
known
or
suspected
of
causing
allergic
contact
dermatitis.
The
Agency
is
also
interested
in
seeking
advice
on
whether
there
are
any
susceptibility
issues
in
the
general
population
with
respect
to
development
of
allergic
contact
dermatitis,
including
any
potential
special
sensitivity
of
children.
The
Agency
will
present
hexavalent
chromium
as
a
case
study
of
a
known
dermal
sensitizer,
and
possible
approaches
to
quantitating
risk
of
allergic
contact
dermatitis
from
exposure
to
hexavalent
chromium.

Dermal
Sensitization
Dermal
sensitization,
also
known
as
allergic
contact
dermatitis
(
ACD),
delayed
contact
hypersensitivity,
or
Type
IV
allergic
contact
dermatitis,
is
defined
by
Marzulli
and
Maibach
(
1996)

as
a
delayed,
immunologically
mediated,
inflammatory
skin
disease
consisting
of
various
degrees
of
erythema,
edema,
and
vesiculation.
Kimber
(
2004)
has
also
defined
sensitization
as
"
stimulation
by
chemical
allergen
(
in
an
inherently
susceptible
individual)
of
an
immune
response
of
the
quality
and
vigor
required
to
permit
the
provocation
of
an
elicitation
reaction
upon
subsequent
encounter
with
the
same
chemical."
ACD
is
typically
characterized
by
two
phases,

termed
induction
and
elicitation.
Induction
occurs
when
there
is
an
exposure
of
sufficient
magnitude
and/
or
duration
to
activate
specific
immune
mechanisms
resulting
in
the
acquisition
of
sensitization,
while
elicitation
occurs
from
a
subsequent
exposure
to
the
same
chemical
allergen.

As
it
is
generally
recognized
that
a
certain
level
of
allergen
exposure
must
be
attained
to
induce
immune
activation
in
susceptible
individuals,
dermal
sensitization
is
characterized
as
a
threshold
5
type
of
response.
However,
dose­
response
relationships
are
observed
for
both
the
induction
and
elicitation
phases
of
ACD
(
Gerberick
and
Robinson,
2000;
Kimber
et
al.,
2003;
Scott
et
al.,

2002),
and
thresholds
for
induction
can
be
reached
following
either
a
single
sufficiently
high
amount
of
exposure
to
the
allergenic
chemical,
or
after
contact
with
a
large
area
of
skin,
or
as
a
consequence
of
repeated
skin
applications
(
Marzulli
and
Maibach,
1996).
Experiments
with
2,4­

dinitrochlorobenzene
(
DNCB)
and
other
sensitizers
have
shown
that
a
single
contact
can
be
sufficient
for
sensitization,
but
less
data
exist
of
the
relationship
between
lower
area
doses
and
repeated
contacts
over
a
longer
time
period
(
Griem
et
al.,
2003).
A
study
summarized
by
Griem
et
al.,
(
2003)
and
conducted
by
Vandenburg
and
Epstein
(
1963)
in
which
previously
non­
sensitized
persons
were
exposed
to
nickel
chloride
suggested
that
subclinical
sensitization
occurred
in
some
of
the
subjects
who
responded
negatively
from
the
first
test,
as
an
increased
percentage
of
nonsensitized
subjects
responded
positively
upon
a
repetition
of
the
test
four
months
later.
More
studies
are
needed
in
this
area.

To
be
capable
of
inducing
an
allergenic
response,
the
chemical
itself
must
possess
certain
characteristics.
Those
chemicals
able
to
cause
sensitization
are
usually
low
molecular
weight
protein­
reactive
substances
that
can
gain
access
to
the
viable
epidermis
via
the
stratum
corneum,

and
are
also
able
to
cause
sufficient
local
trauma
to
induce
cutaneous
cytokines
and
be
inherently
antigenic
and
recognized
by
responsive
T
lymphocytes.

Once
through
the
stratum
corneum,
the
allergen
makes
contact
with
the
Langerhans
cell,
a
member
of
the
bone­
marrow
derived
dendritic
cell
family
whose
function
is
to
act
as
a
`
sentinel'

cell
and
serve
as
a
trap
for
antigens
entering
the
skin
(
Kimber,
in
Marzulli
and
Maibach,
1996).
6
ELICITATION
INDUCTION
LOW
MOLECULAR
WT
ALLERGEN
MIGRATION
TO
LOCAL
LYMPH
NODE
IL­
1
 ,
IL­
6,
IL­
12
IL­
1
 ,
TNF­
 ,
GM­
CSF
T­
CELL
LYMPHOCYTE
PROLIFERATION
ICAM­
1
LANGERHANS
CELL
(
LC)

"
PRIMED"
LYMPHOCYTES
SPECIFIC
INFLAMMATORY
RESPONSE
CYTOKINES,
COSTIMULATORY,
ADHESION
MOLECULES
INCREASE
CELLULAR
INFLUX
*
Illustration
by
D.
Sailstad
EDEMA
AND
ERYTHEMA
Langerhans
cells
then
direct
the
allergen
to
a
regional
lymph
node,
where
interaction
with
T
lymphocytes
occurs,
followed
by
proliferation
of
lymphocytes
that
have
been
`
primed'
to
react
against
the
presented
antigen.
A
subsequent
contact
with
the
allergen
will
result
in
elicitation
of
the
sensitization
response
due
to
the
reaction
of
sensitized
lymphocytes
with
the
allergen.
The
process
is
illustrated
in
the
following
diagram:

From:
Sailstad,
D.
M.
(
2003):
Allergic
Contact
Hypersensitivity:
Mechanisms
and
Methods.
In:
Alternative
Toxicological
Methods.
Harry
Salem
and
Sidney
Katz,
eds.
,
CRC
Press,
Boca
Raton,
Florida.
Pages
193­
205.
7
It
should
also
be
noted
that
in
addition
to
Langerhans
cells,
epidermal
cytokines
and
chemokines
may
also
play
a
role
in
the
development
of
the
sensitization
response.
This
is
based
on
the
observation
that
the
functional
activity
of
Langerhans
cells
and
presumably
other
cutaneous
antigen­
presenting
cells,
is
regulated
largely
by
the
availability
of
cytokines
(
Kimber,
in
Marzulli
and
Maibach,
1996).

