-------
03
0)
'o
c
03
'o
1
.2"
d
£T
i
Q.
&l
>,
?
w
f,
I
•s
8
»-: E
3
-08^2
.851*
III
S E —
Sgfcg
Iff
" O
Q. *^
32
oi-2
c
^p
o^ p
»s
o
&J^»a€-St
ID
; «J .
CO ^
!ffl5«,
Arsenical
Medicinals
(continued)
73
OJ
•tjw
3 S3
d£
0~ Q g.£^ |
If gi.sl"
slllllll
- Ss.Si s
ra
<5
^£
73
u.
in
$
I
63
-------
CO
-2
1
"S
?
£
if
2
1
cc
Study population
I
to
"o
— 2
•o §
3 *~
"c
5
V
C
t— J
«i i
1
The 6 patients treated with arsenical
medicinais exhibited the keratotic lesi
associated with arsenical poisoning.
•a
CD
6 patients being trea
with arsenical
medicinals
Is
"I
_ *•
3 V>t
3 "jg CO <
c -i|-j
CD T3 I
to CD (
^ S
£
CO
tn
CO
O
CO
CO
CO
3 C
> s
5 8
1 i
1 i
to
§
Angiosarcoma of liver developed 25
after therapy.
•o
5
Individual who rece'p
7-month course of
arsenic therapy for
psoriasis
I
£
tn
o
_
15
03
is
O ^^
3 ID
o5 o
Six cases of basal cell carcinoma,
carcinoma in situ, or squamous cell
carcinoma of the skin were reported
of these patients had systemic cano
(breast and colon). All 7 patients sh
keratosis on palms and soles.
•£
'3 e
7 individuals tested
Fowler's solution.
Cases selected fror
clinical files.
a
£
<0
CO
O
in
CO ?-
O 03
J2 g
o -55
3o
03
_Q
All of these patients had previously
treated with Fowler's solution.
CO
03 ^
2 male patients, on
with a reticulosarco
a
03
CD
o
§
o>
1
*
ID
C
"c
o
• — *
en
ja
o
The patient had previously taken Fi
solution.
•a ._:§ ^
-a m
c 1^
co en
64
-------
•o
0)
c
•
65
-------
>£
• CD
l^"lS«i
.S Q *" to « co
- C. CD c- 'f= n^
»J .to S = ~ *
° 8
Arsenical
Medicinal!
ff^nntinilfir
= 0
*S
CO •£
">- 0. .
•i » CD
III
l?s
O •«— o
"w C
« CO Q>
Sg,e
c "S ^
CD O 03
'•==£:>
CO ^ CD
Q. CO -0
8_
Female patient with
nasopharyngeal can
and with palmar and
•e
a
S
O
o>
! 1
1 IE
i Q- T-
11
a|
•i§
i.1
CO
CD "5
'13 o .
Q.U. CO
£•§ s.
plantar keratosis.
Male patient with
hemangioendothelia
sarcoma of the liver
•c
a
CD
O
i
i
CD 00
O)CD
£2
> m
S.
-
co.S
S co
5 to
II
S-Ss
CD .Q O
!
-------
CD
I
1
O)
g>
X
EC
•f-
55
I
0)
I
'ro co CD
IcS
£ O JO-Jr
S P??
§18?
••S •£. c co
CO OJ CO ^
a. CD o -o
"M^'Q.g
CD P S a.
£ ^ c r «
CO 0} O 5*°^ p C
•"
.
o
:§=£
co
g
Q.
CD
O
s»-
E co
OTS
67
-------
-------
Appendix B
Quantitative Estimate of Risk for Skin Cancer
Resulting from Arsenic Ingestion
69
-------
Contents
71
List of Tables 72
List of Figures 73
I. Methodology ?3
II Application to Taiwan Epidemiologic Study
III. Use of the Mexican Data to Evaluate Taiwan's Dose- yg
Response Model
IV. Use of the German Data to Evaluate Taiwan's Dose- ^
Response Model
V. Discussion of the Uncertainties of the Risk Estimates
87
VI. Summary
70
-------
Tables
B-1 _
t risk (skin
in Taiwan"""' ~' "a" a'""|J """ u
-------
Figures
model, linear in dose .......................
Sffi^sjs^c^sss^ -
model, linear and quadratic in dose ......................
ssr-ttss.-trs; » *
of the model, linear in dose ..........................
* Ts
of the model, linear and quadratic in dose ..................
linear and quadratic in dose .....................
linear and quadratic in dose .....................
72
-------
I. Methodology
exposure to arsenic concentration d. Ve models Jhe ^oSing fo'm : "^
F(t,d) = l-exp[-g(d)H(t)]
"• Application to Taiwan Epidemiologic Study
(1) LhlTrtality rate was the same in ^e diseased
cancer) persons as in the nondiseased persons.
The population composition (with respect to the risk
factors of the skin cancer) remained constan over time
Th.s assumption .mplies that there was no cohort effect
The skin cancer was not surgically removed
(skin
(2)
(3)
73
-------
procedure used for estimating the actual number of persons at risk is
* , • «i .•.„.*„*•.^i-ii-\he* tfrlQt ff^llnW
^
dd,a,es, normalized «o ,he reference
were given in *9ur|s BJda™4 show the analogous plots for females.
74
-------
Table B-1.
Estimated Distribution of the Surveyed Male
Portion at Risk (Skin Cancer Cases? by Age Grow
nCentratl"°n °f Arsenic '" WeN Tater "n
Arsenic
concentration
(ppm)
Low (0-0.30)
Medium (0.30-0.60)
High (> 0.60)
Unknown
Total
2.714&
(0)0
1,542
(0)
2,351
(0)
4,933
(0)
11,540
(0)
935
(D
531
(2)
810
(18)
1,699
(3)
3,975
(24)
653
(4)
371
(18)
566
(56)
1,188
(61)
2,778
(139)
236
(11)
134
(22)
204
(52)
429
(64)
1,003
(149)
4,538
(16)
2,578
(42)
3,931
(126)
8,249
(128)
19,296
(312)
• \ ' ' "/ ^*J
"Estimated number of persons at risk
cEstimated number of skin cancer cases observed.
