DRAFT—DO NOT CITE OR QUOTE
NCEA-S-0570
SEPA
TOXICOLOGICAL REVIEW
OF
CHLOROETHANE
(CAS No. 75-00-3)
In Support of Summary Information on the
Integrated Risk Information System (IRIS)
July 1999
NOTICE
This document is a preliminary draft. It has not been formally released by the U.S.
Environmental Protection Agency and should not at this stage be construed to represent Agency
position on this chemical. It is being circulated for peer review on its technical accuracy and
science policy implications.
U.S. Environmental Protection Agency
Washington, DC
-------
DISCLAIMER
This document is a preliminary draft for review purposes only and does not constitute
U.S. Environmental Protection Agency policy. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
7/12/99
11
DRAFT—DO NOT CITE OR QUOTE
-------
CONTENTS—TOXICOLOGICAL REVIEW FOR CHLOROETHANE
(CAS No. 75-00-3)
FOREWORD viii
AUTHORS, CONTRIBUTORS, AND REVIEWERS ix
LIST OF ABBREVIATIONS x
1. INTRODUCTION 1
2. CHEMICAL AND PHYSICAL INFORMATION RELEVANT TO ASSESSMENTS 3
3. TOXICOKINETICS RELEVANT TO ASSESSMENTS 4
3.1. ABSORPTION 4
3.1.1. Gastrointestinal Absorption 4
3.1.2. Respiratory Absorption 4
3.2. DISTRIBUTION, METABOLISM, AND EXCRETION 6
4. HAZARD IDENTIFICATION 15
4.1. STUDIES IN HUMANS—EPIDEMIOLOGY, CASE REPORTS, CLINICAL
REPORTS 15
4.1.1. Oral Exposure 15
4.1.2. Inhalation Exposure 15
4.1.3. Dermal Exposure 16
4.2. ACUTE, SUBCHRONIC, AND CANCER BIOASSAYS IN ANIMALS—ORAL AND
INHALATION 16
4.2.1. Oral Exposure 16
4.2.2. Inhalation Exposure 17
4.2.2.1. Landry Inhalation Studies 17
4.2.2.2. Principal Study Performed by the U.S. National Toxicology Program
18
4.2.2.2.1. NTP acute study 19
4.2.2.2.2. Subchronic study 19
4.2.2.2.3. Chronic study 19
4.2.2.2.3.1. I''344 Rat ToxicologicalResults in the NTP Study 20
4.2.2.2.3.2. B6C3F1 Mouse Toxicological Results in the NTP Study . 23
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES—ORAL AND INHALATION ... 25
4.3.1. Oral Exposure 25
7/12/99
in
DRAFT—DO NOT CITE OR QUOTE
-------
CONTENTS (continued)
4.3.2. Inhalation Exposure 25
4.3.2.1. Principal Study 25
4.3.2.2. Supporting Reproductive or Teratological Studies 30
4.4. OTHER TOXICITY STUDIES 31
4.4.1. Acute Toxicity Studies 31
4.4.1.1. Neurotoxicity 31
4.4.1.2. Immunotoxicity 33
4.4.1.3. Cardiac Sensitization 33
4.4.1.4. Dermal Effects 34
4.4.1.5. Kidney Effects 34
4.4.2. Genotoxicity 35
4.5. SYNTHESIS AND EVALUATION OF MAJOR NON-CANCER EFFECTS AND MODE
OF ACTION—ORAL AND INHALATION 35
4.5.1. Primary Effect 36
4.5.1.1. Reproductive and Developmental Toxicity 36
4.5.2. Secondary Effects 37
4.5.2.1. Weight Loss 37
4.5.2.2. Hepatotoxicity 37
4.5.2.3. Neurotoxicity 38
4.5.3. Mode of Action of Toxic Effects 38
4.6. WEIGHT-OF-EVIDENCE EVALUATION AND CANCER
CHARACTERIZATION—SYNTHESIS OF HUMAN, ANIMAL AND OTHER
SUPPORTING EVIDENCE, CONCLUSIONS ABOUT HUMAN
CARCINOGENICITY, AND LIKELY MODE OF ACTION 39
4.7. OTHER HAZARD IDENTIFICATION ISSUES 44
4.7.1. Possible Structural-Activity Relationships 44
4.7.2. Possible Gender Differences 47
5. DOSE-RESPONSE ASSESSMENTS 48
5.1. ORAL REFERENCE DOSE (RfD) 48
5.2. INHALATION REFERENCE CONCENTRATION (RfC) 48
5.2.1. Choice of Principal Study and Critical Effect - With Rationale and Justification
48
5.2.2. Methods of Analysis 48
5.2.2.1. Principal Study 48
5.2.2.2. Primary Supporting Study 49
5.2.3. RfC Derivation Including Application of Uncertainty Factors (UF) and Modifying
Factors (MF) 49
7/12/99
iv
DRAFT-DO NOT CITE OR QUOTE
-------
CONTENTS (continued)
5.3. CANCER ASSESSMENT 51
5.3.1. Qualitative Cancer Assessment in Animals 51
5.3.2. Quantitative Cancer Assessment in Animals 52
5.3.2.1. Considerations in Quantitative Cancer Assessment 52
5.3.2.2. LMSMethod 53
5.3.2.3. LMS Method Calculation of Cancer Slope 54
5.3.3. Discussion of Confidence in Cancer Quantitative Assessment in Animals 55
6. CHARACTERIZATION OF ASSESSMENTS 58
6.1. ORAL RID 58
6.2. INHALATION RfC 58
6.3. CANCER ASSESSMENT 58
6.4. CHARACTERIZATION OF HAZARD EXPECTED UPON HUMAN EXPOSURE TO
CHLOROETHANE 60
7. REFERENCES 64
8. APPENDIX 71
7/12/99
v
DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF TABLES
Table 1. Distribution of recovered radiolabeled chloroethane in female rats and mice 48 hr after
inhalation exposure 5
Table 2. GSH levels in female B6C3F1 mouse tissues after in vivo inhalation exposure to CE: specific
GSH levels at the completion of exposure and after an 18-hr recovery 8
Table 3. GSH levels in rat tissues after in vivo inhalation exposure to CE-specific GSH levels at the
completion of exposure and after an 18 hr recovery 9
Table 4. The effect of chloroethane exposure on baseline cytosolic GSH-transferase 11
Table 5. Excretory kinetics of S-ethyl-N-acetyl-cysteine in CE-exposed F344 rats and B6C3F1 mice
12
Table 6. Excretory kinetics of S-ethyl-L-cysteine in CE-exposed B6C3F1 mice 12
Table 7. Tumors of F344/N rats at 2 years 22
Table 8. Incidence of tumors in female B6C3F1 mice after exposure to CE for 2 years 24
Table 9. Chloroethane inhalation teratology in CF-1 mice: incidence of fetal alterations among litters of
mice 27
Table 10. Toxicity/carcinogenicity of chloroethane in experimental studies 41
Table 11. Summary and Conclusions of Tumorigenesis in Rats and Mice 43
Table 12. Common metabolic features of chloroethane and chloromethane: potential relevance to
tumor formation in experimental studies 46
Table 13. Quantitative cancer responses in the female B6C3F1 mouse liver and uterus 54
Table 14. Comparison of noncancer and cancer hazard evaluations 61
Table 15. Chemical-specific dose parameters 72
7/12/99
vi
DRAFT-DO NOT CITE OR QUOTE
-------
LIST OF FIGURES
Figure 1. Chloroethane chemical structure 1
Figure 2. Proposed Metabolic Pathways for Chloroethane. Scheme presented in Fedtke et al., 1994b).
6
Figure 3. Proposed metabolic scheme for chloroethane disposition and toxicity in mice and rats following
a high-level inhalation exposure 7
Figure 4. Reductive conjugation of chloroethane with glutathione (GSH) 10
Figure 5. Occurrence of delayed foramina closure in skulls of CF-1 mice 26
7/12/99
vii
DRAFT—DO NOT CITE OR QUOTE
-------
FOREWORD
The purpose of this Toxicological Review (ToxR) is to support the hazard identification
and dose-response assessment for cancer and noncancer effects (the oral reference dose [RfD]
and the inhalation reference concentration [RfC]) from chronic exposure to chloroethane (CE).
Supportive CE subchronic studies also are included. The ToxR is a review and analysis of data
supporting the chemical or toxicological nature of CE and supports the Integrated Risk
Information System (IRIS) Summary document. The ToxR characterizes each relevant study
with regard to overall confidence in the quantitative and qualitative aspects of hazard. This
analysis considers knowledge gaps, uncertainties, and quality of data, while highlighting the
limitations of the individual studies and providing a guide to the risk assessment process.
For other general information about this assessment or other questions relating to IRIS,
the reader is referred to EPA's Risk Information Hotline at 513-569-7254.
7/12/99
viii
DRAFT-DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Chemical Manager/Author
James W. Holder, Ph.D.
National Center for Environmental Assessment-Washington Office
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
The first draft of the ToxR was prepared by TN&A, Inc. under EPA Contract No. 68-C6-0024. Relevant
literature has been reviewed through January 1999.
Internal Reviewers1
Jane Caldwell, Ph.D.
U.S. EPA
OAQPS
Research Triangle Park, NC 27709
Dennis Lynch, Ph.D.
Experimental Toxicology
NIOSH
Cincinnati, OH
Daniel L. Morgan, Ph.D.
Respiratory Toxicology
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Alberto Protzel, Ph.D.
OPPTS/OPP/HED
U.S. EPA
401 M St., S.W.
Washington, DC 20460
Collegial Reviewers1
Femi Adeshina
NCEA-CIN/ORD/U.S. EPA
Cincinnati, OH 45268
Gary Foureman, Ph.D.
HPAG/NCEA-RTP/U.S. EPA
Research Triangle Park, NC 27711
Jennifer Jinot
NCEA-W/ORD/U.S. EPA
401 M St., S.W.
Washington, DC 20460
Gary Kimmel, Ph.D.
NCEA-W/ORD/U.S. EPA
401 M Street, S.W.
Washington, DC 20460
Jennifer Seed, Ph.D.
RAD/OPPT/U.S. EPA
401 M St., S.W.
Washington, DC 20460
G. Daniel Todd, Ph.D.
Environmental Health Scientist
Toxicology Information Branch
ATSDR
Atlanta, GA
1 The contributions and criticisms of all the reviewers are appreciated. Peer review of the IRIS support
document (ToxR) was performed by Internal (U.S. Government) Reviewers listed on the left. These reviewers
were selected without knowledge of the author of this document, whereas the collegial reviewers in the right
column were invited because of their expertise on issues that were particularly of concern in characterizing
chloroethane toxicology. Comments of all 10 reviewers were reconciled with no major outstanding issues.
7/12/99
IX
DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS
ALT Alanine aminotransferase
BE Bromoethane
BM Bromomethane
BMC Benchmark Concentration
CE Chloroethane
CDNB l-Chloro-2,4-dinitrobenzene
CHO Chinese hamster ovary
CUT Chemical Industry Institute
of Toxicology
CM Chloromethane
EROD Ethoxyresorufin O-dealkylase
GD Gestation day
GSH Glutathione
HDT Highest dose tested (in a bioassay)
HEC Human equivalent concentration
HPRT Hypoxanthine
Phosphoribosyltransferase
IM Iodomethane
MDT Maximum tested dose (in a
bioassay; same as HDT)
MTD Maximum tolerated dose
(ascertained in a subchronic
bioassay)
MN Micronucleus
MNL Mononuclear cell leukemia
NPSH Non-protein sulfhydryl
NTP National Toxicology Program
p-NP p-Nitrophenyl hydroxylase
PBPK Physiologically based
pharmacokinetic
PROD Pentoxyresorufin O-dealkylase
RfC Reference concentration
RfD Reference dose
ToxR Toxicological review
SECys S-Ethyl-L-cysteine
SEG S-Ethyl glutathione
SENAYCys S-Ethyl-N-acetyl-cysteine
UDS Unscheduled DNA
synthesis
7/12/99
x
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
1. INTRODUCTION
This document (ToxR) presents a complete compilation and analysis of available
information on the toxicity of chloroethane (CE) in experimental exposure studies to animals.
No human studies are known to exist. CE is a simple halohydrocarbon (Figure 1). In an attempt
to establish relative safe environmental exposure levels, the
quantitative oral reference dose (RfD) and inhalation reference
concentration (RfC) values shall be developed from applicable
non-cancer toxicological responses to CE where the data are
sufficient.
Toxicological analysis of chronic exposure studies leads
to the derivation of the RfD and/or RfC that provide
information on long-term toxic effects other than
carcinogenicity. The RfD assumes that thresholds exist for
certain toxic effects such as cellular necrosis, but not for other
toxic effects such as some carcinogenic responses. The RfD,
expressed in milligrams per kilogram per day (mg/kg/day), is an
approximation of the daily exposure to humans that is likely to
result in no appreciable risk of deleterious effects over a lifetime
of continuous exposure. The inhalation RfC is analogous to the oral RfD and considers toxic
effects to the respiratory system (portal-of-entry) and extrarespiratory, or systemic, effects
expressed in milligrams per cubic meter (mg/m3).
The carcinogenicity assessment provides information on aspects of the carcinogenic risk
assessment for the agent in question which includes the U.S. EPA classification, and quantitative
estimates of risk from oral exposure and from inhalation exposure. The classification reflects a
weight-of-evidence judgment of the likelihood that the agent may be a human carcinogen, or not,
and the conditions under which any potential carcinogenic effects may be expressed. Quantitative
risk estimates are presented in three ways. The slope factor, resulting from the application of a
low-dose extrapolation procedure, is presented as the risk per milligrams per kilogram per day
[(mg/kg/day)"1]. The risk is the quantitative estimate in terms of either risk microgram per liter
[(lig/L)"1] drinking water or risk per microgram per cubic meter [(|ig/m3)"'] air breathed. The
third form in which risk is presented is a drinking water or air concentration providing cancer risks
of 1 in 10,000, 1 in 100,000, or 1 in 1,000,000.
H
i
H
H—C-
I
1
-C—CI
i
1
H
I
H
Figure 1. Chloroethane
chemical structure.
Chloroethane is a small, gaseous,
hydrophobic molecule. C-l is
susceptible to nucleophilic attack
due to the polarity of the C-Cl
bond because of the
electronegativity of the CI atom
relative to the C-l carbon.
7/12/99
1
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
i
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
a
23
24
25
26
27
28
Hazard identification and dose-response assessment for CE follow the general risk
assessment principles for established by the National Research Council (1983).
EPA guidelines used in the development of this assessment include the following:
1. Guidelines for Carcinogenic Risk Assessment (U.S. EPA, 1986)
2. Proposed Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1996a)
3. Guidelines for Developmental Toxicity Risk Assessment (U.S. EPA, 1991)
4. Guidelines for Reproductive Risk Assessment (U.S. EPA, 1996b)
5. Guidelines for Neurotoxicity Risk Assessment (proposed) (U.S. EPA, 1995b)
6. Methods for Derivation of Inhalation Reference Concentrations and application of
Inhalation Dosimetry (U.S. EPA, 1996)
7. Guidelines for Mutagenicity Risk Assessment (U.S. EPA, 1986c)
8. Methods for Derivation of Inhalation Reference Concentrations and Application of
Inhalation Dosimetry (U.S. EPA, 1994)
9. Recommendations for and Documentation of Biological Values for Use in Risk
Assessment (U.S. EPA, 1988a)
10. Use of the Benchmark Dose Approach in Health Risk Assessment (U.S. EPA, 1995a)
11. Science Policy Council Handbook: Peer review (U.S. EPA, 1998b)
The literature search strategy for CE is based on the CASRN and at least one common
name, and includes the following databases: HEAST, RTECS, HSDB, TSCATS, CCRIS,
GENETOX, EMIC, EMICBACK, DART, TOXLINE, CANCERLINE, MEDLINE, and
MEDLINE backfiles. The current IRIS file for this chemical (U.S. EPA, 1998a) and the ATSDR
toxicological profile (ATSDR, 1997) was also used as a resource.
7/12/99
2
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
9
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
2. CHEMICAL AND PHYSICAL INFORMATION RELEVANT TO ASSESSMENTS
Common synonyms of chloroethane (CE) include ethyl chloride, monochloroethane,
Kelene, moriatic ether, narcotile, hydrochloric ether, Chloryl, Chloryl Anesthetic, Dublofix, and
NCI-C06224. Some relevant physical and chemical properties of CE are listed below (U.S. EPA,
1988b):
CE is a colorless gas at room temperature, with a sweet taste and a pungent ether-like
odor. CE is flammable. Even under increased pressure and lowered temperatures it is a volatile
and mobile liquid. The explosion limits are 3.8% up to 14.8% by volume in air, which means air
concentrations of 38,000 ppm or more can ignite. This indicates the upper limit in testing CE in
bioassays. Combusted CE forms phosgene (COCl2), among other products. CE reacts with
steam to form corrosive oxidizing materials. Under ambient conditions, CE is an extreme fire and
explosion risk at higher concentrations.
CE is used primarily as an intermediate in the production of perfumes, tetraethyl lead (a
decreasing use), ethylcellulose, ethylbenzene, alkyl catalysts, and pharmaceuticals. In the past, CE
was used as a general anesthetic (loss of sensation and consciousness) or a narcotic (producing
stupor) (Lawson, 1965; Cole, 1967). In recent times, however, CE's medical application has
become limited to use in topical skin analgesic sprays, for example, for the temporary relief of
sports injuries or the discomfort associated with ear piercing. CE has also been used on a limited
basis as a solvent (e.g., for elemental phosphorous, fats, waxes, acetylene, and a number of resins)
and a refrigerant (ATSDR, 1997).
Empirical formula: C2H5C1
Structural formula: CH3CH2C1
Molecular weight: 64.5
Specific gravity: 0.897 (at 20°C)
Vapor pressure: 1,000 mm Hg (at 20°C)
CASRN: 75-00-3
Vapor density: 2.22 (air = 1.0)
water solubility: 5,710 mg/L
Melting point: -138.7°C
Boiling point: 12.3 °C at 760 mm Hg
Log Kow: 1.43
Chloroethane gas conversion factors:
1 ppm = 2.64 mg/m3, 1.0 mg/m3 = 0.38 ppm
7/12/99
3
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
3. TOXICOKINETICS RELEVANT TO ASSESSMENTS
3.1. ABSORPTION
3.1.1. Gastrointestinal Absorption
Though no information is available on the intestinal absorption of CE in humans, a report
by Dow Chemical Co. (1992) addressed the potential for the compound to induce toxic effects via
the oral route in laboratory animals. CE was administered in a single gavage dose of either 37 or
1,750 mg/kg 14C-CE in corn oil. The animals were sacrificed 48 hr after dosing. An alternative
regimen involved the administration of seven daily doses of the unlabeled compound at 37 mg/kg,
followed on day 8 by 37 mg/kg of 14C-CE, before termination. Overall recovery of radioactivity
was good and ranged 87-93% of the administered dose. Most of the counts (77-89%) were in
exhaled as 14C02 or as unchanged CE. Recovery in feces was only 1.44%. This supports the
conclusion that CE absorption from the gastrointestinal tract is nearly quantitative. This uptake is
consistent with the ability of many lipid-soluble xenobiotics, such as CE, to cross the brush border
(and other biological membranes) with great facility.
3.1.2. Respiratory Absorption
The same physicochemical characteristics that favor absorption of the compound at the
intestinal mucosa might be expected to facilitate absorption at the alveolar membrane. Morgan et
al. (1970) investigated respiratory absorption: they studied a human volunteer who took one
breath of 38Cl-labeled CEvia the mouth, held his breath for 20 seconds, then exhaled. This was
repeated. Only 18% of the counts (radioactivity) was exhaled after two exhalations, thus by
inference 82% was retained and adsorbed. The constituents of the exhaled breath were not
analyzed. It was further noted that an additional 30% of the CE counts were exhaled during the
first hour. Only small amounts were excreted in the urine. Although this experiment was not
quantitative, it shows that pulmonary CE retention in the first hour is > '/2 of the initial inhaled
38Cl-labeled CE counts. This suggests that CE is more likely to be absorbed in the lung than to be
retained within the alveolar lumen.
Respiratory absorption has also been studied in laboratory animals. Groups of 10 female
B6C3F1 mice and 10 female F344 rats were exposed by inhalation to 150 (a low dose) or 15,000
ppm (a high dose) of 14C-CE (0.14-2.25 |iCi/mg CE) for 6 hr (Dow Chemical Co., 1992). No
males were studied. Half the animals were sacrificed immediately after dosing, while the other
7/12/99
4
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
half were maintained in metabolic cages for 48 hr. While these animals were metabolizing CE,
urine, feces, and exhaled gases were collected. Animals were sacrificed after 48 hr and selected
tissues were analyzed for radioactivity. The tissue distribution of CE or CE metabolites is
presented in Table 1.
At 150 ppm, female mice and rats, respectively, exhaled 42% and 54% of the counts as
C02, 35% and 32% in the tissues and carcass, 16% and 10% in the urine, 6% and 3% in the feces,
and < 2% in the breath as unchanged CE. So, at the low dose, a substantial portion of the inhaled
CE input was metabolized by both species, and there were comparable counts in the various
compartments in both species. These data support those of Morgan et al. (1970) by indicating
that CE can be readily absorbed at the alveolar membrane.
However, with exposure at the highest dose treated (HDT), 15,000 ppm, the relative
distributions shifted (Table 1). In both mice and rats, respectively, expired C02 decreased from
"150 ppm levels" to 32% and 19%, tissues and carcass decreased to 16% and 8%, urine increased
to 38%) in the female mouse but showed no change at 9% in the female rat, feces remained
unchanged at 7% in mice but decreased to 2% in the rat, and expired unchanged CE in the breath
Table 1. Distribution of recovered radiolabeled chloroethane
in female rats and mice 48 hr after inhalation exposure3
Mode of
excretion/
deposition
Relative percentage of radioactivity recovered (%)
Female mouse
Female rat
150 ppm
15,000 ppm
150 ppm
15,000 ppm
Expired CE
1.72 + 0.53
6.96+1.75
1.12 + 0.32
62.81 + 1.32
Expired C02
41.76 + 11.35
31.60 + 6.84
53.57 + 2.34
19.17 + 0.98
Urine
15.86 + 4.10
38.37 + 9.13
9.66+1.09
8.68 + 1.22
Feces
6.02+1.91
7.05 + 4.84
3.15 + 0.16
1.60 + 0.66
Tissue/carcass
34.65 + 12.79
16.02 + 2.04
32.03 +2.82
7.64 + 0.97
aValues are the means + SD for five animals in each exposure group. Males were not tested.
