EPA/600/R-95/100
April 1998
Nitrobenzene Carcinogenicity
(CAS No. 98-95-3)
National Center for Environmental Assessment-Washington Office
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
-------
DISCLAIMER
This document has been reviewed in accordance with the U.S. Environmental Protection
Agency policy and approved for publication. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
11
-------
CONTENTS
LIST OFTABLES iv
LIST OF FIGURES ... iv
ABSTRACT v
PREFACE j vi
AUTHORS AND REVIEWERS vii
1. INTRODUCTION 1
2. CARCINOGENICrrY OF NITROBENZENE 7
3. CANCER BIOASSAY RESULTS 9
4. DISCUSSION 4 14
5. REFERENCES 20
6. APPENDIX 24
111
-------
LIST OF TABLES
1. Vapor pressure of comparable compounds (mm Hg) 2
2. Tumor incidence in B6C3F1 mice ; 9
3. Tumor incidence in F344/N rats 11
4. Tumor incidence in Sprague-Dawley (CD) rats . 13
5. Summary of nitrobenzene carcinogenicity results 17
LIST OF FIGURES
1. The chemical structure of the potentially hazardous air pollutant nitrobenzene 1
2. The ring hydroxylated metabolites of nitrobenzene 4
3. The cecal nitroreductase reduction of nitrobenzene by intestinal flora .5
IV
-------
ABSTRACT
Nitrobenzene (NB, CAS No. 98-95-3) oxidizes to p-aniinophenol and p-nitrophenol
\
in animals and humans, while being reduced also to nitrosobenzene, phenylhydroxylamine, and
aniline. The reductants are known to cause methemoglobinemia and anemia. NB is negative for
mutagenicity systems in Salmonella, hepatocyte repair, or sister chromatid exchange assays.
No human studies cancer data are available, therefore this analysis is based on.a
Chemical Industry Institute of Toxicology (CIST) NB inhalation study. The study involves both
sexes of B6C3F1 mice (0,5,25,50 ppm NB) and F344/Nrats (0,1,5,25 ppm NB), but only
male Sprague-Dawley (CD strain) rats (0,1,5,25 ppm NB). All are exposed at 6 hrs/day, 5
days/wk,.for 104 wks. Mouse results show increased benign male tumors in the alveolus,
bronchus, and thyroid, whereas female mice indicated mammary cancers. F344/N rats responded
with male liver cancer and benign responses in the follicular thyroid and kidney, whereas females
had benign endometrial polyps. Male CD rats had benign hepatocellular adenomas. Because NB
causes tumorigenicity at 8 sites, 6 organs, both sexes, among 3 test species, NB is classified as a
B2 carcinogen or a likely human carcinogen by any route by the proposed EPA Cancer
Guidelines (U.S.EPA, 1986)
-------
PREFACE
This hazard assessment of nitrobenzene (NB) carcinogenicity in rodents (rats and
mice) was prepared by the National Center for Environmental Assessment, Washington office.
NB (C6H3NO2) is a nitroarene that is a potentially hazardous air pollutant (HAP) in the United
States. It is listed as HAP in the 1990 Clean Air Act Amendment, Section 112b. This NB
support document for carcinogenicity originally was developed to assist the U.S. EPA Office of
Air.
NB has been presented in summary report form to the Carcinogen Risk Assessment
.. Verification Endeavor (CRAVE) group (12/7/94). The presentation (oral and written) was
reviewed for inclusion into the Integrated Risk Information System (IRIS) database; comments
were taken from all members. This support document was corrected and a final IRIS summary
submission was sent to CRAVE on April 24,1995, for submission to the IRIS database. The
quantitative cancer unit cancer slope has been re-estimated (September 1995); the quantitative
section is presented in the Appendix. The NB support document was reviewed again
administratively by NCEA-W in February 1997, further revised, and finalized in April 1998.
VI
-------
AUTHORS AND REVIEWERS
PRIMARY AUTHOR
James W.-Holder, Ph!D.
Tojycologist
Effects Identification and Characterization Group
National Center for Environmental Assessment
Washington Office
CONTRIBUTING AUTHOR
Jennifer Jinot
Environmental Health Scientist
Quantitative Risk Methods Group
National Center for Environmental Assessment
Washington Office
REVIEWERS
Cheryl Siegel-Scott
National Center for Environmental Assessment
Dharm Singh
National Center for Environmental Assessment
Susan Velazquez
National Center for Environmental Assessment
Vll
-------
1. INTRODUCTION
The purpose of this report is to characterize the carcinogenic hazard potential of
nitrobenzene (NB, CAS No. 98-95-3). NB is a potentially hazardous air pollutant (HAP) listed
in the 1990 Clean Air Act Amendment1 Alternate NB chemical names are nitrobenzol and oil of
mirbane. .
The chemical formula of NB is C6H3N02 and the molecular weight is 123.11. The
structural formula for NB is presented in figure 1. ,NB is a lipophilic bipolar compound that can
be hydrolyzed various nitrophenols and phenol, or it can be reduced to aniline.
X
N
0
Figure 1. The chemical structure of the potentially hazardous air pollutant nitrobenzene.
Nitrobenzene is chemically produced by treating benzene with HzSCh and HMOs.
• NB is a colorless to pale yellow oily liquid that has the odor of almond oil. The flash
point of NB = 89 °C, melting point = 6"C, boiling point = 2lO°C, density = 1.2037, and vapor
pressure = 1.22 mm Hg (at 25 °C). NB volatility is presented in table 1 (number 6) and can be
compared with the volatility of other reference compounds at 20°C. Some of the reference
compounds in .table 1 are themselves potential HAPs. The data in table 1 indicate that NB is .
volatile enough to be a potential inhalation toxicant wherever it is stored, handled, processed, or
disposed.
NB is chemically stable and dissolves in aqueous solution at 0.2% (v/v). It can be
manufactured by a batch process or by a continuous process in big plants. Nitronium ion (+NOj)
is generated from a sulfuric and nitric acid mixture. A nitror.ium ion attack on benzene forms
nitrobenzene (*NO2 + C6H6 - CJHjNOj) among other nitrated benzenes such as the dinitro and
trinitro products. The mononitrated product (mp = 6°C) is separated from the mixture of sulfuric
and nitric acids and other nitrated benzenes (mp = 173°C) by cold crystallization, decanting, and
distillation at 210°C.
The U.S. production volume of NB ranks in the top 75 chemicals at 1.4 billion
pounds/year. At the present-time, the bulk of NB production is used in making chemical
intermediates, the most of which (98%) is used in aniline production. Aniline is manufactured
-------
Table 1. Vapor pressure of comparable compounds (mm Hg)"
Number
1 .
