oEPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 2771 1
EfA-600/2-78-107
May 1978
Research and Development
Estimation of
Risk from
Carcinogens in
Drinking Water
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-78-107
May 1978
Estimation of Risk
from Carcinogens
in Drinking Water
by
Robert W. Handy
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, N.C. 27709
Contract No. 68-02-2612
Task No. 16
Program Element No. 1BB610
EPA Project Officer: Max Samfield
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
This report gives the results of a study aimed at developing a
means for estimating cancer mortality as a function of carcinogen
concentration in drinking water. Cancer risk data for cigarette smokers
was treated by the method of standard additions to provide an estimate
of ambient carcinogen levels in drinking water. A similar treatment
was carried out on lung cancer risk data to give an estimate of carcinogen
levels in ambient air.
This report was prepared for the Industrial Environmental Research
Laboratory-RTF, Environmental Protection Agency, to present results of
the work carried out by RTI under Contract No. 68-02-2612 (Task 16).
This work was performed in the Chemistry and Life Sciences Division of
the Research Triangle Institute.
ii
-------
CONTENTS
Abstract 11
Figures iii
Tables IV
Table of Symbols and Definitions v
Introduction 1
Summary and Conclusions 2
Caveats 5
Methodology 7
Ambient Air 7
Drinking Water 18
References 33
FIGURES
Number Page
1 Lung Cancer Mortality Versus Age and Cigarette Use ±Q
2 Standard Additions Method 14
3 Lung Cancer Mortality Versus Carcinogens Intake of Dorn
Study Smokers 17
4 Lung Cancer Mortality Versus Carcinogens In Ambient Air -
General Population 20
5 Other Site ..Cancer Mortality Versus Carcinogen Intake of
Dorn Study Smokers 25
6 Other Site Cancer Mortality Versus Carcinogens in Drinking
Water at Different Ambient Air Concentrations. General
Population 28
iii
-------
TABLES
Number Pag
1 Carcinogenic Hydrocarbons Isolated From Cigarette Smoke . . 8
2 Respiratory Carcinogen Intake for Smokers as a Function
of Cigarettes Smoked Per Day 9
3 Lung Cancer Mortality as a Function of Smoking Pattern
and Age (From Dorn Study Unless Indicated Otherwise) .... 12
4 Lung Cancer Mortality Versus Carcinogen Intake for Dorn
Study Group, Age 45-84 16
5 Estimated Nonsmoker Lung Cancer Death Rate Versus
Carcinogen Concentration in Ambient Air 19
6 Cancer (Other Than Lung) Mortality as a Function of Smoking
Patterns and Age (From Dorn Study) 22
7 Cancer Mortality and Respiratory Carcinogen Intake for Dorn
Study Group, Age 45-84 23
8 Relationship Between Cancer (Other Than Lung) Mortality
and Systemic (Oral) Carcinogen Burden 24
9 Oral and Respiratory Uptake of Carcinogens as a Function
of Ambient Air and Drinking Water/Food Quality 29
10 Other Site Cancer Mortality as a Function of an Ambient Air
and Drinking Water/Food Quality 30
11 Summary of Carcinogen Concentration In Drinking Water
and Corresponding Cancer Risk 32
iv
-------
TABLE OF SYMBOLS AND DEFINITIONS
A,B,D = Correlation constants,
L = Respiratory carcinogen intake (|Jg/yr or ng/day) ,
MI = Annual lung cancer mortality/100,000,
M = Annual cancer mortality/100,000 at all sites excluding lung
OS
and bronchus,
N = Number of cigarettes smoked/day,
2
r = Correlation coefficient,
S , = Systemic carcinogen uptake from the gastrointestinal tract
from food and drinking water (ng/day),
S, = Systemic carcinogen uptake from drinking water (ng/day),
Sf , = Systemic carcinogen uptake from food (ng/day),
S = Systemic carcinogen uptake from respiratory tract (ng/day),
S , = Systemic carcinogen uptake from both the gastrointestinal and
L.O L3 J_
respiratory tract, S , + S (ng/day),
* J ' oral resp 6 * '
X = Carcinogen concentration in ambient air (ng/m ),
aa
X, = Carcinogen concentration in drinking water (ng/1),
X = Estimated safe drinking water concentration (ng/1),
X- , = Carcinogen concentration in food (ng/kg).
-------
INTRODUCTION
The purpose of this report is two-fold. (1) Lung cancer risk:
Carcinogen exposure data will be evaluated using the method of standard
additions to estimate background ambient air carcinogen levels. (2) A
means of estimating cancer mortality risk as a function of carcinogen
concentration in drinking water will be developed using the same smoker
mortality data that served as the basis for the ambient air treatment.
The EPA has set several maximum concentration exposure levels for
environmental pollutants in both air and water. However, there are many
known toxic materials for which no safe concentration levels have been
established. The main objective of an earlier report was to derive
pollutant hazard criteria for this group of compounds in order that
safe, permissible concentrations in the environment might be estimated.