Source: Dow Chemical Co. (1992).
showed relatively large increases in the mouse, to 7% (4-fold), and to 63% in the rat (56-fold).
On a per microequivalent basis the authors reported a 49-fold increase in nonmetabolized CE in
the breath in the mouse and a 700-fold increase in the rat (Dow Chemical Co., 1992). Thus, the
7/12/99
5
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
compartmental recoveries did not increase in proportion to CE exposure. These disproportions,
and the high amount of parent CE exhaled at the HDT, suggest that CE metabolic disposition is
saturated at 15,000 ppm compared to 150 ppm.
3.2. DISTRIBUTION, METABOLISM, AND EXCRETION
There is no information on
the distribution, metabolism, and
excretion of CE in humans in the
literature. However, toxicological
data that shed light on these issues
have come from animal studies
involving (1) inhalation of
radiolabeled CE, and (2) in vivo
and in vitro experiments in which
the relevance of certain putative
catabolic mechanisms has been
evaluated following challenge with
CE. Taken together, these findings
have identified some CE
intermediates and excretory
products, thereby pointing to the
possible mechanism(s) that may be involved in CE's metabolism (Figure 2).
Dow Chemical Co. (1992) drew a contrast between the disposition of products of CE
metabolism after inhalation of low (150 ppm) versus high (15,000 ppm) concentrations of the
radiolabeled compound by female F344 rats or female B6C3F1 mice (Table 1). The higher
concentration of CE appeared to saturate the metabolic processes, resulting in an increase in the
proportion of unchanged compound that was exhaled. At the HDT, this CE exhalation was
especially marked in rats (62.81% of recovered radioactivity) versus mice (6.96% of recovered
radioactivity). These data in Table 1 suggest that female B6C3F1 mice may have a greater
capacity to metabolize CE at 15,000 ppm than female F344 rats. Whether as C02 or as CE, most
of the exhaled counts and those collected in urine were recovered in the first 24 hrs, thereby
showing rapid CE metabolism. After 48 hr, the primary target tissues appeared to be ovary,
adrenals, and skin (Dow, 1992). By contrast, the Dow research team noted a lack of selective
ch3-CHO
(acetaldehyde)
ch3ch2ci
chloroethane
CYT .
P-450 "
CH2OH-CH2-Cl
(2-chloroethanol)
P-450
CH3CHOH-CI
(1 -chloro-1 -hydro xy-ethane)
-HC1
GSH transferase
\/
gs-ch2-ch3
(S - ethyl - glutathi one)
CH3CH2OH
(ethanol)
CH3-COOH
(acetic acid)
NAcCYS-CH2-CH3
(S-ethyl-N-acetyl-L-cysteine)
A
CyS-CH2-CH3
(S-ethyl-L-cysteine)
Figure 2. Proposed Metabolic Pathways for
Chloroethane. Scheme presented in Fedtke et al.,
1994b).
7/12/99
6
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
retention of counts in the uterus, an organ identified as an important site of potential carcinogenic
responses to the compound in female B6C3F1 mice (Section 4.2.2.2.3.4, p. 23).
Reduction. Dow Chemical Co. (1992) explored the effect of inhalation CE exposure on
tissue glutathione (GSH) content in female F344 rats and female B6C3F1 mice. GSH is a
reducing agent often employed in cells to metabolize xenobiotics (Figure 4). GSH content was
measured by analyzing the non-protein bound free sulfhydryls (NPSH). Dow researchers exposed
female rats and mice to 150, 3,000, 6,000 (mice only), or 15,000 ppm unlabeled CE for 3 or 6 hr.
Effects were seen only at 15,000 ppm CE, which suggests a threshold for GSH depletion
(Table 2). During this exposure period (at 15,000 ppm CE), the GSH decreased in mice and rats
below normal levels. Mice, for example, showed GSH depletions in the following tissues: liver
(21%), kidney (56%), lung (32%), and uterus (55%). Mouse brain and adrenals did not show
GSH decreases. Blood showed the largest absolute decrease, 870 —~ 618 nmol GSH/mg blood.
Blood can account for significant amounts of [GSH] changes in tissues, or could reflect systemic
GSH changes, or both. The above were the only tissues sampled for GSH. Recovery to control
levels and overshooting to excessive GSH tissue concentrations occurred 18 hr after CE exposure
in mice (Table 2). The rat data pointed to GSH decreases in the liver (65%), ovaries {51%) and
adrenals (32%) (Table 3). The authors discussed that a relationship between CE-induced GSH
depletion and the induced toxicity is plausible and a suggested pathway is presented in Figure 3 .
Fedtke et al. (1994a,b) sought to explain
the biochemical mechanism(s) by which CE is
catabolized and the processes by which CE
induces metastatic endometrial uterine tumors in
B6C3F1 mice but not in F344 rats. Using an
analogous protocol to that employed by the NTP
(1989a), these workers exposed groups of male
and female F344 rats and B6C3F1 mice to 0 or
15,000 ppm CE via inhalation, 6 hr/day for 5
days.
CE
(exhaled)
Inhalation
High level Chloroethane
CE-GSH
Urine
C02, Tissues
Altered
CNS
Activity
GSH depletion
(unidentified metabolites)
Hormonal
Effects
Uterine
Tumors
Hyperactivity
Figure 3. Proposed metabolic scheme for
chloroethane disposition and toxicity in mice
and rats following a high-level inhalation
exposure. Source: This scheme is adapted
from a scheme proposed by the Dow Chemical
Co., 1992.
7/12/99
7
DRAFT—DO NOT CITE OR QUOTE
-------
<1
to
Table 2. GSH levels in female B6C3F1 mouse tissues after in vivo inhalation exposure to CE: specific GSH levels at the
completion of exposure and after an 18-hr recovery
Exposure for 6 hr at 15,000 ppm
Tissues
Mouse tissue GSH levels (nmoles/mg tissue)a
Control
Exposed
Time after incubation
0 hr
18 hr
0:18
ratio
0 hr
18 hrb
0:18 ratio
Liver
5.49 +
0.37
4.73 + 1.21
1.22
1.18 + 0.27
5.50 + 1.29***
0.21
Kidney
2.80 +
0.31
3.33 + 0.28*
0.84
1.65 + 0.04
3.62 + 0.16***
0.46
Brain
1.62 +
0.09
1.55 + 0.11
1.05
1.30 + 0.09
1.48 + 0.12
0.88
Lung
1.74 +
0.23
1.85 + 0.12
0.94
0.56 + 0.11
2.33 + 0.05***
0.24
Ovary
1.68 +
0.04
1.60 + 0.03
1.05
0.94 + 0.10
1.59 + 0.17*
0.59
Adrenal
2.00 +
0.47
1.69 + 0.23
1.18
1.36 + 0.13
1.87 + 0.17
0.73
Uterus
1.48 +
0.42
1.20 + 0.65
1.23
0.82 + 0.27
1.67 + 0.25**
0.49
Blood
870.99 +
96.23
999.26+ 132.33
0.87
618.00 + 68.58
872.50 + 39.26***
0.71
00
o
o
o
H
n
HH
H
M
O
P
iO
a
o
H
Exposure to inhaled CE preceded tissue analysis for GSH levels. Recovery tissues were analyzed after exposure and 18 hr of
nonexposure. Values are the means of eight liver samples, four kidney samples, two adrenal and ovary samples, and four samples of
all other tissues.
bStatistical comparisons (0 hr versus 18 hr) were done by Dow using the Student's t-test, * <0.05, ** <0.01, *** <0.001.
-------
Source: Dow Chemical Co. (1992).
7/12/99
DRAFT9-DO NOT CITE OR QUOTE
-------
Table 3. GSH levels in rat tissues after in vivo inhalation exposure to CE-specific GSH levels at
the completion of exposure and after an 18 hr recovery
Exposure for 6 hr at 15,000 ppm
Rat tissue GSH levels (nmol/mg tissue)
Control
Exposed
Time after
incubation
0 hr
18 hr
0:18 ratio
0 hr
18 hr"
0:18
ratio
Liver
5.53+0.41
5.75 + 0.37
0.96
3.58 + 0.38
5.51+0.70***
0.65
Kidney
3.91+0.23
3.72 + 0.24
1.05
3.04 + 0.16
3.81+0.19***
0.80
Brain
1.60 + 0.05
1.56 + 0.10
1.03
1.45 + 0.10
1.44 + 0.11
1.01
Lung
1.68 + 0.08
1.76 + 0.06
0.95
1.32 + 0.07
1.69 + 0.28*
0.78
Ovary
2.56 + 0.18
2.89 + 0.35
0.89
1.45 + 0.39
2.72 + 0.51
0.53
Adrenal
2.42 + 0.02
2.82 + 0.05*
0.86
0.77 + 0.11
3.45 + 0.55*
0.22
Uterus
0.98 + 0.13
1.27 + 0.17*
0.77
0.67 + 0.10
1.01+0.30
0.66
Blood
805.88 + 43.28
N/D
-
1160.00 + 36.74
1006.76 + 84.07*
1.15
aExposure to inhaled CE preceded tissue analysis for GSH levels. Recovery tissues were analyzed after exposure and 18 hr of
nonexposure. Values are the means of eight liver samples, eight kidney samples, two adrenal and ovary samples, and four samples
of all other tissues.
b Statistical comparisons (0 hr versus 18 hr) were done by Dow using the Student's t-test, * <0.05, ** <0.01, *** <0.001.
Source: Dow Chemical Co. (1992).
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
In one study, Fedtke et al. (1994b) examined the ability of GSH to conjugate CE in a
reductive conjugation reaction, an example of which is shown in Figure 4. The enzymatic nature
of the CE-GSH reaction was investigated in an in vitro protocol featuring the addition of cytosolic
preparations from mice or rats to a mixture of CE and GSH. Cytosolic GSH concentrations (as
measured by NPSH) were measured in 105,000 x g supernatant centrifuge preparations from
liver, lung, kidney, and uterus in control and CE-exposed rats and mice. Also measured were the
most likely CE-GSH reaction product, S-ethyl glutathione (SEG), and an enzyme catalyzing the
synthesis of the SEG, GSH-S-transferase. Finally, the appearance of the putative SEG
metabolites, S-ethyl-N-acetyl-cysteine (SENACys) and S-ethyl-L-cysteine (SECys), was
monitored in the urine of control and CE-exposed rats and mice.
GSH-metabolite results were as follows (Fedtke et al., 1994b): In rats, GSH was
CH3CH2-C1 + GSH + [GSH-S-transferase]^ CHgCH^SG + HC1
Figure 4. Reductive conjugation of chloroethane with glutathione (GSH).
The same reaction is thought to take place with other methyl and ethyl halides.
decreased compared to controls in male liver (/KO.O l), female uterus and kidney (/K0.01), and
the lung of both sexes (/K0.01), In mice, significant GSH decreases were observed also in the
uterus and kidney (/K0.01), In rats, similar amounts of SEG were measured in both exposed and
control preparations, thus no effect. In mice elevated levels of SEG were measured in liver
cytosols (/K0.01),
Table 4 summarizes GSH S-transferase enzyme activities in various tissues in mice and
rats that were treated with 15,000 ppm CE or were air controls. The GSH S-transferase is
measured by use of the nonspecific substrate CDNB, which measures total GSH S-transferase
activity in tissue preparations; the use of CDNB may obscure any specific isozymic GSH S-
transferase changes. The comparison of S-transferase total activities showed that rats had
consistently higher activities than mice (Fedtke et al., 1994b). This does not agree with an earlier
Dow study finding that depletions in mice were -80% and in rats were -35% after CE treatment,
suggesting higher transferase activity in mice (Pottenger et al., 1992). When the activities were
compared within species pre- and posttreatment, the only biologically significant changes were in
the livers of female rats and female mice (Fedtke et al., 1994b).
7/12/99
11
DRAFT-DO NOT CITE OR QUOTE
-------
Table 4. The effect of chloroethane exposure on baseline cytosolic GSH-transferase
enzyme acl
tivity in rats and mice
Species/sex
Tissue
Controls
CE exposure
(|imol/min-mg)
(|amol/min-mg)
F-344-m
Liver
0.61+0.04
0.61+0.04
Lung
0.11+0.03
0.10 + 0.01
Kidney
0.19 + 0.02
0.15 + 0.01 a
F-344-f
Liver
0.54 + 0.05
0.74 +0.10 a
Lung
0.17 + 0.01
0.12 + 0.02b
Kidney
0.17 + 0.01
0.16 + 0.01
Uterus
0.26 + 0.04
0.33 + 0.16
B6C3Fl-m
Liver
3.09 + 0.12
3.07 + 0.45
Lung
0.33 + 0.05
0.40 + 0.09
Kidney
0.55 + 0.05
0.52 + 0.04
B6C3Fl-f
Liver
1.14 + 0.23
1.87 + 0.14b
Lung
0.48 + 0.02
0.32 + 0.06a
Kidney
0.70 + 0.02
0.63 + 0.01 b
Uterus
0.62 + 0.06
0.72 + 0.07
*p < 0.05 versus controls.
hp< 0.01 versus controls.
Source: Fedtke et al. (1994b). This paper also tabulated the excretory kinetics of the conjugate SEG.
1
2
3 metabolites, SENACys and SECys, and demonstrated elevated specific amounts (in |imol/kg body
4 weight) of SENACys in the urine of mice compared to rats (Table 5). SECys was undetected in
5 the urine of exposed rats, though the compound was present in the urine of both exposed and
6 control mice (Table 6).
7 Taken together, the data presented by Fedtke et al. (1994b) and Pottenger et al. (1992)
8 make the case that reductive GSH conjugation constitutes at least one important pathway for CE
7/12/99
12
DRAFT—DO NOT CITE OR QUOTE
-------
Table 5. Excretory kinetics of S-ethyl-N-acetyl-cysteine in CE-exposed F344 rats
and B6C3F1 mice a
Urine collection
time interval (hr)
F-344 (m)
(jimol/kg bw)
F-344 (f)
(jimol/kg bw)
B6C3F1 (m)
(jimol/kg bw)
B6C3F1 (f)
(|iimol/k<; bw)
0-7
19.2 + 7.6
27.0 + 5.0
93.4 + 36.7
102.1 + 14.7
7-24
50.1 + 8.6
54.2 + 5.1
44.1 + 33.0
27.1 + 8.9
24-31
17.5 + 12.2
8.1 + 11.5
155.6 + 45.7
58.6 + 5.6
31-48
55.7+16.0
89.7 + 9.2
31.2+16.5
6.3b
48-55
16.4 + 5.6
24.7 + 2.4
169.3 + 30.6
105.5 + 49.4
55-72
41.9 + 8.7
59.8 + 7.7
43.9 + 28.7
11.3 + 10.4
72-79
13.1 + 1.4
20.7+14.5
127.9 + 29.5
99.7 + 47.1
79-96
43.8+16.3
55.0+10.0
35.2+12.9
36.1°
96-103
6.9 + 2.2
5.6 + 6.2
34.8+15.8
75.6 + 38.6
a Data are from Fedtke et al. (1994b) and are the mean + SD.
b Data from one group only.
c Data from two groups only.
Table 6. Excretory kinetics of S-ethyl-L-cysteine in CE-exposed B6C3F1 mice a
Urine collection time interval (hr)
B6C3F1 (m) (jimol/kg bw)
B6C3F1 (f) (jimol/kg bw)
0-7
46.6+19.4
23.9 + 3.5
7-24
42.3 + 33.7
19.5 + 9.7
24-31
112.8+15.0
28.0 + 7.1
31-48
31.6+11.1
8.5b
48-55
46.8 + 24.7
33.7 + 5.6
55-72
28.8+10.3
6.3 + 3.2
72-79
43.5 + 18.2
25.3+2.4
79-96
18.7 + 8.6
8.4 + 6.9
96-103
9.3 + 3.1
17.8 + 8.7
a Data are taken from Fedtke et al. (1994b) and are the mean + SD.
b Data from one group only.
7/12/99
13
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
metabolism (see Figure 1, p. 1). Fedtke et al. (1994b) further discussed their results in the
context of CE's ability to induce uterine tumors in B6C3F1 mice, and speculated that CE-induced
tumor formation might be a consequence of alterations in normal cellular GSH pools, GSH
conjugation, and GSH-related metabolites.
Oxidation. Fedtke et al. (1994b) also examined the potential for CE to be oxidized via
one or more of the cytochrome P-450-dependent metabolic pathways. Liver microsome
preparations from CE-exposed or control F344 rats and B6C3F1 mice were measured for their
ability to oxidatively metabolize CE to acetaldehyde in vitro. Also measured were the specific
activities of p-nitrophenyl hydroxylase (p-NP), a marker enzyme for cytochrome P450IIE1;
ethoxyresorufin O-dealkylase (EROD), a marker for cytochrome P450IA; and pentoxyresorufin
O-dealkylase (PROD), a marker for cytochrome P450IIB. Samples of blood and urine from
exposed and control animals were also analyzed for acetaldehyde.
In the in vivo phase of the study, all animals lost weight after exposure to 15,000 ppm CE
6 hr/day for 5 days, though differences between exposed and control animals were not statistically
significant. Absolute and relative weights of the major internal organs were likewise unchanged
as a result of exposure to CE, except for the uterine weights of CE-exposed female B6C3F1 mice,
which averaged about 65% of controls. The 35% loss in uterine weight is considered
toxicologically significant. The appearance of acetaldehyde in the urine of exposed animals
reflected species-specific differences. In male mice, the urine acetaldehyde concentration ranged
from 15.4 - 70.1 |imol/L urine CE treated versus 7.6-20.3 in air male controls. In female mice,
there was less of an effect: 11.6-17.0 |imol/L for CE-treated versus 0-18.1 in air-treated controls.
In rats, acetaldehyde concentrations in urine were at or below the limit of detection (2 |imol/L).
These results suggested that at 15,000 ppm CE the P450 oxidative is not normally a major CE
pathway in the rat but is employed in the mouse responses.
In experiments that explored the capacity of liver microsomes from exposed or control
animals to break down CE in vitro with concomitant acetaldehyde formation, the presence of an
NADPH-generating system in the incubation mixture was shown to be essential for oxidative
activity (Fedtke et al., 1994a). Research showed that there were significant increases in treated
versus control rates of NADPH-dependent CE oxidative metabolism in microsomal preparations
from female rat liver (/K0.05), male mice (/K0.05), and female mice (/K0.01) compared to their
unexposed controls. The oxidative rates of the treated mice were about twice those for the
treated rats. These general metabolic responses were complemented by the increased specific
rates of P450IIE1 (p-NP activity) in female rats and both sexes of mice (/K0.01), This indicates
that CE induces its own oxidative metabolism. However, the activities of microsomal P450IA
(EROD activity) and P450IIB (PROD activity) either decreased or remained unchanged in
response to CE. The role of liver microsomal cytochrome P450TTE1 in the metabolism of CE was
7/12/99
14
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
confirmed by the use of the specific P45IIE1 inhibitor, 3-amino-l,2,4-triazole. This inhibitor
decreased the in vitro oxidative metabolism of CE by 75% in the rat and 100% in the mouse, and
correspondingly decreased the microsomal reaction of p-NP (i.e., P450IIE1 enzyme) by 57% in
the rat and 62% in the mouse.
Gargas et al. (1990) has described a physiologically based pharmacokinetic (PBPK) model
in male F344 rats for chlorinated methanes and ethanes that included CE. The metabolism of CE
is characterized kinetically as proceeding via a combination of a saturable and a first-order
process: (1) the first-order component might be due to GSH conjugation (Fedtke et al., 1994b),
and (2) the saturable component might be due to the activity of the cytochrome P450IIE1. The
saturable component would be expected to convert CE initially to 2-chloroethanol or 1-chloro-l-
hydroxy-ethane and then on to 2-chloroaldehyde and acetaldehyde (Figure 1, p. 1). Doubts have
been expressed as to whether oxidation is likely to be of specific etiological significance in the
onset of CE-induced tumors in the uterus of female B6C3F1 mice (Fedtke et al., 1994b). These
authors also speculate that other dehalogenation mechanisms might be involved, on the basis of an
earlier report (Van Dyke and Wineman, 1971). The latter observed 36Cl-chloride formation by
dechlorination of 36Cl-labeled CE in rat hepatic microsomes in either the presence or absence of
NADP, and considered the data to indicate the existence of both enzymatic and nonenzymatic
dechlorination mechanisms for CE.
In conclusion, how the CE metabolic pathways are linked to the observed toxicity
(fetotoxicity and uterine cancers) is unknown. The suggestion of metabolic saturation and implied
nonlinear kinetics suggests further dosimetry work on CE toxicity should be fruitful.
7/12/99
15
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
4. HAZARD IDENTIFICATION
4.1. STUDIES IN HUMANS—EPIDEMIOLOGY, CASE REPORTS, CLINICAL
REPORTS
4.1.1. Oral Exposure
No reports have been identified that describe toxicological effects in humans arising from
oral exposure.
4.1.2. Inhalation Exposure
Short-term exposure to CE in human beings has occurred through the compound's use as
a general anesthetic. However, in recent times, the widespread use of CE has been superseded by
more effective and manageable anesthetics.
A considerable amount of information has accumulated on CE's acute neurological and
other pharmacological effects stemming from its former use as a general anesthetic. For example,
Lawson (1965) pointed to the compound's ability to induce rapid anesthesia at a vapor
concentration of 4% (40,000 ppm). Maintenance of anesthesia with CE alone was considered to
be difficult because of the compound's rapid expiration via the lungs (Section 3.1.2, p. 4).
Accordingly, the suitability of its sole use seems to be limited to short operations or procedures.
Cole (1967) discussed his own extensive use of CE as an anesthetic, in which the CE was used
predominantly mixed with nitrous oxide (N20) or as an intermediate agent between fast-acting
intravenous thiopentone and the slower-acting trichloroethylene. Both authors point to the
compound's capacity to induce respiratory stimulation followed by depression, with attendant
fluctuations in systolic blood pressure and pulse rate (Lawson, 1965; Cole, 1967).