While
it
is
generally
accepted
that
the
necessary
exposure
for
induction
of
dermal
sensitization
is
greater
than
the
exposure
needed
to
elicit
sensitization,
it
is
important
to
recognize
that
thresholds
are
largely
determined
by
the
potency
of
the
chemical
allergen
and
that
induction
and
elicitation
thresholds
vary
among
individuals
(
Kimber
et
al.,
2003).
It
is
thus
necessary
to
consider
doseresponse
relationships
in
establishing
"
safe"
levels
for
prevention
of
induction
and
elicitation.
A
recent
investigation
by
Scott
et
al.
(
2002)
examined
the
quantitative
relationship
between
sensitization
and
elicitation
concentrations
and
the
ability
to
elicit
an
ACD
reaction
in
murine
models.
Using
two
established
sensitizers
(
DNCB
and
squaric
acid
dibutyl
ester
(
SADBE),

doseresponse
relationships
were
determined
using
the
LLNA
to
derive
EC3
values
for
both
chemicals.

Then,
for
DNCB,
various
elicitation
concentrations
were
tested
in
mice
that
had
been
induced
to
various
concentrations
of
DNCB.
As
the
sensitizing
concentration
of
DNCB
increased,
it
was
observed
that
the
concentration
required
to
elicit
measurable
sensitization
in
sensitized
mice
decreased.
That
is,
mice
who
had
been
sensitized
to
higher
concentrations
of
DNCB
required
less
DNCB
to
elicit
a
measurable
sensitization
response.
For
SADBE,
two
different
elicitation
8
concentrations
were
compared
over
a
range
of
induction
concentrations.
The
group
given
the
higher
elicitation
concentration
showed
measurable
sensitization
at
lower
induction
concentrations.
The
results
of
this
study
suggested
that,
as
the
induction
dose
was
increased
(
using
DNCB),
the
concentration
required
for
elicitation
was
decreased.
Correspondingly,
as
the
elicitation
concentration
was
increased
(
using
SADBE),
the
concentration
required
for
induction
was
decreased.

Hazard
Identification
It
is
desirable
to
be
able
to
conduct
quantitative
assessments
for
dermal
sensitization
in
order
to
prevent
consumers
especially
from
developing
ACD
when
dermal
contact,
including
repeated
dermal
contact,
could
not
be
completely
avoided
(
Griem
et
al.,
2003).
In
the
case
of
hexavalent
chromium
as
a
component
of
treated
wood,
this
is
important,
as
there
would
be
repeated
dermal
contact
with
the
treated
wood
surface
when
this
treated
wood
is
used
in
residential
decking
and
playground
structures.

There
are
several
accepted
methods
for
hazard
identification
of
dermal
sensitization,
including
the
Buehler
occluded
patch
test,
the
guinea
pig
maximization
test,
and
the
murine
local
lymph
node
assay
(
LLNA).
The
guinea
pig
maximization
test
and
the
Buehler
test,
while
providing
reliable
information
on
whether
a
substance
is
a
skin
sensitizer
or
not,
are
best
suited
only
for
hazard
identification.
9
By
contrast
with
the
Buehler
test
and
maximization
test,
the
Mouse
Local
Lymph
Node
Assay
(
LLNA)
is
a
more
recent
test
method
for
assessing
the
allergic
contact
dermatitis
(
skin
sensitization)
potential
of
chemicals,
specifically
the
induction
phase
of
sensitization.
Using
the
incorporation
of
radiolabeled
thymidine
or
iododeoxyuridine
into
DNA,
the
LLNA
measures
lymphocyte
proliferation
in
the
draining
lymph
nodes
of
mice
topically
exposed
to
the
test
article.

The
stimulation
index
(
ratio
of
lymphocyte
proliferation
in
treated
mice
compared
to
controls)
is
used
as
the
indicator
of
potential
sensitization.
In
1998,
following
review
by
the
FIFRA
Scientific
Advisory
Panel
(
SAP),
the
LLNA
was
incorporated
as
a
screening
test
in
OPPTS
Test
Guideline
870.2600
Skin
Sensitization.
In
1999,
the
Interagency
Coordinating
Committee
on
the
Validation
of
Alternative
Methods
(
ICCVAM)
Immunotoxicity
Working
Group
(
IWG)

recommended
the
LLNA
as
a
stand­
alone
alternative
for
contact
sensitization
hazard
assessment,
provided
that
certain
protocol
modifications
were
made.
Following
additional
studies
to
validate
the
method,
the
FIFRA
SAP
agreed
with
the
Agency
proposal
that
the
LLNA
was
applicable
for
testing
chemicals
to
elicit
contact
sensitization
and
should
be
considered
a
preferred,
stand­
alone
assay.
The
OPPTS
guideline
870.2600
(
Skin
Sensitization)
has
been
revised
to
include
the
LLNA
as
a
stand­
alone
assay
for
appropriate
applications.
The
OPPTS
guideline
has
also
been
harmonized
with
OECD's
Guideline
429
for
LLNA,
which
was
adopted
in
April
2002.