A means for estimating lung cancer mortality risk as a function of
carcinogen concentration in ambient air was developed and presented in
Section V of that report. Since lung cancer mortality data for smokers
was the best human dose response data available, this information was
used to correlate carcinogen exposure and lung cancer death rates.
-------
SUMMARY AND CONCLUSIONS
A previous report by Handy and Schindler analyzed published lung
3
cancer mortality data for smokers (Kahn Study) and derived an estimate
for a carcinogen level in ambient air. The result of this treatment gave
a carcinogen concentration to which the nonsmokers of the study group
were exposed prior to 1962 (the Kahn study covered cancer deaths from
3
1954 to 1962). A value of 4.9 ng carcinogens/m was obtained using a
methodology that did not take into account the ambient air contribution
to the smokers' respiratory carcinogen intake.
This report evaluates the same smoker-lung cancer risk data using a
treatment analogous to the standard additions method used routinely in
the field of analytical chemistry. The result obtained by this approach
3
gives an estimate of 22.9 ng carcinogens/m , higher than the previous
approximation by a factor of approximately five.
From this treatment, it is concluded that the concentration figure
3
of 22.9 ng/m more closely approximates the actual carcinogen concentration
in ambient air prior to 1962 and that amounts in excess of this level
will result in an increased lung cancer risk relative to the pre-1962
base period (see Figure 4) . The lung cancer mortality may be estimated
by the following equation.
X (ng/m3)
M= 362.5 In <
lc . 350.5
when X = 22.9, M. = 22.9
aa lc
A similar treatment of total cancer mortality, excluding lung and
bronchus, gave a nonsmoker systemic carcinogen intake of 92.7 ng/day
from all sources (air, food, and drinking water). Analyzing this "other
site" mortality data within the limitations delineated in the Caveat
-------
Section afforded an expression relating "other site" mortality (total
cancer risk excluding lung and bronchus) and carcinogen concentration in
drinking water. Since this cancer death rate is a function of both oral
and respiratory carcinogen intake, the effect of drinking water carcin-
ogens on cancer risk is depicted graphically (Figure 6) at different
ambient air concentrations.
3
The estimated ambient air concentration of 22.9 ng/m corresponds
to a systemic uptake of 116 ng/day through the respiratory tract. Since
this checks closely with the 92.7 ng/day figure for total systemic
uptake from all sources, it must be concluded that the average carcinogen
intake through the gastrointestinal tract was negligible prior to 1962.
It is also apparent that any combination of carcinogens in ambient
air, food, and water which results in a total systemic carcinogen uptake
in excess of 92.7 ng/day will produce an elevated other site mortality
relative to the pre-1962 base period. This daily systemic carcinogen
3
uptake corresponds to a maximum ambient air concentration of 18.3 ng/m
(no contribution from food and drinking water) or a maximum drinking
water concentration of 30.9 ng/1 (no contribution from ambient air).
The total systemic carcinogen uptake (S , ) is determined by the
following equations.
S = S + S
total (ng/day) resp oral'
where S , ., , = 5.07 X (ng/m3)
resp (ng/day) aa 6/
and S -. f ,, -. — 3X, .
oral (ng/day) dw
The total other site cancer mortality may then be estimated by the
following equation.
M = 43.3 In (- + I)-
os 0.34
Use of Figure 6 allows the estimation of other site cancer mortality
at different carcinogen levels in ambient air and drinking water. For
example, an increase in the drinking water concentration from 4 to 20 ng/£
3
at a constant carcinogen air concentration of 10 ng/m results in an
other site cancer mortality increase from 226 to 251/100,000 (11% increase)
3
-------
3
Similarly, increasing the air concentration from 10 ng to 25 ng/m at a
constant drinking water concentration of 4 ng/£ gives an other site
cancer mortality increase from 226 to 360 (15% increase). Note that
examples reflect an increase cancer risk relative to the pre-1962 base
period.
-------
CAVEATS
The reader should be aware of the following limitations of the
methodology used in this report.
a) Ambient Air-
The term carcinogen refers only to agents responsible for the
initiation and development of lung and bronchial cancers.
It is assumed that a major contributing factor in the inci-
dence of lung cancer death of cigarette smokers and nonsmokers
alike is the inhalation of carcinogenic compounds.
The exposure level of eight (8) hydrocarbon carcinogens,
identified and quantitated in cigarette smoke, is assumed to
be a major contributing factor in lung cancer incidence and a
valid measure of lung cancer mortality of smokers and nonsmokers
The 8-component hydrocarbon mixture, present in cigarette
smoke, possesses a total carcinogenic potency equal to a
mixture of the same materials in ambient air.
The presence of other carcinogenic compounds in ambient air
is assumed. The quantity (concentration) of these compounds
in ambient air may be expressed in terms of an equipotent
amount of the eight cigarette carcinogen mixture.
A number of lung cancer deaths are due to carcinogens in the
ambient air and this is reflected in the lung cancer mortality
of nonsmokers.