2*
3*
4
5*1
6*
7*
8
9
10
11*
12*
13*
14
15
16*
17*
18*
19
20
Compound
Diphenylamine
Phthalic anhydride
Quinoline
C sor
Naphthalene
Nitrobenzene
Ethylene dibromide .
Chlorobenzene .
Acetic acid
Water
Dichloroethane
Benzene . '
Carbon tetrachloride
Hexane
Ethyl ether
Bromoethane
Acetaldehyde
Chloroetharie .
Butane
Propane
Vapor pressure
0.011
0.025
0.181
0.260
0.53
1.22
10.1
12.5
•15.7
17.3
60.6
74.6
76.4
120.0
290.8
475.0
764.3
1002.3
1645.3
6677.8
"Approximate vapor pressures «re presented and are calculated ftom data presented in The Handbook of Chemistry and Physics,
67th edition, CRC Press, 1986-1987, pp. D-212ff. Hie estimating formula for vapor pressure at 20*C is: tog p » -1.78 x 10"1 a
-i- b, where a and b are presented in the handbook. Compounds with asterisks are potential HAPs as presented by U.S_. EPA't
Office of Air. The above table compares HAP nitrobenzene volatility with the volatility of other representative compounds.
-------
from NB in the gaseous state by reduction with H2 and FeCl, or by reduction in solution with
Fe+3 and HC1. In addition to being used as a chemical feeder stock, NB also is used to make
pesticides, shoe polishes, analgesics, dyes, and pyroxylin compounds.
The oral acute LDJO of NB in the rat is * 1 g/kg, which puts NB in the slightly toxic
range of acute oral toxicity. Oral or inhalation exposures can cause red blood cell (RBC) damage
as well as spleen swelling and engorgement and anemia. The toxicity of NB to humans can lead
to death, especially in infants and children. Inhalation exposure of NB is considered more likely
than oral exposure in the United States. Short inhalation exposures cause hepatic, splenic, and
testicular lesions in B6C3F1 male mice. Testicular injury as a target tissue for NB involves
necrosis of primary and secondary spermatocytes and development of giant multinucleated cells.
The Registry for Toxic Effects of Chemical Substances (RTECS) lists rat testicular injury at a
TCU = 5 ppm of inhaled NB. RTECS lists rat fertility effects as occurring at 252 ppm of
exposure. The Occupational Safety and Health Administration has recommended a human
threshold limit of 1 ppm (5 mg/m3).- Nitrobenzene is a skin and eye irritant and is considered a
dermal toxicant . •
The current U.S. EPA reference dose (RfD) is 4.6 mg/kg/day based on a B6C3F1
mouse inhalation study (IRIS database, June 1997). This RfD is based on a 1984 subchronic
inhalation study that included hematologic, adrenal, renal, and hepatic lesions (Cnr, 1993).
After acute high NB exposure, liver injury includes bile stasis, fatty degeneration, centrilobular
necrosis, and hepatocellular nucleolar enlargement Excessive doses of inhaled NB can lead to
central nervous system suppression, headache, nausea, methemoglobinemia, and liver, brain
(e.g., malacia of the cerebellar peduncle), and testes injuries (ATSDR, 1990).
NB is readily absorbed by human or animal skin, oral, or inhalation exposures. At
high inhaled doses, NB is metabolized in a few days to p-nitrophenol and p-aminophenol in
human urine (Bceda and Kita, 1964; Dorigan and Hushom, 1976; Rickert, 1984), whereas at
lower doses only p-nitrophenol is found in volunteer urines (Salmowa et al., 1963). A typical
. animal metabolism study (gavage, rabbit) at high NB dose suggests an array of NB metabolites, in
urine (figure 2) (Robinson et al., 1951; Parke, 1956). The generality of this NB metabolic
response was demonstrated in other animals (rat, guinea pig) and by other routes of exposure
(intravenous, dermal) (Piotrowski et al., 1975; Kiese, 1974; Eyerand Ascherl, 1987; Beauchamp •
etal., 1982). .
Various NB resonance structures are possible due to the nitro group electron
withdrawing properties (figure 2). Predictable P-450 mediated rmg-hydroxylations occur at the
meta position, and to a lesser extent at the o- andp- positions. Epoxidation followed by an
isosceles hydrolytic-cleavage of the three-membered ring explains the presence of both o~ and p-
-------
ringepoxidation
with hydrolysis
p-nitrophcnoI(9S) o-nitrophcnol(0.1%) m-nltrophenol(9S) p-nitrophenylmercapturic acid
p-«mlnophenol(31S) o-«mlnophenol(3%) . m-amlnophenol(4%) nltroqulnol(0.1%>
Figure 2. The ring hydroxyfated metabolites of nitrobenzene. These metabolites
account for most of the nitrobenzene converted to eliminated products in the
excreta.
isomers (Tomaszewski et-al., 1975). NB is catabolized by these microsomal oxygenations to
form phenolic residues that usually conjugate with sulfates or glucuronides to reduce their
toxicky and aid excretion (Beauchamp et al.. 1982). These metabolic conversions generally are
expected to be shared by man and animals (Dorigan and Hushorn, 1976; Beauchamp et al.,
1982).
Considering its small molecular weight, volatility, and limited water solubility
(0.2%), inhaled NB likely has free access to most compartments of the corpus (Mabey et al.,
1982). NB is expected to deposit transiently in fat depots because of its lipophilic nature.
However, NB would not be expected to be a lingering corporal contaminant due to its moderately
low log (octanol/water) coefficient (K^J of 1.87 (Mabey et al., 1982). This K^ is similar to the
for chloroform, thus demonstrating about the same Irpophilicity. • •
The flora of the cecum of the large intestine can metabolically reduce NB to aniline
keda and Kita, 1964; Rickert et al., 1983). It has been observed that aniline-caused
methemoglobinemia does not occur in NB-exposed germ-free (axenic) animals but is present in
normal animals. The normal animals have gut flora to convert NB to aniline by cecum
reductases (Reddy et al., 1976). Presumably then, aniline is an essential part of the NB metabolic
and toxicity profile. Aniline is already classified as a Category B2 carcinogen, a likely human
carcinogen. This is based on the carcinogenicity in the spleen and body cavity in two strains of
rat (U.S. EPA, 1995).
-------
A. Nitrobenzene bacterial reduction mechanism (cecum)
flavins
NADPH
NO
nitrosobenzene (NOB)
-NH2
Aniline (AN)
•N—OH
phenyihydroxylamine/
Bacteria] nitrorcductasc concerted reduction
B. Nitrobenzene microsomai reduction, mechanism
•
lupcnndde free radical Of*
futile reaction (reforms nttrobenxent again)
Nttroreductase
Nitrobenzene (MB)
*
NETREDUCnON
*
\
P-450, Flavin
NADH
1H+
nitrosobenzene
.(NOB)
•N—O-
phenyihydraxylainine hydronhnndde ttet radical
(PH)
Figure 3. A. The cecal nitroreductase reduction of nitrobenzene by Intestinal flora.