Low (0-0.30)
Medium (0.30-0.60)
High (> 0.60)
Unknown
Total
2,65 1&
(0)c
1,507
(0)
2,296
(0)
4,819
(0)
11,273
(0)
' ' — .
1,306
(0)
742
(1)
1,131
(4)
2,373
(2)
5,552
(7)
— — — — _^_
792
(3)
450
(9)
686
(33)
1,440
(13)
3,368
(58)
239
(2)
136
(8)
207
(22)
435
(27)
1,017
(59)
4,988
(5)
2,835
(18)
4,320
(59)
9,067
(42)
21,210
(124)
' \v^/ \ I.
°Estimated number of persons at risk
cEstimated number of skin cancer cases observed.
75
-------
Table B-3. Conversion of Arsenic Dose
for Taiwanese to Equivalent
Arsenic Dose for U.S.
Populationsa
Taiwanese
(ppm)
Males
Females
0.17
0.47
0.80
0.17
0.47
0.80
U.S. person
(yg/kg/day)
10.8
29.9
50.9
6.8
18.8
32.0
. -- - ---
aAssumptions: A U.S. person weighs 70 kg and
drinks 2 L of water daily; a Taiwanese male weighs
55 kg and drinks 3.5 L of water daily; a Taiwanese
female weighs 50 kg and drinks 2 L of water daily.
hypothesis that the coefficient corresponding to & is zero is rejected at p <
SS
estimate calculated on the basis or aai * JP females: (1) the daily water
that calculated on the bas.s ^^{JJ^£SSTto that consumed by
by Taiwanese males {3J5 _uaayj in r particu|ar those
migrate o'u, « Wn.
equivalent to 1.4 yg/
a US. reference person as
The lifetime risk is 8st,ma,«l to
£3.
from
^ ma|es) |or
contaminated «i,n 50 9/L o,
to, the lour age .
respectively, 0.8.0 x 10 ». 67
« «
for'males, and 0, 7.0
^ ^ ^
*e comment made b, MargoHs December
76
-------
Figure B-1. Ob«,vedland predictedskin cancerpreva.enceforTafwanesema.es
" 39e;
.2978
Observed
Prevalence
H = High Dose
M = Medium
L = Low
.0088E-3
Figure B-2. Observed and predicted skin cancer prevalence forTaiwanese m
~
ales
.0053E-3
Observed
Prevalence
H = High Dose
M = Medium
L = Low
H/
/
/
77
-------
Figure B-3.
use of the model, linear in dose.
.1422
10
use of the model, linear and quadratic in dose.
.1515
|
.0750
Observed
Prevalence
H = High Dose
M = Medium
L = Low
78
-------
Table B-4.
Results of Model
Data
Linear
Males:
Doses (d): 0, 10.818,
g(d) = (0.302525 x 10-7)d
H(t) = (1-6.931)2-935
In L = -614.551
Unit risk (probability of skin cancer
in lifetime due to 1 ng/kg/day
of arsenic)
= 5.0x10-3
Fitting to Taiwan Skin Cancer
Quadratic
29.909, 50.909 ug/kg/daya
g(d) = (0.124707 x10-7)d
+ 0.404871 x 10-9)d2
H(t) = (t - 6.873)2.950
In L = -610.088
Unit risk (probability of skin cancer
in lifetime due to 1 ng/kg/day
of arsenic)
= 2.3 x 10-3
Females:
Doses (d): 0, 6.8,
g(d) = (0.682262 x 10'8)d
H(t) = (t - 9.0)3.225
In L = -348.041
Unit risk (probability of skin cancer
in lifetime due to 1 iig/kg/day
of arsenic)
= 3.4x10-3
18.8, 32.0 ug/kg/daya
g(d) = (0.157281 x 10-8)d
+ 0.204076 x 10-9)d2
H(t) = (t - 9.0)3.231
In L = -344.365
Unit risk (probability of skin cancer
in lifetime due to 1 jig/kg/day
of arsenic)
= 1.0 x 10-3
ono f°r U'S- persons
-------
Figure B-5.
rhk for .U.S.
linear and quadratic in dose
.01
Figure B-6.
Environmental Doses (yug/kg/day)
linear and quadratic in dose.
.0046
.7
Environmental Doses (/vg/kg/day)
80
-------
ra<,,ea.5.
0^0.3, P^ing to Taiwan SKn Cancer Data, Adiustec, for
Linear
Quadratic
Doses (d): 0, 10.818, 29.909, 50.909 ug/kg/dayc
9(d) = (0.351576 X10-7)d 9(d) = (0.106619 X10-7)d
H(t) = (t-6.934,2.885 Hm + (°-**°64x10-9)d2
, , ' H(t) = (t - 6.867)2.903
In L = -596.744 , ,
In L = -590.501
Unit risk: 4.0 x 10-3 (yg/kg/day)-'
Females:
Unit risk: 1.6 x 10-3 (ng/kg/day)-i
Doses (d): 0, 6.8, 18.8, 32.0 yg/kg/dayc
from Tseng et al., 1968
cDose estimate for U.S. persons (see Table B-3).
Chinese
years. However, since the
differed significantly ?n the
cancer in this age group js
in format on
"n S Syss
' and * 60
°™ years °ld
°f Skin
Table B-6. Since th re
Taiwan data for both genders were
in yg/kg/day for the rSSSTTs
same reason, it was necessary to c
of the reference U.S. person This
.. s
male (female) weighs 60 (55) kg and drinks 3 1
et al., 1983). If there were an equal
reference Mexican person would
experience,
'S Pr°Vided '"
, Sender-specific, the
^rS'ff tt°.d°Se e^ival^ts
^° the model- For the
d°S6 0Stimate to that
Mex''Can
wate'' da.ly (Cebrian
andfemales. »e
.
