2/
26/
98
MEMORANDUM
SUBJECT:
Assessment
of
the
Dietary
Cancer
Risk
of
Hexachlorobenzene
and
Pentachlorobenzene
as
impurities
in
Chlorothalonil,
PCNB,
Picloram,
and
several
other
pesticides.
DP
Barcode
D243499.
Chemical
codes
061001
(
Hexachlorobenzene)
&
081901
(
Chlorothalonil).

FROM:
William
Smith,
Chemist
Chemistry
And
Exposure
Branch
I
(
CEB­
I)
Health
Effects
Division
(
7509C)

THROUGH:
F.
B.
Suhre,
Branch
Senior
Scientist
Chemistry
And
Exposure
Branch
I
(
CEB­
I)
Health
Effects
Division
(
7509C)

TO:
Mary
Clock,
Chemical
Review
Manager
Risk
Characterizaton
and
Analysis
Branch
Health
Effects
Division
(
7509C)

This
memorandum
is
primarily
in
response
to
issues
arising
related
to
reregistration
of
chlorothalonil
and
registration
of
some
new
uses
for
the
same
chemical.
To
date,
in
assessment
of
human
health
risks
associated
with
chlorothalonil,
the
Agency
has
addressed
the
cancer
risk
due
to
the
presence
of
hexachlorobenzene
(
HCB)
as
a
minor
impurity
in
the
formulated
products.
Hexachlorobenzene
is
considered
to
be
a
B2
possible
human
carcinogen
with
a
Q*
of
1.02
(
mg/
kg/
day)­
1.
This
memorandum
is
being
written
in
response
to
requests
from
RD
and
SRRD
to
provide
a
cumulative
cancer
risk
for
the
dietary
exposure
to
HCB.
Four
pesticides
that
are
used
on
food/
feed
crops
have
been
assessed
for
cancer
risk
due
to
contamination
with
HCB.
These
are
chlorothalonil,
dacthal,
picloram,
and
pentachloronitrobenzene
(
PCNB).

Pentachlorobenzene
(
PCB)
has
also
been
found
from
use
of
PCNB.
HED
has
concluded
that,
based
on
the
similarities
of
the
chemical
structures
and
toxicities
of
HCB
and
PCB,
it
is
reasonable
to
assume
that
the
carcinogenic
potential
of
PCB
is
comparable
to
HCB.
Thus,
for
the
PCNB
dietary
risk
assessment
the
Q*
for
PCB
has
been
assumed
to
be
equal
to
that
for
HCB.

The
Agency
has
also
issued
a
DCI
(
10/
92)
for
analytical
data
on
the
possible
presence
of
HCB
or
2
1.
See
discussion
section
below
for
source
of
assessment
and
any
new
assumptions
or
calculations
contained
in
this
consideration.

2.
Q*
for
HCB
and
PCB
assumed
to
be
1.02
(
mg/
kg
bwt/
day)­
1
3.
Uses
of
this
chemical
may
be
voluntarily
canceled
in
the
near
future.
Cancer
risks
that
include
the
potential
contribution
from
dacthal
are
enclosed
in
brackets
4.
Five
pesticides
that
are
currently
used
on
food/
feed
crops
and
were
identified
in
the
product
chemistry
DCI
of
10/
92
as
containing
either
HCB
or
PCB.
The
chemicals
are
endosulfan,
chlorpyrifos­
methyl,
atrazine,
simazine,
and
clopyrilid.
PCB
in
26
other
pesticidal
active
ingredients.
Based
on
data
reviewed
to
date,
one
or
both
of
these
impurities
have
been
detected
in
five
active
food­
use
pesticides.

In
the
discussion
provided
below
we
are
providing
estimates
of
dietary
exposure/
risk
for
HCB
and
PCB
in
nine
chemicals.