Dobkin and Byles (1971) drew attention to the capacity of CE to form explosive mixtures
in air at concentrations in the effective pharmacological and anesthetic range. Similarly, the blood
concentrations achieved during anesthesia appeared to be too close to those associated with
respiratory failure (20 to 30 mg % versus 40 mg %). The danger of CE overdose in anesthesia is
great. Other potential side effects are the fall in blood pressure, considered to occur through
depression of vasomotor centers, and the peripheral vasodilation of blood vessels (Dobkin and
Byles, 1971). Subsequent vagal depression causes tachycardia, with bradycardia being a sign of
overdose. More moderate effects of CE-induced anesthesia include moderate salivation and, on
recovery, nausea and vomiting.
7/12/99
16
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Reports of the toxicological consequences of exposure to subanesthetic concentrations of
CE center on case studies of persons deliberately sniffing the compound for hallucinogenic
purposes. The amounts of CE involved in such cases are ill-defined. A 28-year-old woman who
had sniffed 200 to 300 mL of CE from her coat sleeve for 4 months showed the following
neurological symptoms: ataxia, tremors, nystagmus (involuntary movements of the eyeball),
scanning dysarthria (speech difficulties), diadochokinesis of the arms (alternate extension and
flexion of each of the arms back and forth, or pronation and supination of the arms), sluggish
lower limb movements, and hallucinations (Hes et al., 1979). Similar symptoms were described
for a 52-year-old man who had a history of abusing solvents, barbiturates, and alcohol over 30
years (Nordin et al., 1988). In the period immediately before hospitalization, he was reported to
have inhaled about 100 mL CE on a daily basis over a 4-month period. Despite suffering a
dramatic fall in blood pressure and a grand mal seizure 12 hr after admission, the patient was able
to recover from all symptoms (short-term memory loss, visual hallucinations, neuropathy of the
lower extremities, plus some clinical chemistry fluctuations) during a 6-week period. The authors
attributed the neurological symptoms to the abuse of CE and a response to subsequent
withdrawal.
4.1.3. Dermal Exposure
CE has been used as a pain-killing spray for such conditions as fibrositis, dysmenorrhoea,
causalgia, and renal colic because it can cause a local rapid lowering of temperature, thereby
acting as a surface analgesic (Lawson, 1965). CE has been used in sports such as American
football to relieve local traumatic pain. In a recent report, Bircher et al. (1994) described a
patient with an allergic contact reaction to CE, with sensitization to dichlorodifluoromethane
(Freon 12). Immunohistochemical analysis identified responses that were consistent with a T-cell-
mediated allergic reaction.
4.2. ACUTE, SUBCHRONIC, AND CANCER BIOASSAYS IN ANIMALS—ORAL AND
INHALATION
4.2.1. Oral Exposure
A 7- or 14-day oral CE palatability study was conducted in F344 rats; it investigated acute
toxicology of aqueous CE (Pottenger et al., 1995). The F344 rats were administered at either 0
or 0.57 g CE/100 g water (570,000 ppm), which is at the practical solubility limit of CE in water
at room temperature. Toxicology parameters investigated were body weights, body weight gain,
7/12/99
17
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
food and water consumptions, gross pathology, selected organ weights, histopathology, clinical
chemistries, and hematology. Rats (5/sex) consuming water at this high dose for 7 days were only
modestly affected (within 15% of controls), showing little effect on palatability. At 14 days water
consumption (10 rats/sex) was decreased to 81% of controls for males and 76% for females thus
showing palatability effects. At 14 days feed consumption and body weight decreases were noted,
but were within 10% of control values. All other parameters were normal. Thus, consumption of
CE at high water concentrations (0.57g/100 g) for 14 days did not produce significant subchronic
toxicological effects. The NOEL for CE dissolved in water may be estimated to be 297 mg/kg
bw/day for male rats and 361 mg/kg bw/day for female rats (Pottenger et al., 1995). No oral
chronic CE study exists.
4.2.2. Inhalation Exposure
4.2.2.1. Landry Inhalation Studies
A report by Landry et al. (1982) described the acute inhalation exposure of six F344
rats/sex/group and two male beagle dogs/group to CE for 6 hr/day, 5 days/week for 2 weeks.
Concentrations applied were 0, 1,590, 3,980, or 9,980 ppm and duration-adjusted values are 0,
800, 1,900, and 4,700 mg/m3. Landry et al. (1982) observed daily clinical signs and measured
serial body weights before, during, and after the 2-week exposure period. Initial and terminal
blood and serum samples were measured for routine hematological and clinical chemistry
parameters in rats. Dogs were measured before and after CE exposure for hematology. All
animals received a complete gross pathological examination at necropsy, with a full range of
tissues and organs processed for histopathological evaluation.
Other than transient behavioral excitement, there were few compound-related effects in
the dogs due to CE at these exposures. Similarly, except for a slight lethargy in the high-dose
rats, there were no clinical signs, body weight changes, gross necropsy, or histopathological
effects due to treatment. Hematology, urinalysis, and clinical chemistry fluctuations were
unremarkable in male rats. There was, however, a statistically significant decrease in BUN in
female rats at the two highest exposures, but it is not interpretable to any toxic effect because of
the lack of any associated histopathological changes in the kidneys. There were increases in the
relative liver weights of male rats at the two highest concentrations. The authors considered the
observed changes to be minor, and to probably represent adaptive rather than toxicological
changes. The subchronic study identified the highest level of exposure (9,890 ppm) as a free-
standing NOAEL, equivalent to a NOAEL(ADJ) for extrarespiratory effects of 4,700 mg/m3.
7/12/99
18
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
In a separate section of the study (Landry et al., 1982), six male F344 rats/group were
subjected to a single 6-hr exposure at nominal CE concentrations of 0, 1,600, 4,000, or 10,000
ppm to analyze the effects of CE on liver NPSH concentration. Five B6C3F1 mice/group were
exposed to 0 and 4,000 ppm CE only. For mice and rats decreased cellular NPSH (GSH) was
observed. Levels of 88% and 89% of control at 4,000 and 10,000 ppm CE were observed for
rats, and 64% of controls were observed at 4,000 ppm CE for mice. Statistical significance was
observed at 4,000 and 10,000 ppm.
In an unusual protocol, Landry et al. (1989) exposed seven B6C3F1 mice/sex/group to
actual concentrations of 0, 250, 1,247, or 4,843 ppm CE for 23 hr/day for 11 days (duration-
adjusted exposure values are 0, 630, 3,200, and 12,200 mg/m3). Animals were observed daily for
clinical signs; on day 12 a blinded neurobehavioral observation battery was conducted. Terminal
body weights were measured, then blood samples were collected to measure hematological and
clinical chemistry parameters. At sacrifice, animals were subjected to a gross pathological
examination. Slides of sections of brain, heart, liver, kidney, thymus, and testes from the control
and high-dose groups were examined for histopathological lesions.
In general, for doses 250-4,843 ppm there were no clinical signs of exposure,
neurobehavioral manifestations, body weight changes, clinical chemistry, or hematological
responses at any of the CE concentrations tested. Apparent compound-related effects were
limited to increases in the relative liver weights in both sexes exposed at the highest CE
concentration (4,843 ppm). This change has been associated with an increase in the size of the
liver noted in some animals in this group, and with an increase in the incidence of hepatocellular
vacuolization evident in 4/7 mice/sex exposed to this concentration. The authors did not consider
any of the observed histopathological or relative weight changes in the liver to be correlated to
CE. Accordingly, the Landry study defined a free-standing NOAEL of 4,843 ppm. It is perhaps
notable that exposures of 250 ppm and 1,247 ppm do not show any effects, neurological or
clinical.
4.2.2.2. Principal Study Performed by the U.S. National Toxicology Program
The most comprehensive study on the inhalation toxicology of CE in mice and rats is that
sponsored by the U.S. National Toxicology Program (NTP, 1989a). Groups of F344 rats and
B6C3F1 mice of both sexes were exposed to CE vapor (whole body) for periods of 2 weeks
(acute), 13 weeks (subchronic), or 2 years (chronic). Other acute studies are described that
usually explored the anesthetic properties of CE.
7/12/99
19
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
4.2.2.2.1. NTP acute study. A single exposure experiment (19,000 ppm for 4 hr) was part of the
range-finding exercise that resulted in concentrations of 0 and 15,000 ppm being chosen for the
chronic portion of the study. In this acute study, all rats and mice (5/sex/group) survived the
single exposure for 4 hr at 19,000 ppm CE with no concurrent or subsequent clinical signs.
Similarly, those animals (5/sex/group) exposed for 10 days at 19,000 ppm CE and held 2 weeks
survived for the duration of the study. Among the rats, there were no compound-related effects
of weight gain, whereas for the mice, body weights of exposed animals were greater than those of
controls. Overall, no mice or rats in this portion of the study displayed clinical signs, and gross
necropsy and histopathological findings indicated an absence of CE-related effects. Hence, in the
NTP study, CE produced no apparent acute effects at a high dose of 19,000 ppm.
4.2.2.2.2. Subchronic study. The subchronic portion of the NTP study featured the
administration of 0, 2,500, 5,000, 10,000, or 19,000 ppm CE to 10 F344 rats and B6C3F1
mice/sex/group, 6 hr/day, 5 days/week for 13 weeks (NTP, 1989a). Duration-adjusted exposures
in units of mg/m3 were 0, 1,180, 2,360, 4,710, and 8,950 mg/m3, respectively. No exposure-
related clinical signs or gross or histopathological lesions were evident in either rats or mice in this
study.
Possible compound-related consequences of exposure were limited to comparatively
minor fluctuations in body and liver weights. Thus, for both males and females in 13 weeks, slight
decreases in body weights were noted at HDT, i.e., 19,000 ppm CE. Statistically significant
increases in relative liver organ weights were observed in male rats (+14%) and female mice
(+18%) exposed at the HDT (8,950 mg/m3), however, male mice exposed to 4,710 mg/m3 CE
displayed a significant decrease in liver weight. Based on the increases in relative liver weight in
male rats and female mice, this study identified a NOAEL(HEC) of 4,710 mg/m3 and a
LOAEL(HEC) of 8,950 mg/m3 (HDT). It is notable that 2,500 ppm produces no significant
effect. Benchmark concentration modeling was not conducted on liver weight because it was not
excessive and monotonic increases with concentration were not observed.
4.2.2.2.3. Chronic study. Male and female F344 rats and B6C3F1 mice (50/sex/group) were
exposed to 0 or 15,000 ppm CE (39,570 mg/m3) for 6 hr/day, 5 days/week for 103 weeks (rats)
or 100 weeks (mice). The time-adjusted dosage is 39,570 mg/m3 x 6/24 x 5/7 = 7,070 mg/m3.
The particular concentration of 15,000 ppm was chosen and was based on an apparent lack of
toxicity in the subchronic portion of the study, and on concerns for potential flammability and
explosion at higher concentrations. Clinical signs were observed daily, while body weights were
recorded weekly for the first 12 weeks, then monthly. A complete histological examination was
carried out on all animals dying prematurely and on those animals surviving to term.
7/12/99
20
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
4.2.2.2.3.1. F344 Rat Toxicological Results in the NTP Study
Male F344 rat survivals in control (16/50) and CE-exposed (8/50) groups were low after
103 weeks, with no statistically significant difference between control and treated group. The
NTP authors suggested that an unusually high incidence of mononuclear cell leukemia in both
groups likely contributed to poor male rat survivals (NTP, 1989a). In contrast, female rats
showed good survival in control (31/50) and CE-treated groups (22/50) at study termination; and
there was no statistical difference between the groups. A slight decrease in mean body weight gain
(4-8%) in the male rats compared to controls was observed after wk. 33 of chronic exposure, and
the mean body weights of female rats were 5-13% lower than controls from wk. 11 to the end of
the study (NTP, 1989a). At termination, the mean body weight of exposed female rats was
reduced byl0% compared with concurrent controls. No remarkable clinical signs were observed
in the exposed animals, and no CE-induced nonneoplastic lesions were observed even at this high
dose. This level of weight loss is not considered to to be a critical toxicological effect.
A number of uncommon skin tumor types were observed in exposed male F344 rats
(Table 7). The total tumor response in male F344 rat skin seems to show that skin and certain
skin appendages are displaying a cancer response. Because skin under the fur is exposed to CE in
the inhalation chamber during the 102 weeks, there is some dermal exposure.
When compared with the concurrent control incidence, that is, 5/49 (10%) versus 8/46
(17%>), the male rat malignant whole skin response is not statistically increased (p=0.23). The
first skin tumor, a subcutaneous fibroma, occurred at 79 weeks in the treated group. Moreover,
the rates are not significantly increased at 15,000 ppm CE when adjusted for animals dying before
the first skin tumor. The comparison, in this case, is 5/42 (12%>) versus 8/42 (19%>), p=0.27.
When the male rat skin tumors of the treated group are compared with those of the historical
inhalation controls from the same testing laboratory, there is a statistically significant increase in
epithelial cancers: 2/300 (0.7%) versus 8/46 (17.4%>),/>=2 x 10"6. Similarly, when NTP controls
from noninhalation historical experiments are compared with the treated group (28/1,936 [1.4%]
versus 8/46 [17.4%],/>=8 x 10"5), there is also a statistically significant increase in epithelial skin
tumors.
Historical incidence rates can be characterized. For example, tumor incidences may be
subjectively ranked: (1) incidence rates <0.5% are rare, (2) incidences occurring >0.5% but <2%
may be considered uncommon, and (3) incidences >2% are generally common to aging test
rodents. These definitions are operational, not absolute, and they represent expert judgment. In
this bioassay, the historical malignant skin tumor incidence is 0.7%, and NTP incidence is 1.4%
where both are designated as uncommon tumor incidences. On the other hand, the observed
control skin incidence is 10% (5/49) (Table 7). Comparing either the observed or historical
control incidences to the treated group incidences leads to different conclusions: there is a
7/12/99
21
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
statistically significant increase when historical skin controls are considered, but not when the
study concurrent control is considered as the reference control.
In the female rats, brain astrocytomas occurred at a low incidence of 3/50 (6%) (Table 7).
In analyzing the significance of this low-incidence brain tumor, it is known that astrocytomas are
not common in most strains of rat or in humans. So, even low incidences could be a sign of
carcinogenicity. There is extra concern when astrocytomas do occur because such a tumor type
in the brain has fatal implications in rodents and humans. When compared statistically with the
concurrent control (0/50 [0%] versus 3/50 [6%]), the response yields statistical insignificance
(p=0.12), which suggests that there may be no effect. The same may be stated when the adjusted
rates are examined by subtracting the number of animals dying before the first astrocytoma
appears (52 weeks): 0/46 versus 3/49,p = 0.12.
When rare tumors occur, the tumor rates require special consideration. Uncommon or
rare tumor incidences may not indicate a statistical increase when compared with their respective
concurrent control incidences. This is because the number of trials (i.e., the number at risk in the
control and treated groups) is small, ~50/sex/group, and a larger number of animals (in this case,
at the 95% level of confidence, «150/sex/group) is needed to statistically score a rare tumorigenic
event. Accordingly, when the observed incidence (3/50) is compared with historical pooled
control incidence (1/297) from the same testing laboratory (Battelle Pacific Northwest
Laboratories), the statistically significant increase in astrocytomas isp=0.01 (Table 7). Note that
the larger denominator affects the statistical inference in the case of uncommon or rare tumors.
Similarly, when the observed 3/50 astrocytomas in female F344/N rats are compared with the
incidence of all experimentally discovered astrocytomas inNTP studies (23/1,969), the statistical
significance isp=0.02 (Table 7).
The 3/50 (6%) astrocytoma response in female F344/N rats is statistically significant when
compared with historical controls, but not when compared with the concurrent controls. The
observed and historical control incidences present different conclusions; that is, a statistically
significant increase in astrocytomas is seen when historical controls are considered, but not when
the study concurrent control is considered.
Further analysis shows, however, that Battelle Pacific Northwest Laboratories had a
singular prior incidence of 3/50 (6%) astrocytomas in a female concurrent control group of
F344/N rats. This singular control brain tumor incidence happens to be commensurate with the
brain response in the 15,000 ppm CE group (Table 7). Thus, if a past concurrent control
incidence can reach as high as 3/50 (6%), the apparent statistical significance of the dosed group
7/12/99
22
DRAFT—DO NOT CITE OR QUOTE
-------
Tab
e 7. Tumors of F344/N rats at 2
years
Sex
Controls
15,000 ppm chloroethane
Estimate of
p value3
Males
Keratoacanthoma = 4/49 (8%)
Fibroma = 1/49 (2%)
Total = 5/49 (10%)
Basal cell carcinomas = 3/46 (7%)
Keratoacanthoma = 2/46 (4%)
Squamous cell carcinoma = 1/46
(2%)
Trichoepithelioma = 1/46 (2%)
Lip, squamous cell carcinoma =
1/46 (2%)
Total = 8/46 (17%)
0.23
Adjusted to first appearance of
tumor (79 weeks)
(42 males)
Tumor incidence = 5/42 (12%)
Adjusted to first appearance of
tumor (79 weeks in treated group)
(42 males)
Tumor incidence = 8/42 (19%)
0.27
Skin
Historical controls = 2/300
(inhalation) (0.7%)
See above, 8/46
2.0 x 10"6b
Skin
Historical controls = 30/1,936
(noninhalation) (2%)
See above, 8/46
1.3 x 10"6b
Females
Astrocytomas = none in controls
astrocytomas = 3/50 (6%)
0.12
Adjusted to animals on test at 0
weeks (46 females)
Tumor incidence = 0/50 (0%)
Adjusted to first appearance of
tumor at 52 weeks (49 females)
tumor incidence = 3/49 (6.1%)
0.12
Historical astrocytoma controls =
1/297 (inhalation studies) (0.3%)
See above, 3/50
0.01 b
Historical astrocytoma controls =
23/1,969 (all studies) (1.1%)
See above, 3/50
0.02 b
3 The p value is the likelihood (probability) that the assumption of a positive cancer effect is in error.
Usually p<0.05 is taken as a reasonably significant level of certainty to continue to assume there is a
positive cancer effect.
b Designates statistical significance in a Fischer's exact test comparison. Data taken from NTP report no.
346 (NTP, 1989a).
7/12/99
23
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
response—also an incidence of 3/50 (6%)—becomes less important. Moreover, in past NTP
studies, the average astrocytoma incidence is 0.9% (18/1,969) and the range is 0% to 6% in
female F344/N rats. Here, too, it is observed that an incidence level of astrocytoma cancers as
high as 6% may be observed in concurrent controls.
It is determined, then, that this female rat astrocytoma effect may be real, but if so is
marginal. Sensitivity analysis indicates that only one more rat with an astrocytoma would have
shifted the concern to a significant response. Therefore, the female rat brain response is
designated as equivocal evidence for CE carcinogenicity.
4.2.2.2.3.2. B6C3F1 Mouse Toxicological Results in the NTP Study.
In male B6C3F1 mice, survivals were significantly reduced compared to controls after
wk.42. The same was true for female mice after week 81. Thus, because of the low survival
rates the NTP mouse study was terminated earlier than protocol called for. The mice were
terminated at the 100th week, at which time survivals were 28/50 control versus 11/50 CE in
exposed males and 32/50 control versus 2/50 CE in exposed females. The high mortality in the
male mice was attributed to a greater than normal incidence of nonneoplastic urogenital lesions
observed in both the control and exposed males, although the exposed mice were more severely
affected. In female mice, the majority died as a result of CE-induced carcinomas of the uterus,
endometrium, myometrium, and complications of metastasis, as further discussed in Section
4.2.2.2.3.4, p. 23.
Female mice exhibited a characteristic hyperactivity during the daily periods of exposure,
a transient response to treatment because activity returned to normal at the end of each exposure
period (NTP, 1989a). There was no effect on body weight in either sex, and no other exposure-
related clinical signs or nonneoplastic lesions were observed. Based on this absence of
noncarcinogenic toxic effects, the single concentration tested (7,070 mg/m3) was a NOAEL for
female mice. Benchmark concentration modeling could not be conducted because only one
exposure level was tested.
Because of poor survivals in mice, the murine portion of the study was terminated at week
100. Many of the male mice died prematurely from urogenital infections, thereby reducing the
power of the male group results to detect late-developing neoplasms (NTP, 1989a). Survival
until termination was 28/50 in male controls and 11/50 in 15,000 ppm males. Notwithstanding
male mouse results, there were no significant cancer increases—except possibly an increase in
lung adenomas and/or carcinomas. Lung cancer incidence was 5/50 in controls versus 10/48 at
15,000 ppm (p = 0.11). The poor survivals in the male B6C3F1 mice force the conclusion that
the male mouse is inadequate to determine carcinogenicity.
7/12/99
24
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
1
2
3
4
5
6
7
8
9
Survivals in female mice were 32/50 in controls and 2/50 in 15,000 ppm. The study
diagnosis was that female mice died early because of aggressive carcinogenicity (NTP, 1989a).
Life-shortening is a primary element in assessing carcinogenicity. Treated female mice had a high
incidence of primary tumors in the uterine endometrium (Table 8). High incidence is another
primary element in assessing carcinogenicity. These lesions occurred in almost all females tested:
43/50 of CE-exposed female B6C3F1 mice
Table 8. Incidence of tumors in female B6C3F1 mice after exposure to CE for 2
years
Effect
Incidence of hy
jerplasia/tumors
Pa
Controls
CE-exposed
Uterine hyperplasia and
cytstic hyperplasia
41/49
6/50
3 x 10"5 b
Uterine carcinoma
1/49
43/50
<10"8
Uterine carcinoma0
1/46
43/48
<10"8
Uterine lymphomas
Systemic lymphomas
1/49
5/49
7/50
10/50
0.03
0.14
Hepatic combined adenomas
and carcinomas
3/49
8/48
0.025
aAs determined using Fischer's exact test.
bNegative correlation biologically. Uterine lining in aging females normally show hyperplasia, but CE exposure
demonstrates an obliteration of the normal hyperplasia due to the dispersed metastatic carcinomas.
Corrected for time to first tumor at 67 weeks which as a uterine tumor.