2
Any mathematical relationship (r >.95) between carcinogen in-
take by smokers and the lung cancer death rate corresponding
to these smokers may be extrapolated to yield valid mortality
data for low carcinogen intake values.
Ambient air concentrations have been calculated assuming a
3
total breathing volume of 3700 m /year for the average adult.
-------
b) Drinking Water
The term carcinogen refers only to agents responsible for the
initiation and development of cancers at all sites, excluding
lung and bronchial tissue.
The 8-component hydrocarbon mixture, present in cigarette
smoke, possesses a total carcinogenic potency equal to a
mixture of the same materials in drinking water.
The presence of other carcinogenic compounds in drinking water
and food is assumed. The quantity (concentration) of these
compounds in drinking water and food may be expressed in terms
of an equipotent amount of the eight cigarette carcinogen
mixture.
A number of cancer deaths are due to carcinogens in drinking
water and this is reflected in the cancer mortality of non-
smokers .
Carcinogens ingested orally in the form of food and drinking
water exert a negligible effect on the risk of lung cancer,
but a major contributing factor in cancers at all other sites.
Fifty percent of inhaled carcinogens are released from the
respiratory tract and into the systemic circulation.
Food and drinking water contain the same weight percent
concentration of carcinogens.
It is assumed that the average adult consumes two liters of
drinking water and one kilogram of food per day.
2
Any mathematical relationship (r >.95) between systemic carcino-
gen body burden and the cancer mortality rate at all sites,
excluding lung, may be extrapolated to yield valid
mortality data for low carcinogen burdens.
-------
METHODOLOGY
Ambient Air
Cigarette smoke has been analyzed for carcinogens components by
several investigators. The amounts of eight (8) hydrocarbon carcinogens
present in cigarette smoke have been reported by Wynder and Hoffman and
2
are listed in Table 1 with a ranking of their relative carcinogenicity.
Knowing the carcinogen content in cigarette smoke, a respiratory
carcinogen intake rate (L ) is readily determined for individuals
ITGSp
smoking a different number of cigarettes/day for varying lengths of time
using the following expression:
a 10. 8 (N) 365
resp 100
L = 39. 4N
resp
where L = respiratory carcinogen intake (yg/yr)
10.8 = carcinogen content (yg) per 100 cigarettes (See Table 1)
N = number of cigarettes smoked/day
Values of L for different groups of cigarette smokers are shown
in Table 2.
Lung cancer death rates as a function of age and individual smoking
patterns have been reported in the Dorn Study of Smoking and Mortality
3
Among U.S. Veterans. This study was conducted over an eight one-half
year period with a population of over 293,000 military veterans holding
Government life insurance policies. The study group was composed of
practically all white males drawn from the middle and upper socioeconomic
classes.
The Dorn study gives the cause of death for individuals classified
into ten-year age brackets (45-54, 55-64, 65-74 and 75-84). In correlating
age and lung cancer mortality (see Figure 1) these groupings are identified
-------
Table 1. CARCINOGENIC HYDROCARBONS ISOLATED
FROM CIGARETTE SMOKE
Relative ^
carcinogenicity
Hydrocarbon
Benzo-a-pyrene
Dibenz-a,h-anthracene
Benzo-b-fluoranthene 4+
Benzo-j-fluoranthene ++
Benz-a-anthracene +
Chrysene +
Benzo-e-pyrene 4-
Indeno-1,2,3c,d-pyrene +
Total yg/100 cigarettes
* Carcinogenicity determined on mouse skin.
** Isolated from cigarette smoke.
Micrograms per
100 cigarettes**
2.5
0.4
0.3
0.6
0.3
6.0
0.3
0.4
10.8
-------
Table 2. RESPIRATORY CARCINOGEN INTAKE FOR
SMOKERS AS A FUNCTION OF CIGARETTES SMOKED PER DAY
Carcinogen intake from cigarettes (L )
Cigarettes smoked/day (N) yg/yr ng/day
5 197 540
15 591 1620
30 1182 3240
50 1970 5400
-------
1,000
500 L_
o
o
o
cT
o
LJJ
I
<
LU
Q
CC 100
LU
O
z
<
o
50
10
Cigarettes Smoked/Day
0- Nonsmoker
O- 5 (1-9)
A- 15 (10-19)
D- 30 (20-39)
A- 50 (39+)
40
50
60 70
AGE, YEARS
80 100
Figure 1. Lung cancer mortality versus age and cigarette use.
10
-------
by their median age (50, 60, 70, and 80). Dorn classified smoking
patterns as never or occasionally smoked, 1-9, 10-20, 21-39 and more
than 39 cigarettes/day. These groupings are indicated in the present
report as follows: Nonsmokers, 5, 15, 30 and 50 cigarettes/day.
The data used in this evaluation was taken from the Dorn study as
presented in a paper by Kahn on pages 30, 38, 40, 42 and 44 under the
3
cause of death listing of cancer of lung and bronchus. The death rate
in each category was determined by dividing the number of deaths by man-
years of observation and relating this value to a rate per 100,000.