81
-------
Experience, Both Genders Combined
A f,s* ttmt tr\ (\t(mrfi\
Control town
UL (observed) 0/201 a (0)t>
PK (observed) 0/201 (0)
Exposed town
unobserved) 0/187(0)
PK (observed) 0/187 (0)
UL (predicted) 0.08/1 87 (0.04)
•
aoata from Cebrian et al., 1983.
bprevalence in percentages.
0/73 (0)
0/73 (0)
1/68(1.5)
8/68 (1 1 .8)
0.7/68(1.0)
_ ..
0/29 (0)
0/29 (0)
2/27 (7.4)
6/27 (22.2)
1.2/27(4.4)
0/15(0)
0/15(0)
1/14(7.1)
1/14(7.1)
Table B-7. Conversion of Arsenic
Dose for Mexicans to
Equivalent Arsenic Dose
for U.S. Persons^
Mexican person
(ppm)
• i
0.005
0.411
U.S. person
(pg/kg/day)
0.26
21.63
aAssumptions: A U.S. person weighs 70 kg
and drinks 2 L of water daily; a Mexican
person weighs 57 kg and drinks 3 L of water
daily.
response data for both genders
equivalents for the reference U S . pe son anc re better
model, with linear and quadra^ ° ten™ in dose, prov a 9 |jkelihood
^ th°e°comybined (i.e., sex-b.ind) data
are:
g(d) = (0.564398 x 10-8)d + (0.435613 x 10-»)d2
82
-------
and
H(t) = (t - 8.0)3.028
' «*
o, 8 years
carcinomas, but for which no E Ln 9 -of ePldermo'd or basal cell
diagnosis of ulceraU lesions n ^MeS'Ttudv T ^'f 8' The
d.agnosis of skin cancer in the Taiwan study V corresP°nds to the
reference's0^" (fe^fne'dosf1-63-11^^ 3S the dose rate fer the
1.5%, and 7.4%. The differences
resPectively: 0.0%,
* ,
was conducted by Rerz ( 965? (See M ?, f ™C) SetWeen
study.) uaos;. (bee II.A.3. for a description of this
ww« as
examined (total 262) by to?al dose '" CanCef (total 21> out of those
affect the response
_n »e study (e.g
differences are likely to
rates at equivalent total osea
for each dose rate and exposure comb.naflEJSCg
study
r6Sp°nse
83
-------
Follow-Up Study3
Fowler's solution Crucle__
(milliliters)
0-50
50 - 100
100 - 150
150 -200
200 - 250
250 - 300
300 - 350
350 - 400
400 - 450
450 - 500
500 - 600
600 - 700
700 - 1 ,000
1,000 - 1,500
1,500
respoiioB
0/24 ( 0.0)
2/45 ( 4.4)
2/24 ( 8.3)
1/12(8.3)
1/14(7.1)
1/31 ( 3.2)
1/17 ( 5.9)
2/11 (18.2)
2/11 (18.2)
0/7 ( 0.0)
1/18 ( 5.6)
1/14(7.1)
2/13 (15.4)
4/15 (26.7)
1/5 (20.0)
—
Adjusted
response*3
0/24 ( 0.0)
2/45 ( 4.4)
6/98(6.1)
6/98 (6.1)
6/98(6.1)
6/98(6.1)
6/98 (6.1)
6/61 ( 9.8)
6/61 ( 9.8)
6/61 ( 9.8)
6/61 ( 9.8)
6/61 ( 9.8)
2/13 (15.4)
5/20 (25.0)
5/20 (25.0)
• —
aResponse is given as no. oaroinomas/no. patients at risk, and, in
SOURCE: Fierz, 1965.
dose rate by the total number of exposure da^ Assuming , an avenge
iUweight of 70 kg and a weight o 7.6 mg «™% **^ to obtain an
solution, we multiply the total dose '" £9'Kg ,ytj (FS).1 The prevalence
ff anlS i. L, read w, to.
for 20
study, the prevaience rate ,s
mg arsenic/mL FS) - 9.2 x 10- mL
FS.
84
-------
B-8)
* .
V. Discussion of the Uncertainties of the Risk Estimates
Ih/Sp'eciarSrt' [g°J S^°sf 4«M *e ™ «*»»»« pressed in
elsewhere in that document ) In tht t?«'S T" a"eady been Bussed
' Sc= a
«ne quan,ltafce rlsk
(1) the mortality rate was the same in the diseased /skin
cancer) mdividuals as in the nondiseasec
-.._ . ,v, i-ji_.^,woovj II KJIVIUUolS.
(2) the population composition (with respect to the risk
factors of the skin cancer) remained constant over time
(3) the skin cancer was not surgically removed
mortality on the risk estimates. S6SS the 'mpact of differential
PO = the skin cancer prevalence at age x
PI = the skin cancer prevalence at age x +1
i the nondiseased persons in the age-
h =
the^mortality rate in the diseased persons in the age-interval
the age-specific skin cancer rate in the age-interval (x, x + 1)
F(t)= 1 -exp[- \ h(x)dx]
">
of differential mortality on the
85
-------
It is shown (Podgor and Leske, 1986) that the age-specific incidence rate,
h. satisfies the following equation.