CONCLUSIONS:

Table
1
contains
estimated
dietary
cancer
risks
from
all
known
pesticidal
sources
of
HCB
and
PCB.
The
cumulative
estimated
dietary
cancer
risk
for
HCB
is
1.34
x
10­
6.
An
additional
0.46
x
10­
6
may
be
attributed
to
PCB.
If
this
total
is
adjusted
for
the
potential
cancellation
of
dacthal
(
personal
communication,
Jill
Bloom,
SRRD),
then
the
cumulative
dietary
cancer
risk
is
estimated
to
be
1.1
x
10­
6.
The
assumptions
and
methods
that
are
incorporated
in
these
assessments
can
be
found
in
the
discussion
section
that
follows.

Table
1.
Estimated
Dietary
Cancer
Risk
of
the
Pesticide
Impurities,
HCB
and
PCB1.

Source
Pesticide
Oncogenic
Risk2
HCB
PCB
Combined
Chlorothalonil
2.4
x
10­
7
None
2.4
x
10­
7
PCNB
1.6
x
10­
7
4.3
x
10­
7
5.9
x
10­
7
Picloram
1.5
x
10­
7
None
1.5
x
10­
7
Dacthal3
[
7.1
x
10­
7]
None
[
7.1
x
10­
7]

Five
Other
Chemicals4
7.5
x
10­
8
2.75
x
10­
8
1.0
x
10­
7
Total
6.3
x
10­
7
[
1.34
x
10­
6]
4.6
x
10­
7
1.1
x
10­
6
[
1.81
x
10­
6]
3
DISCUSSION:

General
Considerations:
For
purposes
of
this
cumulative
assessment
we
have
added
the
individual
assessments
for
picloram,
dacthal,
PCNB
and
chlorothalonil,
each
with
its
individual
set
of
data
and
assumptions.
For
the
five
additional
chemicals
obtained
from
responses
to
the
DCI
of
10/
92
we
made
a
first
cut
screening
assessment
based
on
the
most
recent
DRES
run
for
the
active
ingredient.
The
assumption
was
made
that
the
impurities
would
occur
on
food
commodities
at
the
same
ratio
to
the
active
ingredient
as
was
present
in
the
formulation
applied
to
crops.
Further
refinement
of
anticipated
residues
was
not
attempted
for
the
five
additional
chemicals
because
of
limited
resources
available
at
this
time.
It
is
the
judgement
of
this
reviewer
that
given
the
low
levels
of
HCB
and
PCB
in
the
DCI
chemicals
relative
to
picloram,
PCNB,
chlorothalonil,
and
dacthal,
any
errors
in
the
simple
screening
approach
taken
here
would
be
insignificant
compared
to
the
overall
dietary
exposure
assessment
including
all
nine
chemicals.

The
major
problem
in
attempting
to
estimate
the
dietary
exposure
to
HCB
is
that
it
is
generally
present
in
food
commodities,
if
at
all,
at
levels
that
are
below
the
limit
of
detection
of
analytical
methods.
In
such
situations
the
general
rule
for
exposure
assessment
has
been
to
make
the
assumption
that
residues
are
present
at
one
half
the
limit
of
quantitation
or
one
half
the
limit
of
detection.
Given
the
high
Q*
of
HCB,
this
assumption
can
lead
to
the
appearance
of
an
unacceptable
risk,
depending
on
the
sensitivity
of
the
analytical
method.
In
some
cases
levels
of
HCB
have
been
estimated
based
on
levels
of
the
active
ingredient
in
food
commodities.
In
these
cases
it
was
assumed
that
the
impurity
would
dissipate
from
the
food
commodity
at
an
equal
or
greater
rate
than
the
active
ingredient.
There
are
some
data
available
from
exaggerated
use
rates
of
chlorothalonil,
picloram
and
DCPA
that
support
this
approach.

The
sources
of
data
for
each
chemical
and
any
new
refinements
introduced
in
this
memo,
are
summarized
below.