B. The systemic reduction of nitrobenzene by P-450 enzymes to aniline. Note the
production of the free radical intermediates and nitrosobenzene and
phenyhydroxlamine.
-------
The steps by which NB is reduced to aniline go through the intermediates
nitrosobenzene and phenylhydroxylamine (Goldstein and Rickert, 1984) (see figure 3). The
nitroanion free radical is formed by a le transfer and is a relatively stable free radical. If oxygen
is sufficiently present in the tissue, then a reaction takes place that produces the superoxide free
radical, that is, the nitroanion free radical reacts with 02 reproducing NB, a futile reaction, plus
creating the superoxide free radical Of (figure 3B) (Mason and Holtzman, 1975b; Sealy et al.,
1978; Levin et al., 1982)'. The futile reaction can account for a number of the toxic and . .
carcinogenic actions of NB based on the known carcinogenic properties of O2i (Kensler et al.,
1989a, b; Flohe" et al., 1985; Guyton and Kensler, 1993; Feig et al., 1994; Cerutti, 1994; Dreher
and Junod, 1996). In excess of antioxidant capacity, this radical can cause cancer. If the NB
exposure escapes the intestinal reduction—as inhalation exposure does—then the P*450 system'
and possibly the mitochondria! systems reduce NB to aniline (Mason and Holtzman, 1975a).
Nitrosobenzene and phenylhydroxylamine are produced (figure 3) as well as free radical
intermediates' of which the nitroanion free radical is the most stable.
These intermediates are known to bind to hemoglobin and are the direct mechanism
of observed methemoglobinemia activity from NB exposure (Goldstein and Rickert, 1985).
Because of their chemical reactivity, they likely bind to many other cellular components such as
DNA. Almost all nitrosoamine and hydroxylamine compounds are thought to be potential
carcinogens (Miller, 1970; Weisburger and Weisburger, 1973). However, a literature search did
not find any carcinogenicity studies specifically on nitrosobenzene or phenylhydroxylamine. The
amount of NB exposure determines if these potentially pernicious reduction intermediates exceed
the metabolic capacity to remove them. If not quenched, the free radicals generated from NB
exposure can cause tissue damage, which is also a dose-dependent process (Kehef, 1993). That
is, the free radicals could exceed the tissue's ability to quench them by antioxidants such as
vitamins C and E (Netke et al., 1997; Kimmick et al., 1997; Primiano et al., 1997).
-------
2. CARCINOGENICITY OF NITROBENZENE
NB. mutagenicity was reviewed and was found to be negative in Salmonella test
systems with reductase activation (Chiu et al., 1978; Suzuki et al.f 1987) and in V79 cells
(Kuroda, 1986). Negative results also were observed in primary hepatocyte repair assays and in
vivo sister chromatid exchange assays. Thus, with negative mutagenicity data, no current cancer
bioassay data, and no human carcinogenicity information, NB previously has been classified as a
Category D carcinogen (IRIS database 1993). The purpose of this current review is to determine
the carcinogenicity of NB based on the latest information available.
The Chemical Industry Institute of Toxicology (CUT) began a 2-year NB inhalation
carcinogenicity bioassay in 1983, and the results of NB bioassay data became available in 1993.
The results of this bioassay were reported in full in January 1993 by James A. Popp, study
director, and his colleagues (CUT, 1993). The report provides the basis of this assessment of the
carcinogenicity of NB. The inhalation study involved both sexes of B6C3F1 mice (0,5,25,50
ppm NB), F344/N rats (0,1,5,25 ppm), and Sprague-Dawley (CD strain) rats (6,1,5,25 ppm)
(1.0 ppm = 5.12 mg NB/m3 or 1 mg/m3 = 0.20 ppm). Exposures were 6 hours/day x 5 days/week
x 104 weeks. The results were later published (Cattley et al., 1994 #52). NB was introduced
into the dosing chamber as a vapor at >99% purity. The chamber concentration did not vary
significantly in target concentrations. NB was chemically stable during the test
At the study end, the tumor incidences in each treated group was pair-wise compared
with tumor incidence of the vehicle air-dosed control animals to determine.any change in
tumorigenicity. A number of organs were examined for tumorigenicity. The necropsied tissues
were nose (anatomical regions i to 4), adrenal glands, brain, ear canal, bronchial lymph'nodes,
clitoral or preputial gland, cecum, urinary bladder, esophagus, gallbladder, trachea, tissue masses
with regional lymph nodes and any gross lesions, heart, thymus, thyroid, ileum, jejunum, rectum,
kidneys, spleen, sternebrae, salivary glands, larynx, liver, lungs, bronchi, mammary gland,
mandibular lymph nodes, snout, pancreas, parathyroid glands, pituitary, prostate, testes,
epididymis or ovaries, and bone. Tumor discovery was either at autopsy after adventitial death or
death by other pathologic means or at planned mid- or terminal-necropsy.
In some cases there were no changes in body weights, and because the remainder
body weights decreased only 4% to 8% compared with controls, the mouse and rat body weights
were only minimally affected by NB during the 2-year dosing. The maximum tolerated dose
(MTD) was achieved in this study in both species based on the toxic effects observed: nasal
epithelial degeneration, induced methemoglobinemic anemia, and hepatic enlargement All these
toxicities were graded as significant effects. Compared with controls, the mouse and rat
survivals were unaffected by NB inhalation (CUT, 1993).
-------
Inflammatory degeneration occurred in the nasal and nasal olfactory passages in all
test animals due to chronic inspiration of NB. Decreased RBC counts and hematocrit and
hemoglobin levels were observed as a result of NB exposure in mice and rats. Hematopoietic
I .
dysfunction also was noted in both B6C3F1 mice and F344/N rats based on RBC interaction and
platelet parameter changes. NB caused methemoglobinemic anemia in all test species (both
sexes).
Increases in bone marrow cells also suggested hemotoxic effects, but no erythroid
stem cell carcinogenesis or leukemias were observed. Most of the hemotoxic effects caused by
exposure can be explained by the hemotoxic actions of aniline, a NB metabolite. Nitrosobenzene
and phenylhydroxylamine, both NB metabolites, also produce hemotoxic effects (Weisburger
and Weisburger, 1973 #13; Kiese, 1974 #31; Beauchamp et al., 1982 #9; Kiese, 1966 #11).