= P. exp(-m ) +
°
1-P,
(1 _pQ)h[exp(- TO,) - exp(-mQ-Kj\
mQ—
h
From this equation, it is possible to investigate the effect of differential
increase ranging from about 2/0 to ^/° _w ,,«. re|atjve
two (rm = 2mo) is assumed; from about 2 ^ to _49 /o tjons are
«?
hr7eH- thtLCarrSenicS fntake fro^ ^lood consumption" can affect the risk
86
-------
Fabte B-9. Age-Specific Incidence Rates Calculated from Age-
Specific Prevalence with Equal and Diffferential
Mortalities
Exposure
groupb
Low-dose
Mid-dose
Low-dose
Skin
Age
20-39
40-59
60-69
20-39
40-49
60-69
20-39
40-59
60-69
Cancer Age-Specific 1
Observed Equal
skin cancer mortality
prevalence m1 = m0
1.07x10-3
6.13x10-3
4.66x10-2
3.77x10-3
4.85x10-2
1.64X10-1
2.22x10-2
9.89x10-2
2.54x1 0-1
1.07x10-3
5.94x10-3
4.16x10-2
3.78x10-3
4.59x10-2
1.29X10-1
2.25x10-2
8.17x10-2
1.89X10-1
ncidencea
Differential mortality^
m-) = 2mo
1.09x10-3
(2)
6.04x10-3
(2)
4.84x10-2
(16)
3.85x10-3
(2)
5.30x10-2
(2)
1.57X10-1
(22)
2.29x10-2
(2)
8.39x10-2
(3)
2.34x10-1
(24)
n^ = 3m0
1.11x10-3
(4)
7.06x10-3
(2)
5.56x10-2
(34)
3.91x10-3
(3)
6.06x10-2
(3)
1.85X10-1
(43)
2.33x10-2
(4)
8.61x10-2
(5)
2.81X10'1
(49)
Inn n°Jl ty rateS f?r those without skin cancer "B assumed to be 0 035 0 26
and 0.25 respectively for the age-intervals 20 to 39, 40 to 59, and 60 to 69
bForthe low exposure group, P0 = o, PI = i.07xl0-3 for the age-interval 20 to
fi3xi°o-3~P 2t'K^U=, 6-!3x10"3 for the age-interval 40-59; P0 =
fiQ?Fnrn;h 1 = 4'66x10"2 for the age-interval 60* (assumed to be 60 to
69). For other exposure groups, P0 and P-t are similarily defined
The parenthesized values are the ratio (xlOO) of age-specific skin cancer
purpose of discussion we will assume that a man in the study population ate
one cup of dry rice and two pounds of potatoes per day and thaHhe amW
of water required to cook the rice and potatoes was about L UndeTtnTs
f f8™?™,1 ,etr'Sk °alCU!f6d bef°re is overestimated by about 30% ('T J3 5
L). This caculation considers only the water used for cooking- the arsenic
content m the r,ce and potatoes that might have been absorbed from -
arsemc ,s not considered because of the lack of information. For a
would need
of the diet
certainly
VI. Summary
This section presents a dose-response analysis for skin cancer from
exposure to inorganic arsenic in drinking water. Results base^Ton The
multistage theory of carcinogenesis have been obtained from the Taiwan
87
-------
epidemiologic study and are compared to two studies in other environments
(Mexico- Cebrian et al., 1983; and Germany: Fierz, 1965). Compatibility of
results across studies (1) suggests the conclusion that arsenic exposure is the
likely causal factor in the increased prevalence of skin cancers in these
studies- (2) provides additional statistical evidence for refinement of statistical
estimates; and (3) helps to identify potential sources of variability and
environmental factors, or patterns of exposure, that may be influential.
None of these studies contains all of the details needed for an ideal
statistical analysis, such as: ages at times of initial exposure, termination of
exposure, and first appearance of skin cancer; similar information on lesions
that may frequently precede appearance of skin cancer; number of subjects
with cancer at multiple sites; locations of cancers; and prior disease including
those that lead to the use of Fowler's solution. Consequently, it is important to
glean what information is available from each study for purposes of
complementarity as well as comparison. . .
Analysis of the Taiwan data required estimation of the number at risk in
each dose/aqe category because only response rates and marginal totals by
aqe qroups are provided. The estimated values, which fit the marginal data
closely make possible the estimation of dose-response for the generalized
multistage model by means of maximum likelihood. The cancer response is
well described by a quadratic polynomial in dose (with positive linear
coefficient) for both male and female data. The minimum tumor induction time
is estimated at 7 and 9 years for males and females, respectively; in both
cases the cancer response for time-to-tumor is best described by time of
observation (minus induction time) to the third power. The observed data in
the Mexican study, taken at only one concentration of arsenic in well water,
but collected for different exposure intervals, are consistent with predictions
from the model using the Taiwan data.
The data from the study in Germany consist of the response of former
dermatology patients who had been treated with Fowler's solution (a 0.5 /=
solution of arsenic trioxide, which is a relatively toxic form). Patients were
treated for up to 26 years (many for apparently a much shorter period) in
intermittent dosing patterns specific to the prescribed treatment This is m
contrast to exposure to arsenic-contaminated well water which is likely to be
consumed at a reasonably uniform rate over time.
The published data do not include much information that could be useful
for risk assessment. Except for a few specific cases cited here, the data were
summarized by response for total dose. When compared to predictions from
the model for Taiwan with total dose held fixed at values equivalent to total
doses in the German study, and then varied over a wide range of possible
exposure durations in the Taiwan data, the skin cancer prevalence values in
the German study exceeded the values predicted.