Supplemental data was taken from a similar study conducted by
4
Hammond. This study was carried out over a 40-year period and con-
sisted of over one million men and women. The study results were re-
ported separately by sex. The male cohort of nonsmokers under age 55
was much larger than the comparable Dorn group and, as a result, pro-
vided a more reliable estimate of lung cancer risk for that category.
The lung cancer mortality figures from both studies have dubious
significance for age groups under 45 years. The reported number of
deaths in these study subgroups was too low (less than 5) to warrant use
in this report. Any age-smoking mortality rates derived from less than
5 reported deaths were considered invalid. These values were established
by alternative means as described below. Hammond set a similar criterion
in evaluating his study results.
The annual death rates used in this treatment for the different
age-smoking categories is shown in Table 3.
The operations that lead to the final expression describing the
relationship between lung cancer risk and ambient air carcinogen levels
are summarized below.
a) construction of log-log plot of lung cancer mortality vs. the
number of cigarettes smoked per day (also expressed as [Jg
3
carcinogen intake/year and annual mean ng carcinogens/m air)
for the composite age group 45-84.
b) The relationship between lung cancer mortality and carci-
nogen intake was represented by a straight line, described
11
-------
Table 3. LUNG CANCER MORTALITY AS A FUNCTION OF SMOKING PATTERS
AND AGE (FROM DORN STUDY UNLESS INDICATED OTHERWISE)
Cigarettes Lung cancer Person-years Lung cancer
Asa smoked/day deaths of observation death rata/ICO .000
45-54
55-64
65-74
75-84
MS
5
15
30
50
NS
5
15
30
50
MS
5
15
30
50
IIS
5
15
30
50
1
(1)
(9)
10
(3)
25
31
183
245
63
.49
44
239
194
50
4
5
15
7
(2)
15,134
3,129
'16,392
12,839
1,928
213,858
45,217
151,664
103,020
19,649
171,211
37,130
101,731
50,045
8,937
8,489
1,923
3,867
1,273
232
6.4*
23.0
57.0
77.9
160.0
11.7
68.6 :
120.7
237.8
320.6
28.6
113.5
234.9
3S7.7
559.9
47.1
260.0
387.9
549.9
8*2.0
From Hammond Study (ref. 4).
Deaths in parenthesis were calculated from a death rate obtained by
extrapolation (Fig. 1).
12
-------
Y
mathematically in the form Mn = A(L ) where L =
J Ic resp resp
respiratory carcinogen intake, A and Y = constants.
c) M, was set equal to the value given for nonsmokers and the
corresponding carcinogen intake (L ) calculated for this
subgroup.
d) Since the Dorn study population consisted only of white males,
a correction factor was derived so that the mathematical
expressions would apply for the general population,
Since the only source of respiratory carcinogen intake for nonsmokers
is ambient air, the value of L obtained in (c) was a measure of the
resp 3
cigarette carcinogen concentration in ambient air, X =4.9 ng/m .
3 3
However this is only an approximation. Since cigarette smokers also
inhale ambient air, their carcinogen intake values would have to be
increased by this background level. This would result in a new series
of data points, a different linear representation and a revised ambient
air estimate. A different approach to this problem is described in the
following section.
The technique called the method of standard additions is commonly
used by the analytical chemist to determine the quantity of a particular
component in a complex matrix. The procedure involves the addition of
known amounts of the "component of interest" to the complex matrix
"original sample" and plotting the changes in instrument response as a
function of the amount of added "component of interest". The line that
describes the data points is drawn and extrapolated to the negative X--
axis. The absolute value of the X-intercept corresponds to the quantity
of the "component of interest" in the original sample. The technique is
shown graphically in Figure 2.
The estimation of cigarette carcinogens in ambient air is amenable
to this kind of treatment. The situation is analogous to the general
analytical case described above when making the following substitutions:
Instrumental Response = M, ,
Original Sample = Nonsmoker,
13
-------
I I
Instrumental
Response
j i i
0
Amount of Component of Interest
Added to Original Sample
x-lntercept = Amount of Component of Interest in Original Sample
Figure 2. Standard Additions Method
14
-------
Component of Interest = Cigarette carcinogens,
Standard Additions = Quantity of carcinogen intake by different
groups of cigarette smokers,
X-Intercept = Quantity of cigarette carcinogen intake by nonsmoker;
To perform a standard additions treatment it is necessary to
replot the Dorn study smokers' carcinogen intake and M, data on Cartesian
coordinates. Several different curve-fitting regression programs were
attempts, on the data in Table 4. The one which yielded the highest
2
correlation coefficient (r = .9968) was an exponential of the following
form:
BM..