Enlarged and multinucleated hepatocytes were seen in male and female mice and in female rats.
Testicular atrophy and epididymal hypospermia were viewed in male B6C3F1 mice and CD rats
(Bondetal., 1981 #51; McLaren etal., 1993 #78; Levin etal., 1988 #71). It is known that
Fischer rats always have old-age testicular effects and are not adequate to determine testicular
cancer, so the CD-I rats were added to this study to appropriately assess the carcinogenicity in
testes.
-------
3. CANCER BIOASSAY RESULTS
Increased tumor incidences were observed at various organ sites and are presented in
tables 2 through 4. The statistical methods used are as follows, The adjusted incidences at risk
are corrected according to Kaplan-Meier (K-M) methods. The life table method corrects for
intercurrent mortality, and the incidence method corrects for tumor occurrence, assuming the
tumors were only incidentally found and were not the cause of death. The Cochran-Armitage
probability is the test for trend of tumor incidence versus dose. The Fisher exact test compares
the paired increase of tumor incidence at each dose with the incidence of the concurrent control
at that site. Probability values iO.05 and *0.01 are considered significant, and values <0.01 are
considered quite significant.
Table 2 presents data on tumor incidence in male and female B6C3F1 mice.
Significant tumor increases were observed in the alveolus and bronchus, thyroid (follicular cells),
and mammary gland. •
Table 3 presents data'on tumor incidence in male and female F344/N rats. Significant
tumor increases were observed in liver, thyroid (follicular cell), kidney, and endometrium.
Table 4 presents data on tumor incidence in male Sprague-Dawley (CD) rats. A
significant tumor increase was observed in the liver of the Sprague-Dawley (CD) rat. '
Table 2. Tumor incidence in B6C3F1 mice
Male B6C3F1 mouse lung (alveolar/bronchial) adenomas or carcinomas*
Dose
Adenomas, A/B
Carcinomas, A/B
Total incidence
Adjusted incidence
First occurrence (days)
Life table
Incidental tumor
Cochran-Armitaxg
Fisher exact
0 ppm
7/68 (10)
4/68(6) .
9/68 (13%)
20.5%
507
0.059
0.035
0.017
not applicable
5 ppm
12/67(18)
10/67 (15)
21/67 (31%)
42.6%
536
0.017
0.031
_
0.010
25 ppm
15/65 (23)
8/65(12)
21/65 (32%)
45.6%' .
711
0.015
0.017
— '
0.007
• 50 ppm
18/66 (27)
8/66 (12)
23/66 (35%)
46.0%
•704
0.011
0.010
_•
0.003
-------
Table 2. Tumor incidence in B6C3F1 mice (continued)
Male B6C3F1 mouse thyroid follicular cell adenomas
Dose
Overall incidence
Adjusted incidence
First occurrence (days)
Life table
Incidental tumor
Cochran-Armitage
Fisher exact
Oppm
0/65 (0%)
0%
__
0.022
0.019
0.015
not applicable
5 ppm
4/65 (6%)
9.1%
724 T
0.075
0.07J
^_
0.06*0
25ppm
1/65 (2%)
2.2%
724 T
0.519
0.519
_
0.500
50 ppm
7/64(11%)
14.7%
659
0.075
0.014
_
0.006*
Female B6C3F1 mouse mammary gland adenocarcinomas
Dosek
Overall incidence
Adjusted incidence
First occurrence (days)
Life table
Incidental tumor
Cochran-Armitage
Fisher exact
Oppm
0/48 (0%)
0% .
_
——
_»
__ •
not applicable
5ppm
not examined
_
__
^ .
_
H '
—
25ppm
not examined
__
_
_
M
_
—
50 ppm
5/60(8%)
11.9%
609
0.048
0.069
—
0.049
' Tumor incidence, K-M adjusted tumor incidence, and first occurrence of tumor at a particular site are presented above the
double line for each tumor site showing carcinogenicity. Below the double line, statistics are presented in italics; the;?-values
below the zero dose are for the trend test for each accounting method of tumor observation. The other p-values are Fisher exact
tests compared with the zero dose concurrent control.
b Only mammary gland control and high dose examined. Low and mid doses were not examined, and there was no rationale
presented why these groups were not analyzed for carcinogenicity. Because this is a two-dose examination, no trend can be
assessed by the Cochran-Armitage method.
Source: CUT, 1993.
10
-------
Table3. Tumor incidence in F344/N rats
Male F344/N rat hepatocellular adenomas/carcinomas"
Dose
Adenomas, ]iver
Carcinomas, liver
Total incidence
Adjusted incidence
First occurrence (days)
Life table
Incidental tumor
Cochran-Armitaxe
Fisher exact
Oppm
1/69(0
0/69 (0)
1/69 (0%)
2.3%
714 T
<0.001
-------
Table 3. Tumor incidence in F344/N rats (continued)
Male F344/N rat kidney tubular adenomas/adenocarcinomas
Dose
Adenomas, kidney
Carcinomas, kidney
Total incidence
Adjusted incidence
First occurrence (days)
Life table
Incidental tumor
Cochran-Armitage
Fisher exact
Oppm
0/69 (0)
0/69 (0)
0/69 (0%)
0%
__
<0.001
-------
Table 4. Tumor incidence in Sprague-Dawley (CD) rats
Male Sprague-Dawley (CD) rat hepatocellular adenomas or carcinomas*
Dose
Overall Incidence
Adjusted incidence
First occurrence (days)
Life table
Incidental tumor
Cochran-Armitaxe
Fisher exact
Oppm
2/63 (3%)
5.3%
592
0.002
0.003
0.002
• not applicable
Ippm
1/67(1%)
3.3%
696 • '
0.449 N
0.446N
__
0.447N
5ppm
4A70(6%)'
9.5%
4571
0.409
0.370
«
0.391
25ppm
9/65 (14%)
30.8%
609
0.036
0.046
_
0.031
' Tumor incidence, K-M adjusted tumor incidence, and first occurrence of tumor at a particular rite are presented above the
double line for each tumor site showing carcinogenicity. Below the double line, statistics are presented in italics; the p-values
below the zero dose are for the bend test for each accounting method of tumor observation. The others-values are Fisher exact
tests compared with the zero dose concurrent control.
Source: CHT. 1993.
13
-------
4. DISCUSSION
The CIIT study was started in November 1983 but was not finalized as.a full report until
January 1.993 (CIIT, 1993) because of a backlog in the CHT histopathology laboratory.