Picloram:
The
most
recent
DRES
analysis
(
4/
12/
94,
J.
Bazuin)
yielded
an
estimate
of
oncogenic
risk
of
0.6698
x
10­
6.
The
anticipated
residues
for
HCB
from
picloram
were
included
in
the
Residue
Chemistry
Chapter
of
the
RED
(
W.
Smith,
3/
7/
94).
The
estimated
cancer
risk
in
the
diet
from
uses
of
picloram
is
essentially
all
from
the
consumption
of
meat
and
milk
and
was
based
on
a
number
of
conservative
assumptions
including
tolerance
level
picloram
on
feed
items
and
worst
case
animal
diets.
Also,
the
transfer
factors
used
in
estimating
residues
in
meat,
milk,
poultry
and
eggs
were
more
conservative
than
those
used
for
similar
dietary
exposure
assessments
for
HCB
in
dacthal,
PCNB
and
chlorothalonil,
even
though
the
factors
were
derived
from
the
same
HCB
feeding
study.
These
factors
are
shown
below.
The
lower
numbers
in
the
third
column
were
derived
from
a
linear
regression
analysis
of
multidose
feeding
study
data
while
the
numbers
used
in
the
picloram
assessment
were
rounded
up
based
on
visual
inspection
of
the
same
data.
In
the
interest
of
consistency
the
same
transfer
factors
are
taken
into
account
here
as
that
were
used
for
assessments
of
HCB
in
animal
commodities
due
to
use
of
chlorothalonil,
dacthal
and
PCNB
on
feed
items
(
Table
2,
column
3).
4
Table
2.
Transfer
factors
used
in
estimating
anticipated
residues
of
HCB
in
meat
milk,
poultry
and
eggs.

Commodity
Factor
used
in
picloram
RED
for
deriving
anticipated
residues
of
HCB
(
DRES
run
of
4/
12/
94)
Factor
used
for
deriving
anticipated
residues
of
HCB
from
uses
of
dacthal,
chlorothalonil,
and
PCNB
Cattle
fat
10x
4x
Milk
0.2x
(
with
residues
confined
to
fat)
0.08x
(
with
residues
confined
to
fat)

Cattle
meat,
liver
and
kidney
0.5x
0.2x
Poultry
fat
10x
6x
Poultry
liver
2x
2x
Poultry
meat
0.2x
0.1x
Egg
yolk
3x
2x
Egg
white
0.01x
0.004x
Based
on
the
revised
factors
shown
in
Table
2
refinements
can
be
made
in
the
cancer
risk
assessment
for
HCB
in
picloram.

The
contribution
of
HCB
residues
in
beef
fat
to
the
cancer
risk
in
the
DRES
run
of
4/
12/
94
was
0.2856
x
10­
6.
Since
this
value
is
directly
related
to
anticipated
residue
and
we
can
reduce
the
anticipated
residue
by
a
factor
of
4/
10
(
Table
2)
,
a
refined
risk
due
to
these
food
items
is
0.1142
x
10­
6.

In
the
cancer
assessment
of
4/
12/
94,
milk­
non­
fat­
solids
contributed
0.1343
x
10­
6.
This
should
be
removed
from
the
cancer
risk
because
the
anticipated
residues
of
HCB
are
restricted
to
milk
fat
only.
The
risk
due
to
milk­
fat
solids
was
0.1938
x
10­
6.
This
can
be
refined
by
a
factor
of
0.08/
0.2
(
Table
2).
Thus
a
revised
estimate
for
risk
from
milk
fat
products
would
be
0.07752
x
10­
6.

The
revisions
are
summarized
in
the
following
table.
5
Table
3.
Revisions
in
the
assessment
of
cancer
risk
from
HCB
in
Picloram
based
on
changes
in
transfer
factors
for
estimating
residues
in
meat
and
milk.