L = Ae - D
resp
1 n , resp + DN
M - ln - -
lc B A
where A, B, D = constants
The values of the constants were determined to give the following
expression:
.0024M - 3.785
L = 3.553 e
resp
When Mlc = 0, X = 3.553-3.785 = -0.232 (X- intercept)
Nonsmoker carcinogen intake = 232 ng/day
The smooth curve in Figure 3 describes the above equation and shows
the agreement with the original data points. The Dorn study nonsmoker
carcinogen intake value corresponds to 84.7 pg/year which is approximately
5 times higher than the 17.9 pg/year value previously reported in
Section V of reference 1. The former figure is equivalent to an annual
3
mean carcinogen concentration in ambient air of 22.9 ng/m (see eqn. 73
in Ref. 1 for conversion calulation) . This baseline level is equivalent
to 2.15 cigarettes/day.
The graph in Figure 3 may be corrected for the carcinogen level in
ambient air by shifting the Y-axis to the X-intercept value of 232
ng/day. The equation describing the curve is modified as follows:
15
-------
Table 4. LUNG CANCER MORTALITY VERSUS CARCINOGEN INTAKE
FOR DORN STUDY GROUP, AGE 45-84
Smoker carcinogen Lung Cancer
intake, ng/day (L ) Mortality/100,000 (M, )
Cigarettes & J resp ' Ic
NS (232)* 19.1
5 540 92.7
15 1,620 163
30 3,240 273
50 5,400 384
* Estimated by method of standard additions
16
-------
400 r-
-1.0
Regression Equation
LRESP= 3.553e-0024Mic-3.785
(r2 = .9968)
x = Dorn Study Data Points
1.0
ZO
3.0
4.0
5.0
6.0
I LRESP SMOKER CARCINOGEN INTAKE, jug/day
x-Intercept = 0.232
Figure 3. Lung Cancer Mortality Versus Carcinogen
Intake of Dorn Study Smokers
17
-------
resp
.0024M1
-0.232 = 3.:
.0024M
L (ng/day) = 3,553(e C -1)
resp
.0024M
Xaa (ng/m3) = 350.5 (e C - 1)
M = 416.7 In
It was shown in Section V, part G of Ref. 1 that the MI(, of the
general population was 87% of the Dorn study group. Correcting for this
fact results in the following expressions applicable to the general
population:
.0024M
L (ng/day) = 3,088(e -1)
.0024M
X (ng/ni ) = 304.6(e -1)
3.3.
x
r5_3.
Mlc=362.51n (35^5 + !)
Table 5 shows estimated lung cancer mortality for the nonsmoker in
the general population as a function of the carcinogen concentration in
ambient air. The same data is plotted in Figure 4.
Drinking Water
Carcinogens that are present in drinking water (and food) also
represent a biological risk to body tissues. Oral ingestion and subsequent
transport of these agents increase systemic carcinogen levels. The
purpose of this treatment is to establish a correspondence between this
body burden and cancer mortality with the view of assessing the risk
associated with carcinogens in drinking water.
Carcinogens which are introduced via the respiratory tract come
into direct contact with lung tissue and are then potentially available
for transport to other tissue sites. It has been shown that approximately
50% of a respiratory carcinogen dose is retained in lung tissue. The
18
-------
Table 5. ESTIMATED NONSMOKER LUNG CANCER DEATH RATE
VERSUS CARCINOGEN CONCENTRATION IN AMBIENT AIR
Age 45-84
Carcinogen cone.
Daily carcinogen
intake (general
Lung
Cancer mortality
in ambient air (X , ng/m )
5.3.
100
50
22.9
10
5
2
ocoulation) CT. , na/dav)
• • reso
1,010
507
124
101
51
20
100,300
91
43
22
10
5
?
(Mlc)
.0
.3
_Q
.2
.1
.1
* Ambient air baseline level for general population based on Dorn Study
data this report
X
"ic
362.5 In (•
350.5
+ 1)
19
-------
175
Ambient Air (Dorn Study)
20 30 40 50 100 150
Xaa CARCINOGEN CONCENTRATION IN AMBIENT AIR, ng/m3
250
Figure 4. Lung Cancer Mortality Versus Carcinogens
In Ambient Air - General Population
20
-------
remaining portion passes into the systemic circulation and exerts its
biological action at other body sites. This 1:1 partition is assumed
during this treatment.
Orally ingested carcinogens pass through the gastrointestinal tract
and are then presented to other body sites via absorption into the
systemic circulation. Complete carcinogen uptake (absorption) is assumed.
Several carcinogen distribution studies have shown that amounts ranging
from 0.38 to 2-3% of an orally administered dose of polycylic hydrocarbons
are found in lung tissue. ' Kotin and coworkers reported a maximum of
2% of the dose in the respiratory tract. The fraction of orally admini-
stered carcinogens concentrated by lung tissue is assumed negligible and
is not considered further in this report.
The Dorn study report itemizes mortality figures for cancers at all
sites. These risk data were compiled for all cancers, other than lung,
and categorized by age and smoking history (see Table 6). The same
imformation is shown in Table 7 for the combined 45-84 age group.