Subsequent to the drafting of this EPA report, the CUT study scientists and sponsors involved
with the bioassay published the results of their study.(Cattley et al., 1994). The hazard
evaluation analysis presented in the published paper closely agrees with the independent hazard
evaluation work reported here. ,
The CUT study was apparently well conducted, used good laboratory practices, and was
well reported. The report has adequate statistical analysis except that historical control averages
and variations were not presented for all significantly increased tumor sites. The MTD was
obtained in this study showing respiratory, testicular, and hemopoietic effects. .Even though the
rodent fur/skin was exposed in the chambers with NB for 2 years, no skin cancer resulted.
Hyperplasia of the alveolus and bronchus was observed in males but not in females. This
could be a predisposing factor in the lung epithelial tumor causation in male mice, and in fact,
increases in B6C3F1 mcuse alveolar and bronchial (A/B) tumors were observed in male but not
in female mice (table 2). The increased lung incidence was significant by all statistical measures:
life table analysis, incidental analysis, Cochran-Armitage trend, and Fisher pairwise comparisons
of treated incidences versus control incidence. The effect is driven by the adenomas, that is,
there is a trend with adenomas but not with carcinomas. This makes the effect tumorigenic
rather than carcinogenic. Because there is no indication that carcinomas (evidence of cancer)
were increased with NB in the pool of increased hyperplastic cells and adenomas in 2 years of
exposure, this A/B tumor evidence is benign and therefore limited.1 ,
B6C3F1 mouse thyroid follicular cell adenomas increased also (table 2). This effect too
was tumorigenic, not carcinogenic. The thyroid effect was statistically increased by all measures
of statistical analysis (listed above). This thyroid adenoma evidence is a benign response in 2
years of exposure^ Because each male B6C3F1 mouse response demonstrated limited evidence,-
the male mouse response is not sufficient by itself for a determination of carcinogenicity. The
male B6C3F1 mouse cancer responses, however, do factor into the overall determination of NB
carcinogenicity (see below).
In the female B6C3F1 mouse, increases were observed in malignant mammary
adenocarcinomas (table 2), with some having squamous metaplasia. Unfortunately, only the
' A "limited" response is not a sufficient response; it is something less. A sufficient response is a carcinojenlcity dose
response that is qualitatively applicable from animals to humans and statistically significant by trend and pair-wise comparisons
(FEXT). Limited responses can occur in various degrees of relevance and concern in carcinogenicity considerations. Expert
judgment is necessary to factor in the relative weights of each respective limited response.
14
-------
high-dose group and the control groups were examined histologically. This is a data gap. The
high-dose incidence of cancers was 5/60 (8%) (corrected incidence of 11.9%) versus hone in
controls. The first tumor of this kind was discovered late in the study at 609 days. The historical
control average (n = 1,791) is 1.76% with standard deviation = 2.6%. This makes the observed
response between the 20 and 3o limits (6.95% < 8.0% < 9.55%). The whole observed historical
control range for mammary adenocarcinomas incidence is 0% to 12%, which may include
outliers (Haseman et al., 1985). These references suggest that the incidence magnitude is
marginal and could be in the population of normally occurring tumors of the mammary gland.
The corrected incidence of 11.9% is almost out of the range of experience for control animals.
Moreover, the observed incidence of 8% is sufficiently greater than the historical control average
(1.8%) to suggest but not prove a positive effect. When no additional information is known, as
in this case, the concurrent control (0/48 = 0%) is taken as the valid reference. Therefore, this
effect in the mouse mammary gland is determined to be positive and marginally sufficient (see
footnote 1) to determine carcinogenicity.
In the male F344/N rat, a cancer response was observed in the liver (table 3). The cancer
response was mostly adenomas (trend p < 0.001), with a modest increase in hepatocellular
carcinomas (trend p = 0.040). This suggests that the adenomas may serve as a pool for
developing into carcinomas, making this a malignant effect The combination of these tumors
(benign and malignant) is increased by all statistical measures (vide ante). This carcinogenic
effect was discovered only at terminal sacrifice and therefore did not cause these animals to die
early because of the liver tumors. This liver cancer effect in male F344/N rats is sufficient to
determine carcinogenicity.
Significant increases of thyroid adenomas or adenocarcinomas were observed in male
F344/N rats. These results are presented in table 3. It is notable that the thyroid follicular cells
were affected in both the male F344/N rat and the male mouse (see above). Because there were
more carcinoma-bearing male rats than benign follicular cell adenoma-bearing'male rats, the rat
thyroid response is considered carcinogenic rather than tumorigenic. There was a'trend
(Cochran-Armitage) of adenomas and adenocarcinomas with increasing NB dose (p = 0.01), but
the other statistical measures were not significant. Therefore, the certainty of the cancer response
is similar for male F344/N rats and male B6C3F1 mice: both show limited evidence for
carcinogenicity. ' '
The male rat kidney showed increased tubular adenomas and adenocarcinomas (table 3).
It was a high-dose effect only. Of the 6 tumors observed in .70 male F344/N rats, 5 were
adenomas and 1 was an aderiocarcinoma. Thus, this is primarily a benign tumorigenic effect
The tumors were observed at day 714, which was the terminal sacrifice; the tumors were not life
threatening and were statistically increased by all statistical measures (vide ante). This kidney
15
-------
response shows some evidence of carcinogenicity.
The female F344/N rats had an endometrial polyp response to NB (table 3). This benign
response is statistically increased, at the high dose only, over a high' background concurrent
control response of 11/69 (16%). These tumors were first discovered at day 611 or 86% through
the study. This benign endometrial response provides limited evidence for carcinogenicity.
A second rat species, the Sprague-Dawley (CD) rat, was tested for NB carcinogenicity
because the F344 rat shows high testicular tumors normally in old age. Therefore, Sprague-
Dawley (CD) rats were added to cover this tissue. No tumor increases were observed in testes,
but as shown in table 4, there were increases in hepatocellular adenomas or adenocarcinomas.
Liver is the only site in Sprague-Dawley (CD) rats for which tumor incidence increased.
Although the response was flat for liver carcinomas (2/63,0/67,2/70,2/65), the adenomas
showed an increased trend (1/63,1/67,2/70,7/65) that was determined to be significant (trend p
= 0.003). The combination of adenomas and carcinomas is quite significant by all statistical
measures (table 4). Because this is a mostly benign response, it is determined that Sprague-
Dawley (CD) rats have some evidence for carcinogenicity.
In summary, there were eight tumorigenic or carcinogenic rodent organ sites responding
to NB exposure in the CHT inhalation study. Significant oncogenic increases in organ sites are
summarized in table 5.