Although
the
LLNA
has
not
been
validated
for
determination
of
sensitization
potency
of
chemical
allergens
and
sensitization
to
metals
has
not
been
extensively
studied
,
approaches
for
determination
of
quantitative
assessment
of
sensitization
induction
thresholds
have
been
proposed
in
the
scientific
literature
using
LLNA
data
(
Gerberick
2000,
2001;
Griem
et
al.,
2003).
It
may
10
therefore
be
useful
to
consider
data
from
the
LLNA
as
a
starting
point
for
quantitation
of
induction
thresholds,
as
dose
response
data
can
be
generated
from
this
assay,
and
a
NOAEL
can
potentially
be
identified
(
Felter
et
al.,
2003).
However,
while
the
LLNA
has
been
validated
and
accepted
as
a
stand­
alone
test
method
for
assessment
of
dermal
sensitization
potential
by
the
Interagency
Coordinating
Committe
on
the
Validation
of
Alternative
Test
Methods
(
ICCVAM
,

1999),
the
test
itself
has
not
been
validated
for
its
utility
in
dermal
risk
assessment
(
Felter
et
al.,

2003).
It
has
been
proposed
to
group
sensitizing
chemicals
according
to
relative
potency
as
determined
in
the
LLNA,
and
then
compare
these
categories
with
the
relative
sensitization
potency
in
humans
on
the
basis
of
clinical
experience
and/
or
prevalence
of
ACD
in
the
population
(
Griem
et
al,
2003).
While
a
good
correlation
between
LLNA
results
and
sensitization
potency
in
humans
has
been
reported
(
Griem
et
al.,
2003),
there
is
not
yet
general
agreement
on
categories
and
ranges
that
should
be
used
in
classification
of
relative
potency.

Induction
Threshold
Methods
Gerberick
(
2000,
2001)
proposed
a
methodology
for
determination
of
a
`
sensitization
reference
dose'
for
sensitizers
in
consumer
products.
The
lower
boundary
of
the
potency
category
for
a
sensitizing
chemical
is
used
as
the
starting
point,
with
application
of
uncertainty
factors
for
interindividual
variability,
product
matrix
effects,
and
use
pattern.
This
approach
was
applied
to
the
fragrance
component
cinnamic
aldehyde
and
the
preservative
methylchloroisothiazolinone/
methylisothiazolinone
for
which
both
LLNA
and
human
sensitization
potency
were
available
(
Griem
et
al.,
2003).
11
A
use
for
the
LLNA
in
quantitative
risk
assessment
was
investigated
by
Griem
et
al
(
2003).

Identification
of
known
human
sensitizing
chemicals
for
which
both
an
EC3
value
from
an
LLNA
test
and
a
NOAEL
or
LOAEL
from
human
repeat
insult
patch
tests
(
HRIPT)
or
the
human
maximization
test
(
HMT)
were
available
were
identified.
The
reported
concentrations
were
converted
into
specific
and
molar
area
doses.
Comparison
of
the
area
doses
of
the
LLNA
and
human
test
results
indicated
that
sensitization
thresholds
were
similar
in
mice
and
humans
despite
the
fact
that
the
area
doses
for
different
chemicals
ranged
over
several
orders
of
magnitude
(
Griem
et
al.,
2003).
It
was
concluded
from
this
analysis
that
the
LLNA
EC3
value
is
a
useful
measure
of
sensitizing
potency
in
humans,
and
that
the
EC3
value
can
be
used
as
a
surrogate
value
for
the
human
NOAEL
which,
in
turn,
can
be
used
as
a
starting
point
in
quantitative
risk
assessment.

Elicitation
threshold
Methods
The
above
approaches
describe
methods
that
may
be
used
to
estimate
thresholds
for
safe
area
doses
that
are
considered
protective
against
induction
of
sensitization.
There
are
also
proposed
approaches
for
estimation
of
safe
area
doses
that
are
considered
protective
against
elicitation
of
sensitization
in
already
sensitized
persons.
By
inference,
protection
against
elicitation
would
also
be
protective
of
induction,
as
thresholds
for
induction
are
generally
higher
than
those
for
elicitation
(
Kimber
et
al.,
2003).

An
approach
to
estimate
an
acceptable
area
dose
for
protection
against
elicitation
is
the
concept
of
the
Minimum
Elicitation
Threshold,
or
MET.
This
approach
has
been
discussed
in
12
previous
publications
(
Nethercott
et
al.,
1994;
Zewdie,
1998;
NJDEP,
1998;
Basketter
et
al.,

2003)
specifically
with
respect
to
hexavalent
chromium.
The
concept
behind
the
MET
is
that
there
is
an
`
elicitation
threshold'
below
which
no
sensitization
reaction
is
expected;
thus,
the
MET
is
analogous
to
an
RfD
(
Horowitz
and
Finley,
1994).
The
estimation
of
an
MET
is
usually
based
on
the
results
of
tests
in
previously
sensitized
individuals;
thus,
the
MET
is
considered
protective
of
elicitation
reactions.
However,
there
has
not
been
an
extensive
discussion
of
the
criteria
for
employing
this
concept
for
purposes
of
risk
assessment.
Nethercott
(
1994)
calculated
a
value
of
0.089
ug/
cm2
as
a
10%
MET
(
i.
e.
the
concentration
at
which
10%
of
the
study
group
responded)

based
on
results
of
a
human
repeat
insult
patch
test
in
54
chromium­
sensitized
volunteers
and
claimed
that
this
value
should
be
protective
against
ACD
for
hexavalent
chromium
in
at
least
99.99%
of
the
population
exposed
to
contaminated
soil.
Paustenbach
(
1992)
estimated
a
10%

threshold
response
of
54
parts
per
million
for
hexavalent
chromium
in
soil,
but
no
discussion
of
the
relevance
of
the
10%
response
level
was
presented.
Two
states
(
Massachusetts
and
New
Jersey)
have
proposed
soil
cleanup
standards
based
on
an
ACD
endpoint
using
either
the
human
patch
test
data
of
Nethercott
(
1994)
or
the
human
forearm
water
exposure
data
of
Fowler
(
2000).