From this summary it is readily apparent that sites other than lung
are responsible for the majority of cancer deaths, particularly among
nonsmokers, and that lung cancer risk increases much more rapidly than
cancers at other sites with increased frequency of smoking.
As noted above one-half of the carcinogens introduced through the
respiratory tract are released into the systemic circulation. Thus,
sites other than lung are exposed to 50% of the carcinogens contained in
the cigarettes smoked each day. The relationship between cigarettes
smoked/ day, the corresponding systemic carcinogen intake (S ) , and
rG s L)
the resultant other site cancer mortality (M ) is given in Table 8.
o s
If one assumes that the carcinogen level in ambient air, drinking
water, and food was constant for all Dorn study subjects, the cigarette
smoker's increased other site risk must be due to his higher systemic
burden (S ) as shown in Table 8. A standard additions treatment of
resp
systemic carcinogen uptake (Sreg ) vs. other site mortality (Mog)
values was carried out. A plot of these data is shown in Figure 5. The
regression equation which best describes the relationship between M
and S is given below.
ircsp
21
-------
Table 6. CANCER (OTHER THAN LUNG)
Age
45-54
55-64
65-74
75-84
MORTALITY AS
AND
Cigarettes
A FUNCTION OF SMOKING
AGE (from Dorn Study)
Cancer (other
smoked/day than luris) deaths
NS (nonsraokers)
5 (1-9)
15 (10-20)
30 (21-39)
50 (> 39)
NS
5
15
30
50
NS
5
15
30
50
NS
5
15
30
50
7
1
8
16
0
463
112
452
370
79
589
183
518
250
59
85
19
42
22
1
PATTERNS
Person-years
of observation
15,134
3,129
16,392
12,339
1,928
213,858
45,217
151,664
103,020
19,649
171,211
37,130
101,731
50,045
8,937
8,489
1,923
3,867
1,273
232
M
OS
Cancer(other
than lung) death
rate/100~OCO
46.3
32.0
48.8
125
-
217
243
298
359
402
344
493
509
500
600
1,001
983
1,086
1,723
431
22
-------
Table 7
Respiratory
Cigarettes carcinogen intake
smoked/day tig/yr ng/day
NS Ambient Air
5 197 540
15 591 1,620
30 1,182 3,240
50 1,970 5,400
CANCER MORTALITY AND RESPIRATORY CARCINOGEN INTAKE
FOR DORN STUDY GROUP AGE 45-84
Cancer (other than
than lung) deaths
1,144
315
1,020
709
139
Person-years
of observation
408,692
87,399
273,554
167,177
30,746
M
OS
Cancer (other
than lung) death
rate/ 100, 000
280
360
373
424
452
M
Mlc
Lung Cancer
death rate/100,000
19.1
92.7
163
273
384
-------
Table 8. RELATIONSHIP BETWEEN CANCER
(OTHER THAN LUNG) MORTALITY AND
SYSTEMIC (ORAL) CARCINOGEN BURDEN
Cigarettes
smoked/ day
NS
5
15
30
50
*-ty o uciii-LL. (_ca.j- (_ j_iiugcii
uptake, no/day (S )
resp
Ambient Air
270
810
1620
2700
Cancer (other than lung)
mortality/100,OOP
resp
280
360
373
424
452
= 50% of respiratory carcinogen intake
24
-------
o
o
cc
o
cc
LU
O
o
LU
K
CO
cc
111
I
o
450 ,-
400 _
350 _
300
250,
200 _
Regression Equation
SRESP= 0.0003e-020Mos-0.093
(r2 = 0.9692)
x = Dorn Study Data Points
I
•0.5
0.5 1.0 1.5 2.0 2.5
SMOKER SYSTEMIC CARCINOGEN INTAKE,
3.0
RESP
x-lntercept = 0.0927
Figure 5. Other Site Cancer Mortality Versus Carcinogen
Intake of Dorn Study Smokers
25
-------
.020M 9
S = 0.34 °S - 93.0 (r = .9692)
resp
When M = 0, S = -92.66 (X-intercept)
os resp
Thus the Dorn study nonsmoker is subjected to a daily systemic
carcinogen dose of 92.7 ng. This value includes carcinogen burden from
all sources (air, food, and drinking water).
Adjusting for this background level gives an expression for total
systemic burden, a term that includes carcinogen contributions from all
sources.
S , = S + S -
total resp oral
S - = 0.5 L + S, + S,, ,
total resp dw food
.02M
S ,-92.66 = 0.34e °S - 93.0
total
.02M
Stotal -
Mos - 50 1
As noted in the caveats stated at the beginning of this report a
daily consumption of two liters of drinking water and one kilogram of
food is assumed. The assumption is also made that drinking water and
food contain equal concentrations (wt %) of carcinogens. Thus, two
thirds of the carcinogens ingested orally are present in drinking water.