Foods
Cancer
risk
(
4/
12/
94)
Cancer
risk
based
on
present
revision
in
anticipated
residues
Beef
fat
0.2856
x
10­
6
0.1142
x
10­
6
Milk­
non­
fat
0.1343
x
10­
6
­

Milk­
fat
0.1938
x
10­
6
0.07752
x
10­
6
Sum
0.6137
x
10­
6
0.19172
x
10­
6
The
difference
in
estimated
cancer
risk
from
beef
fat
and
milk
is
0.42198
x
10­
6.
Therefore,
adjusting
the
total
cancer
risk
for
these
revisions
yields
(
0.66980
x
10­
6)
­
(
0.42198
x
10­
6)
=
0.2478
x
10­
6.
Finally,
the
assessment
of
4/
12/
94
used
a
Q*
for
HCB
of
1.7,
which
has
subsequently
been
revised
to
1.02.
Thus
the
cancer
risk
can
be
adjusted
by
a
factor
of
1.02/
1.7
resulting
in
a
revised
value
of
0.15
x
10­
6.

Pentachloronitrobenzene:
A
refinement
of
anticipated
residues
for
HCB
and
PCB
was
completed
by
HED
in
1994
(
W.
Smith,
D203453,
5/
25/
94);
however,
a
dietary
risk
assessment
had
not
been
performed.
The
latest
anticipated
residue
data
were
entered
into
the
DRES
for
this
assessment
(
B.
Steinwand,
2/
18/
98)
and
the
oncogenic
risk
calculated.
These
data
yield
revised
risks
of
1.6
x
10­
7
for
HCB
and
4.3
x
10­
7
for
PCB.
It
should
be
noted
that
in
deriving
these
anticipated
residues
it
was
assumed
that
HCB
is
present
in
PCNB
at
levels
not
to
exceed
0.1
%.
Since
that
time
the
two
producers
of
PCNB
have
certified
that
levels
of
HCB
will
not
exceed
0.05%
and
that
levels
of
PCB
will
not
exceed
0.01%.

Chlorothalonil:
The
latest
refinements
in
the
dietary
risk
assessment
for
HCB
in
chlorothalonil
were
performed
in
conjunction
with
the
RED.
An
oncogenic
risk
of
2.4
x
10­
7
has
been
estimated
assuming
that
all
chlorothalonil
used
on
food
or
feed
crops
contains
0.05%
HCB.
The
major
producer,
ISK
has
certified
that
the
level
is
actually
lower
than
0.004%.
It
has
been
argued
by
the
producer
that
if
one
considers
the
relative
market
share
of
different
chlorothalonil
products
chronic
risk
assessments
using
the
0.05%
level
result
in
a
considerable
overestimate
of
the
risk.

The
most
recent
risk
assessment
for
chlorothalonil
included
the
anticipated
risk
from
all
proposed
new
tolerances
also.

Dacthal:
The
latest
dietary
risk
assessment
for
HCB
in
dacthal
resulted
in
an
oncogenic
risk
of
7.1
x
10­
7.
It
has
been
indicated
that
uses
of
dacthal
will
soon
be
canceled.
At
present
this
chemical
has
the
highest
allowable
level
of
HCB
(
0.3
%)
of
any
of
the
pesticides
known
to
contain
HCB.
6
Responses
to
the
DCI
of
10/
92:

The
10/
92
DCI
was
sent
to
producers
of
pesticides
that
had
been
identified
as
being
potentially
contaminated
with
HCB
or
PCB,
based
on
knowledge
of
the
manufacturing
process.
A
total
of
26
pesticide
active
ingredients
were
included
in
the
DCI.
The
producers
of
these
pesticides
were
required
to
submit
analytical
data
on
possible
contamination
using
an
analytical
method
capable
of
quantifying
residues
at
levels
as
low
as
0.1
ppm
(
1
x
10­
5%).
If
quantifiable
levels
of
either
contaminant
were
found
then
certified
upper
limits
were
required
for
its
presence
in
the
active
ingredient
involved.
Several
of
the
responses
to
the
DCI
indicated
the
presence
of
either
HCB
or
PCB
but
at
levels
considerably
lower
than
those
assessed
for
the
four
food­
use
chemicals
above.
There
are
five
additional
food­
use
chemicals
that
have
been
identified
as
contaminated
with
HCB
or
PCB
at
levels
not
to
exceed
those
indicated
in
Table
4
below.
These
levels
are
considerably
lower
than
those
previously
certified
for
pesticides
(
0.01%­
0.3%).
The
potential
dietary
risk
from
these
additional
pesticides
was
estimated
based
on
the
most
recent
DRES
runs
for
the
active
ingredient.
It
was
assumed
that
the
dietary
exposure
to
the
contaminant
would
be
directly
correlated
to
the
estimated
exposure
to
the
active
ingredient
based
on
the
level
of
the
contaminant
in
the
formulation
applied
to
food/
feed
crops.
It
was
assumed
that
the
contaminant
would
be
present
in
all
formulations
at
the
maximum
concentration
reported
or
at
the
certified
limit
set
for
that
active
ingredient.
See
Table
4
for
the
dietary
risk
estimates.
7
1.
These
pesticides
were
identified
in
the
DCI
of
10/
92
as
containing
detectable
levels
of
either
HCB
or
PCB.
The
highest
level
detected
in
the
given
pesticide
technical
grade
was
used
for
this
assessment.