NB inhalation exposure produces an array of tumor types, some of which could be
influenced by the toxicity caused by chronic NB exposure for 2 years. Of the eight sites covered
in tables 2 through 4, significant carcinogenicity evidence exists for one site: liver .
carcinogenicity in the male F344/N rat Also, limited rat responses occur in the male rat kidney
and thyroid and the female F344 rat endometrium. These rat responses collectively suggest a •
sufficient response in the F344 rat The evidence for carcinogenicity in the B6C3F1 mouse is not
as clear as in the rat. The weight of evidence of the three mouse sites—male lung/bronchus and
thyroid follicular cell and female mammary gland—suggests collectively a sufficient response in
the B6C3F1 mouse.. The Sprague-Dawley (CD) rats show a benign, limited liver response but at
a site concordant with the other rat species tested (F344/N). The Sprague-Dawley (CD) rat does
not respond with testicular tumors, although there are known toxic effects of NB with male
reproduction and sperm production.
16
-------
Table 5. Summary of nitrobenzene cardnogenicity results
Site of increased
tumorigenldty
Sex
Evidence of
carcinogenidty
Comments
B6C3F1 Mouse
Lung:
Alveolus and bronchus
Thyroid:
Follicular cell
Mammary gland
M
M
F
Limited
Limited
Marginally
sufficient
F344/NRat
Liver:
Hepatocellular
Thyroid:
Follicular cell
Kidney:
Tubular cell
Endometrial polyp
M
M
M
F
Sufficient
Limited
Limited
Limited
Benign tumor increase only;
carcinomas spread evenly over dose
groups (no trend)
Benign tumors only with dose trend
Historical and concurrent controls
suggest malignant but low-level
cancer incidence in a two-dose
examination (0 and 50 ppm dose
groups).
Clear evidence of malignancy and
exposure related to cancer incidence in
the liver
Another thyroid follicular response;
marginal statistics: only a trend with
dose and only a suggestion of
.malignancy
High dose only; benign tumorigenic
response
Benign response
Sprague-Dawley (CD) rat
Liver:
Hepatocellular
M
Limited
Benign response (unlike the male
F344/Nrat) .
17
-------
NB is metabolically related to aniline by nitroreduction, that is, aniline can be made from
NB in the gastrointestinal tract by resident flora nitroreductases (Mirsalis et al., 1982; Bceda and
Kita, 1964; Rickert et al., 1983; Reddy et al., 1976). After NB oral ingestion and after NB passes
through the acid stomach, aniline forms in the cecum (figure 3A). This bacterial nitroreductase
activity is found in rodents as well as in humans. Aniline already is classified by EPA as a
Category B2 carcinogen. A compound such as NB that generates aniline also should be
considered a B2 carcinogen if comparable exposure and metabolic conversion conditions occur.
In the CUT study exposure is by inhalation. The aniline metabolite may not be formed in
the nasal passage or lungs of CIIT rodents because the nasal pH is neutral in the nasal passages,
and there is a lack of nitroreduction flora in the passageways. However, rodents consistently
groom each other, and because the fur most likely has bound NB residues from the chronic
inhalation exposure, rodents can obtain oral doses of the parent compound (NB' from licking the
fur during grooming. Therefore, aniline can be formed from the licked and ingested NB parent
compound. The amount of aniline produced by this mechanism is uncertain. It is clear that NB
exposure also can be metabolized systemically, as other exocylic nitroaromatics, to the amine (in
this case aniline) passing through reactive free radicals and nitrosoamine and
phenylhydroxylamine (figure 3B). These are likely carcinogenic compounds participating in the
cancer mechanism of NB.
Dinitrobenzene is a chemical analogue of NB. However, there is inadequate cancer
information for the ortho or para isomers of the chemical analogue dinitrobenzene; these isomers
are classified as Category D carcinogens. Benzene is a rather remote chemical analogue of NB
and is a Category A carcinogen. Also, in tissues where oxygen is sufficiently available
superoxide O2a can form, and O2* is thought to be carcinogenic (Dreher and Junod, 1997; Kensler
et al'., 1989a; b; Guyton and Kensler, 1993). Therefore, based on the potential presence of the
above reactive metabolites, NB may be considered carcinogenic based on metabolic mechanisms.
The mutagenicity of NB is not positive and thus does not support the carcinogemcity
classification by a direct DNA-altering mechanism. It is not certain whether NB or
NB/metabolite combinations cause the mu'ltisite and two species carcinogenesis. ]
In conclusion, inhaled NB is toxic, causing anemia and methemoglobinemia, at a number
of noncarcinogenic sites (Cattley et al., 1994). Whether this type of toxicity occurs at
carcinogenic sites and is related to the mechanisms of carcinogemcity is unknown;-nonetheless,
NB is carcinogenic in rats and in mice, a fact that is likely related to the redox reactions of NB.
In the male F344/N rat, evidence for liver carcinogemcity is sufficient, and limited
carcinogenicity occurs at rat thyroid, kidney, and endometrium sites. There also is limited liver
tumorigenic evidence in a second rat strain, Sprague-Dawley (CD). In the mouse, NB is
sufficiently carcinogenic in the mammary gland at the high dose (lower and mid doses were not
18
-------
histologically examined). There are two additional limited mouse sites, thyroid and lung. The
systemic excessive tumors, compared with no-dose controls, suggest that inhaled MB causes
internal cancer effects that stem from absorbed NB and/or its metabolites (Cattley et al., 1994).
The NB parent, aniline, nitrosobenzene, and phenylhydroxylamine are all carcinogenic candidate
causal agents, and it could be a combination of these agents—in the proper organ deposition—that
actuates the initiation process followed by tumor promotion and, in the F344/N rat liver, tumor
progression to malignancy. This mixture could be responsible for the spectra of oncogenicity
both in various organ sites and degree of tumorigenicity severity. Furthermore, aniline,
nitrosobenzene, and phenylhydroxylamine can cause methemoglobinemic anemia.
Methemoglobinemic anemia was observed in all three species tested in the CIIT study (1993)
and has been seen in humans too (Kiese, 1974). Therefore, the presence of methemoglobinemic
anemia suggests the corporal presence of these metabolites of NB in NB exposures.
It is presumed that due to similarity in metabolism between animals and man that both
species are at risk by either inhalation, dermal, or ingestion routes of exposure. It is expected,
however, that the inhalation and dermal routes are the primary means by which humans could be
exposed to NB in the United States. The weight of evidence for NB carcinogenicity is based on
an adequate CIIT inhalation study (1993) that indicates, according to the 1986 Guidelines for
Cancer Risk Assessment (U.S. EPA, 1986), that NB is a Category B2 carcinogen: NB is
probably carcinogenic to humans. The B2 categorization is supported by (1) metabolism and (2)
structure-activity relationships, but not by the mutagenicity information available. This report
recommends the upgrading of the current carcinogenicity Category D (memorandum from J.