Xdw (ng/1) = XfoQd (ng/kg)
Sdw (ng/dav) = 2(l)Xdw (ng/1)
Sfood (ng/day) - l (kg) Xfood
S; = 2Sf A
dw food
Soral= 3Sfood = 1>5Sdw
26
-------
The Dorn study group consisted of adult white males. As a result,
data derived from this study group cannot be directly applied to any
geographical segment of the general population. The technique used to
convert the derived equations to the form necessary for general appli-
cation is given below. An equal number of males and females is assumed
Q
in addition to the following recent mortality figures.
1975 Cancer Death Rates
200 per 100,000 nonwhite male
157 per 100,000 white male
119 per 100,000 nonwhite females
107 per 100,000 white females
Assuming nonwhites make up 15% of the population, the annual
cancer mortality for the total population is 136/100,000 or 86.6% of the
white male (Dorn study subjects) death rate.
200(.075) = 15.000
157(.425) = 66.725
119(.075) = 8.925
107(.425) = 45.475
Overall Cancer Death Rate 136.125
Dorn study nonsmoker M =280 (from Table 8)
OS
Adult population nonsmoker M =243
r r os
The equation that expresses other site cancer mortality as a
function of total systemic carcinogen burden and is applicable to the
adult general population is shown below.
g
M = 50.0(.866) In (
os .
S
M = 43.3 In ( *°l, + 1)
os 0.34
The systemic burdens for various ambient air and drinking water
carcinogen concentrations are given in Table 9 and the other site cancer
mortality values for different ambient air and drinking water combinations
are listed in Table 10. The data from these tables is plotted in Figure 6
27
-------
OOO'OOl/AinVldOW U3DNVD 31IS U3H1O
Figure 6. Other Site Cancer Mortality Versus Carcinogens
in Drinking Water at Different Ambient Air
Concentrations - General Population
28
-------
Table 9. ORAL AND RESPIRATORY UPTAKE OF CARCINOGENS AS A FUNCTION OF AMBIENT AIR AND DRINKING
WATER/FOOD QUALITY
Adult General Population
a) Carcinogens in Drinking Water/Food
Concentration In drinking water, X. (ng/1)* 0 1 5 10 25 50 100
uw
Concentration In food, X. . (ng/kg)
Total oral (systemic) carcinogen uptake, S (ng/day)**
° 0 3 15 30 75 150 300
Other site cancer 'mortality, M 0 99 165 195 234 264 294
OS
to * Equal wt.% concentration of carcinogens In drinking water and food.
^O
** Two liters of drinking water and one kilogram of food consumed per day.
b) Carcinogens in Ambient Air
Concentration in ambient air, Xaa(ng/m ) 0 2 5 10 25 50 100 500
Respiratory carcinogen uptake, L (ng/day) 0 20.2 50.7 101 253 507 1014 5070
Release to systemic circulation, S ^(ng/day) 0 10.1 25.3 50.7 127 253 507 2535
Oilier site cancer mortality, M , 0 149 187 217 257 287 316 386
• 10.14X
aa
** 50% Lung retention
S
M =43.3 In (~S£ + D
os 0.34
-------
Table 10, OTHER SITE CANCER MORTALITY AS A FUNCTION OF A AMBIENT AIR AND DRINKING WATER/FOOD QUALITY
Adult General Population
Other Site Cancer Mortality, M
Ambient Air Cone.,
0
2
5
10
25
50
100
500
oral
M = 43
OS
w - /. Q
, Systemic Uptake from
X (ng/m ) Ambient Air, S (ng/day)
33 ITGSp
0
10.1
25.3
50.7
127
253
507
2535
3X,
dw
S . + S
-, ,„ / oral resP , I-,
. j in (. . -t- L)
i ^„ r c.otal a. 11
Xdw(ng/l) = 0
0
149
187
217
257
287
316
386
1
99
160
192
219
258
287
317
386
5
165
187
207
228
261
289
318
386
10
195
207
219
237
266
291
320
387
25
234
240
246
257
277
298
323
387
50
204
267
271
277
290
307
327
388
100
294
295
297
300
309
321
337
391
-------
as risk of other site cancers per 100,000 versus drinking water carci-
nogen concentration (ng/1) at various ambient air carcinogen levels.
In 1971 the World Health Organization established a maximum permis-
sible concentration for six polycyclic aromatic hydrocarbons, taken
a
collectively, at 200 ng/1. Four of the six compounds used in this
estimation are known carcinogens.
Harrison and co-workers have prepared a comprehensive review on the
concentrations of carcinogenic polycylic aromatic compounds found in
different types of water streams. Ground water levels ranged from 1
to 81 ng/1 (average 60 ng/1). Several studies have been carried out on
the effectiveness of polycylic carcinogen removal by conventional sewage
treatment processes. ' Treated effluents contained from 7 to 54 ng/1
(in most cases less than 30 ng). Untreated river water may contain ten
times this amount and if polluted by nearby petroleum-related activities,
will probably exceed that concentration.