Source: NTP (1989a).
versus 0/49 in controls. These endometrial tumors showed a remarkable capacity for
metastasizing. Aggressive metastasis is another primary element of carcinogenesis evolving into
the malignant state of cancer. Secondary cancer sites (16 total) included, out of 50 starting
female mice, lung (23), ovary (22), lymph nodes (18), kidney (8), adrenal gland (8), pancreas (7),
urinary bladder (7), mesentery (7), spleen (5), heart (4), colon (2), and stomach, gall bladder,
small intestine, ureter, and liver (1 each).
Other carcinogenic effects of treatment included increased incidences of combined
adenomas and carcinomas in the livers of female mice (8/48 versus 3/49 in controls; p=0.025).
There were also increases in hematopoietic cancer involvement with CE treatment, including
7/12/99
25
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
increases in a number of white cell types in bone marrow, lymph nodes, spleen, and thymus.
Though these effects were difficult to differentiate from the metastatic impacts of the primary
carcinogenic effect, they lend support to the powerful carcinogenic effects of CE in female
B6C3F1 mice. It has been concluded that there is clear evidence of the carcinogenicity of CE in
female B6C3F1 mice (NTP, 1989a; Holder, 1998).
4.3. REPRODUCTIVE/DEVELOPMENTAL STUDIES—ORAL AND INHALATION
4.3.1. Oral Exposure
No reports of studies were identified that addressed the reproductive/developmental
toxicity of CE when administered via the oral route.
4.3.2. Inhalation Exposure
4.3.2.1. Principal Study
Scortichini et al. (1986) reported the findings of an inhalation reproductive
toxicity/teratological study on 120 pregnant female CF-1 mice exposed to nominal exposure
concentrations of 0, 500, 1,500, or 5,000 ppm CE for 6 hr/day on gestation days (GD) 6-15.
This is 10 days of CE exposure for the dams. All animals were sacrificed on GD 18. Mean
concentrations were found to be 0, 491 + 37 ppm, 1,504 + 84 ppm, or 4,946 + 159 ppm which
convert to 0, 1,300, 4,000, and 13,000 mg/m3, respectively. These values were not duration
adjusted, in accordance with current EPA practice. As indicators of the compound's potential
maternal toxicity, the dams were observed for clinical signs: body weights and food and water
consumption were measured every 3 days, and at necropsy, dam liver weights and gravid uterine
weights were recorded. As indicators of possible developmental toxicities and teratological
effects of CE, fetal observations included the number and position of fetuses in utero, the number
of live and dead fetuses and the number of resorption sites, the weight and sex of each fetus, and
the incidence of any gross external alterations or cardiac abnormalities. Half of each fetal litter
were necropsied to look for visceral abnormalities and skeletal alterations. Serial sections of the
head were made in a subset of fetuses.
Observations showed no maternal toxicity from CE inhalation exposure as measured by
clinical signs, food and water consumption, body weight, and liver weight. Nor were there any
CE-related changes in reproductive performance: pregnancy rate, resorption rate, litter size, fetal
7/12/99
26
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
sex ratio, or fetal body weights. By contrast, in examining the possible teratological effects of
CE, a number of effects appeared sporadically. Cervical ribs appeared in the exposed fetuses
1/257, 1/299, 6/311, and 4/242 (p trend = 0.13). On a per litter basis (2/22, 1/25, 5/26, 4/22), the
response of cervical ribs was not statistically significant (p = 0.31). Exposures to higher levels of
CE might have produced significant rib malformation. There was, however, an increase in the
incidence of delayed fetal foramina closure (DFFC) of the CF-1 mouse skull bones (Scortichini et
al., 1986). This developmental delay, viewed at at GD 18, is a retardation of a small frontal area
of ossification of the skull. This is not to imply that the foramina will not ultimately close in
exposed CF-1 mice. Thus, this is a fetotoxic effect—not a teras—and possibly represents a
CE-induced skeletal variation. In
Table 9, the data show that at
4,946 ppm that 5 fetuses (4%) were
affected from a total of 5 litters (23%),
compared with 1 fetus in 1 litter for
each of the lower exposure groups,
including the control group. The
average historical control incidence of
this DFFC variance is 0.2% with a
range of 0-1.2% in CF-1 mice (Figure
5). At the HDT, the
incidence of fetotoxicity of 5/116
fetuses (4.39%>) falls outside this range
. , , _ „ Figure 5. Occurrence of delayed foramina closure in skulls
of historical controls Comparison of ofC|., mice Dams were treatetl with 0 49|> 1>504i or 4946
the HDT incidence to the upper ppm CE by inhalation for 6 hrs/day on days 6-15 of gestation.
historical control incidence (3/245) Similar statlsltcal results were obtained when incidence on a
litter basis was considered rather than on an individual basis as
yields/? = 0.074; the marginality of this shown in this flgure
HDT effect is also indicated by the
pairwise comparison to control incidence (1/126) by the Fischer's exact test,/? = 0.077 (Figure 5).
However, supporting the concept of DFFC as a CE-related effect was the statistically significant
trend in skull foramina (/K0.05), using various nonparametric trend tests. The apparent effect is
weak in intensity. This study should hve been followed by a similar study in the 2,000-12,000
ppm CE range to see if there was a DFFC dose response.
Although there is insufficient evidence to unequivocally resolve the dose-response issue,
this is the lowest-dose critical effect suggested by CE exposure of all noncancer studies reviewed
in this document. Therefore, the middle dose or subthreshold exposure concentration of 3,970
Scortichini Study on Foramina Closure (DOW Study)
-
p
p(pa
(trend) =0.048
irwise Fisher's)
>ound Limit on Hi
=0.077
itorical Control
-
Mean Value fc
r Historical Contrc
Is in CF-1 mice
1 1
1 1
i i
i i
i i i
0 1000 2000 3000 4000 5000
Chloroethane Exposure (ppm)
7/12/99
27
DRAFT-DO NOT CITE OR QUOTE
-------
Table 9. Chloroethane inhalation teratology in CF-1 mice: incidence of fetal
alterations among litters of mice
Chloroethane concentration (ppm)
0
500
1,500
5,000
Number of fetuses (number of litters) examined
External examination
258 (22)
299 (25)
311 (26)
242 (22)
Soft-tissue examination
132 (22)
156 (25)
164 (26)
126 (22)
Skeletal examination
257 (22)
299 (25)
311 (26)
242 (22)
Bones of the skull
126 (22)
142 (24)
147 (25)
116 (22)
Percent affected (numbers affected)
External observations
Cleft palate
Fetuses
0.4(1)
1(3)
0.3 (1)
0.4(1)
Litters
5 (1)
12 (3)
4 (1)
5 (1)
Exencephaly
F
0
0
1 (2)
0
L
0
0
8 (2)
0
Micrognathia
F
0
0
0
0.4(1)
L
0
0
0
5 (1)
Chloroethane concentration (ppm)
0
500
1,500
5,000
Microphthalmi
a
F
0.4(1)
0
0
0
L
5 (1)
0
0
0
Soft tissue observations
Dilated renal
pelvis and
ureter
F
0
1(1)
0
0
L
0
4(1)
0
0
Pale spleen
F
0
0
1 (1)
0
L
0
0
4 (1)
0
Pale foci on
liver
F
1 (1)
0
0
0
L
5 (1)
0
0
0
7/12/99
28
DRAFT—DO NOT CITE OR QUOTE
-------
Table 9. Chloroethane inhalation teratology in CF-1 mice: incidence of fetal
alterations among litters of mice (continued)
Chloroethane concentration (ppm)
0
500
1,500
5,000
Dilated
ventricles
of brain
F
0
1(1)
0
0
L
0
5(1)
0
0
Intraventricular
hemorrhage
F
1(1)
0
0
0
L
5(1)
0
0
0
Skeletal observations
Skull -delaved
ossification
F
2(3)
6(9)
4(6)
2(2)
L
9(2)
29 (7)
20 (5)
9(2)
Foramina
F
1(1)
1(1)
1(1)
4 (5)a
L
5(1)
4(1)
4(1)
23 (5)
Irregular
pattern
of ossification
F
0
0
0
1(1)
L
0
0
0
5(1)
Vertebrae
Delayed
ossification
F
1(3)
0.3 (1)
1(2)
0
L
14 (3)
4 (1)
8(2)
0
Centra delayed
ossification
F
0
0.3 (1)
0.3 (1)
0
L
0
4 (1)
4 (1)
0
Atlas forked
F
1(2)
1 (2)
1(2)
1(2)
L
9(2)
8 (2)
8(2)
5(1)
Fused
F
0
0
0.3 (1)
0
7/12/99
29
DRAFT—DO NOT CITE OR QUOTE
-------
Table 9. Chloroethane inhalation teratology in CF-1 mice: incidence of fetal
alterations among litters of mice (continued)
Chloroethane concentration (ppm)
0
500
1,500
5,000
L
0
0
4(1)
0
Ribs
Delayed
ossification
F
0
0.3 (1)
0
0
L
0
4 (1)
0
0
Forked
F
0
0.3 (1)
0
0
L
0
4 (1)
0
0
Fused
F
0
0.3 (1)
0
0
L
0
4 (1)
0
0
Cervical
F
1(2)
0.3 (1)
2(6)
2(4)
L
9(2)
4 (1)
19(5)
18(4)
Sternebrae
Delayed
ossification
F
4(11)
6(18)
4(12)
2 (5)
L
27 (6)
48 (12)
23 (6)
14 (3)
Fused
F
5(12)
6(18)
7 (23)
5(13)
L
46 (10)
28 (7)
54 (14)
36 (8)
Staggered
F
0
0.3 (1)
0.3 (1)
0.4(1)
L
0
4 (1)
4(1)
5(1)
Irregular
pattern
of ossification
F
1(2)
0
0
0
L
9(2)
0
0
0
Misshapen
F
0.4(1)
0
0
0
L
5(1)
0
0
0
"p<0.()5 using a censored Wilcoxon test.
Source: Scortichini et al. (1986).
7/12/99 30 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
mg/m3 (1,504 ppm) is characterized as a NOAEL for the fetotoxic foramina effect, with the HDT
= 13,000 mg/m3 (4,946 ppm), which is the LOAEL.
4.3.2.2. Supporting Reproductive or Teratological Studies
In an earlier reproductive/teratological study 8-10 pregnant female CF-1 mice were
exposed to 0, 5,000, 10,000, or 15,000 ppm CE for 6 hr/day on GDs 6-15 (Dow Chemical Co.,
1985). Among the responses investigated were the number of litters, number of implantation
sites/dam, number of live fetuses/litter, resorptions/litter, percentage implantations resorbed, and
the ratio of resorptions to litters with resorptions. Most exposed mice displayed stereotypical
behavior characterized by repetitive running, and significant decreases in body weight on GD 16
and decreased body weight gains on GDs 10-16 were observed at all CE doses. However, there
were no compound-related reproductive, developmental, or teratological effects in any treatment
group.
Breslin et al. (1988) reported an estrous cycling study in B6C3F1 mice. A vaginal lavage
technique measured estrous cyclicity, following CE exposure of two groups of 10 female mice to
0 or 15,000 ppm for a minimum of 14 consecutive days (three estrous cycles). Before exposure,
the groups had been acclimated to the inhalation chambers until their estrous cycles stabilized to a
regular estrus periodicity. All animals were also monitored for clinical signs, body weight
changes, and reproductive pathology and histopathology at termination. No effects on behavior,
gross pathology, or histopathology were observed in the 15,000 ppm group, but mean body
weight gain was significantly increased (/K0.05), The mean length of the estrous cycle in exposed
mice was 5.6 days, significantly longer than the pre-exposure duration for the same group (5.0
days) and for the corresponding control group (4.5 days). The authors noted that, in some
animals, the estrous phase was lengthened, while in others it was the diestrous phase that was
affected. Consequently, they attributed the observed effects to a generalized stress reaction rather
than to any specific reproductive CE effect, but a direct exposure-related effect of CE on
neuroendocrine function cannot be ruled out. Thus, assuming that CE does have the ability to
disrupt the estrous cycle of mice, these data would point to a duration-adjusted free-standing
LOAEL of «7,071 mg/m3 = LOAEL(HEC).
Bucher et al. (1995) sought to explain why CE induces a lengthened estrous cycle in
B6C3F1 mice. Because CE (Section 4.2.2.1.3.4) and bromoethane (BE) (Section 4.7.1) both
cause murine uterine tumors, an uncommon B6C3F1 tumor, it was decided to look for a
hormonal basis for the chemical carcinogenesis. Serum levels of estradiol and progesterone were
measured in haloethane-exposed and control female mice. Female mice (30/group) were exposed
to 15,000 ppm CE, 400 ppm BE, or filtered air as controls for 6 hr/day over a duration of 21 days
7/12/99
31
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
to monitor hormone levels. Vaginal smears were determined and daily cell cytology was done.
Body weights of animals in the study were recorded once a week, and, at termination, blood
samples were obtained via cardiac puncture for hormone analysis. At necropsy, the liver, uterus,
pituitary, adrenal glands, and ovaries were removed, the liver and uterus were weighed, and the
organs were fixed for histopathological examination.
In line with the data reported by Breslin et al. (1988), Bucher et al. (1995) observed a
slight but statistically significant increase (+ 0.4 days) in the mean duration of the estrous cycle in
mice exposed to CE. However, there were no consistent concomitant hormone treatment-related
changes in serum estradiol or progesterone. Likewise, there were no CE-induced clinical signs,
body weight gains, or changes in uterine weight. The latter observation is in contrast to that of
Fedtke et al. (1994a), who reported an overall 35% reduction in uterine weight as a result of a
similar level of CE exposure in this animal model. Taken with the absence of any consistent
compound-related effects on the duration of individual estrous stages, the lack of any changes in
the serum concentrations of estradiol or progesterone due to CE or BE exposure suggests that the
minimal alteration of the estrous cycle described by Breslin et al. (1988) and Bucher et al. (1995)
is unlikely to represent a major mechanism by which the haloethanes perturb uterine metabolism
to cause cancer.
4.4. OTHER TOXICITY STUDIES
4.4.1. Acute Toxicity Studies
4.4.1.1. Neurotoxicity
In many older acute or subchronic inhalation experiments, narcotic or anesthetic doses of
CE gas were administered and the doses were often uncertain. In such experiments, use was
often made of a saturated substrate (e.g., cotton) that generates a high, but unknown, flow to the
nose. In one such CE inhalation study, the rat cerebral cortex demonstrated decreased
respiration, but the thalamus and white matter did not appear affected upon gross examination
(Seller, 1938). Rats, mice, and rabbits were each anesthetized with CE; acetylcholine was then
extracted from the respective frozen brain tissues. Each showed increased acetylcholine levels as
a result of CE anesthesia (Sayers et al., 1929). Mice were administered 30,000 or 60,000 ppm
CE for up to 1 hr via inhalation (Neal et al., 1964). After 25 minutes, 17% of the mice in the
60,000 ppm group had become anesthetized, but no anesthesia occurred in the 30,000 ppm group.
7/12/99
32
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
It was generally found that > 35,000 ppm CE causes primary CNS and circulatory effects
(Lazarew, 1929; Henderson, 1930).
CE anesthesia for 60 min caused a decreased sedimentation rate of RBCs from rabbits,
followed by a period of accelerated sedimentation rate, reaching its maximum 3 hr after anesthesia
was initiated and then normalizing in 12-14 hr (Hinko, 1934). Rats were anesthetized with
54,000 ppm CE for 5 min, and subsequently 02 consumption and C02 excretion were decreased
significantly and the body temperature fell by 2.5 °C (Hattori, 1957). Rats were anesthetized for
2 hr with CE, after which occurred a disappearance of liver glycogen, a decrease in acid
phosphatase levels, and increases in alkaline phosphatase and succinic dehydrogenase levels
(Heller et al., 1966). A two-hr inhalation LC50 value of 60,632 ppm for rats and mice has been
reported (Troshina, 1966).
Guinea pigs were exposed to CE via inhalation at concentrations ranging from 10,000 to
240,000 ppm for times ranging from 5 to 810 min (Sayers et al., 1929). An unsteady gait
appeared after 25 min at 20,000 ppm. Some deaths occurred at exposures of 15,300 ppm and
higher. At exposures >20,000 ppm, pulmonary congestion, hemorrhage, and edema were
observed in gross pathology. At 87,000 ppm for 130 minutes, violent shaking occurred in one
pig, and after 270 minutes, rales were heard in several guinea pigs. At 127,000, 142,000, and
153,000 ppm for 1 minute, there was complete loss of equilibrium, a running movement, and
scratching. Abdominal walls seemed distended and convulsion of the intestines was observed.
After 15 to 20 min, struggling became less violent; respiration became shallow, rapid, and
convulsive; and death occurred in 30 to 40 min. At 232,000 and 240,000 ppm, there was loss of
equilibrium in 30-60 seconds, and in 5 min. animals lost consciousness.
The effects of CE on feline brain blood flow were studied in the cortex and the medulla
oblongata (Tokita, 1953). CE gas increased feline brain blood flow. The avoidance flexion reflex
was tested, during administration of CE, on a super-maximal single electrical stimulation to the
hind limb of intact rats (Hiraiwa, 1952). Changes in the flexion reflex curves were observed.
Dogs and rats were examined for neurological behavior at 0, 1,600, 4,000, and 10,000
ppm CE inhaled 6 hr/day, 4 days/wk, for 2 wk (Landry et al., 1982). Dog examinations were
performed 2 days prior to exposure and at exposure end. Dogs were examined for gait, posture,
mental status, cranial nerve reflexes, postural reactions, spinal cord reflexes, muscle tone, and pain
perception. An ophthalmoscopic exam was also performed. No reactions were seen in dogs
except for some hyperactivity. Only hyperactivity in exposed rats was observed by Landry et al.
(1982). Although these were acute observations, longer exposure to CE than 2 weeks may have
produced different results.
The well-described capacity of CE to induce anesthesia in human beings (Lawson, 1965;
Cole, 1967) and case reports of the abuse of the compound for hallucinogenic purposes at
7/12/99
33
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
subanesthetic concentrations (Hes et al., 1979; Nordin et al., 1988) may be consistent with
evidence of the neurotoxicity of CE that has accumulated from animal studies. For example,
when sublethal CE concentrations (0, 5,000, 10,000, or 15,000 ppm CE for 6 hr/day on GDs 6-
15) were explored in neurotoxicity experiments, increased physical activity in female mice was
observed at all doses (Dow Chemical Co., 1985). Most exposed mice displayed stereotypical
behavior characterized by repetitive running. Similarly, hyperactivity was reported in female
B6C3F1 mice exposed to the high dose of 15,000 ppm CE for 6 hr/day, 5 days/week for 2 years
(NTP, 1989a). Hyperactivity was also observed by Pottenger et al. (1992); the observed
depletion of GSH pools (using buthionine sulfoximine) blocked hyperactivity, thus showing it was
GSH mediated (Pottenger et al., 1992). The hyperactivity was not apparent in mice exposed to
the similar concentrations in the subchronic portion (13 weeks) of the NTP study, raising doubt as
to whether the response was compound-related or a more generalized or uncontrolled response to
stress.
4.4.1.2. Immunotoxicity
There have been a number of studies in animals that do not produce immunotoxicological
results. For example, 438 ppm for 10 days of inhalation was negative (Schmidt et al., 1972).
Likewise, 10,000 ppm for 2 weeks' murine and canine exposures was immunotoxicologically
negative (Landry et al., 1982). Moreover, the 1988 NTP studies of 19,000 ppm for 2 weeks or
13 weeks, and 15,000 ppm for 104 weeks were all negative immunotoxicologically. There is one
study where 5,305 ppm CE for 8.5 weeks, as well as a lower exposure of 216 ppm for 24 weeks,
caused reduced leukocytic phagocytes in rats (Troshina, 1966). This study did not report
sufficient experimental details and has not been validated.
4.4.1.3. Cardiac Sensitization
CE-anesthetized dogs showed increased cardiac sensitivity to epinephrine, as
demonstrated by ventricular tachycardia (Morris et al., 1953). In CE-anesthetized dogs, cardiac
irregularities observed are asystole, ventricular standstill, and ventricular tachycardia (Haid et al.,
1954). Dogs were either anesthetized with CE only or in combination with atropine, an
anticholinergic drug (Bush et al., 1952). They observed electrocardiographic changes that
suggested two mechanistic CE effects on the heart: (1) a direct depression of cardiac tissues and
(2) a cardiac inhibition resulting from vagus nerve stimulation. Beagle dogs were exposed while
conscious to high concentrations of CE 5 minutes after an intravenous injection of 0.008 mg/kg
epinephrine. The treatment resulted in an exacerbated incidence of epinephrine-induced
7/12/99
34
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
arrhythmias, marked by ventricular fibrillation and tachycardia. Exposures to 40,000-50,000 ppm
CE were not well tolerated, as the dogs entered the excitatory stages of anesthesia. These
investigators concluded that the dogs were susceptible to cardiac sensitization that was induced
by CE. CE-induced cardiac sensitivity has not been thoroughly tested in the dog with a
chemically-related series of chlorinated solvents. CE has been generally classified as a weak
cardiac sensitizer in the dog, at least at high doses (Reinhardt et al., 1971).
4.4.1.4. Dermal Effects
Eye irritation has been reported by exposing human volunteers to short exposures of
40,000 ppm CE but not to 20,000 ppm (Sayers et al., 1929). The eye may be the only surface
tissue that responds adversely to CE. Histopathological effects of the dermis (canine and murine,
10,000 ppm or 19,000 ppm for 2 weeks) were negative (Landry et al., 1982; NTP, 1989a). It has
been noted: 1) after 48 hr inhalation exposure to 15,000 ppm CE (saturating metabolic
conditions), the B6C3F1 mice show more metabolism than F344 rats, and 2) the primary target
rat tissues appeared to be ovary, adrenals, and skin (Dow, 1992). If there are acute dermal
toxicological effects in the rat, these may build up during chronic exposure. NTP F344 rats
treated with CE by inhalation for 2 years did have more skin tumors: basal cell carcinomas = 3/46
(7%), keratoacanthoma = 2/46 (4%), squamous cell carcinoma = 1/46 (2%), trichoepithelioma =
1/46 (2%), lip and squamous cell carcinoma = 1/46 (2%) (NTP, 1989a). This total is 8 skin
cancers in 46 rats (17%) in the CE inhalation group versus 5 in 42 (12%) in concurrent control
versus 2/300 in historical controls.