2.
Daily
dietary
intake
is
based
on
the
most
recent
DRES
analysis
for
the
active
ingredients
listed.
The
TMRC
represents
the
daily
dietary
intake
for
the
U.
S.
population
­
48
states
assuming
tolerance
level
residues
and
100%
crop
treated.
The
ARC
values
are
based
on
the
most
recent
refinement
of
anticipated
residues
available.

3.
Estimated
as
the
TMRC,
or
ARC,
multiplied
by
the
percent
of
impurity
and
the
Q*
(
1.02
(
mg/
kg/
day)­
1)
for
HCB.

4.
NA
indicates
not
available.
Table
4.
Estimation
of
cancer
risk
from
the
presence
of
hexachlorobenzene
or
pentachlorobenzene
in
pesticides
identified
in
data
call
in
of
10/
92.
The
"
anticipated
residues"
of
the
impurities
are
estimated
as
being
directly
proportional
to
their
level
in
the
pesticide
formulation
applied
to
food
crops,
i.
e.,
residues
of
the
active
ingredient
and
the
impurity
dissipate
at
the
same
rate
on
all
food
commodities.

Pesticide1
Daily
Dietary
Intake2
[
TMRC/
ARC]
(
mg/
kg/
day)
Level
of
Impurities
in
Technical
(%)
Estimated
Oncogenic
Risk3
HCB
PCB
HCB
PCB
Endosulfan
0.012144/
0.001227
7
x
10
­
5
0.0006
8.7
x
10­
9
8.8
x
10­
10
7.4
x
10­
8
7.5
x
10­
9
Ch
l
o
r
p
y
r
i
f
o
s
­
methyl
0.012672/
NA4
­
3
x
10­
5
None
3.9
x
10­
9
Atrazine
0.0009902/
0.0001981
0.0017
2.5
x
10­
5
1.7
x
10­
8
3.4
x
10­
9
2.5
x
10­
10
5.1
x
10­
11
Simazine
0.002506/
0.000113
0.002
­
5.1
x
10­
8
2.3
x
10­
9
None
Clopyrilid
0.008295/
NA
0.0008
0.00019
6.8
x
10­
8
1.6
x
10­
8
Sum
of
estimated
dietary
risk
based
on
TMRC
Sum
of
estimated
dietary
risk
using
ARC
when
available
1.5
x
10­
7
7.5
x
10­
8
9.4
x
10­
8
2.75
x
10­
8
Total
estimated
dietary
risk
(
sum
of
HCB
and
PCB
based
on
TMRC)
2.4
x
10­
7
Total
estimated
dietary
risk
(
sum
of
HCB
and
PCB
using
ARC
where
available)
1.0
x
10­
7
8
cc:
W.
Smith
(
CEB­
I),
R.
Kearns
(
RD),
J.
Bloom
(
SRRD),
HCB
SF,
Chlorothlonil
SF,
Chlorothalonil
Reg
Std
File,
RF
7509C:
CEB­
I:
WOS:
wos:
Rm805A:
CM2:
305­
5353:
02/
24/
98
RDI:
ResChemTeam
(
02/
24/
98)
FSuhre
(
02/
26/
98).