Holder to V. Dellarco, Office of Health and Environmental Assessment, dated September 28,
1992)toaB2. •
In summary, there were eight NB-caused tumorigenic organ sites spread over different
species, sexes, and doses. The specific mechanism of NB carcinogenicity is still not completely
understood but likely proceeds by redox mechanisms in various tissues. NB is classified by any
human route as Category B2 according to the 1986 cancer guidelines (U.S. EPA, 1986);
according to the April 23,1996, proposed cancer guidelines (U.S. EPA, 1996),, NB is classified
as a likely human carcinogen by any route of human exposure.
NB should be examined further for human metabolic and cancer results. A literature
search did not produce any relevant epidemiologic carcinogenicity studies on NB that could
substantiate the results of the animal studies presented here in support of the carcinogenicity
classification of NB.
19
-------
5. REFERENCES
Agency for Toxic Substances and Disease Registry (ATSDR). (1990) Toxicological profile for
nitrobenzene. Public Health Service, U.S. Department of Health and Human Services, Atlanta,
GA.TP-90-19.
Beauchamp, RO; Irons, RD; Rickert, DE; Couch, DB; Hamm, TE. (1982) A critical review of
the literature on nitro benzene toxicity. CRC Crit Rev Toxicol 1133-83.
Bond, JA; Rickert, DE; Popp, JA. (1981) Induction of hepatic and testicular lesions in Fischer F-
344 rats by single oral doses of nitrobenzene. Fund Appl Toxicol 1:389-394.
Cattley, R; Everitt, JI; Gross, EA; Moss, OR; Hamm, TE; Popp, JA. (1994) Carcinogenicity and
toxicity of inhaled nitrobenzene in B6C3F1 mice and F344 and CD rats. Fund Appl Toxicol
22:328-340.
Cerutti, PA. (1994) Oxy-radicals and cancer. Lancet 344:862-863.
Chemical Industry Institute of Toxicology (CUT). (1993) A chronic inhalation toxicity study of
nitrobenzene in B6C3F1 mice, Fischer 344 rats and Sprague-Dawley (CD) rats. CIIT, Research
Triangle Park, NC.
Chiu, CW; Lee, LH; Wang, CY; Bryan, GT. (1978) Mutagenicity of some commercially
available nitro compounds for Salmonella typhimurium. Mutat Res 58:11-22.
Dorigan, J; Hushorn, J. (1976) Air pollution assessment of nitrobenzene. Hazard evaluation
prepared for the U.S. Environmental Protection Agency, Washington, DC. Available from:
National Technical Information Service, Springfield, VA, PB-257-776.
Dreher, D; Junod, AF. (1996) Role of oxygen free radicals in cancer development. Eur J Cancer
.32A:30-38."
Eyer, P; Ascherl, M.' (1987) Reactions of para-substituted nitrosoamines with human
hemoglobin. Biol Chem Hoppe-Seyler 368:285-294.
Feig, DI; Reid, TM; Loeb, LA. (1994) Reactive oxygen species in tumorigenesis. Cancer Res
54:1890s-1894s.
Flohe", L; Beckmann, R; Giertz, H; Loschen, G. (1985) Oxygen-centered free radicals as
mediators of inflammation. In: Oxidative stress, Sies, H, ed. NewYork: Academic Press, pp. 403-
435. .
I
Goldstein, RS; Rickert, DE. (1984) Macromolecular binding of [MC]nitrobenzene in the
erythrocyte and spleen of rats and mice. Chem Biol Interactions 50:27-37.
20
-------
Goldstein, RS; Rickert, DE. (1985) Relationship between red blood cell uptake and
methemoglobin production by nitrobenzene and dinitrobenzene in vitro. Life Sci 36:121-125.
Guyton, KZ; Kensler, TW. (1993) Oxidative mechanisms in carcinogenesis. Brit Med Bull
49:523-544.
Haseman, J; Huff, J; Rao, GN; Arnold, JE; Bobrman, G; McConnell, EE. (1985) Neoplasms
observed in untreated and corn oil gavage control groups of F-344/N rats 'and (C57BL/6N x
C3H/HeN)F, (B6C3F1) mice. JNCI 75:975-983.
Dceda, M; Kita, A. (1964) Excretion of ^-nitrophenol andp-aminophenol in the urine of a patient
exposed to nitrobenzene. Br J Ind Med 21:210-213.
Keher, JP. (1993) Free radicals as mediators of tissue injury and disease. Crit Rev Toxicol 23:21-
4S.
Kensler, T; Egner, P; Taffe, B; Trush, M. (1989a) Role of free radicals in tumor promotion and
progression. Prog Clin Biol Res 298:233-248.
Kensler, TW; Egner, PA; Taffe, BG; Trush MA. (1989b) Role of free radicals in tumor
promotion and progression. In Skin carcinogenesis: mechanisms and human relevance, New
York: Alan R. Liss, p..233.
Kiese, M. (1966) The biochemical production of ferrihemiglobin-forming derivatives from
aromatic amines, and mechanisms of ferrihemiglobin formation. Phannacol Rev 18:1091-1161.
Kiese, M. (1974) Methemaglobinemia: a comprehensive treatise. Cleveland, OH: CRC Press.
Kimmick, GG; Bell, RA; Bostick, RM. (1997) Vitamin E and breast cancer a review. Nutri
Cancer Int 127:109-117.
Kuroda, Y. (1986) Genetic and chemical factors affecting chemical inutagenesis in cultured
mammalian cells. In: Antimutagenesis and anticarcinogenesis mechanisms, Shankel, DM;
Hartman, PE; Kada, T; Hollaender, A, eds. New York: Plenum Press; pp. 359-375.
Levin, AA; Dent, JG. (1982) Comparison of the metabolism of nitrobenzene by hepatic
microsomes and cecal microflora from Fischer-344 rats and the relative importance of each in
vivo. Drug Metab Disposit 10:450-454.
Levin, AA; Bosakowski, T; Earle, LL; Butterworth, BE. (1988) The reversibility of
nitrobenzene-induced testicular toxicity. Toxicology 53:219-230.
Mabey, W;!Smith, JH; Podoll.'R. (1982) Aquatic fate process data for organic priority pollutants.
Prepared for the Office of Water Regulations and Standards, U.S. Environmental Protection
Agency, Washington, DC: EPA/444/4-81/014.
21
-------
Mason, RP; Holtzman, JL. (1975a) The mechanism of microsomal and mitochondria!
nitroreductase. Electron spin resonance evidence for nitroaromatic free radical intermediates.
Biochemistry 14:1626-1632.
Mason, RP; Holtzman, JL. (1975b) The role of catalytic superoxide formation in the 02
inhibition of nitroreductase. Biochem Biophys Res Com.m 67:1267-1274.