In conclusion, the lifetime risk of skin cancer for a 70-kg person who
consumes 2 liters per day of water contaminated with 1 ug/L of arsenic is
calculated to range from 3 x 10-5 (on the basis of Taiwanese females) to 7 x
10-5 (on the basis of Taiwanese males); equivalent^, the lifetime risk due to
1 pg/kg/day of arsenic intake from water ranges from 1 x 10-J to 2 x 10--*-
88
-------
Appendix C
Internal Cancers Induced by Ingestion Exposure to
Arsenic
89
-------
Internal Cancers Induced by Ingestion Exposure to
Arsenic
As noted in the Technical Panel's Special Report on Ingested Inorganic
Arsenic areenic ingestion has been associated with cancer of internal organs
Chronic Senic ingestion has been reported to be asso-c.aec ' with cancer of
ar^a^^^
IB^ ' b^StS^^1^:^^ 51:
985, r98P6P)?nasopharynx (Prystowsky et al., 1978) OWney (Chen e a , 985;
Nurse, 1978), and other internal organs (Rosset, 1958; Rey™ann.^' 1 97n8J
Chen et al 985). Many of these references are case reports, however, and
do not dlserve the attention given a well-designed ep.demiolog.c study.
The Technical Panel felt it important to summarize the stud.es of Chen et
al 1985 1986) since these studies have been referred to in the text of the
Ss^^^
zrstfn
^
stud? of lung, bladder, and liver cancer mortality cases and random y
Sampled controls from the endemic area. They found odd ratios ; that were
sianHicantly (p < 0.05) elevated, and rema.ned much the same when
adiusS i for other risk factors including cigarette smok.ng. Chen et al. (1985)
fnSed a positive correlation between the SMRs of those cancers which
wee sianfficantly elevated and Blackfoot disease prevalence rates. Also,
SMRs we "greater in villages where only artesian wells were used « > the
^^SST±t?!^
s
andTnte^al cancer risk; however, the data is not sufficient to assess the
3ose-"?sPonse For this purpose, it is necessary to have the indiv.dua^
studied bv Chen grouped by well-water arsenic concentra .on and age.
Thesf data quite likely do (or did) exist, because they were ava.lable to Tseng
II ?i (1968) Tor the skin cancer study. EPA is currently trying to obtain these
data.
90
-------
Appendix D
Individual Peer Review Comments on Essentiality
91
-------
Individual Peer Review Comments on Essentiality
This appendix seeks to clarify some uncertainty in the workshop report of
arsenic essentiality."1
The Subcommittee also
nutritional essentiality. The
establishing essentiality as including: mnHp|o
1) performance of empirical observations in animal models
to establish the plausibility of nutritional essentiality,
2) establishment of a reproducible syndrome through the
use of chemically defined diets in animal models;
3) definition of biochemical lesions to characterize the
specificity of the lesions;
4) establishment of specific biochemical functions
absolutely dependent on the factor being investigated.
position on tnis 'f^ue- Arsenic (U S EPA, 1987) presents all of the post-
workshop comments in full.
I Comments on Plausibility of Arsenic Essentiality in Animals
A Post-Workshop Comments on Essentiality [page numbers
refer to the summary report of the Peer Rev/ew Workshop
on Arsenic (U.S. EPA 1987)]
The section [in the peer review draft] on [essentiality] of
Irsenlc should be rewritten with a more positive emphasis
on the probable [essentiality] of arsenic. . . (p. E-17).
1 Report of the EPA Risk Assessment Forum Peer Review Workshop on Arsenic,
December 2-3, 1986.
92
-------
Mushak: . . .the overall conclusion would seem to be that it is
premature to conclude that essentiality is established (p E-
uth be en°U9h ^'mental evidence
suggest that in some animals, diets low in arsenic
,
what would be found in the normal human diet
Further, the supplementary arsenic is all inorganic whereas
oraaaniSceTh,',n 1? human d'et is' '" a" 'ikelihL, almost S
d M ^Jn r' ^ a,m°Unt °f inor9a"ic arsenic in the human
diet (excluding drinking water) is really quite small (perhaos
a few yg/day), but there are no apparent health effects S
have been observed in humans. The relevaricTofthe anima
experiments to humans is therefore not at all clear Sdtt
seems unrealistic to believe that arsenic is needed in
B. Oral Comments Drawn from EPA Notes of Meeting-
'
and histidine were cited as f^'^&^O^S^
'
'
absence of a clear definition of a reproducible syndrome [general]
II. Estimation of a Human Nutritional Requirement for Arsenic
2 General discussion. Individual attribution uncertain
Arsenic,
93
-------
A Post-Worfcshop Comments [page numbers refer to the
summa% report of the Peer Review Workshop on Arsenic
(U.S. EPA 1987)J
Menzel- the development of the estimate for the human daily
reauirement H rf and carefu, dehneatlon of he
S should be included. . . .uncomfortable about, providing
a single estimate and would encourage the provision of a
range of values citing the uncertainties in the methods ; o
estimation and the interactions between arsen.c and methyl
donor. . .availability in the diet (p. E-17).
Straver I feel that a certain tone could be struck by the report to
StraV ndicate that evaluating the question of lower imits fo
arctic in drinking water is not so much a matter of direct
pS of essentiSy in any species. Rather, the fact that the
Dossibi itv of essentiality has been raised by workers in
Sy Disparate species and settings should deter us from
Sting very low limits even if proof of its essentially in man
is not forthcoming (p. E-30).
B Oral Comments Drawn from Observers Notes of Meeting
'
III.
c
er to make direct weight comparisons or to use
conversions); lack of a biochemical mechanism, and lacK or
Sedge of arsenic requirements as a function of age. [general]
Discussants reached a consensus that development of an order-
" ofTaSudeSmate of intake requirements is possible However^
SeTSt that the factors influencing the uncertainty of such an
assessment (as listed above) should be spelled out. [general,
subcommittee report 4]
Use in Risk Assessments
he sL cancer, and this is 'inappropriate. The nskrf
skin cancer is unlikely to be influenced by the possfce
essentfality of arsenic. The use of the risk model to regulate
Sfshould take into account such a P^W'^JJ^
does not appear to be a basis for doing so at this time (p.
E-6).
4 Report of the EPA Risk Assessment Forum Peer Review Workshop on Arsenic,
December 2-3, 1986.