4.4.1.5. Kidney Effects
Kidney responses in the rat show no effects at low doses (438 ppm for 10 days) (Gohlke
and Schmidt, 1972). Other than a decreased BUN there were no renal effects at 4,000 ppm and
10,000 ppm by inhalation for 11 days (Landry et al., 1982). NTP showed no adverse kidney
effects at 19,000 ppm for 13 weeks (1989a). Moreover, exposure at 9,625 ppm for 6V2 months
showed no kidney histological effects (unrefereed study by Adams et al., 1939).
In guinea pigs at high levels (40,000 ppm for 9 hr), CE shows kidney congestion and
degeneration (Sayers et al., 1929). Exposures at 15,000 ppm for 2 years seem to promote mouse
tubular regeneration and glomerulosclerosis, albeit mild, while rats were without renal effects
(NTP, 1989a).
7/12/99
35
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
4.4.2. Genotoxicity
As described in NTP (1989a), a methodological variation is necessary to quantitatively
examine the effects of volatile chemicals such as CE in the Ames test. NTP solved this problem
by introducing CE into sealed desiccators through the vacuum valves, thereby gassing the plates
of S. typhimurium bacteria tester strains. Another innovation involved the use of a gas sampling
bag as an exposure vessel (Araki et al., 1994). Using these techniques NTP reported CE-induced
gene reversion in the S. typhimurium base substitution strain TA1535, with or without S-9
metabolic activation. Negative results were obtained in strains TA100 or TA98 (NTP, 1989a).
The positive result was confirmed for TA1535 as well as a related strain, TA1537 (Araki et al.,
1994). These authors also observed CE-induced gene reversion by CE in E. coli WP2 uvrA.
A difference between the genotoxicity of CE in vitro versus in vivo test systems was
demonstrated by Ebert et al. (1994), who compared CE's effects in a hypoxanthine
phosphoribosyltransferase (HPRT) test using Chinese hamster ovary (CHO) cells, an in vivo/in
vitro unscheduled DNA synthesis (UDS) assay in female B6C3F1 mice, and an in vivo
micronucleus (MN) test in male and female B6C3F1 mice. Positive evidence of CE's
genotoxicity was obtained in the in vitro test system, but the compound appeared to have no
ability to induce UDS or MN in vivo. This result caused the authors to question whether CE
possesses clastogenic potential and to speculate on what other combination of mechanisms (other
than genotoxicity) might be involved in the compound-induced carcinogenicity of the uterus.
Therefore, based on the totality of the genotoxicity/mutagenicity evidence, CE may be
considered to be a positive mutagen based on its strong gene reversion effects in certain strains of
S. typhimurium and E. coli. However, the absence of positive genotoxic effects of CE in vivo
leaves open the question of the compound's carcinogenic mechanism in animal studies.
4.5. SYNTHESIS AND EVALUATION OF MAJOR NON-CANCER EFFECTS AND
MODE OF ACTION—ORAL AND INHALATION
Data gaps limit the toxicological (and carcinogenic) evaluations of CE. For example, first,
there is no published information on the toxicity of CE when chronically administered via the oral
route. Second, there is no two-generation CE exposure reproduction study. Thirdly, most of the
well-documented toxicological effects of CE, that have been described, have resulted from
frequent exposures to comparatively high concentrations, i.e., 15,000 ppm CE, but no inhalation
studies identified effects at concentrations lower than 250 ppm (660 mg/m3) (Landry et al., 1989).
It is plausible to infer that CE is not very toxic at low exposure levels because the noncancerous
7/12/99
36
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
CE effects observed at high doses appear to be limited to marginal changes in fetotoxicity, body
weight loss, mild nephrosis, and changes in uterine and liver weights. It is reasonable to
hypothesize that any toxicological effects of the compound at the intermediate to lower CE levels
(and so far untested) would be even more mild, and perhaps cease. This hypothesis cannot be
validated however, with the present data. The interpretation of noncancer effects did not follow
the Agency's risk assessment guidelines for developmental toxicity.
4.5.1. Primary Effect
4.5.1.1. Reproductive and Developmental Toxicity
Reproductive organ and fetal developmental CE effects have been shown in various
experiments. For example, at higher CE concentrations in the dog, the uterus has been noted to
respond with decreased muscle tone and lessened contraction force (Van Liere et al., 1966).
Moreover, in an acute study Fedtke et al. (1994a) reported a 35% decrease in relative uterine
weight in B6C3F1 mice, but not in F344 rats, exposed to 15,000 ppm (39,570 mg/m3) CE 6
hr/day for 5 days. These workers also found decreased GSH levels in the uterus of CE-exposed,
mice and rats. An NTP chronic cancer bioassay has demonstrated quite clearly in mice that the
uterus is the primary organ site for carcinogenesis at 15,000 ppm CE; male rats had skin tumors
and female rats had brain tumors, both marginal (NTP, 1989a). The above findings are consistent
with the hypothesis that the uterus is a primary CE target tissue in rats and mice.
Breslin et al. (1988) observed a statistically significant lengthening of the estrous cycle
(+0.6 days) in B6C3F1 mice exposed to 15,000 ppm (39,600 mg/m3) for 6 hr/day for a minimum
of 14 consecutive days (3 estrous cycles), although no single phase of the cycle appeared to be
uniquely affected. Bucher et al. (1995) also found a statistically significant increase (+0.4 days) in
the mean duration of the estrous cycle in B6C3F1 mice exposed to 15,000 ppm (39,600 mg/m3) 6
hr/day for 3 weeks.
Evidence points to a perturbation of fetal skeletal development in pregnant CF-1 mice
(Table 9, p. 26; Scortichini et al., 1986). These authors reported an apparent statistically
significant (by trend only,/? =0.048) increased incidence of delayed foramina ossification closure
in the skulls of fetal CF-1 mice, but only at the highest exposure of 4,946 ppm CE (13,000
mg/m3) and not at lower doses of 0, 491, or 1,504 ppm CE (Figure 5, p. 25). This HDT effect
likely represents a weak but true fetotoxic response to CE exposure because it was manifested in
the absence of maternal toxicity by all measures. It is the lowest dose (4,946 ppm) and shortest
time representing a critical toxicological effect for CE. Therefore, the fetuses of CF-1 dams
exposed by inhalation during organogenesis to 4,946 ppm CE (13,200 mg/m3) represent a
7/12/99
37
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
sufficiently significant response in delayed fetal foramina closure (DFFC) and are employed herein
as the basis for deriving the noncancer RfC.
4.5.2. Secondary Effects
4.5.2.1. Weight Loss
As noted in Section 4.2.2.1.3.1, a slight (4-8%) decrease in mean body weight gain in
male rats treated at 15,000 ppm CE compared with controls was observed after week 33 of
chronic exposure, with the mean body weights of female rats 5-13% lower than controls from
week 11 to the end of the study (NTP, 1989a). At bioassay termination, the mean body weight of
exposed female rats was reduced by 10% compared with controls. No food consumption data
were described in the study report. Neither clinical signs nor other CE-induced nonneoplastic
lesions were observed. This suggests that the observed weight loss may have been compound-
related and not simply a consequence of food aversion. Because the extent of the weight loss
(10%>) is at the threshold that EPA considers to be toxicologically significant, the response is thus
considered a secondary noncancer effect. On the basis of decreased body weight in female rats at
the single exposure level tested, this study identified a LOAEL(ADJ) for CE of 7,070 mg/m3, a
concentration that would represent a NOAEL in males.
4.5.2.2. Hep ototoxicity
Increased relative liver weight in response to CE exposure at 19,000 ppm for 13 weeks
was observed in both sexes of B6C3F1 mice (NTP, 1989a). In addition, slight increases in mean
relative liver weights with a possibly related increase in the degree of hepatocellular vacuolization
were reported by Landry et al. (1989) in mice exposed to 5,000 ppm for 23 hr/day for 11
consecutive days. Similarly, statistically significant increases in relative liver weights were also
observed in male rats exposed to 4,000 or 10,000 ppm for 6 hr/day, 5 days/week for 2 weeks
(Landry et al., 1982). However, these liver changes appeared to be unaccompanied by any
evidence of compound-related histopathology.
Combining these inferential findings of CE's hepatotoxicity with the observation of a
moderate elevation of the activity of alanine aminotransferase (ALT) in the serum of the 52-year-
old man who had a history of CE sniffing along with other substance abuse activities (Nordin et
al., 1988) suggests that the liver may be a CE target organ at high exposures, although, in general,
few instances of CE-related histopathology of the liver or changes in clinical chemistry
components have been identified.
7/12/99
38
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
CE effects were alluded to in a Russian report in which rats were exposed daily (4 hr/day)
to 570 mg/m3 (220 ppm) for 6 months (Troshina, 1966). This exposure was reported to result in
perturbed hepatic function and lipid degenerative changes, along with decreased arterial blood
pressure and some dystrophic changes to the lungs. However, these findings remain
uncorroborated by other workers and have generally been discounted because of inadequate
reporting (ATSDR, 1997; U.S. EPA, 1998c).
4.5.2.3. Neurotoxicity
A range of neurotoxicological responses to CE have been reported in human beings and in
laboratory animals, with a wide range of doses. Thus, CE can induce anesthesia in humans at
higher doses (Cole, 1967; Dobkin and Byles, 1971; Lawson, 1965), and hyperactivity in mice
(NTP, 1989a), even as low as 5,000 ppm (Dow Chemical Co., 1985). However, these transient
effects may not be appropriate as the basis for developing an RfC because of the acute reversible
nature of the responses. More sensitive measures of central nervous system effects have not been
observed, although Landry et al. (1989) conducted a neurobehavioral observation battery for mice
exposed to concentrations of CE up to 5,000 ppm for 23 hr/day for 11 consecutive days. In
general, mechanisms of anesthesia are not well understood, though it is likely that the observed
effects are due to the direct action of high concentrations of the parent compound on nervous
tissue.
4.5.3. Mode of Action of Toxic Effects
It is difficult to determine the exact nature of CE toxicity in each responding tissue. It also
is difficult to know if there is a interconnecting mode of action among tissues. Section 3.2
discusses what is known on CE metabolic issues, that may underlie the mode of toxic action.
Many haloethanes and halomethanes are conjugatively reduced by GSH. Specifically, it is
known that CE binds GSH to form SEG because the conjugate (SEG) has been directly
measured, as have GS-synthetase enzyme activities of the GS-ethyl formation reaction (Fedtke et
al., 1994b). Other SEG-induced metabolites SENACys and SECys have been demonstrated in
elevated amounts following CE exposure that is a further indication of the reductive conjugative
pathway. At high doses, such as 15,000 ppm CE, the metabolism by GSH conjugation (Figure 4)
can become saturating. When this happens, oxidation of CE to acetaldehyde (and other oxidation
products) occurs by the P450 metabolic route (Ivanetich and Van Der Honert, 1981). This
oxidation occurs more in the mouse than the rat (cf. p. 13). The uterus is a target organ of CE in
the mouse, among others, and may respond by lowering GSH to below normal levels, thereby
7/12/99
39
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
depleting GSH pools, which in turn leads to oxidation with acetaldehyde. The oxidative products
are known to react with cellular macromolecules (Morris, 1997; Behrens et al., 1988). This in
turn can lead to toxicity.
4.6. WEIGHT-OF-EVIDENCE EVALUATION AND CANCER
CHARACTERIZATION—SYNTHESIS OF HUMAN, ANIMAL AND OTHER
SUPPORTING EVIDENCE, CONCLUSIONS ABOUT HUMAN CARCINOGENICITY,
AND LIKELY MODE OF ACTION
Although no data exist that document a tumorigenic effect of CE in human beings, there
are consistent lines of evidence indicating that CE is a carcinogen in animal test systems. These
include:
(1) Chemical carcinogenesis. The clear-cut demonstration of CE's causality in uterine
carcinogenicity in female B6C3F1 mice (uterine incidence: 43/50 = 86%), which is
relevant because uterine cancer is rare to uncommon in B6C3F1 mice occurring at an
uncommon rate of 4/1,371 (0.29%) in a limited population (NTP, 1989a; IARC, 1992).
(2) Mutagenesis. CE is mutagenic in it's capacity to induce gene reversion in certain strains
of S. typhimurium.
(3) Structural Activity Relationship. BE, a structural CE analogue, induces similar
tumorigenic effects in the uterus of B6C3F1 mice as does dichloroethane.
(4) Metabolism. CE's ability to lower GSH pools, similar to methyl halohydrocarbons
CH3C1 and CH3CH2Br, and oxidative metabolism proceeding via acetaldehyde. CM may
be a carcinogen too forming the renal cortex and papillary cystadenocarcinomas in male
mice exposed. BE causes uterine tumors and marginally other tumors: lung,
pheochromocytomas of the adrenals, and brain.
Section 4.2.2.1.3 gives a detailed summary of the principal carcinogenesis study (NTP,
1989a). As noted, the primary effect was the high incidence of uterine carcinomas in female
B6C3F1 mice (43/50 at 15,000 ppm vs. 0/49 in controls). The tumors were of endometrial origin
and showed a profound capacity for metastasizing. First the cancers moved to the neighboring
myometrial tissue and from there disseminated to such secondary tissue sites as lung, ovary,
lymph nodes, kidney, adrenal gland, pancreas, mesentery, urinary bladder, spleen, heart and, to a
lesser extent, colon, stomach, gall bladder, liver, small intestine, and ureter. The complications
arising from these CE-induced tumors are considered to be the cause of the poor survival in these
female B6C3F1 mice (NTP, 1989a). Thus, in addition to metastasis, life-shortening tumor effects
7/12/99
40
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
were observed in female mice, providing further emphasis on the severity of CE chemical
carcinogenesis.
Survival was poor in the male B6C3F1 mice and tumorigenic responses could not be
inferred because of low statistical power. There was a borderline tumorigenic response in the
lung, but the B6C3F1 tumor incidence data in male mice are considered inadequate to determine
potential carcinogenicity in humans.
As noted in Section 4.2.2.1.3.3, marginal increases in some uncommon skin tumors in
male F344 rats were not persuasive enough to unequivocally designate the compound as a
carcinogen in the rat animal model, but the rat skin response is suggestive of tumorigenesis. The
female F344 rat brain astrocytoma response, an uncommonly occurring response, was also
equivocal because of its low incidence (3/49) versus controls (0/46), but the rat brain response is
suggestive. Table 11 summarizes all the relevant tumor-forming effects of CE from the NTP
(1989a) study in F344 rats and B6C3F1 mice. It is concluded: CE is clearly carcinogenic in female
B6C3F1 mice, but the evidence for CE carcinogenicity in male and female F344 rats is equivocal.
Taken as a whole, the mutagenicity and metabolic data that have been amassed for CE are
consistent with the findings of carcinogenicity but some information is lacking (Figure 2). For
example, well-documented positive point mutations have been obtained in Ames tests (NTP,
1989a; Zeiger et al., 1992; Araki et al., 1994), but CE does not induce in vivo clastogenic
responses in the same strain of mice (B6C3F1) in which the uterine carcinomas were described
(Ebert et al., 1994). GSH links with the ethyl group of CE in an elimination reductive conjugation
pathway. If CE is dosed high enough, GSH pool levels likely become limiting for further
ethylation of CE. GSH pools likely become rate-limiting for other detoxification reactions too
(Figure 2, p. 6). In this way the excess CE then would be forced to flow through oxidative
metabolism (oxidation in mice is twice that of rats) with acetaldehyde being an intermediate
(Figure 1, p. 1). This can cause cancer at high internal intermediate mutagenic doses; whether
excessive systemic SEG or acetylaldehyde are involved in the mode of carcinogenic action, or
both, is not known. CE induces its own metabolism, likely p-450 enzyme P450IIE1, which
happens primarily in the liver.
7/12/99
41
DRAFT-DO NOT CITE OR QUOTE
-------
<1
to
to
Table 10. Toxicity/carcinogenicity of chloroethane in experimental studies
Species/strain
Sex/number
Route of
exposure
Dosing regimen
Principal effects
NOAEL /
LOAEL
Reference
Carcinogenicity
Rats/F344
M and F
50/group
Inhalation
0 or 15,000 ppm, 6
hr/day for 2 years.
Uncommon skin tumors
in males at incidence
8/46; malignant
astrocytomas in females
(3/50).
N/A
NTP (1989a)
Mice/B6C3F1
M and F
50/group
Inhalation
0 or 15,000 ppm, 6
hr/day for 2 years.
Uterine tumors in
females (43/50).
N/A
Noncancer
toxicity
B6C3F1
Mice
F
10/group
3 groups
Inhalation
0 or 15,000 ppm
CE for 6 hr/day for
5 days.
Reduction in relative
and absolute (-35%)
uterine weights in
females.
39,570 mg/m3
Fedtke et al.
(1994a)
Rats/F344
M and F
50/group
Inhalation
0 or 15,000 ppm, 6
hr/day, 5 days/week
for 2 years.
Reduction in body
weight gain in females.
39,570 mg/m3 (L)
NTP (1989a)
Mice/F344
M and F
50/group
Inhalation
0 or 15,000 ppm, 6
hr/day, 5 days a weeks
for 2 years.
No significant toxic
effects other than
transient hyperactivity
during dosing.
39,570 mg/m3 (N)
Mice/CF-1
F, 3 0/group
Inhalation
0, 491, 1,504, or
4,946 ppm, 6 hr/day
on GDs 6-15.
Marginal increase in
delay of foramina
closure in the fetal skull.
4,000 mg/m3 (N)
Scortichini et
al. (1986)
Table 10 is continued on the following page
-------
-J
K>
4^
u>
Table 10. Toxicity/carcinogenicity of chloroethane in experimental studies
Species/strain
Sex/number
Route of
exposure
Dosing regimen
Principal effects
NOAEL /
LOAEL
Reference
Rats/F344
Males only,
not females,
10/group
Inhalation
0, 2,500, 5,000,
10,000, or 19,000
ppm, 6 hr/day, 5
days/week, for 13
weeks.
Increases in relative
liver weight in males at
the highest dose.
50,122 mg/m3 (N)
NTP (1989a)
Mice/B6C3F1
Females
only,
not males,
10/group
Inhalation
0, 2,500, 5,000,
10,000, or 19,000
ppm, 6 hr/day, 5
days/week, for 13
weeks.
Increases in relative
liver weight in females
at the highest dose.
50,122 mg/m3 (N)
NTP (1989a)
Rats/F344
M and F
6/group.
Inhalation
0, 1,590, 3,980, or
9,980 ppm, 6 hr/day, 5
days/week, for 2
weeks.
No biologically
significant effects at any
dose level.
26,300 mg/m3 (N)
Landry et al.
(1982)
Mice/B6C3F1
M and F
7/group
Inhalation
0, 250, 1,247, or
4,843 ppm, 23 hr/day
for 11 days.
No biologically
significant effects.
12,200 mg/m3 (N)
Landry et al.
(1989)
Mice/B6C3F1
F
10/group
Inhalation
0 or 15,000 ppm, 6
hr/day for 14 days.
Elongation of the
estrous cycle.
39,570 mg/m3 (L)
Breslin et al.
(1988)
Mice/B6C3F1
F
3 0/group
Inhalation
0 or 15,000 ppm, 6
hr/day for 21 days.
Elongation of the
estrous cycle.
39,570 mg/m3 (L)
Bucher et al.
(1995)
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Table 11. Summary and Conclusions of Tumorigenesis in Rats and Mice
Sex
F344 rats
B6C3F1 mice
Males
Marginal evidence
! Skin tumors
Inadequate evidence
Females
Equivocal evidence
! Brain tumors
Clear evidence
! Uterine tumors that metastasized to
16 secondary organ sites.
! Liver response (weak).
! Hematopoietic response in a number
of tissues and lymph nodes.
(Profound life shortening associated
with the primary effect)
Source: NTP (1989a). Two year bioassay for chemical carcinogenesis in B6C3F1 mice and F344 rats.
Evidence in support of the carcinogenicity of CE is also provided by similar long-term
experimental studies that were carried out on its structural analogue, BE (NTP, 1989b). When
challenged with concentrations of BE at 100, 200, and 400 ppm, female B6C3F1 mice responded
with the formation of uterine squamous cell carcinomas, adenomas, and carcinomas, in direct
analogy to CE. The uterine responses were 4/50, 5/47, and 27/48 for 100, 200, and 400 ppm BE
exposure groups, respectively, versus an incidence in controls of 0/50. Although not statistically
significant, 1,2-dichloroethane administered by gavage produced adenocarcinomas of the uterus in
3/49 mice at 148 mg/kg and 4/47 mice at 229 mg/kg in 78 weeks (NTP, 1978). Related
chlorohydrocarbons that do not cause increased uterine tumors are 1,1-dichloroethane,
1,1,1 -trichloromethane, 1,1,2-trichloroethane, 1,1,1,2-tetrachloroethane,
1,1,2,2-tetrachloroethane, pentachloroethane, and hexachloroethane (all referenced in NTP,
1989a). Other analogues such as methyl chloride cause only cystadenomas and adenomas of male
mice. Methylbromide seems not to be carcinogenic; dibromoethane does produce uterine cancers
and also produces alveolar/bronchiolar carcinomas in male and female mice, as well as
hemangiosarcomas, fibrosarcomas in the subcutaneous tissue, nasal carcinomas, and mammary
adenocarcinomas in female mice.
The similar carcinogenic responses in female B6C3F1 mice to some structural analogues
of CE, i.e., BE, and 1,2-dichloroethane in separate assays, provide support for two concepts: (1)
uterine effects associated with these compounds are unlikely to have come about by chance alone,
and (2) these effects may be brought about by metabolically related mechanisms. Although the
present database for CE carcinogenicity is limited in animals, evidence supporting carcinogenicity
classification are adequate to classify CE.
7/12/99
44
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Based on the criteria set forth in the current Guidelines for Carcinogenic Risk Assessment
(U.S. EPA, 1986), a weight-of-evidence classification of B2 is indicated. That is, CE is a
probable human carcinogen based on no evidence in human beings and adequate evidence in
animals with medium confidence. Categorizing CE according to the weight of evidence approach
proposed by the Guidelines for Carcinogenic Risk Assessment (U.S. EPA, 1996) would derive a
likely human carcinogen by the inhalation route classification for CE.