McLaren, TT; Foster, PMD; Sharpe, RM. (1993) Effect of age on seminiferous tubule protein
secretion and the adverse effects of testicular toxicants in the rat Int J Androl 16:370-379.
Miller, JA. (1970) Cafcinogenesis by chemicals: an overview--GHA. Clowes Memorial Lecture.
Cancer Res 30:559-576. '
Mirsalis, JC; Hamm, TE; Sherrill, M; Butterworth, BE. (1982) The role of gut flora in the
genotoxicity of dinitrotoluenes. Nature 295:322-323.
Netke, SP; Roomi, MW; Tsao, C; Niedzwiecki, A. (1997) Ascorbic acid protects guinea pigs
from acute aflatoxin toxicity. Toxicol Appl Pharmacol 143:429-435.
Parke, DV. (1956) Detoxication. LXVin. The metabolism of C14-nitrobenzene in the rabbit and
guinea pig. Biochem J 62:339-346.
Piotrowski, J; Dutkiewicz, B; Popinska, E. (1975) Excretion of p-aminophenol and p-nitrophenol
in rat urine after intravenous and dermal treatment with nitrobenzene. Bromatol Chem Tokskol 8:
271-272.
Primiano, T; Sutler, TR; Kensler, TW. (1997) Antioxidant-inducible genes. Adv Pharmacol
38:293-328.
Reddy, BG; Pohl, LR; Krishna, G. (1976) The requirement of the gut flora in nitrobenzene-
induced methemoglobinemia. Biochem Pharmacol 25:1119-1122.
Rickert, DE; Bond, JA; Long, RM; Chism, JP. (1983) Metabolism and excretion of nitrobenzene
by rats and mice. Toxicol Appl Pharmacol 67:206-214.
Rickert, DE. (1984) Toxicity of nitroaromatic compounds. Washington, DC: Hemishere
Publishing Corp.
Robinson, D; Smith, JN; Williams, RT. (1951) Studies in the detoxic'ation. 40. The metabolism
of nitrobenzene in the rabbit, o-, m-, p-nitrophenols and 4-nitrocatecol as metabolites of
nitrobenzene. Biochem J 50:228-235.
Salmowa, I; Piotrowski, J; Neuhorn, U. (1963) Evaluation of exposure to nitrobenzene. Br J Ind
Med 10:41-49.
Sealy, RC; Swartz, HM; Olive, PL. (1978) Electron spin resonance-spin trapping. Detection of
22
-------
superoxide formation during aerobic microsomal reduction of nitre-compounds. Biochem
Biophys Res Comm 82:680-684.
Suzuki, J; Takahashi, N; Kobayashi, Y; Miyamae, R; Ohsawa, M; Suzuki, S. (1987) Dependence
on Salmdnella typhimurium enzymes of mutagenicities of nitrobenzene and its derivatives in the
presence of rat-liver S9 and norharman. Mutat Res 178:187-193.
Tomaszewski, JE; Jerina, DM; Daly, JW. (1975) Deuterium isotope effects during formation of
phenols by hepatic monooxygenases. Evidence for an alternative to the arene pathway.
Biochemistry 14:2024-2031,
U.S Environmental Protection Agency. (1986, Sept. 24) Guidelines for carcinogen risk
assessment. Federal Register 51(185):33992-34003.
U.S. Environmental Protection Agency. (1995) Integrated Risk Information System (IRIS).
Electronic database in the National Library of Medicine Toxline, current listing, Online. National
Center for Environmental Assessment, Cincinnati-, OH.
U.S. Environmental Protection Agency. (1996, April 22) Proposed guidelines for carcinogen risk
assessment. Federal Register 61(79):17960-18011.
Weisburger, JH; Weisburger, EK. (1973) Biochemical formation and pharmacological,
lexicological, and pathological properties of hydroxylamines and hydroxamic acids. Pharmacol
Rev 25:1-66.
23
-------
6. APPENDIX
This section is abstracted from the 1995 CRAVE presentation as submitted to the
Integrated Risk Information System (IRIS) committee. It has been approved for IRIS. Questions
concerning these calculations should be addressed to Jennifer Jinot, NCEA-Washington office.
QUANTITATIVE ESTIMATE OF CARCINOGENIC RISK FROM INHALATION
EXPOSURE
1. SUMMARY OF RISK ESTIMATES
• Inhalation unit risk: 3E-5 per ug/m3
• Inhalation slope factor: 1E-1 permg/kg/d
• Extrapolation method: linearized multistage (GLOBAL86), extra risk
• ED 10: 1 mg/kg/d (5E+3 ug/m3)
• Air concentrations at specified risk levels:
Risk Level Concentration
E-4 (1 in 10,000) = 3 ng/m3
E-5 (1 in 100,000) = 0.3 ug/m3
E-6 (1 in 1,000,000) = 0.03
2. DOSE-RESPONSE DATA
• Tumor type: hepatocellular, thyroid follicular cell, or kidney tubular cell adenomas or
carcinomas
• Test animal: male F344/N rat
• Route: inhalation
• Reference: A chronic inhalation toxicity study of nitrobenzene in B6C3F1 mice, Fischer
344 rats, and Sprague-Dawley rats (CUT, 1-22-93)
Administered Human equivalent Adenomas + carcinomas
exposure fppm) exp (ug/cu.m.) liver thyroid kidney
0 0 1/60 2/60 0/60
1 9.1E+2 4/60 1/60 0/60
5 4.6E+3 5/60 5/60 0/60
25 2.3E+4 16/60 8/60 6/60
24
-------
3. ADDITIONAL COMMENTS
• The unit risk is considered to be a plausible upper bound on the increased cancer risk
from lifetime inhalation of nitrobenzene.
• The human equivalent exposures were based on an assumption of ppm equivalence in
exposure across species and were calculated by converting ppm nitrobenzene to ug/m3 (1
ppm = 5.12 mg/mj) and adjusting for partial exposure 6 hours/day, 5 days/week.
• The risk estimates were derived for the proportions of tumor-bearing animals, that is,
those animals with any individual tumor or combination of tumors from the three
significant tumor sites Giver, thyroid, and kidney), as determined from the individual
animal data provided by CITT. The proportions of tumor-bearing animals were 3/60,
5/60,10/60, and 26/60 for 0,1,5, and 25 ppm exposure, respectively.
• The tumor incidence denominators reflect the original number of animals per dose group
(excluding those for interim sacrifice). No adjustments were made for the number of
animals at risk at the time of the first tumor because these were late-occurring tumors (the
first tumor at one of these sites was a thyroid tumor at day 563), and there was no
substantial early mortality (only one rat died in the first year).
25
------- |