Griem
et
al.
(
2003)
also
proposed
an
approach
for
derivation
of
a
safe
area
dose
at
which
ACD
would
not
be
manifest
in
sensitized
individuals.
As
several
of
the
factors
that
influence
induction
of
sensitization
(
skin
penetration,
uptake
by
antigen­
presenting
cells,
metabolism)
are
also
relevant
for
elicitation,
it
was
postulated
that
a
correlation
between
the
induction
potency
and
elicitation
potency
of
a
chemical
could
be
established.
However,
comparison
of
induction
and
elicitiation
area
doses
from
limited
data
in
humans
did
not
show
an
obvious
correlation.
While
induction
13
threshold
doses
spanned
five
orders
of
magnitude,
values
for
elicitation
were
mainly
within
one
order
of
magnitude.
Using
the
log
transformation
of
the
ratio
of
induction/
elicitation
to
elicitation,
it
was
proposed
that
the
ratio
of
the
induction/
sensitization
threshold
could
be
predicted
on
the
basis
of
an
established
sensitization
(
induction)
threshold.
Examples
of
the
derivation
of
safe
area
doses
for
three
chemicals
(
MCI/
MI,
a
preservative
in
many
cosmetics
and
household
products;
cinnamic
aldehyde,
a
fragrance
and
flavor
ingredient;
and
nickel)
were
presented.
For
strong
sensitizers
such
as
MCI/
MI,
it
was
demonstrated
that
the
safe
area
doses
for
induction
and
elicitation
were
close
together,
while
for
relatively
weaker
sensitizers
such
as
nickel,
the
safe
area
doses
for
induction
and
elicitation
were
farther
apart,
consistent
with
the
mathematical
relationship
of
the
ratio
of
induction/
elicitation
threshold
vs.
sensitization
threshold.

That
is,
for
lower
potency
sensitizers,
a
relatively
high
area
dose
may
be
needed
to
cause
induction
of
sensitization,
but
elicitation
may
be
possible
with
much
lower
area
doses,
while
for
a
potent
sensitizer,
the
area
dose
needed
for
induction
is
close
to
the
area
dose
that
will
elicit
the
reaction
in
a
sensitized
individual.

Uncertainty
After
the
appropriate
level
of
concern
has
been
identified
(
e.
g.
NOAEL
determination,
for
example),
areas
of
uncertainty
need
to
be
considered
in
extrapolating
the
result
to
conditions
relevant
to
the
human
exposure
of
interest.
Areas
of
uncertainty
that
have
been
identified
for
dermal
risk
assessment
include
(
1)
interspecies
variability/
susceptibility
(
i.
e.
extrapolation
from
animals
to
humans);
(
2)
intra­
species
variations
in
response
within
humans;
(
3)
vehicle
or
product
matrix
effects;
and
(
4)
exposure
considerations
(
i.
e.,
area
of
the
body
exposed,
repeated
14
exposures).
Briefly,
and
as
discussed
in
more
detail
by
Felter
et
al.
(
2002)
and
Griem
et
al.

(
2003),
the
inter­
individual
variation
in
response
to
induction
and
elicitation
of
dermal
sensitization
must
be
taken
into
account,
as
there
may
be
differences
in
response
based
on
age,

sex,
and
genetic
factors,
or
health
status
of
the
skin.
In
addition,
formulation
of
chemical
allergens
into
product
matrices
that
may
result
in
an
enhancement
or
inhibition
of
ACD
must
be
considered
in
the
risk
assessment
paradigm,
as
must
also
extrapolating
from
experimental
conditions
to
realworld
exposure
conditions,
i.
e.
site
of
the
body
exposed
,
effects
of
occlusion,
and
environmental
conditions
(
temperature,
humidity,
and
repeated
dermal
exposures).
For
each
of
the
4
areas,
a
range
from
1­
10
has
been
suggested
for
uncertainty
in
each
area.

Populations
of
Concern
Approaches
to
estimating
safe
area
doses
for
ACD
have
been
proposed
using
EC3
values
derived
from
the
murine
LLNA
(
Gerberick
and
Robinson,
2000;
Felter
et
al.,
2003,
and
MET
values
from
human
repeat
insult
patch
tests
(
Nethercott
et
al.,
1994;
Basketter
et
al.,
2001;
Hansen
et
al.,

2003).
Approaches
using
the
results
of
murine
LLNA
data
are
intended
to
estimate
area
doses
that
are
protective
against
induction
of
ACD,
while
use
of
the
MET
approach
is
intended
to
be
protective
against
elicitation
of
ACD
in
sensitized
persons.
Griem
et
al.
(
2003)
have
also
proposed
an
approach
for
calculation
of
safe
area
doses
designed
to
be
protective
against
both
induction
and
elicitation
of
ACD
through
use
of
appropriate
uncertainty
factors
applied
to
the
results
of
murine
LLNA
or
human
patch
test
study
results.
This
proposal
is
worth
examining,
as
it
has
been
acknowledged
in
the
past
for
sensitizing
chemicals
such
as
nickel
and
hexavalent
15
chromium
that
it
is
difficult
to
protect
individuals
who
are
already
sensitized
(
Felter
et
al.,
2003;

USEPA,
1998).