94
-------
Menzel:
As a consequence of the agreement of the workshoo
participants on the probable essentiality of arsen^ a new
section will have to be added to deal with [the] problem [ol
essentiality versus toxicity] EPA should ace the
tPhP±rh0f,Hhehn0'threShold treatment of oncogenesis and
the threshold phenomenon of essentiality
I see no need to abandon the no-threshold treatment for
StTaT Tn *?, '^ arsenic °r Other "^nera'smfght £
essential. To not face this ,ssue directly will only encouraqe
Mushak. ft ,s premature to factor essentiality into risk assessment
models for arsen.c exposure in human populations ThSe
is no inherent limitation on the use of linear extrapoaX
models for, e.g., skin cancer, because of any thTeshdd
implicit in a daily required intake (p. E-21). ""^snoia
95
-------
-------
Appendix E
Metabolic Considerations
Prepared by:
Dr. William Marcus
Dr. Amy Rispin
97
-------
Contents
99
List of Tables ' ' ' ' 99
List of Figures 100
,!; Sp'Su^ve.s'of Arsenic; Chemica. Forms'and AvaNability ... 100
A. Drinking Water 101
B. Ambient Air '.'.'.... 101
C. Food 103
D. Occupational Exposed Groups 1Q3
E. Total Daily Body Burden
HI. Metabolism, Bioavailability, and Toxicity ^
A Toxicity of Arsenic Chemical Species ^
B. Absorption, Distribution, and Elimination • ^
C Detoxification via Methylation
D Human Metabolism and Enzyme Kinetics
IV. Pharmacokinetics of Arsenic Metabo.ism and Its Implications.for ^
Oncogenicity
98
-------
Tables
E-1 Percentage of inorganic arsenic in food: a preliminary analysis
E-2 Daily arsemc body burden feg/day) in the United States
102
104
E-1
E-2
E-3
E-4
Reproduction of arsenic
lypoic acid
Figures
I forms by
Role of s-adenosylmethionine in methylation of
arsenic III ...
107
E-5 Urinary excretion of arsenic (As) and its
107
110
99
-------
I. Introduction
of the rt*. associated «» human
"
11 Exposure Levels of Arsenic; Chemical Forms and Availab.lity
arsenic are via mgeshon of /°odj"°. 7^r' •' auamented in proportion to
(Weiler, 1987; IARC, 1986).
100
-------
A. Drinking Water
SaltS in the tri'
, ,
'
S. Ambient Air
C. Food
101
-------
Table E-1. Percentage of Inorganic
Arsenic in Food: A
Preliminary Analysis3
Food
Percentage of
Inorganic Arsenic
Milk and dairy products
Meat - beef and pork
Poultry
Fish - saltwater
- freshwater
Cereals
Rice
Vegetables
Potatoes
Fruits
75
75
65
0
10
65
35
5
10
10
oecaon of the arsenic content of basic food
g?S b^sed on preliminary data from the Ontano
Research Foundation and other sources.
SOURCE: Weiler, 1987.
102
-------
D. Occupationally Exposed Groups
E. Total Daily Body Burden
MS«
III. Metabolism, BioavailabilSty, and Toxicity
A. Toxicity of Arsenic Chemical Species
103
-------
Table E-2. Daily Arsenic Body
Burden (ug/day) in the
United States
Source
Usual
Unusual
Water
Air
Food
Smoking
TOTAL
5
0.09
50d
55.09
100a
1.5 -45&
68°
50
2 -6e
up to 224
aAt the ODW maximum containment level
(see Part II.A).
bNear industrial use sites such as smelter or
cotton gins (see Part II.B).
^Occupational exposure.
dSee Part II.C.
e2 pg arsenic/package (Weiier, 1987;
I ARC, 1986).
while arsenate may interfere with phosphorylation reactions due to its
cheS^^^
B Absorption, Distribution, and Elimination
104
-------
*
and control subjects, but eels of
comParable for workers
exposure to arsenic in toharm
that some smokersln the ?9Ss
each day. Although thec
known and the duration of exoosurp
completely defined he Brune et a?
inhaled arsenic binds irreveSy to
hJWTuSS fiffi ±S
water ranging from 6 ng/L fc TSa /L
concentrations increased in urine and
in concentrations in drink ig™t™
biood unti, drinking water co'nceSions
Ple>
a function of chronic
(1986>
is not
° -9!?UpS Of retirees is ™t
^ ** * P°rtion of
blood- urine- and
Proportion to i
arsenic
in
the body via sweating and desquamo'n "' oTfhT llf'T ""? inated from
excessively exposed to inorganic arsenic SIP hL«^ r humans not
arsenic is generally found in skin hat and nfi ,.?,!* ft!sue concenfation of
Kagey et al. (1977Ulsc ^studied womenTn ?t2" ^5ntd ftS andSith' 1968)'
umbH,cal cord levels of arsenic were similar to ma Sna? SSs
's-Scof as ^^?
ss?^rr&^
(inorganic) arsenic and used whole Sv rad ionSnh,Te ^- radiolabeled
and clearance. Initial concet^^ y * L djstribution
apatite crystals in bone can aSc ^ ^^ *"* PhosPnate in the
hair, and upper gastrointestinaTtrarTS M »? c^umulatlon of arsenic in skin,
keratin (Goyer 1^rointestinal tract * «s binding of sulfhydryl groups of
vivo (Vahter and Marafante 1983) The
oral administration (Vahter et al 984? In
different pattern from that shown '
r the tissues
°tained
105
-------
discussed above. The highest
found in the kidneys lungs ^ 9|s
the longest retention time were me l
Tissue retention of arsenic ;njhe
methylate arsenic was much ' •J^f
methylate arsenic (Vahter and Marat ante
injection with inorganic arsenic a most 60 /° or .