For the sake of comparison, the permissible concentration of carcino-
genic polycyclic hydrocarbons was determined using Method IIIC (see
Section V, page 62 of reference 1). The TLV of the benzene soluble coal
3 12
tar pitch volatiles is 0.2 mg/m . Since approximately 10% of this
material consists of polycyclic hydrocarbons (assumed carcinogenic), the
3 13
TLV for this fraction may be estimated at 0.02 mg/m .
METHOD III C (T=365 days)
X = 1.14 x 10~3 TLV
e
X = 1.14 x 10~3 (.02)
e
X = 0.0228 ug/1 (22.8 ng/1)
e
X = Estimated safe drinking water concentration
A summary of these different water carcinogen levels (X, ) and the
calculated other site mortality values (M ) is shown in Table 11.
o s
31
-------
Table 11. SUMMAKY OF CARCINOGEN CONCENTRATION IN DRINKING WATER AND CORRESPONSINC CAHCKR RISK
Carcinogen Concentration in Othe-r Site Cancer Mortality
„ . , ., , , Drinking Water, X, (ng/1) per 100,000 M
Basis for Estimation _ dw os
X = 2 10 50 ng/
aa
1. General adult population based on Dorn Study
nonsmoker cancer mortality 27.2 2/i3
13.6 - 243
2. Method III C - TLV for coal tar pitch vola-
tiles (0.2 mg/m ) and assuming 10% carcino-
gen polycylic hydrocarbon content 22.8 ^'!(> 254 297
3. WHO - maximum permissible concentration;
based on six polycylics, four of which are
known carcinogens 200 Tl'i 327 339
228 330 332 343
4. Method III C - 11V for coal tar pitch vola-
tiles (0.2 mg/m ), assume 100% carcinogen
content
5. Typical effluents as reported in
Reference 7
a) Average groundwater levels 60 274 282 310
b) typical maximum treated river water
levels 30 246 261 300
c) minimum treated river water levels 7 196 232 290
0.5 L + S
M = 43.3 In ( -*£%—"" + «
os U. J4
5.07X + 3X,
M = 43.3 In ( -jiV—* + D
os 0.34
M =43.3 In (-4-l'r1 + 1)
os 0.34
-------
References
1. "Estimation of Permissible Concentrations of Pollutants for
Continuous Exposure", R. Handy and A. Schindler, EPA 600/2-
76-155 (1976).
2. Wynder and Hoffman, Tobacco and Tobacco Smoke, Academic Press,
New York, N. Y., 1967-
3. H. A. Kahn, The Dorn Study of Smoking and Mortality Among U.S.
Veterans: Report on Eight and One-Half Years of Observation,
National Cancer Institute Monograph No. 19, 1966.
4. E. C. Hammond, Smoking in Relation to the Death Rate of One
Million Men and Women, National Cancer Institute Monograph No. 19,
1966.
5. P. Kotin, H. L. Falk and R. Bussea, J. National Cancer Institute,
23, 541 (1959).
6. C. Heildelberger and H. B. Hardin, Cancer, 252 (1948).
7- P- M. Daniel, 0. E. Pratt and M. M. L. Prickland, Nature, 215,
142, (1967).
8. Testimony by Mrs. Dorothy P- Rice, director of the National Center
for Health Statistics to the House Intergovernmental Relations and
Human Resources on June 14, 1977 as reported in the June 15 issue
of the News and Observer, Raleigh, North Carolina.
9. World Health Organization, "International Standards for Drinking
Water", 3— ed., Geneva, Switzerland, 1971.
10. R. M. Harrison, R. Perry and R. A. Wellings, Water Research, 9^,
331 (1975).
11. R. M. Harrison, R. Perry and R. A. Wellings, Environ. Science and
Technol., 10, 1151 (1976).
12. "Documentation of Threshhold Limit Values for Substances in
Working Room Air", American Society of Governmental Industrial
Hygenists, 3rd ed., 1971, 2— printing 1974.
13. E. Sawicki, et. al., Am. Ind. Hyg. Assn. J., 23, 482 (1962).
33
-------
TECHNICAL REPORT DATA
(Please read Inuructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-78-107
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Estimation of Risk from Carcinogens in Drinking
Water
5. REPORT DATE
May 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Robert W. Handy
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, North Carolina 27709
10. PROGRAM ELEMENT NO.
IBB 610
11. CONTRACT/GRANT NO.
68-02-2612, Task 16
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 5-12/77
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTESlERL.RTp
541-2547.
officer fe Max Samfleld, Mail Drop 62, 919/-
16. ABSTRACT
The report gives results of a study aimed at developing a means for estimating
cancer mortality as a function of carcinogen concentration in drinking water. Cancer
risk data for cigarette smokers was treated by the method of standard additions to
provide an estimate of ambient carcinogen levels in drinking water. A similar treat-
ment was farr/ed out on lung cancer risk data to give an estimate of carcinogen
levels in otmbient air.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
Pollution
Potable Water
Carcinogens
Malignant Neoplasms
Pulmonary Neoplasms
Mortality
Estimating
Tobacco
Pollution Control
13B
08H
06E
05;:
14B
06C
B. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
39
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
34
------- |