In
addition,
consideration
should
be
given
to
whether
there
are
potentially
susceptible
subpopulations
who
may
be
more
susceptible
to
the
induction
and/
or
elicitation
of
ACD.

Paustenbach
et
al.
(
1992)
and
Felter
et
al.
(
2002)
have
discussed
the
issue
of
whether
children
are
more
or
less
at
risk
for
development
of
ACD.
Paustenbach
et
al.
addressed
this
issue
specifically
for
hexavalent
chromium,
and
it
was
concluded
that
risk
to
children
ages
3
to
8
is
not
likely
to
be
greater
than
risk
to
adults
as
there
is
no
evidence
that
repeated
exposures
to
hexavalent
chromium
places
a
person
at
greater
risk
of
sensitization.
Felter
et
al.
suggested
that
infants
and
children
may
actually
be
at
lower
risk
for
development
of
ACD
based
on
data
gathered
from
dinitrochlorobenzene
and
pentadecylcatechol
(
poison
ivy
allergen).
These
views
are
somewhat
counter
to
the
opinion
of
Griem
et
al.
(
2003)
who
suggested
a
possible
higher
sensitizing
potency
of
a
chemical
upon
repeated
exposures.
This
would
make
sense
in
the
case
of
hexavalent
chromium,
as
the
significant
irritancy
of
the
chemical
could
lend
itself
to
an
increased
sensitizing
potency
by
allowing
more
chemical
to
penetrate
the
stratum
corneum.

Case
Study­
Hexavalent
Chromium
As
noted
in
the
1998
IRIS
Toxicological
Review
for
hexavalent
chromium
(
USEPA,
1998)
as
well
as
in
numerous
publications,
hexavalent
chromium
is
one
of
the
most
common
and
potent
contact
sensitizers.
Exposures
to
hexavalent
chromium
occur
in
a
number
of
occupational
settings
including
including
chrome
plating
baths,
chrome
colors
and
dyes,
cement,
tanning
agents,
wood
preservatives,
anticorrosive
agents,
welding
fumes,
lubricating
oils
and
greases,
16
cleaning
materials,
and
textiles
and
furs
(
USEPA,
2003).
Non­
occupational
exposures
to
hexavalent
chromium
have
also
been
noted
in
household
detergents
(
Basketter
et
al.,
2003;
Stern
et
al.,
1993)
as
well
as
in
cement.

Elicitation
thresholds
in
persons
sensitized
to
hexavalent
chromium
have
been
described
in
the
literature
(
Nethercott,
1994;
Fowler,
2000).
However,
there
are
no
recent
data
on
induction
thresholds
for
hexavalent
chromium.
Some
investigation
has
been
performed
on
the
question
of
induction
thresholds
in
general,
as
it
has
been
stated
(
Marzulli
and
Maibach,
1996;
Griem
et
al.,

2003)
that
repeated
dermal
contact
over
a
longer
time
period
may
also
result
in
a
threshold
for
induction.
Although
more
work
is
needed
in
this
area,
Griem
(
2003)
proposed
an
uncertainty
factor
be
applied
for
repeated
dermal
contact
with
chemical
allergens,
as
there
may
be
a
higher
sensitizing
potency
of
a
chemical
upon
repeated
exposure.

The
Antimicrobials
Division
of
OPP
is
concerned
with
the
risk
from
dermal
exposure
that
may
occur
from
dermal
contact
with
hexavalent
chromium
on
the
surface
of
wood
treated
with
a
wood
preservative
product
containing
hexavalent
chromium,
as
hexavalent
chromium
is
known
to
be
a
potent
dermal
irritant
and
sensitizing
agent.
The
Antimicrobials
Division,
using
existing
science
policies
in
OPP,
currently
performs
hazard
and
risk
assessments
for
non­
cancer
endpoints
through
selection
of
a
level
of
concern
(
e.
g.,
a
NOAEL
or
LOAEL
value)
and
compares
this
level
of
concern
to
estimated
or
actual
exposures
to
derive
a
Margin
of
Exposure.
The
Margin
of
Exposure
is
then
weighed
against
the
"
acceptable"
Margin
of
Exposure,
which
takes
into
account
uncertainties
in
the
risk
assessment
(
e.
g.,
inerspecies
differences
in
response,
intraspecies
17
differences
in
sensitivity).

Hazard
Identification
for
hexavalent
chromium
Murine
LLNA
hazard
data
reported
by
Kimber
et
al.
(
1995)
from
five
different
laboratories
reported
EC3
values
for
hexavalent
chromium
using
potassium
dichromate
as
the
test
substance.

These
data
are
shown
below:

laboratory
A
B
C
D
E
Avg.
US
Avg
country
UK
US
UK
US
US
area
dose
µ
g/
cm2
5.12
11.56
13.24
10.77
11.2
10.36
11.15
Several
published
studies
(
Nethercott
et
al.,
1994;
Basketter
et
al.,
2001;
Hansen
et
al.,
2003)

have
reported
elicitation
thresholds
in
persons
previously
sensitized
to
chromium.
In
the
study
by
Nethercott
et
al.,
a
10%
MET
of
0.089
µ
g/
cm2
was
reported
from
results
of
patch
testing
using
54
human
volunteers
known
to
be
sensitized
to
hexavalent
chromium.
The
lowest
dose
tested
in
this
study,
0.018
µ
g/
cm2,
also
showed
evidence
of
elicitation
in
one
subject.