4.7. OTHER HAZARD IDENTIFICATION ISSUES
4.7.1. Possible Structural-Activity Relationships
As discussed in Section 4.6, BE and chloromethane (CM) are structural analogues of CE.
Bromoethane can deplete GSH, much like CE and CM, and therefore is normally detoxified
primarily by reductive conjugation to GSH (Khan and O'Brien, 1991). Excessive amounts of
deplete GSH and become oxidized to unresolved acetaldehyde (from BE and CE) and
formaldehyde (from BM and CM). "Unresolved" refers to the lack of metabolic steady state and
a buildup of the toxicant intermediate. The toxicity of these analogues may be used to explore
and illuminate the toxicity and mechanism of action of CE.
For BE, the toxic process has been thoroughly documented in an NTP study of the
combined acute, subchronic, and chronic toxicology and carcinogenicity (NTP, 1989b).
Consistent with the lethality data at high doses (5,000-10,000 ppm) of BE, subchronic exposure
to lower concentrations of BE induced profound indications of noncarcinogenic toxicity,
including clinical signs such as tremors, atrophy in some major organs and tissues (e.g., the thigh,
skeletal muscle, and lung), and degeneration of the sex organs (uterus in rats; ovary in mice)
(NTP, 1989b). Interestingly, the histopathologic findings of the 14-week subchronic studies at
single exposures of 1,600 ppm were not observed at lower doses and frequent exposure at 400
ppm in the 2 year study, which suggests a threshold for these high-dose (1,600 ppm) events.
Inhalation concentrations of 0, 100, 200, and 400 ppm were chosen for the 2-year BE study in
contrast to 15,000 ppm for CE.
Though BE displays greater toxicity than CE, especially in the nasal passages and lungs, a
comparison of the carcinogenic endpoints of the two halogenated hydrocarbons implies a domain
of commonality in toxicity mechanisms. That is, the most sensitive carcinogenic response to BE
was the incidence of uterine cancers of female B6C3F1 mice: 0/50 at 0 ppm, 4/50 at 100 ppm (p
= 0.06), 5/50 at 200 ppm (p = 0.03), and 28/48 ppm (p < 10~8) at 400 ppm BE. This same uterine
carcinogenic response was also reported for mice exposed to CE: 0/50 at 0 ppm versus 43/50 (p
7/12/99
45
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
< 10 s) at 15,000 ppm (NTP, 1989a). Other notable cancer sites for BE were lung,
pheochromocytomas, bronchiolar, nasal, and low-level granular-type tumors and gliomas in the
brain; other cancer sites for CE included the liver. The tumor spectra share the uterus, but the
other carcinogenic sites are different. The conclusion of carcinogenicity drawn for BE supports
the chemical carcinogenesis hypothesis for CE.
A recent review by Bolt and Gansewendt (1993) has collated much of the available
information on the toxicity, carcinogenicity, and underlying metabolism of potential CE structural
analogues, e.g., CM, BM, and iodomethane (IM). Examination of available data for CM may be
expected to shed light on the toxic potential of CE if a sufficiently similar pattern of toxic
responses and possibly related metabolic processes are revealed. In this context, CM appears to
behave in a similar manner to CE in the Ames test, inducing positive responses +/- S9 in the S.
typhimurium strain TA 1535 (and TA 100). Though CM failed to induce genotoxic responses in
a number of in vivo tests, a positive dominant lethal response was observed in male F344 rats.
The key information linking the carcinogenic potential of CE and CM comes from an
unpublished study carried out at the Battelle Memorial Institute on behalf of the Chemical
Industries Institute of Toxicology (CUT), which was cited by Bolt and Gansewendt (1993). As
described in the review, Battelle exposed 30 F344 rats and B6C3F1 mice/sex/group to 0, 50, 225,
or 1,000 ppm CM via inhalation, 6 hr/day, 5 days/week for 2 years. Though characterized by
poor survival among the groups, the study revealed an increase in tumors consisting of
cystadenomas and adenomas of the renal cortex and papillary cystadenocarcinomas in male mice
exposed to the highest CM concentration.
There appear to be striking analogies between the metabolic fates of CM and CE. For
example, following rapid uptake via the lungs, important metabolic products of CM were
identified as formaldehyde, formic acid, and carbon dioxide, with part of the material being
incorporated into the Crpool (tetrahydrofolic acid) of intermediate metabolism (Kornbrust and
Bus, 1982; Landry et al., 1983). An important outcome of these experiments was the
demonstration that the incorporation of 14C from CM into major structural macromolecules such
as DNA occurred as a consequence of normal protein synthesis rather than as a result of
methylation (Kornbrust et al., 1982).
Bolt and Gansewendt (1993) combined four lines of evidence into an argument that may
explain the carcinogenic consequences of CM in male B6C3F1 mice in biochemical terms:
(1) demonstration of GSH-linked sulfhydryl derivatives in the urine of
CM-exposed rats (Landry et al., 1983)
(2) the depletion of NPSH content in the liver, lung, and kidneys of exposed F344
rats (Dodd et al., 1982)
7/12/99
46
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1
2
3
4
5
6
(3) the demonstration of both P450-mediated oxidative and GSH-mediated
reductive pathways for the metabolism of CM (Kornbrust and Bus, 1982)
(4) the inhibition of the acute toxicity of CM in male B6C3F1 mice by GSH depletion
(Chellman et al., 1986).
Bolt and Gansewendt (1993) used the latter inhibition data to develop a cancer
mechanistic hypothesis for CM. They assumed a relationship existed between GSH depletion and
cancer because the exposure CM levels for the onset of kidney tumors in male B6C3F1 mice and
the depletion of kidney GSH levels are comparable. Decrementing GSH can switch CM from the
reductive pathway, which is used principally for CM metabolism, to also include oxidative
pathway, i.e., catabolism by P450 of CM to formaldehyde, formate, and C02. Excessive
formaldehyde can cause cancer (Morgan ,1997; Morris, 1997; Monticello and Morgan, 1997).
GSH depletion can likewise cause a paucity of the cofactor GSH for formate dehydrogenase, the
enzyme inactivating formaldehyde. This effect can promote the formation of DNA-protein
crosslinks in susceptible target organs.
As shown in Table 12, lines of evidence that draw parallels between the metabolism of CE
and CM include (1) the duality of CE metabolism, in which reductive metabolism through binding
to GSH and oxidative metabolism mediated by cytochrome P450IIE1 are featured,
Table 12. Common metabolic features of chloroethane and chloromethane:
potential relevance to tumor formation in experiment
tal studies
Mechanisms
Chloroethane
Chloromethane
GSH conjugates
S-ethyl-N-acetylcysteine (SENACys)
S-ethyl-cysteine (SECys)
S-methyl-cysteine (SMCys)
S-methylglutathione
Oxidative metabolism
Cyt P450IIE1
CytP450IIEl
Oxidative products
Acetaldehyde
Formaldehyde
Tumor sites
Uterus in B6C3F1 ? mice
Liver in B6C3F1 9 mice
Kidney inB6C3Fl a* mice
route of exposure
Inhalation
Inhalation
LOAEL
15,000 ppm
1,000 ppm
Tumor ineffective doses
Not tested at lower doses
0, 50, 225 ppm
(2) the depletion of GSH that occurs in target tissues in response to CE exposure, and (3) the
formation of the formaldehyde homologue acetaldehyde during the oxidative metabolism of CE.
If CE and CM share similar mechanisms for tumorigenesis, the occupational exposure
information that has accumulated for CM in humans may also be of relevance to the potential
carcinogenicity of CE. In general, CM data show a marked diversity in the ability of persons to
7/12/99
47
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
metabolize CM, and in the extent of the toxic response elicited by the compound. Bolt and
Gansewendt (1993) discuss a number of findings that point to the existence of different
population subgroups defined by their ability to metabolize CM. Persons with lower rates of GSH
conjugation might be expected to be at greater risk of tumor formation arising from either CM or
CE exposure.
4.7.2. Possible Gender Differences
A report by Griesemer and Eustis (1994) summarized the findings of NTP with regard to
the sex- and tissue-specific onset of carcinogenicity observed throughout their series of 2-year
toxicity/carcinogenicity studies. A total of 1,760 untreated control groups from 440 studies using
F344 rats and B6C3F1 mice (approximately 88,000 animals, from which 3.5 million tissues were
examined microscopically) have contributed data on the gender-specific background rates of
tumor formation. These data apply both to major organs such as lung, liver, and kidney, and to
gender-specific organs, such as uterus, ovary, and testis. Of particular interest to the potential
carcinogenicity of CE (and BE) is the markedly low background frequency of uterine tumor
formation in B6C3F1 mice (0.3%). Based on the high incidence of uterine tumors at the single
CE concentration tested (15,000 ppm), and on the dose-dependent increase in response to
challenge with BE, the conclusion may be reasonably drawn that the occurrence of these uterine
tumors has direct etiological association with the target compounds. Because CE and BE cause
female uterine cancers to the greatest extent of their carcinogenicity, the mode of action of CE
and BE may be highly gender specific.
7/12/99
48
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
5. DOSE-RESPONSE ASSESSMENTS
5.1. ORAL REFERENCE DOSE (RfD)
A chronic RfD can not be determined for CE in water because no chronic oral CE studies
exist
At water-saturating concentrations (0.57 g/100 g) CE oral intake ad libitum for 14 days
did not demonstrate significant toxicological effects (Pottenger et al., 1995). The acute NOEL
for water is 297 mg/kg bw/day for F-344 male rats and 361 mg/kg bw/day for female rats.
5.2. INHALATION REFERENCE CONCENTRATION (RfC)
5.2.1. Choice of Principal Study and Critical Effect - With Rationale and Justification
As discussed in Section 4.5, the noncancer effects of CE exposure in experimental studies
were fetotoxicity, weight loss, neurotoxicity, hepatotoxicity, and immunotoxicity. Of these
effects, the fetotoxicity effect (Scortichini et al., 1986) occurs at the lowest level of CE exposure
in the current animal database and is the "critical" toxic effect, the designation of critical effect
being a judgment step in EPA's RfC risk assessment methodology.
5.2.2. Methods of Analysis
5.2.2.1. Principal Study
Scortichini et al. (1986) reported a statistically significant increase in the delay of frontal
foramina closure (DFFC) of the progeny of CF-1 mice exposed to a mean concentration of
4,946 ppm CE. This fetotoxicity is covered in more detail in Section 4.3.2.1, p. 24. At 4,946 ppm
CE there were 5 affected skulls/116 skulls examined (incidence = 4.3%), representing 5 litters/22
litters examined (incidence = 22.7%). It is notable that the effect was scattered in five different
litters. The lower CE exposures produce responses at 1,500 ppm (1/147) and 491 ppm (1/142)
that were the same as the control (1/116 = 0.9%). The Scortichini study fetotoxicity at 4,946
ppm CE (13,057 mg/m3) is the noncancer critical effect LOAEL for CE, and 1,500 ppm CE
(3,970 mg/m3) is the NOAEL.
7/12/99
49
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
5.2.2.2. Primary Supporting Study
F-344 rats and B6C3F1 mice were both exposed for 5 days to 15,000 ppm or air controls
(Fedtke, 1994a). There was a loss in body weight in both species, 3.7% in rats and 16.4% in
mice. Because the body weight differences between air and CE exposure were not significant, the
weight losses were considered stress related. At autopsy at study end the liver, lung, and kidney
were normal in weight and appearance. Remarkably, the CE-treated mice, but not the rats, had
decreased uterus weights (mean absolute and mean relative). The decrease was about 35%.
Moreover, the uterus is the site for carcinogenesis in mice, but not in rats. GSH pool reduction in
mice, but not rats, can be the basis of the mode of action causing significant acetaldehyde oxidant
intermediate to cause toxicity. Therefore, the 35% uterine weight decrease at 15,000 ppm for
5 days is the primary supporting noncancer effect of CE.
5.2.3. RfC Derivation Including Application of Uncertainty Factors (UF) and Modifying
Factors (MF)
In developing RfCs from observed NOAELs in experimental studies, human equivalent
concentrations (HEC) for extrarespiratory effects are derived by factoring the time-adjusted
NOAELs with the ratio of the animal/human blood gas partition coefficients (XA /Ah). For human
exposure, it is assumed that in time equilibrium is attained for blood/air (b/a) concentrations.
When blood gas partition coefficients are unavailable for an experimental animal, or when Aa>
Ah, then a default ratio of 1 is used (U.S. EPA, 1994). Human blood:air partition coefficients of
1.9 (Morgan et al., 1970), and 2.69 (Gargas et al., 1989) have been reported. A rat blood:air
partition coefficient of 4.08 has been reported (Gargas et al., 1988; 1989). Because both reported
values for humans are lower than the rat partition coefficient, a default ratio of 1 is used to
calculate the HEC. Thus, LOAEL(HEC) = LOAEL(ADJ) x (Aa/Ah) = 13,057 mg/m3 x (1) =
13,057 mg/m3. Thus, for the fetotoxic effect, the LOAEL(HEC) = 13,057 mg/m3, and
NOAEL(HEC) is 3,970 x 1 = 3,970 mg/m3.
The above data set, 1/126, 1/142, 1/147, and 5/116 (on a skull-examined basis) is not
corrected for continuous exposure. A reasonably good fit (p=0.87) at doses 0, 500, 1,500, and
5,000 ppm was obtained using software designed to estimate the benchmark dose employing the
Weibull model (U.S. EPA, 1998d). The Weibull model, p(d) = 1 - exp{-a - P*(d)y, was used.
This dichotomous model predicted that, on the probability of a fetus being affected and for a
benchmark response (BMR) of 10% incidence, the BMC is 17,832 mg/m3 (6,754 ppm CE). This
BMC for 10%) is just above a LOAEL and provides a dependable reference concentration (Allen
et al., 1994a). The BMDL (benchmark dose lower limit) at 10% is 13,421 mg/m3 (5,084 ppm
CE).
7/12/99
50
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
The litter quantal model reduces all fetal incidence data to a question of whether any
fetuses in a litter are affected, while the above fetal model allows the use of fetal data grouped by
dose. Because there were no interlitter biases reported and the dams are the units that are treated,
the litters are considered the representative biological units of CE-induced fetotoxicity. The litter
data 1/22, 1/24, 1/25, and 5/22 (Table 9) were also modeled using the probability of a litter being
affected. Good fits were obtained at 0, 491, 1,504, and 4,946 ppm CE (p-values 0.88). The
BMC for 10% extra risk was chosen for litters and is based on the results for generic quantal
models in Allen et al. (1994b). The BMC for litters is estimated to be 10,634 mg/m3 (4,028 ppm)
using the Weibull model; the lower 95% confidence limit (BMLD) is 4,240 mg/m3 (1,606 ppm
CE). The polynomial model [P(d) = 1 - exp{-q0- q,*(d) - q2*(d)2...-qk*(d)k}] was little different in
results. These litter results are more conservative than the above per-skull basis results. The
BMLD for litters (4,261 mg/m3) may be used as a substitute for a NOAEL when a NOAEL
cannot be estimated. There is a NOAEL of 3,970 mg/m3 but the BMLD will be used to estimate
the RfC, as will the NOAEL. Both results will be compared below.
To establish an RfC, a number of uncertainty factors must be accounted for. The NOAEL
or the BMLD is divided by a complex factor and therefore the acceptable inhalation level is
lowered to accommodate the areas of uncertainty. The factor includes the major areas of
uncertainty that necessitate accommodation. The net result is the establishment of an acceptable
inhalation exposure level that is the highest concentration that takes the combination of these
factors conservatively into account.
An uncertainty factor of 10 is considered for variations in sensitive subpopulations within
populations, and a further factor of 10 is used for interspecies extrapolation. An uncertainty
factor of 3 is used for extrapolating from a LOAEL to a NOAEL. A full factor of 10 is used for
database deficiencies to account for the lack of a multigeneration reproductive study, and because
no evaluation of reproductive function following long-term exposure is available. These
uncertainty components combine (10 x 10x3 x 10) to an overall uncertainty factor (UF) of
3,000.
No modifying factor for this noncancer fetotoxic effect is proposed, therefore, the RfC is
obtained directly as follows:
• Method I RfC = BMLD - 3,000
RfC = 4,261 mg/m3 - 3,000 = 1.4 mg/m3 (0.54 ppm CE)
RfC = 1.4E0 mg/m3
• Method II RfC = NOAEL/3,000
RfC = 3,970/3,000 = 1.32 mg/m3 (0.50 ppm CE)
RfC = 1E0 mg/m3
7/12/99
51
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
The traditional method of estimating a dose to which one may reference as relatively safe
(RfC) is to factor down the noncancer NOAEL by the UF. With the fetotoxic effect, the RfC =
1E0 mg/m3 (rounded off) and will be the recommended value. The RfC in Method I employing a
fit to the data, establishment of the 95% LCL on dose, and factoring down by the UF yields
essentially the same answer (which is a good check).
5.3. CANCER ASSESSMENT
5.3.1. Qualitative Cancer Assessment in Animals
No human cancer data exists on CE. Indirect evidence for the carcinogenicity of CE was
observed in laboratory animal studies in a single NTP study (NTP, 1989a). In the bioassay 50
B6C3F1 mice and 50 F344 rats/sex/group were exposed via inhalation to only one dose: 15,000
ppm CE, 6 hr/day, 5 days/week for 2 years. This is a nonstandard protocol because normally
there are 2 or 3 doses. Tumorigenic responses in male B6C3F1 mice were compromised by poor
survival and the onset of urogenital infections. The B6C3F1 female mice responses were quite
remarkable because of the strong carcinogenic response: uterine carcinogenicity in 43 female mice
of 50 put on test (86%). This incidence is relevant because uterine cancer is an uncommon cancer
site in B6C3F1 mice. The human historical incidence for uterine cancer is approximately 0.006
%, making it the seventh most common female human cancer (Parkin et al., 1999). The historical
rate for B6C3F1 mice is 0.29%, an uncommon but not rare cancer in mice bioassayed so far
(NTP, 1989a; IARC, 1992). The uterine incidence ratio of the CE-treated B6C3F1 group to the
historical group is 86%/0.29% = 297-fold response.
The primary endometrial tumors metastasized to 16 secondary organs or tissue sites. The
females died early compared to concurrent controls due to tumors indicating aggressive
carcinogenic progression. These considerations represent clear evidence of CE's carcinogenicity
in female B6C3F1 mice (NTP, 1989a). In addition, the F344 rat incidence of marginally
tumorigenic responses in males (various skin tumors) and females (astrocytoma brain tumors) was
suggestive of CE's broader spectrum of possible animal carcinogenic responses.
Supporting CE's chemical carcinogenesis is the structural analogue evidence of
carcinogenicity comparing CE to BE, particularly because of the same organ site specificity of
primary uterine tumors for the haloethanes. This structure-activity relationship between the
haloethanes lends credence to the weight-of-evidence classification of CE's carcinogenicity. Also,
the comparison of CE to CM suggests similar mode of action likely leading to carcinogenicity.
Thus, even though the cancer in female B6C3F1 mouse is in only 1 sex of 1 species, and
not in the male mouse or in either sex of rat, the response is nonetheless very high in incidence,
7/12/99
52
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
malignancy, and life-shortening effects. This constitutes compelling carcinogenicity evidence in
B6C3F1 mice. Accordingly, by combining all of the evidence for CE's mutagenicity, animal
carcinogenicity, and similar animal carcinogenic responses of its structural analogues, a weight-of-
evidence classification for human hazard potential can be inferred. CE is a probable human
carcinogen (Category B2) based on no evidence in humans, and a sufficient evidence in animals
(U.S. EPA, 1986). Categorizing CE according to the newer Proposed Guidelines for
Carcinogenic Risk Assessment would designate CE as a likely human carcinogen by inhalation
(U.S. EPA, 1996).
5.3.2. Quantitative Cancer Assessment in Animals
5.3.2.1. Considerations in Quantitative Cancer Assessment
The absorption and distribution of metabolized CE seems to be nonlinear at high doses of
CE (cf. p. 6). The metabolism of the halohydrocarons CE, BE, dichloroethane, and EDB likely
proceeds under normal circumstances by a reductive conjugation pathway mediated by GSH and
various specific glutathionetransferases (Fedtke, 1994a; Commandeur et al., 1995). Methyl
chloride and methyl bromide are metabolized similarly by GSH (Bolt and Gansewendt, 1993).
When exposures of these halohydrocarbons exceed the capacity of the reduction pathway
enzymes, a lesser used pathway, oxidation, becomes more predominant. Specifically, CE is
oxidized via a P450 pathway through acetaldehyde (a toxic compound itself when in excess),
acetic acid, and finally to C02 and H20 as terminal oxidation products (Fedtke, 1994b). It is
difficult to hypothesize that this change in metabolism, in any combination, occurs by a linear
dose-response process, yet the present database is not informative enough to discern
nonlinerarity. The shape of the total metabolic curve, and perhaps the coupled carcinogenicity, is
unknown. Experiments studying CE cellular binding sites, GSH depletion kinetics, and
acetaldehyde kinetics could be useful.
Based on positive animal studies, the derivation of a potential human cancer risk is based
on two aspects of extrapolation: 1) the extrapolation from high animal doses in the observable
range to low animal doses, and 2) the inference that humans will react metabolically similar to the
chemical as the test animals. In the first extrapolation, curve fitting models are used that are
appropriate to the kind of data in the bioassay (U.S. EPA, 1996a, p. 17,992). Extrapolation to
low environmental ranges, commensurate with human exposure, is done on the fitted curve of the
test animal dose-response. The second extrapolation assumes the route of exposure, comparative
metabolism, and target organ mode of action are similar for test animals and humans. Ideally, the
selection of an extrapolation dose-response model is guided by the mode of metabolic action. For
CE the absorption and distribution of metabolized CE seem to be nonlinear at high CE (cf. p. 6).
7/12/99
53
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
More kinetic data points are needed to establish this issue. Further, not enough is known about
the relationship of CE metabolism, whatever it's true kinetic response with dose, and the apparent
CE cancer outcome in the uterus. Unless it is known that current metabolic evidence is measuring
the responsible metabolic factors which initiate and promote CE chemical carcinogenesis, then a
mode of action can only be speculation. A nonlinear kinetic model would allow direct estimation
of dose-response, or if such a model were not available but there are sufficient kinetic data to
determine nonlinearity in cancer-causing metabolic effects, a Margin of Exposure (MOE) method
may be possible if proper exposure and kinetic data and rational were to be presented (U.S. EPA,
1996a).