The major single binding site wasJne"' !? "intestinal tract. To the extent that
was also retained in the kidney and gas • °£tesj™ ™ stribution and tissue
the marmoset monkey may ^.fS^^the normal detoxification
that arsenic is efficiently absorbea tnrouy y (eve|s Qf exposure
a
ntestinal walls, and lens.
k hich doesn't
^Yn species which
Seventy-two hours atter
y tjssues
s^Dorj inal dose. Arsenic
«.
sites.
C Detox/f/cat/on Via Metfiy/at/on
'Methylation of inorganic ^arsenic is
arsenic methy,ation
and detoxification are summarized Im
Methylation appears Mn ^ta
(Vahter and Envall.1 983). Bas
al. (1984) hypothesized that me hylation
. ^
model compounds, Cullen et
arsenic III requires s-
a on the
arsenate was •fn\itro cellular uptake of the two
o at physiol°9ic PH' arsenite is not
ionized, whereas arsenate is charged sequences of
In order to understand react: on m echamsnv s inMcubations Of
methylation, Buchet and auwery .PTte°r^thylating capacity of red
arid ,dney r 05.^^
activity; and s-
106
-------
Source: Cullen etal., 1984.
HS
HS
HS ,
OH
.
1
MeAs
Figure E-2. R-Jaf «dsr.osy.methionine in Cation of arsenic ,„.
Source: Cullen etal., 1984.
S-Adenosylmethionine -- As
S-Adenosylhomocysteine
MeAs
concentrations, DMA formation was InhlbHeTiXe
107
-------
accumulate in the system, showing that formation of DMA is a rate-limiting
rabbits. Perlodate-oxidized jjenosme ^PAD) an tratjon Qf the
lead to decreased excretion of DMA m tneur . deficiencies lead to
the lungs, skin and liver. '" ^J*0"'.^mes These results in animals
108
-------
°f
Distribution,
D. Woman Metabolism and Enzyme Kinetics
eliminate arsenic at levels of concern numans can handle. detoxify, and
109
-------
Figure E-3.
Urinary concentrations of arsenic and its metabolites.
Source: Adapted fromBuchet et al., 198Z.
1400
100
725
250 50°
Micrograms As per Day
1000
saturation pattern would require that EPA obtain the raw data from Buchet's
£!2f of loTarS ffl%% in people of adequate methy.ating capacrty
(Fi?nrt9853)Lovell and Farmer monitored urine for arsenic metabolites
fonowinf ySn of rg^J-c ^o se^of i^^-n, ^ P ^
110
-------
Figure E-4.
Source: Tarn et al., 1979
1.5
Day
Total arsenic
Inorganic arsenic
X Monomethylarsenic compound
A Dimethylarsinic acid
111
-------
MMA as observed by urmary excret,on_ ooes bQd Qr
workers exposed to high eve
Urinary concen rations o araenic and .s m However_ whep high
with time nearly to that of the control popuh a inorganic
exposure (Figure E-5). Fu 'h«™°™:,t<,asf Ration in concentration of
evening sampling *°0n8f concentration o. its methylated
-echan
o? excre«on o, metabolites indicates a
arsenic metabolites
beon
112
-------
*""*
Source: Foa et al., 1984.
As exposure
er>d resumption
I t
' ArSen'° Metabo"sm
ff
is melhylated enzymatically h the
'« implications
Koestzcor^.
' In!lr9a"'c
certain target tissues, namely he ^ ver i.fn lon9:terrT deposition in
gastrointestinal tract name|y tne l.ver, lung, skin, bladder, and
or pro,*. d9ficfent
113
-------
Engine to the manifestation of cardnogenic ™sponM in
studies of certain highly expo^ groups (US EPA. «8^
to arsenic (in excess of 200 pg/day) "*cate Wat^^ d t to chronjcally
be complete. However, even , rh.human body accomm ^ tQ much
elevated arsen.c levels,theJ"^"3'^3.^ Of time Furthermore, the ability of
more inorganic arsen.c over long^periods or^>me^ constitute a
the human organism to handle more than,500^ or buu ng y yhomeostatic
stress to the body. .^ ™^^dftuen^rd0S-response assessment.
mechanisms is cnt.cal to -mprovrng thei cancer^^aos H t^ ^^ |ungj
Appendix C summanas date on elevate^raes o^ tumQrs jn ^
and bladder in Taiwan and also notes the occur d fi , t animalSi one
Rerz study. Extrapolatng from ^ stud,es an pr protein-deficient
SSSi'SSK SK ^maTSAether the deposition patterns
arematehed by confirmed incidence of internal cancer.
114
-------
IX. References
Albores, A.; Cebrian, M.E.; Tellez I •
'
Publica " °
Proisction
<— » -
°f
Tech.
endemic
Astrup P. (,968) Blackfoot disease. Ugeskr. Ueger 130 1807-1815
^^
and death
115
-------
Biagini, R.E, Quiroga, G.C, Elias, V. (1974) Chronic hydroarsenism in ururau.
S
1 1 «5 FPA Headauarters Library, Washington, D.C.
Headquarters Library, Washington, D.C.
116
-------
microsomeson ,he
, MU,». Re,
,,
biologic,, ^wion iAS^™*™** *'
Cuz,k, J.; Evans, S.; Slllman. M.; Price Evans D IIQnsi i/i' ,
internal mali9nancies. Br. J. Cancer ^904.^' ed'Crnal a'56nfc an<1
' «°ru.ne. xylenes and
v ' . B.W,
arsenobelain. the onj, t of* ° a,"d SV"""='S of
Tetrahedron Letters 13 1543-lS? *e W8stern rock
hepatic angiosarcoL Am J fnd
b
117
-------
Sever. L. (-898, "bar d,, chronlschen
U9"
Syph. 43:221-283 (transited
oondi"ons '
,or
Organization.
France: World
Angiosarcoma of the liver
: Morpho,. l^azsagugyi Or.. Sz.