Basketter
et
al.
(
2001)
reported
elicitation
reactions
to
potassium
dichromate
in
17
volunteers
giving
fully
informed
consent
using
closed­
patch
and
open
appliation
techniques.
Skin
pre­
treated
with
0.2%
sodium
lauryl
sulfate
showed
reaction
to
potassium
dichromate
at
1
ppm,
with
a
doseresponse
evident
at
the
higher
concentrations.
Using
open
application
techniques,
3
of
15
18
volunteers
reacted
to
a
level
of
5
ppm
potassium
dichromate.
To
protect
those
already
sensitized
to
chromium
as
well
as
to
prevent
development
of
additional
chromium­
sensitive
subjects,
a
chromium
contamination
level
of
1
ppm
was
suggested
on
the
basis
of
this
study.

Hansen
et
al.
(
2003)
compared
the
10%
MET
for
both
hexavalent
and
trivalent
chromium
in
18
volunteers
known
to
have
chromium
allergy.
The
results
of
this
study
indicated
a
10%
MET
of
0.03
µ
g/
cm2
for
hexavalent
chromium
and
0.18
µ
g/
cm2
for
trivalet
chromium,
and
suggest
that
both
trivalent
and
hexavalent
chromium
should
be
taken
into
consideration
when
characterizing
chromium
skin
allergy.

The
data
reported
above
using
human
volunteers
are
from
subjects
previously
sensitized
to
hexavalent
chromium
and
give
some
indication
of
an
elicitation
threshold,
while
the
LLNA
data
reported
in
Kimber
et
al.
(
1995)
show
induction
thresholds
in
the
murine
LLNA
test.

Estimation
of
`
safe'
area
doses
for
protection
from
hexavalent
chromium
ACD
can
be
performed
in
both
cases
using
published
methdologies.
For
the
LLNA
study
results,
uncertainty
factors
for
interspecies
extrapolation,
intraspecies
variation,
product
matrix,
and
exposure
considerations
should
be
taken
into
account.
For
the
interspecies
uncertainty
factor,
Griem
et
al.
(
2003)

proposed
a
factor
of
3
based
on
comparison
of
human
and
murine
data
showing
that
sensitizing
doses
are
within
a
factor
of
3
of
each
other,
and
that
skin
penetration
tends
to
be
higher
in
rodents
than
in
humans.
An
uncertainty
factor
of
10
is
proposed
for
intraspecies
variation
in
humans,
as
there
are
few
data
on
induction
thresholds
in
humans.
A
product
matrix
uncertainty
factor
of
10
is
proposed
for
hexavalent
chromium.
The
LLNA
assay
is
performed
using
an
acetone/
olive
oil
vehicle,
while
the
wood
preservative
formulation
containing
hexavalent
chromium
is
likely
more
19
complex
and
could
affect
the
availability
and
potency
of
the
allergen.
An
exposure
uncertainty
factor
of
10
is
proposed
based
on
uncertainty
regarding
the
repeated
dermal
exposure
that
could
occur
to
the
treated
wood
and
how
this
would
affect
the
potency
of
hexavalent
chromium
as
a
dermal
sensitizer.
The
total
uncertainty
factor
from
this
analysis
is
3000
applied
to
the
calculated
area
dose
from
the
average
of
5
LLNA
assays
to
yield
a
`
sensitization
reference
dose.'

This
is
illustrated
below:

laboratory
A
B
C
D
E
Avg.
US
Avg
country
UK
US
UK
US
US
area
dose
µ
g/
cm2
5.12
11.56
13.24
10.77
11.2
10.36
11.15
interspecies
extrapolation
UF
3
3
3
3
3
3
3
intraspecies
variation
UF
10
10
10
10
10
10
10
matrix
UF
10
10
10
10
10
10
10
exposure
UF
10
10
10
10
10
10
10
S­
RfD
µ
g/
cm2
0.0017
0.0038
0.0044
0.0035
0.0037
0.0034
0.0038
The
S­
RfD
can
then
be
compared
to
estimated
or
measured
human
exposure
to
determine
if
the
Margin
of
Safety
is
adequate.
If,
as
suggested
by
Griem
et
al.
(
2003)
that
the
maximum
uncertainty
factor
should
be
no
greater
than
1000,
then
the
S­
RfD
values
shown
above,
using
the
maximum
uncertainty
factor
of
1000,
would
be
0.005,
0.01,
0.013,
0.01,
and
0.01
µ
g/
cm2
respectively
for
the
five
studies,
with
an
average
U.
S.
value
of
0.01
µ
g/
cm2.
20
A
similar
approach
can
be
applied
to
the
MET
values
from
the
Nethercott
et
al.,
Hansen
et
al.,

and
Basketter
et
al.
studies.
As
the
data
are
from
human
studies,
the
interspecies
extrapolation
factor
could
be
reduced
to
1.
An
intraspecies
uncertainty
factor
of
3
is
proposed
based
on
the
use
of
sensitized
persons,
as
elicitation
thresholds
have
been
found
to
be
less
variable
than
induction
thresholds.
An
uncertainty
factor
of
3
is
also
applied
for
the
use
of
LOAEL
values,
as
the
studies
were
not
designed
for
specific
determination
of
a
NOAEL.
An
uncertainty
factor
of
1
is
proposed
for
exposure
considerations,
based
on
the
use
of
a
sensitized
study
group.
The
total
uncertainty
factor
of
10
applied
to
the
reported
human
LOAEL
valuesof
0.018
µ
g/
cm2
,
0.01
µ
g/
cm2,
and
0.03
µ
g/
cm2,
results
in
`
safe'
area
doses
of
0.0018,
0.001,
and
0.003
µ
g/
cm2
for
persons
previously
sensitized
to
hexavalent
chromium.