In cases where there are insufficient rationale to determine the shape of the curve, a
default model is employed. Currently a linearized multistaged model (LMS) is implemented by
Global86 software (U.S. EPA, 1986a; Crump, 1982; Crump 1996). LMS assumes basically
one-critical hit followed by a multistaged process. An 95% upper limit of carcinogenicity risk is
estimated, reported and used until a suitable biological-based dose response model is derived.
The CE cancer risk determination is particularly problematic in that only one dose was
tested by NTP (15,000 ppm), nonetheless that exposure demonstrated a very high tumor
incidence (43/50 = 86%) in the uterus of female B6C3F1 mouse (cf. Section 5.3). Adjusted to
the time of first tumor, which was a uterine carcinoma on week 67, the incidence is 43/49 = 90%.
Because the 1989 NTP bioassay employed a nonstandard protocol with only one point, the LMS
derived cancer slope likely is of low dependability. Thus, other approaches to cancer slope
estimation are considered in the discussion of confidence (section 5.3.3.) as a check to the default
method.
5.3.2.2. LMS Method
The default method of cancer quantitative risk estimation in the U.S. EPA is the linearized
multistaged model (LMS) (Andersen, 1983; U.S. EPA, 1986). This model assumes functional
continuity in the probability-dose function, P f(de), where de are animal experimental doses in
the bioassay and incidence (P) > 0 for all d > 0. The probability-dose function is specifically
assumed to be linear in the human environmental range (d «de) with incidence P = (q| S) x
(d) where qx* is the unit slope. The cancer potency is found by fitting the test animal incidence
data and then finding the upper 95% UCL slope (q| S) of the q, term in the multistage equation
(Crump, 1996). The LMS procedure uses Global86 software to extrapolate the fit of the high-
dose animal data to expected human low-dose incidence (Crump, 1982). The LMS procedure
places an upper limit on risk that is considered to be a plausible upper bound on the increased
cancer risk from lifetime inhalation of CE. However, the range of true risk extends from the 95%
7/12/99
54
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
UCL estimated risk [P = (q| S) x (d)] down to and including zero risk (P = 0). The Agency makes
no true risk presumption.
5.3.2.3. LMS Method Calculation of Cancer Slope
The cancer response at 15,000 ppm CE (7,070 mg/m3) is 43/49 uterine cancers and 8/49
liver cancers for a combined response of 44/49 (90%) (Table 13). The denominator in the tumor
incidence was corrected to include only those animals alive at the time of the first observed tumor,
which was a uterine carcinoma on week 67. This is compared to 3/46 (6.5%) in the concurrent
controls, which were all liver cancers, but no uterine cancers, and below normal aging B6C3F1
hepatocellular adenoma or carcinoma incidence. The combined tumor ratio of the CE-treated
group to the control group is 90%/6.5% ~ 14-fold response.
Table 13. Quantitative cancer responses in the female B6C3F1 mouse liver and uterus
Administered
exposure (ppm)
Human equivalent
exposure
mg/m3
Uterus
incidence
Liver
incidence
Combined
incidences
0
0
0/46
3/46
3/46
15,000
7,070
43/49
8/49
44/49
The denominators in the tumor incidences were corrected to include only those animals alive at the time of the
first observed tumor. First tumor was a uterine carcinoma on week 67. The human equivalent doses were based
on the assumptions that are presented in Section 3.2, converting ppm chloroethane exposure to mg/m3 by a
factor 1 ppm = 2.64 mg/m3, and then adjusting for the specific exposure duration of 6 hr/day (factor: 6 hr/24
hr/day), 5 days/week (factor: 5 days/7days/week).
The cancer risk estimation is based on the responses presented in Section 4.6, p. 38, and
Section 5.3.1, p. 50, and the data in Table 13. The shape of the curve of CE carcinogenicity is not
knowable from the incidence datum in the NTP bioassay. Therefore, a linear model at all doses is
assumed, including the 7070 mg/m3 dose point. The default LMS method, as applied by Global86,
is U.S. EPA policy to determine 95%UCL cancer potency. A Global86 estimate of ED10 is
estimated to be 100 mg/kg/d or 300 mg/m3. For the combined incidence of uterine and liver
cancers (Table 13), Global86 estimates the inhalation unit risk be 4E-4/mg/m3. This unit risk is
approximately equivalent to an inhalation slope factor of q,* = 1.14E-4/mg/kg/day, assuming 20
m3 air breathed/day and body weight of 70 kg. Using the inhalation unit risk of 4E-4/mg/m3,
various CE risk levels may be estimated: at i = 10"4, 300 |ig/m3; at i = 10"5, 30 |ig/m3; and at i =
10"6 = 3 |ig/m3. Using the inhalation slope factor qx* = lE-4/mg/kg/day, then at 10 6 risk, for
example, an exposure rate over a lifetime could not exceed 1E-2 mg/kg/day (10 |ig/kg/day), i.e.,
at 10 |ig/kg/day lifetime CE exposure one may expect, with 95% confidence, no more than a 10 6
7/12/99 55 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
risk in humans based on animal studies, and the risk can be less, considerably less, and even zero
(P = 0).
5.3.3. Discussion of Confidence in Cancer Quantitative Assessment in Animals
Any CE cancer potency estimate, such as a cancer unit slope, made is necessarily not
certain because of the datum on which the estimate is based. Accordingly, the LMS method is
thus not certain because of the one dose-response point and is insufficient for deriving any
estimate of the shape of the dose-response curve. In practice Global86 connects this one point to
the origin and estimates the 95% UCL on this straight line. Another uncertainty is that the
exposed group had nearly 100% tumor incidence (90%), and it is unknown whether such a
saturation of effect would have occurred at an even lower dose, in which case, the proposed
inhalation slope factor could be an underestimate, the degree of which would be unknown. It is
assumed from experience that the plateau on which the response sits is likely < threefold wide
(Gaylor, 1989). It is also unknown whether there are any sublinearities in the dose-response
relationship in the normally observable response range, which could result in the proposed slope
factor being an overestimate. Another issue of quantitative uncertainty concerns the fact that the
study was terminated early (termination week 100), because there was substantial early mortality
in the exposed female mouse group resulting from the tumors (e.g., only 50% of the exposed
group were still alive at week 90 compared with 90% in controls). This time component was not
taken into account in the risk calculations because animals were dying from uterine carcinogenesis
not competing toxicity. Conceivably, a lower dose could have resulted in the same tumor
incidence along with later-occurring tumors in life, in which case there would be an
underestimation of the proposed slope factor; the time-to-tumor issue may be relatively trivial
with respect to the other uncertainties outlined above.
Because of the uncertainties of the LMS method, alternative methods were examined to
gain additional perspective on the upper bound of CE cancer potency. One of the first
nonparametric methods was a procedure taking the lowest dose (7,070 mg CE/m3) and incidence
(i =0.90), and estimating the upper 95% confidence limit incidence (~ 1.00), and then define the
straight line connecting this point to the 0,0 point. The cancer slope of the line « 1/7070 mg/m3 =
1.4E-4/mg/m3 (Gaylor and Kodell, 1980).
Another type of estimate can be made which is enabled by the unusually high incidence
(i =0.90). In the case of CE, the bioassay produces almost the maximum theoretical response of
100%) incidence and can be considered an approximation of the MTD. This assumption is based on
the following: 1) many of the female mice died prematurely because of the tumor load (only 2/50
survived until termination at week 100), and 2) the observation of a maximum cancer response
7/12/99
56
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
with some toxicity at this dose but not overt noncancer toxicity. As with many chemicals, more CE
likely would have been too toxic (i.e., > MTD) from the start of the study and would reduce
cancer because sick or morbid animals do not yield tumors beyond certain doses but rather become
sickly and/or die before cancer evolution. Just as likely, less CE (just how much less is unknown
but assuming in the dose-response range) would have decreased CE carcinogenicity because of less
coupling to the reaction sequence (mode of action) causing cancers. Thus, 7070 mg/m3 can be a
crude estimate of the MTD for CE.
It has been found for most chemicals tested so far, that certain dosimetric relationships
exist among the parameters MDT, TD50, and the! 0 6 Risk Dose even though the TD50 (potency)
varies over eight orders of magnitude among the chemicals (Gaylor and Gold, 1995; Krewski et
al., 1993; Bernstein et al., 1985; Shlyakhter et al., 1992). The Risk Dose is a "low" dose on the
curve presumed to be in the linear range. One of these relationships is that k = (MTD)/(10 6 Risk
Dose) « 740,000 where "k" a geometric average of 317 diverse structured chemicals with only 14
falling outside a 10-fold interval, i.e, an approximate constant (k) exists among different chemical
(for further details see Appendix). Thus, 10 6 Risk Dose ~ MTD/740,000. Because cancer slope
= incidence/dose, then slope = (10 6 incidence)/(10 6 Risk Dose), hence one may estimate an upper
limit simulating the LMS q,* linear cancer slope value (Gaylor, 1989).
The nonparametric estimation of 10 6 Risk Dose involves an empirical factor associated
with the observed median (MDT)/( 10 6 Risk Dose) ratio (i.e., k ~ 740,000) (Gaylor and Gold,
1995). Thus, 10~6 Risk Dose - 7,070 mg/m3/740,000 = 9.55E-3 mg/m3(3.6E-3 ppm) (Table 14).
For inhalation, the CE concentration of the ! 0 6 Risk Dose is ~ 1E-2 mg/m3 or ~ 4 ppb for a lifetime
of CE exposure. Next, the unit cancer slope 95%UCL estimation is 10 6/9,55E-3 mg/m3 = 1.05E-
4/mg/m3. This method compares with the above nonparametric method (1.4E-4/mg/m3) as well as
estimated LMS slope of 4E-4/mg/m3. These cancer unit slope values appear to be comparable
within the limits of error of the methods—see below for reliability. Because the nonparametric
methods did not attempt to model the datum, but rather used dosimetric relationships and
comparison to 317 previously tested carcinogens, the nonparametric cancer slope estimates lend
support to the LMS value 4E-4/mg/m3.
The slope estimation using the 10 6 Risk Dose may be compared with historical controls of
mice and humans to estimate the Margin of Safety (MOS). Thus, an i = 10"6 is less than the NTP
historical control incidence of 0.0029 in female B6C3F1 mice, hence i = 10"6 is conservative level
in mice. However, i = 10"6 is somewhat less than the world-wide human incidence of 59 xlO"6 (5.9
diagnosed cases of uterine cancer per 100,000 females), hence i = 10"6 is somewhat conservative in
humans. Because the spontaneous frequency of uterine cancer is normally low among female
B6C3F1 mice but higher than the assumed 10 6 risk, the nonparametric method using k = 740,000
is sufficiently conservative in the mouse by 29,000-fold (2.9E-3/1E-6). Thus, even though there is
7/12/99
57
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
uncertainty in the MTD estimation, and hence the 10 6 Risk Dose derived from it, there is an ample
margin of safety (MOS) of 29,000 that suggests that inhalation exposures yielding < 10~6 Risk Dose
levels of CE will not likely add to, or exceed, the spontaneous levels of uterine cancers in rodents.
However, humans apparently have some MOS at i = 10"6: 59xl0"6/10"6 = 59. The human MOS is
less than the rodent MOS because the human cancer rates are normally less than the rodent rates
so a smaller numerator in the preceding calculation. Human exposures are concerning too because
uterine cancers are relatively common being the 7th highest cancer occuring in females world-wide.
So, CE carcinogenicity could add to background.
In conclusion, all three upper-bound cancer slope estimates should be considered uncertain
because of the one-point bioassay on which they are based. It is reasoned that a numerical
assessment is prudent, however, because of the striking animal response (i = 0.9), low spontaneous
occurrence of uterine tumors in mice, carcinogenic SAR of BE and 1,2-dichloroethane at the same
organ site, and metabolic comparisons to CM. Notwithstanding, the development of a cancer
potency estimate does not effect the qualitative assessment of CE carcinogenicity (section 6.3,
below).
7/12/99
58
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
6. CHARACTERIZATION OF ASSESSMENTS
6.1. ORAL RfD
An acute oral CE palatability study was conducted in F344 rats (Pottenger et al., 1995).
F344 rats were administered either 0 or 0.57 g/100 g water for 7- or 14-days. This is at the
practical solubility limit of CE in water at room temperature. Toxicology parameters investigated
were comparable between treated and control groups. The acute NOEL of CE in water then is
297 mg/kg bw/day for male rats and 361 mg/kg bw/day for female rats.
A necessary chronic oral study was not located to set a RfD for CE, therefore a oral RfD
cannot be estimated at this time.
6.2. INHALATION RfC
The CE inhalation RfC of 1E0 mg/m3 is based on a Dow Chemical Co. teratology study
(Scortichini et al., 1986). There is low to medium confidence in this study as this is a fetotoxic
effect only in the high-dose group (4,946 ppm CE). Most noncancer effects in the CE database
occurred at the higher exposure level tested of 15,000 ppm CE (with none lower than 4,946 ppm
or in between). Increases in menstrual periods, decreases in uterine weight, and uterine cancers
(see below) are effects (at 15,000 ppm) may support a hormonal mode of action possibly related to
fetal development, but changes in blood estrogen and progesterone were examined and were not
observed. So the hormonal issue is unresolved. The basis of the single fetotoxic effect in the mouse
skull (delayed foramina closure) is not understood.
6.3. CANCER ASSESSMENT
The CE carcinogenic response is highly specific to the female mouse uterus @ 15,000 ppm
CE compared to a low spontaneous uterine cancer incidence in concurrent controls (0%) and
historical controls (0.29%). The CE mouse uterine response compares the average historical
control rate 88%/0.29% = 303-fold, a large increase in incidence that is unlikely due to chance.
The significantly increased mouse uterine cancer response to CE seems to be biologically relevant
because U.S. uterine historical control rates in humans is relatively common in North America
(15.01/100,000) which is about l/6th the breast cancer incidence, the most common, and V2 the
incidence of female lung cancer, the 2nd most common, in the same region. Thus, CE exposure
7/12/99
59
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
19
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
37
38
39
40
41
could add to ongoing the uterine cancer rate in the human population. On the other hand, the
female rat in the NTP bioassay is not affected with these tumors @ 15,000 ppm CE, only
borderline astrocytomas. Bromoethane—a close chemical analogue to CE—supports the mouse
CE carcinogenic uterine response in that it too causes uterine tumors in B6C3F1 female mice
(i = 28/48 = 0.58 @ 400 ppm). The CE-treated mice present an exceptionally large (i = 43/49 =
0.88 @ 15,000 ppm) uterine cancer response, or any other organ site response for that matter,
compared to other chemical carcinogens that the Agency has reviewed to date. The degree of
carcinogenicity is exceptional too in that the primary tumors are very aggressive, metastasizing to
16 diverse organ sites in female B6C3Fi mice and killing them early due to tumor load.
By comparison to mouse historical controls, then, chemical carcinogenesis from CE may be
inferred in humans, but in comparison to rats it may not. It is Agency policy to assume the worst
case until a sufficient mode of action is known that may delineate between the test species. Thus,
the human applicability is not assured but a concordant cancer hazard in humans is inferred by the
Agency from the powerful mouse responses to CE and BE in the uterus.
Therefore, outlining the elements of CE's carcinogenicity:
(1) Exceptionally strong cancer incidence in the female mouse uterus (and some liver
cancers). Uterine cancers progressed from the endometrium to the adjacent
myometrium and from there to 16 secondary malignant organ sites, and female
B6C3F1 mice were killed early due to tumor load,
(2) Structural analogues BE and 1,2-dichloroethane cause similar uterine cancer
responses,
(3) CE metabolizes similarly to CM: saturation of reductive GSH metabolism and induction
of excessive P-450 oxidative metabolism at high CM. Because CM causes renal
cortex cystadenomas and adenomas and papillary cystadenocarcinomas in male
mice and because CE causes uterine and liver cancers in female mice, the
carcinogenicity of both may be linked to their metabolic similarity.
(4) CE's mutagenicity evidence and the prospect that CE can be an alkylating agent under
the correct activating conditions.
Thus, because of the striking mouse cancer response, similarity to BE and EDC chemical
carcinogenesis, uncommon occurrence of the tumor type, mutagenic and potential alkylating
properties, CE exceeds a C Category weight of evidence usually reserved for one-species
responses. The weight-of-evidence supports the choice of B2 carcinogenicity classification for CE,
7/12/99
60
DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
i.e., a probable human carcinogen based on no evidence in human beings and sufficient evidence
for carcinogenicity in animals (U.S. EPA, 1986). CE is a likely carcinogen by the inhalation route
of exposure using the Proposed Guidelines for Carcinogenic Risk Assessment (U.S. EPA, 1996).
Confidence in the carcinogenic categorization is medium based on: (1) the high incidence
of uterine tumors in female B6C3F1 mice but none in F344 female rats, (2) the aggressive nature
of the cancer proliferation from the endometrium to the myometrium then to many secondary
cancers, (3) comparably low historical control rates in mice, (4) the consistency in tumorigenic
responses between CE and its structural analogues, BE and EDC, and (5) the metabolic
comparison to CM and CE that relates to GSH conjugation and P-450 oxidation, which could
relate to CE and CM toxicity and "coupled" carcinogenicity. The mechanistic coupling is shown in
the parentheses:
[CE or CM -~ ^toxicity ( biochemical and cellular steps in time) ->-~ carcinogenicity]
The coupling relates exposure and toxicity to a mode of action via a kinetics model. In time, a loss
of cellular growth control results. The coupling reactions for CE are not certain at this time.
Remaining data gaps include: (1) the lack of dose-response data sufficient to determine the tumor
incidence rate at intermediate CE exposure levels (100-4,500 ppm CE), (2) the absence of any
detailed information on the triggering site or the target organ-specific biochemical processes that
link CE exposure, or intermediate, and response, and (3) any hormonal link that may explain the
mouse uterine tumors from CE and BE and whether this applies to humans, (4) specific
comparisons of mouse and human metabolic patterns and kinetic for CE.
6.4. CHARACTERIZATION OF HAZARD EXPECTED UPON HUMAN EXPOSURE TO
CHLOROETHANE
The most robust toxic effect of CE is the inhalation malignant cancer effect observed in
female B6C3F1 mice @ 15,000 ppm, not the noncancer fetotoxic effect @ 4,900 ppm or the 10%
weight loss or 35% uterine weight loss @ 15,000 ppm (Table 10). Chronic CE oral toxicity studies
are lacking,
7/12/99
61
DRAFT-DO NOT CITE OR QUOTE
-------
g Table 14. Comparison of noncancer and cancer hazard evaluations
Critical qualitative
effect
Dose 11
Qualitative
Cancer or Non-cancer Potency Estimation
(ppm
)
assessment
factors
Method of Estimation b
Reference dosesc
NON-
Delayed
fetal
foramina
closure
(DFFC)
in CF-1
mice skulls
4,946
Weak to mild
fetotoxic effect;
1-point response;
threshold; medium
confidence
RfC = NOAEL/(10 x 10 x 3 x 10)]
RfC = 3,970/3,000 = 1.32 mg/m3
(0.50 ppm CE); RfC = 1E0 mg/m3
RfC = 1.32 mg/m3
RfC = 1E0 mg/m3
(500 ppb)
CANCER
BMC = 4,028 ppm @ 10%; BMLD = 1,606 ppm
RfC = BMLD/3,000 = 1.4 mg/m3 = 1E0 mg/m3
RfC =1.4 mg/m3
RfC 1E0 mg/m3
(535 ppb)
CANCER
Uterine
cancer
production
15,000
Strong effect in
B2C3F1 mice;
malignancy;
LMS method
qx* = inhalation cancer slope = 1E-4/ mg/kg/day
(using Global86 & assuming linearity in the low dose range)
10~6 Risk Dose =
1E-4 mg/kg/day
( * 1 ppb)
On
with
subsequent
aggressive
metastasis,
death due to
cancer burden;
BE, an analogue
of CE, causes
LMS method
cancer unit risk = 4E-4/ mg/m3
(using Global86 & assuming linearity in the low dose range)
10~6 Risk Dose =
2.5E-3 mg/m3
(* 1 ppb)
and
finally,
death
due to
tumor
load
similar cancer
pattern;
category B2;
high
confidence
Nonparametric method 1
95%UCL incidence on lowest significant point above controls is ::
1.0: slope « 1/7070 mg/m3 d cancer slope = 1.4e-4/mg/m3
10~6 Risk Dose =
0.7E-2 mg/m3
0 3 PPb)
a
H
Nonparametric method 2
10 6 Risk Dose = MDTd divided by k, where k 7.4E+5.
Thus, 10 6 Risk Dose = 9.95E-3 mg/m3
cancer slope == 10"6/9.95E-3 mg/m3 =1.05E-4/mg/m3
cancer slope = lE-4/mg/m
10~6 Risk Dose =
1E-2 mg/m3
H ppb)
1
a
o
*
o
H
O
HH
H
M
O
P
a Lowest effect dose; however, in both noncancer and cancer experiments only one dose group produced a critical response.
b Details and assumptions of the calculations should be referred to in the text of the document. Conversion: 2.64 mg/m3/l ppm CE. Cancer slopes for the LMS
and nonparametric methods are given with the cancer calculations and are presented here in bold.
c These are reference doses and not implied to be safe in the case of carcinogenic effects; it has been U.S.EPA policy to assume no safe dose exists for
carcinogens. For the noncancer effects a threshold may be assumed so an RfC is presented.
d Mouse exposure = 15,000 ppm chloroethane gas for 6 hrs/day and 5 days/wk. for 100 wks. adjusted to 7,070 mg/m3 (human) see text for assumptions.
O
c
o
H
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
but acute studies suggest a lack of toxicological activity of oral CE dissolved in water. Firstly,
hazard by CE inhalation is caused by GSH pool depletions: this may not only cause more CE
exposure to be metabolized improperly but also other impending or extant carcinogens in the body
at the time of CE exposure. This GSH depletion is followed by excessive production of oxidants,
such as acetaldehyde, which often can not be eliminated fast enough to prevent initiation and
promotion of cancer events. This decreased protection by reductive conjugation and unresolved
oxidation is the likely mode of action. Specific kinetics of these elimination reactions is lacking in
the CE database. The carcinogenic process is likely indirect, however, because CE itself does not
seem to accumulate in the mouse uterus.