Can-
E,feo,s
o? inorganic arsenic.
supply. Environ
Johnson. R.D.;
.
u«id°e"''in adu«"oSi diet samples. August
/A soToS. Ana,. Chen, 6W.154.,66.
118
-------
1' (1972) Leukemla '
Knoth, W. (1966) Arsentahandlung. Arch. Klin. Exp. Derm. 227228-234
n^'^""^- RJ' Sumner' KW- D«. C.: Boswell DC Arch
1975) Ang,osarcoma o( the liver associated with '
(potassium arsenlte). Gastroenterology es-l^-JS
Leonard, A.; Lauwerys, R.R. (1980) Carcinogenicity teratooeniritv anH
mutagenicity of arsenic. Mutat. Res 75-49-62 terat°9en'city, and
H-.('9681 Ess'"8al and nonessential trace elements A
«*
M>csss
of BAL therapy, din. Toxicol. 18-1067-1075
«
119
-------
sr ss
Pharmacol. Toxicol. 58 Supplement II.
-------
Poma, K.'f Degraeve, N.; Kirsch-Volders, M. (1981) A combined action of
arsenic and ethyl methanesulfonate (EMS) in somatic and germ cells of
mice. Mutat. Res. 85:295.
Popper, H.; Thomas, L.B.; Telles, N.C.; Falk, H.; Selikoff IJ (1978)
Development of hepatic angiosarcoma in man induced by vinyl chloride
thorotrast, and arsenic. Am. J. Pathol. 92:349-376.
chronic hyPerkeratosis in Pisagua (as cited in
Prystowsky, S.D.; Elfenbein, G.J.; Lamberg, S.I. (1978) Nasopharyngeal
t Ion9"term arsenic in9est'°n- Arch. Dermatol.
rrnmf,f ' " ' 'F' (1968> Hemangioendothelial
sarcoma of liver from chronic arsenic intoxication by Fowler's solution.
~
.
Reymann, F.; Moller, R.; Nielsen, A. (1978) Relationship between arsenic
mteke and internal malignant neoplasms. Arch. Dermatol. 114-378-381
Reynolds, E.S. (1901) An account of the epidemic outbreak of arsenical
SnH9 oc
-------
qrhmidt
- Anke M- Groppel. B.; Kronemann, H. (1984) Effects of As-
dcien^ on skeletal muscle, myocardium and liver: a histochemical and
ultrastructural study. Exp. Pathol. 25:195-197.
-------
U.S. Environmental Protection Agency (EPA). (1986) Guidelines for
carcinogen risk assessment. Federal Register 51 •33992-34003
U'S wEonrShnn iTL^**" A9e"Cy (EPA)" (1987> Risk Assessment Forum
W86 Available fTm ' 1?%"^"*?" °f * WOrkshop held in December
1986. Available from: U.S. EPA Headquarters Library, Washington DC
1. ^ielSen> RH' (1985) Effects in chicks of ^senic, arginine and
zinc and their interaction on body weight, plasma uric acid plasma uSa
and k,dney argmase activity. Biological Trace Element Research 7™-
Uthus, E.O.; Cornatzer, W.E.; Nielsen, F.H. (1983) Consequence of arsenic
fepnvatoon m laboratory animals. In: Lederer W.H, Fensterhemlj
?o?k NY-"^ in2UStrial'Hb'D0medical' and environmental perspectS, Net
York, NY. Van Nostrand Remhold, pp. 173-189
Vahter, M. (1983) Metabolism of arsenic. In: Fowler, B.A. ed Bioloqical and
environmental effects of arsenic. Amsterdam: Elsevier Chapter 5
^^^ exposijre to '*«***>
°f ***** ln m'Ce and <****•
r;; Ma';afante' E- <1983) Intercellular interaction metabolic fate of
arsenite and arsenate m mice and rabbits. Chem. Biol. Interact. 4729-
Vahter, M.; Marafante, E. (1985) Reduction and bindinq of arsenatP in
marmoset monkeys. Arch. Toxicoi. 57:119-124 arsenate in
!' E'; Dencker' L <1984) Tissue' distribution and retention
10 acid in mice and rats- Arch- Environ- Co
' °979) ArSeniC levels in human blood-
eXP°SUre Vla drinking Water- Environ-
Wagner, S.L ; Maliner, J.; Morton, W.E.; Braman, R.S. (1979) Skin cancer and
arsenical intoxication from well-water. Arch. Dermata? 115 1205-1207
S°nG1977 A6 indce Curve for cancer- Pr^- Natl. Acad. Sci.
87 ?8R4snn8n7)nSPUTiShed data" Ministry of the environment, report no
87-48-45000-057, Toronto, Ontario. Available from- U S EPA
headquarters Library, Washington, DC.
Welch, K; Higgins, I.; Oh, M.; Burchfield, C. (1982) Arsenic
in
Whittemore, A^; Keller, B. (1978) Quantitative theory of carcinoqenesis
Society Ind. Appl. Math. Review 20:1-30. <-arcmogenesis.
World Health Organization. (WHO, 1981) Environmental health criteria 18-
A^en,c2:g,nternational programme on chemical safety aTnevalpp 63,"
Yamashita, N ; Doi, M.; Nshio, M.; Hojo, H.; Masato, T. (1972) Current state of
Chld°ned MorinaOa
123
-------
*aO»fiQ9-fift7
Yeh, s! (1973) Skin cancer in chronic arsenicism. Human Pathol. 4(4): 469-
™™'
*
chronic arsenicism in Xinjiang. Chin. Med. J. 98(3).21 9-222.
Zaldivar R (1974) Arsenic contamination of drinking water and foodstuffs
causing endemic chronic poisoning. Beitr. Pathol. 151:384-400
" 8 Guillier A (1977) Environmental and clinical investigation on
demic^hfonic'arinic poisoning in infants and childr^
Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. 1. Ong. Re.he B.
243.
124
-------