Comparison
of
the
values
derived
from
the
induction
studies
using
the
LLNA
and
the
patch
test
data
shows
under
the
current
scheme
that
induction
and
elicitation
does
for
hexavalent
chromium
do
not
differ
appreciably,
consistent
with
the
fact
that
hexavalent
chromium
is
a
potent
sensitizer
and
that
induction
and
elicitation
doses
will
not
differ
widely,
as
discussed
by
Griem
et
al
(
2003).
This
factor
could
change
from
application
of
uncertainty
factors
of
differing
magnitude;
a
reasonable
case
has
been
presented
using
the
available
data.

Environmental
Exposures
Application
of
the
experimental
data
to
environmental
exposures
is
also
a
significant
aspect
of
the
risk
assessment,
as
for
hexavalent
chromium,
there
will
be
dermal
contact
not
only
with
the
treated
wood
product,
but
with
soil
in
contact
with
or
in
proximity
to
the
treated
wood
structure.
21
As
the
experimental
hazard
dose
metric
(
i.
e.,
the
area
dose)
for
induction
and/
or
elicitation
of
dermal
sensitization
may
differ
according
to
the
matrix
of
exposure,
it
is
desirable
to
characterize
as
accurately
as
possible
the
influence
of
the
exposure
matrix
variables
on
determination
of
an
acceptable
area
dose
level.
For
contact
with
a
wood
matrix
into
which
a
chemical
allergen
is
incorporated,
the
Office
of
Pesticide
Programs
estimates
a
safe
area
dose
from
available
scientific
data
and
compares
that
level
of
concern
to
the
estimated
or
measured
level
of
exposure
to
calculate
an
acceptable
Margin
of
Exposure.
Variables
that
influence
the
calculation
of
the
exposure
to
the
chemical
on
the
surface
of
wood
include
transfer
efficiency
of
the
chemical
from
the
wood
to
the
skin,
number
of
dermal
contact
events,
surface
area
of
skin
exposure,
and
level
of
the
chemical
on
the
surface
of
the
wood.
Approaches
to
assessing
treated
wood
exposures
have
been
considered
recently
by
the
FIFRA
SAP
in
a
December,
2003
meeting
in
which
the
SHEDS
model
was
presented
and
considered
as
a
probabilistic
approach
to
assessing
exposures
to
arsenic
and
chromium
in
treated
wood,
including
children's
exposures.

For
determination
of
an
acceptable
area
dose
from
contact
with
a
chemical
in
a
soil
matrix,
many
of
the
variables
are
similar
to
the
wood
matrix.
However,
other
soil
matrix
properties
may
have
a
greater
impact
on
the
transfer
of
the
chemical
from
the
soil
to
the
skin.
Additional
variables
are
taken
into
account
in
calculating
skin
contact
dose,
such
as
skin
surface
area
of
contact,
soil
adherence
to
skin,
contact
frequency
(
important
for
determination
of
children's
potential
hazard
and
risk),
bioavailability
of
the
chemical
in
soil,
physio­
chemical
properties
of
the
soil
(
i.
e.

moisture
content,
soil
type),
valence
or
complex
state
of
the
chemical,
and
chemical
solubility
in
sweat.
Guidance
for
determination
of
acceptable
soil
levels
for
Superfund
cleanup
sites
is
found
22
at
the
following
web
address:
http://
www.
epa.
gov/
superfund/
resources/
soil/
ssgmarch01.
pdf.
It
is
noted
here
that
soil
cleanup
values
estimated
for
dermal
contact
in
the
Superfund
program
are
based
upon
systemic
effects
resulting
from
dermal
exposures,
although
dermal
sensitization
effects
should
also
be
considered
in
the
assessment.

Water
also
represents
a
matrix
of
exposure
that
is
different
than
wood
or
soil,
in
that
it
represents
a
three­
dimensional
matrix
of
contact
(
i.
e.
µ
g/
cm3)
vs.
a
two­
dimensional
matrix
of
contact
(
i.
e.
µ
g/
cm2).
Activities
such
as
showering
represent
exposure
scenarios
where
potential
dermal
sensitization
hazard
may
need
to
be
characterized,
but
as
with
soil,
other
variables
may
influence
the
estimation
of
a
safe
area
dose
in
this
matrix.
Thus,
while
the
experimental
hazard
data
for
hexavalent
chromium
indicate
that
low
levels
are
adequate
to
induce
and/
or
elicit
sensitization,
the
actual
concentration
necessary
for
such
reactions
may
differ
according
to
the
matrix
in
which
the
chromium
is
found.

REFERENCES
23
Basketter
D.,
et
al.,
(
2001)
Investigation
of
the
Threshold
for
Allergic
Reactivity
to
Chromium,

Contact
Dermatitis,
2001,
44:
70­
74.

Basketter
DA,
et
al.,
(
2001)
Commentary
Skin
Sensitization,
Vehicle
Effects
and
the
Local
Lymph
Node
Assay,
Food
and
Chemical
Toxicology
39
(
2001)
621­
627.

Basketter
DA,
et
al.,
(
2003)
Nickel,
Chromium
and
Cobalt
in
Consumer
Products:
Revisiting
Safe
Levels
in
the
New
Millennium,
Contact
Dermatitis,
2003:
49:
1­
7.

Dean,
J.
H
et
al.
(
2001):
ICCVAM
Evaluation
of
the
Murine
Local
Lymph
Node
Assay.
II.

Conclusions
and
Recommendations
of
an
Independent
Scientific
Peer
Review
Panel.
Regulatory
Toxicol.
Pharmacol.
34:
258­
273.

Felter
SP,
et
al.,
(
2002):
A
Review
of
the
Scientific
Basis
for
Uncertainty
Factors
for
use
in
Quantitative
Risk
Assessment
for
the
Induction
of
Allergic
Contact
Dermatitis,
Contact
Dermatitis,
2002,
47:
257­
266.

Felter
SA.,
et
al.,
(
2003):
Application
of
the
Risk
Assessment
Paradigm
to
the
Induction
of
Allergic
Contact
Dermatitis,
Regulatory
Toxicology
and
Pharmacology
37
(
2003)
1­
10.

Gerberick,
F.,
et
al.,
(
2000):
A
Skin
Sensitization
Risk
Assessment
Approach
for
of
Contact
Dermalitis
Evaluation
of
New
Ingredients
and
Products.
American
Journal
of
Contact
24
Dermalitis,
Vol.
II,
No.
2
(
June),
2000:
pp.
65­
73.

Griem
P.,
et
al.,
(
2003):
Proposal
for
a
Risk
Assessment
Methodology
for
Skin
Sensitization
Based
on
Sensitization
Potency
Data,
Regulatory
Toxicology
and
Pharmacology
38
(
2003)
269­
290.

Hansen,
M.
B.,
Johansen,
J.
D.,
and
Menne,
T.
(
2003):
Chromium
Allergy:
significance
of
both
Cr(
III)
and
Cr(
VI).
Contact
Dernatitis
49:
206­
212.

Interagency
Coordinating
Committee
on
the
Validation
of
Alternative
Methods
(
ICCVAM)

(
1999).
The
Murine
Local
Lymph
Node
Assay:
A
Test
Method
for
Assessing
the
ACD
Potential
of
Chemicals/
Compounds.
NIH
publication
99­
4494.
National
Institute
of
Environmental
Health
Sciences,
Research
Triangle
Park,
NC.

Kimber
I.,
et
al.,
(
1995)
An
International
Evaluation
of
the
Murine
Local
Lymph
Node
Assay
and
Comparison
of
Modified
Procedures.
Toxicology
103
(
1995)
63­
73.

Kimber,
I
(
1996):
The
Skin
Immune
System.
In:
Dermatotoxicology,
5th
edition
(
Marzulli
and
Maibach,
eds.).
Taylor
and
Francis,
Washington
D.
C.,
pages
131­
141.

Kimber
I.,
et
al.,
(
2003)
Classification
of
Contact
Allergens
According
to
Potency:
Proposals.

Food
and
Chemical
Toxicology
41
(
2003)
1799­
1908.
25
Marzulli,
F.
N
and
Maibach,
H.
I.,
eds.
(
1996)
Dermatotoxicology,
fifth
edition.
Taylor
and
Francis,
Washington
D.
C.,
pages
143­
146.

Nethercott
J.,
et
al.,
(
1994)
A
Study
of
Chromium.
Induced
Allergic
Contact
Dermatitis
With
54
Volunteers;
Implications
for
Environmental
Risk
Assessment,
Occup.
Environ.

Med..
1994;
51:
371­
380.

NIOSH
(
2001):
Occupational
Dermatoses.
A
Program
for
Physicians.
Available
at:

http://
www.
cdc.
gov/
niosh/
ocderm.
html
NJDEP,
1998:
Summary
of
the
Basis
and
Background
of
the
Soil
Cleanup
Criteria
for
Trivalent
and
Hexavalent
Chromium.
Available
at:
http://
www.
nj.
gov/
dep/
srp/
news/
1998/
9812_
04.
htm
Robinson
MK.,
et
al.,
(
2000)
The
importance
of
Exposure
Estimation
in
the
Assessment
of
Skin
Sensitization
Risk,
Contact
Dermatitis,
2000,
42:
251­
259.

Sailstad,
D.
M.
(
2003):
Allergic
Contact
Hypersensitivity:
Mechanisms
and
Methods.
In:

Alternative
Toxicological
Methods.
Harry
Salem
and
Sidney
Katz,
eds.
,
CRC
Press,
Boca
Raton,
Florida.
Pages
193­
205.

Scott
AE,
et
al.,
(
2002)
Insight
into
the
Quantitative
relationship
Between
Sensitization
and
challenge
for
Allergic
Contact
Dermatitis
Reactions.
Toxicology
and
Applied
Pharmacology
26
183,
66­
70
(
2002).

Stern
AH.,
et
al.,
(
1993)
Risk
Assessment
of
the
Allergic
Dermatitis
Potential
of
Environmental
Exposure
to
Hexavalent
Chromium,
Journal
of
Toxicology
and
Environmental
Health,
40:
613­

641,
1993.

USEPA,
1998:
Toxicological
Review
of
Hexavalent
Chromium.
In
Support
of
the
Integrated
Risk
Information
System
(
IRIS).
Available
on
the
Internet
at
www.
epa.
gov/
iris/
toxreviews
USEPA,
2001:
Supplemental
guidance
for
developing
soil
screening
levels
for
Superfund
sites.

Available
on
the
Internet
at
http://
www.
epa.
gov/
superfund/
resources/
soil/
ssgmarch01.
pdf
Zewdie,
T.
(
1998):
Evaluation
of
the
Risk
Assessment
Methodologies
Based
on
the
Hexavalent
Chromium
[
Cr(
VI)]­
Elicited
Allergic
Contact
Dermatitis
(
ACD)
in
Sensitized
Population
and
MADEP's
Recommendations.
Available
at:
http://
www.
state.
ma.
us/
dep/
ors/
files/
crviderm.
doc