The absorption and distribution of CE appears to be nonlinear in the female B6C3F1 mouse
at high doses (Dow Chemical Co., 1992). The system seems metabolically saturated for GSH
conjugation at 15,000 ppm and oxidation, a lesser used pathway for halohydrocarbons, is likely
expressed in addition to reductive conjugation. The net catabolism via reduction plus oxidation is
likely not linear in CE exposure, but the true net curvature is unknown. There is a decrease in
noncancer effects below 15,000 ppm to quantitatively weak effects, like the fetotoxic effect and
body weight loss, and little other remarkable toxic effects. All the toxicity evidence for CE
suggests that the slope from an anesthetic dose (19,000 ppm), to a carcinogenic dose (15,000
ppm), to mild fetotoxicity (5,000 ppm), is a steep slope of biological interactions. This sharp
declination suggests a lack of toxicological activity at lower CE exposures, but because of a
paucity of information this cannot be demonstrated. The historic use of short-term human gaseous
anesthetic doses, up to 40,000 ppm, as well as dermal topical applications for temporary pain relief
has not produced evidence of chronic toxicity, though a systematic study has not been done either.
It has not been demonstrated if CE actually causes human uterine cancer, or any human
cancers for that matter. It is notable that CE spray has been used as a human topical anaesthetic.
For example, in contact sports in the United States, CE has been used in considerable amounts to
temporarily alleviate pain. Also veternarians have until recently used CE sprays for topical animal
surgeries. None of these uses have produced reports of adverse effects. This does not mean there
are no topical carcinogenic responses, merely that none have been reported. Nonetheless, a human
cancer hazard is thought likely from CE chronic inhalation exposure on the basis of rodent studies.
The cancer unit risk for CE is q,* = 4E-4/mg/m3. This cancer unit risk is based on an upper
bound estimate but the true unit risk could be less, even down to zero. The use of this unit risk has
uncertainty based on limited cancer and mode-of-action data and not having a CE PBPK model.
Given these limitations, an upper limit cancer risk to a population with chronic exposure may be
approximated from P(d) = qx* d which can be restated as risk = 4E-4 d where "d"is exposure in
mg/m3 over a 70-year lifetime. Limited or intermittent CE exposures can be evaluated on a case-
by-case basis.
7/12/99
63
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
7. REFERENCES
Adams, E; Rowe, V; Spencer, H. (1939) Experimental investigation of the toxicology of ethyl
chloride. Dow Chemical Co. as cited by Landry: Fundam Appl Toxicol 2:230-234.
Allen, BC; Crump, KS; Shipp, A. (1988) Correlation between carcinogenic potency of chemicals in
animals and humans. Risk Anal 8(4):531-544.
Allen, BC; Kavlock RJ; Kimmel, CA; et al. (1994a) Dose response assessments for developmental
toxicity: II. Comparison of generic benchmark dose estimates with NOAELs. Fundam Appl
Toxicol 23:487-495.
Allen, BC; Kavlock RJ; Kimmel, CA; et al. (1994b) Dose response assessments for developmental
toxicity: III. Statistical models. Fundam Appl Toxicol 23:496-509.
Andersen, EL. (1983) Quantitative approaches in use to assess cancer risk. Risk Anal. 3:277-295.
Araki, A; Noguchi, T; Kato, F; et al. (1994) Improved method for mutagenicity testing of gaseous
compounds by using a gas sampling bag. Mutat Res 307:335-344.
ATSDR (Agency for Toxic Substances and Disease Registry). (1997) Toxicological Profile for
Chloroethane. Public Comment Draft. U.S. Department of Health and Human Services. Agency
for Toxic Substances and Disease Registry, Atlanta, GA.
Behrens, UJ; Hoerner, M; Lasker, JM; et al. (1988). Formation of acetaldehyde adducts with
ethanol-inducible P-450IIE1 in vivo. Biochem Biophys Res Comm 29:584-590.
Bernstein, L; Gold, LS; Ames, BN; et al. (1985) Some tautologous aspects of the comparison of
carcinogenic potency in rats and mice. Fundam Appl Toxicol 5(l):79-86.
Bircher, AJ; Hampl, K; Hirsbrunner, P; et al. (1994) Allergic contact dermatitis from ethyl chloride
and sensitization to dichlorofluoromethane. Contact Dermatitis 31:41-44.
Bolt, HM; Gansewendt, B. (1993) Mechanisms of carcinogenicity of methyl halides. Crit Revs
Toxicol 23:237-253.
Breslin, WJ; Berdasco, NM; Phillips, JE; et al. (1988) Ethyl chloride (EtCl): effects on estrous
cycling in B6C3F1 mice. Final report with cover letter dated 11/21/88. Dow Chemical Co. EPA
Document #86-890000040.
Bucher, JR; Morgan, DL; Adkins, BJR; et al. (1995) Early changes in sex hormones are not
evident in mice exposed to the uterine carcinogens chloroethane and bromoethane. Toxicol Appl
Pharmacol 130:169-173.
Bush, OF; Bittenbender, G; Adriani, J. (1952) Electrocardiographic changes during ethyl chloride
and vinyl ether anesthesia in the dog and man. Anesthesiology 13:197-202.
7/12/99
64
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Chellman, GJ; White, RD; Norton, RM; et al. (1986) Inhibition of the acute toxicity of methyl
chloride in male B6C3F1 mice by glutathione depletion. Toxicol Appl Pharmacol 86:93-104.
Cole, WH. (1967) A re-evaluation of the pharmacology of ethyl chloride. Med J. Aust 1:853-855.
Commandeur, J; Stijntjes, G; Vermeulen, PE. (1995) Enzymes and transport systems involved in
the formation and disposition of glutathione S-conjugates. Pharmacol Rev 47(2):271-330.
Crump, K. (1982) An improved procedure for low-dose carcinogenic risk assessment from animal
data. J Environ Pathol Toxicol 5(4/5):339-348.
Crump, KS. (1996) The linearized multistage model and the future of quantitative risk assessment.
Hum Exper Toxicol 15:787-798.
Dobkin, AB; Byles, PH. (1971) Pharmacodynamics of divinyl ether, ethyl chloride, fluroxene,
nitrous oxide and trichloroethylene. Textbook Vet Anaesth 94-104.
Dodd, DE; Bus, JS; Barrow, CS. (1982) Nonprotein sulfhydryl alterations in F-344 rats following
acute methyl chloride inhalation. Toxicol Appl Pharmacol 62:228-236.
Dow Chemical Co. (1985) Ethyl chloride: inhalation teratology probe in CF-1 mice (final report)
with cover letter dated 102591 (sanitized). EPA Document #88-920000184S.
Dow Chemical Co. (1992) Disposition and metabolism of female F344 rats and B6C3F12 mice
following inhalation exposure with cover letter dated 012893. EPA document #86-930000120.
Ebert, R; Fedtke, N; Certa, H; et al. (1994) Genotoxicity studies with chloroethane. MutatRes
322:33-44.
E.I. Dupont de Nemours and Co. (1971) Initial submission: Cardiac sensitization by inhalation
exposure in dogs with cover letter dated 101592. Document #88-920009728.
Fedtke, N; Certa, H; Ebert, R; et al. (1994a) Species differences in the biotransformation of ethyl
chloride. I. Cytochrome P450-dependent metabolism. Arch Toxicol 68:158-166.
Fedtke, N; Certa, H; Ebert, R; et al. (1994b) Species differences in the biotransformation of ethyl
chloride. II. GSH-dependent metabolism. Arch Toxicol 68:217-223.
Gargas, ML; Seybold, PG; Andersen, ME. (1988) Modeling the tissue solubilities and metabolic
rate constant (Vmax) of halogenated methanes, ethanes and ethylenes. Toxicol Lett 43:235-256.
Gargas, ML; Burgess, RJ; Voisard, DE; et al. (1989) Partition coefficients of low-molecular
weight volatile chemicals in various liquids and tissues. Toxicol Appl Pharmacol 98:87-99.
Gargas, ML; Clewell, HJ; Andersen, ME. (1990) Gas uptake inhalation techniques and the rates of
metabolism of chloromethanes, chloroethanes, and chloroethylenes in the rat. Inhal Toxicol 2:295-
319.
7/12/99
65
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Gaylor, DW. (1989) Preliminary estimates of the virtually safe dose for tumors obtained from the
maximum tolerated dose. Reg Toxicol Pharmacol 9:101-108.
Gaylor, DW; Kodel, RL. (1980) Linear interpolation algorithm for low dose risk assessment of
toxic substances. J Environ Path Toxicol 4:305-312.
Gaylor, DW; Gold, LS. (1995) Quick estimate of the regulatory virtually safe dose based on the
maximum tolerated dose for rodent bioassays. Regul Toxicol Pharmacol 22:57-63.
Gohlke, JM; Scmidt, P. (1972) Subacute action of low concentrations of chlorinated ethanes on
rats with and without additional ethanol-treatment. II. Histological, histochemical and
morphometrical studies. Int Arch Arbeitsmed 30:298-312.
Gold, LS; Slone, TH; Bernstein, L. (1989) Summary of carcinogenic potency and positivity for
492 rodent carcinogens in the Carcinogenic Potency Database. Environ Health Perspect 79:259-
272, cf. Table 3.
Goodman, G; Wilson, R. (1991) Predicting the carcinogenicity of chemicals in humans from rodent
bioassay data. Environ Health Perspect 94:195-218.
Griesemer, RA; Eustis, SL. (1994) Gender differences in animal bioassays for carcinogenicity. J
Occup Med 36:855-859.
Haid, B; White, JM; Morris, LE. (1954) Observations of cardiac rhythm during ethyl chloride
anesthesia in the dog. Curr Res Anesth 33:318-325.
Hattori, L. (1957) Effects of inhalation of anesthetics on the gaseous metabolism in rats. Nippon
YakwigakuZasshi 53:136-144.
Heller, S; et al. (1966) Changes in the histochemical pattern of rat liver in prolonged narcosis.
Folia Morphol 25:9-20 (1966).
Henderson, VE. (1930) Aesthetic toxicology. Arch Int Pharmacodyn Ther 38:150-165.
Hes, JP; Cohn, DF; Streifler, M. (1979) Ethyl chloride sniffing and cerebellar dysfunction (case
report). Isr Ann Psychiatr Relat Discip 17:122-125.
Hinko, M. (1934) The effect of some narcotics upon the sedimentation of red blood cells. Tohoku
J Exp Med 23:279-297 (Japanese).
Hiraiwa, K. (1952) Effect of narcotics on the flexion reflex on rats. Osaka Daigaku lgaku Zasshi
5:95-100 (Japanese).
Holder, JW. (1998) Chloroethane carcinogenicity (CAS No. 75-00-3). U.S. Environmental
Protection Agency, National Center for Environmental Assessment, Washington, D.C.
EPA/600/R-95/099. Available from : http://www.epa.gov/ncea and the National Technical
Information Service, Springfield, VA, PB-98-143274.
7/12/99
66
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
International Agency for Research on Cancer. (IARC) (1992) Cancer incidence in five continents.
Volume VI. IARC Scientific Publication No. 120, Lyon, France.
Ivanetich, KM; Van Der Honert, LH. (1981) Chloroethanes: their metabolism by hepatic
cytochrome P-450 in vitro. Carcinogenesis 2:697-702.
Khan, S; O'Brien, PJ. (1991) 1-Bromoalkanes as new potent nontoxic glutathione depletors in
isolated rat hepatocytes. Biochem Biophys Res Commun 179(1):436-441.
Kornbrust, DJ; Bus, JS. (1982) Metabolism of methyl chloride to formate in rats. Toxicol Appl
Pharmacol 65:135-143.
Kornbrust, DJ; Bus, JS; Doeijer, G; et al. (1982) Association of inhaled [14C]methyl chloride with
macromolecules from various rat tissues. Toxicol Appl Pharmacol 65:122-134.
Krewski, D; Gaylor, D; Soms, A; et al. (1993) An overview of the report: Correlation between
carcinogenic potency and the maximum tolerated dose: implications for risk assessment. Risk Anal
13 (4):383-398.
Landry, TD; Ayres, JA; Johnson, KA; et al. (1982) Ethyl chloride: a two-week inhalation toxicity
study and effects on liver non-protein sulfhydryl concentrations. Fundam Appl Toxicol 2:230-234.
Landry, TD; Gushow, TS; Langvardt, PW; et al. (1983) Pharmacokinetics and metabolism of
inhaled methyl chloride in the rat and dog. Toxicol Appl Pharmacol 68:473-486.
Landry, TD; Johnson, KA; Phillips, JE; et al. (1989) Ethyl chloride - 11-day continuous exposure
inhalation toxicity study in B6C3F1 mice. Fundam Appl Toxicol 13:516-522.
Lawson, JI. (1965) Ethyl chloride. Br J Anaesth 37:667-670.
Lazarew, NW. (1929) Uber die narkotische Wirkungskraft der Dample der Chlorderivate des
Methans, des Athans und des Athylens. Naunyn-Schmiedeberg's Arch Exp Pharmacol 141:19-24.
Monticello, T. M. and Morgan, K. T. (1997). Chemically-induced nasal carcinogenesis and
epithelial cell proliferation: a brief review. Mutation Res. Fundamental and Molecular Mechanisms
of Mutagenesis 380: 33-41.
Morgan, A; Black, A; Belcher, DR. (1970) The excretion in breath of some aliphatic halogenated
hydrocarbons following administration by inhalation. Ann Occup Hyg 13:219-233.
Morgan, K. T. (1997). A brief review of formaldehyde carcinogenesis in relation to rat nasal
pathology and human health risk assessment. Toxicologic Pathology 25: 291-307.
Morris, JB. (1997) Dosimetry, toxicity and carcinogenicity of inspired acetaldehyde in the rat.
Mutation Res. Fundamental and Molecular Mechanisms of Mutagenesis 380:113-124.
7/12/99
67
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Morris, L; Noltensmeyer, MH; White, JM. (1953) Epinephrine induced cardiac irregularities in the
dog during anesthesia with trichloroethylene, cyclopropane, ethyl chloride, and chloroform.
Anesthesiology 14:153-158.
Neal, MJ; Rodson, JM. (1964) The analgesic properties of subanesthetic doses of anesthetics in the
mouse. Br J Pharmacol Chem 22:596-603.
NTP (National Toxicology Program). (1978) Bioassay of 1,2-dichloroethane for possible
carcinogenicity. NTP TRNo. 55. U.S. Department of Health, Education, and Welfare, Public
Health Service, National Institutes of Health, Research Triangle Park, NC.
NTP. (1989a) Toxicology and carcinogenesis studies of chloroethane (ethyl chloride) (CAS No.
73-00-3) in F344/N rats and B6C3F1 mice (Inhalation Studies). NTP TR No. 346. U.S.
Department of Health and Human Services, National Institutes of Health, Research Triangle Park,
NC.
NTP. (1989b) Toxicology and carcinogenesis studies of bromoethane in F344/N rats and B6C3F1
mice. NTP TRNo. 363. U.S. Department of Health and Human Services, National Institutes of
Health, Research Triangle Park, NC.
Nordin, C; Rosenqvist, M; Hollstedt C. (1988) Sniffing of ethyl chloride — an uncommon form of
abuse with serious mental and neurological symptoms. Int J Addict 23:623-627.
Parkin, D.M; Pisani, P.; Ferlay, J. (1999) Global cancer statistics. CA 49 (1): 33-64.
Pottenger, LH; Nieusma, JL; Landry, TD; et al. (1992) Ethyl chloride: disposition and metabolism
in female Fischer 344 rats and B6C3F1 mice following inhalation exposure. Part A. Dow Chemical
Co. R&D Report. EPA document #86-930000120.
Pottenger, LH; Cosse, P; Quast, J; et al. (1995) Ethyl chloride: palatability and 14-day drinking
water toxicity study in Fischer 344 rats. Dow Chemical Co. R&D Report. EPA Document
Number: 86-99000022.
Reinhardt, CF; Azar, A; Maxfield, ME; et al. (1971) Cardiac arrhythmias and aerosol "sniffing."
Arch Environ Health 22:265-279.
Sayers, RR; et al. (1929) Physiological response attending exposure to vapors of methyl
bromide, methylene chloride, ethyl bromide, and ethyl chloride. U.S. Public Health
Bulletin No. 185. U.S. Government Printing Office, Washington, DC.
Schmidt, P; Binneweis, S; Gohlike, R. (1972) Subacute action of low concentrations of chlorinated
ethanes on rats with and without additional ethanol treatment. I. Subacute and chronic toxicity
studies with 1, 1,2,2-tetrachloroethane. Int Arch Arbeitsmed 38:283-298.
Scortichini, BH; Johnson, KA; Momany-Pfruender, JJ; et al. (1986) Ethyl chloride: inhalation
study in CF-mice. Dow Chemical Co. EPA Document #86-870002248.
7/12/99
68
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Seller, C. (1938) Action of hypnotics. Arch Exp Pathol Pharmakol 188:699-713.
Shlyakhter, A.; Goodman G.; Wilson, R. (1992). Monte Carlo Simulation of Rodent
Carcinogenicity Bioassays, Risk Analysis 12: 73-82.
Tokita, N. (1953) Influence of various narcotics on cerebral circulation. Tohoku J Exp Med
59:149-158 (Japanese).
Troshina, MN. (1966) Determination of maximum permissible concentration of ethyl chloride in
the atmosphere of work premises. Gig Truda i Prof Zabolegania 10:37-42. [Russian with English
abstract],
U.S. EPA. (1986a) Guidelines for carcinogenic risk assessment; notice. Federal Register
51:33992-34003.
U.S. EPA. (1986b) Guidelines for the health risk assessment of chemical mixtures. Federal
Register 51(185):34014-34025.
U.S. EPA. (1986c) Guidelines for mutagenicity risk assessment. Federal Register 51(185):34006-
34012.
U.S. EPA. (1988a) Recommendations for and documentation of biological values for use in risk
assessment. Prepared by the Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Cincinnati, OH, for the Office of Solid Waste and Emergency
Response, Washington, DC. EPA/600/6-87/008.
U.S. EPA. (1988b) Summary review of health effects associated with monochloroethane: health
issue assessment. Environmental Criteria and Assessment Office. Office of Health and
Environmental Assessment. Research Triangle Park, NC. EPA/600/8-88/080.
U.S. EPA. (1991) Guidelines for developmental toxicity risk assessment. Federal Register
56:63798-63826.
U.S. EPA. (1994a) Interim policy for particle size and limit concentration issues in inhalation
toxicity: notice of availability. Federal Register 59(206):53799.
U.S. EPA. (1994b) Methods for derivation of inhalation reference concentrations and application
of inhalation dosimetry. Prepared by the Office of Health and Environmental Assessment, Research
Triangle Park, NC. EPA/600/8-90/066F.
U.S. EPA. (1994c) Peer review and peer involvement at the U.S. Environmental Protection
Agency. Signed by the U.S. EPA Administrator, Carol M. Browner, dated June 7, 1994.
U.S. EPA. (1995a) Proposed guidelines for neurotoxicity risk assessment. Federal Register
60:52032-52056.
7/12/99
69
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
U.S. EPA. (1995b) The use of the benchmark dose approach in health risk assessment. Prepared
by the Risk Assessment Forum, Washington, DC. EPA/630/R-94/007.
U.S. EPA. (1996a) Proposed guidelines for carcinogenic risk assessment. Federal Register
61:17960-18011.
U.S. EPA. (1996b) Reproductive toxicity risk assessment guidelines. Federal Register
61(212): 56274-56322.
U.S. EPA. (1997) Risk characterization: a practical guidance for NCEA-Washington risk
assessors, (draft). Prepared by the Office of Research and Development, Washington, DC.
U.S. EPA. (1998a) Bench mark dose software. Beta version 1.1b. Developed under the direction
of Dr. Jeff Gift, U.S. EPA/HERL Research Triangle Park, NC., (919)541-4828.
U.S. EPA. (1998b) Science Policy Council Handbook: Peer Review. Office of Science Policy,
Office of Research and Development, Washington, DC; EPA/100/B-98-001. Available from
http://www.epa.gov/ncepihom/catalog/EPA100B98001.html.
U.S. EPA. (1998c) Integrated Risk Information System. Online. Office of Health and
Environmental Assessment, National Center for Environmental Assessment, Cincinnati, OH.
Van Dyke, RA; Wineman, CG. (1971) Enzymatic dechlorination: dechlorination of chloroethanes
and propanes in vitro. Biochem Pharmacol 20:463-470.
Van Liere, EJ; Mazzocco, TR; Northrup, DW. (1966) The effect of cyclopropane, trichloroethane,
and ethyl chloride on the uterus of the dog. Am J Obstet Gynecol 94(6):861-874.
Zeiger, E; Anderson, B; Haworth, S; et al. (1992) Salmonella mutagenicity tests: V. Results from
the testing of 311 chemicals. Environ Mol Mutagen 19(Suppl. 21):2-141.
7/12/99
70
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
8. APPENDIX
Nonparametric Maximum Tolerated Dose Method of Cancer Potency Estimation
An alternative method to fitting parametrically a number of dose response points was
sought because of the uncertainty inherent in a one dose point bioassay. Although dose-response
incidence points of a carcinogen usually can be parametrically fitted, it is problematic to
extrapolate very far from the actual experimental points down to environmental exposures. This
has been the subject of a number of model fitting methods and proposals in the last 20 years. It
also has been suggested that a nonparametric technique could be used that would
nonpresumptively assess cancer potency at environmental exposures "d" that were much less than
the experimental exposures de (Gaylor and Kodel, 1980). A nonparametric method of interpolation
may be used where the lowest experimental dose (de) that is significantly increased, statistically and
biologically, may be determined. A 95%UCL of the incidence point may be estimated at de. This
95%UCL point is connected to the origin to create a straight line which has been interpreted as an
estimate of the upper bound limit on risk (Gaylor and Kodell, 1980). Because no threshold is
assumed in P f(de), this method allows risk estimates even at low doses. This interpolation
method of Gaylor and Kodell was found to agree with a multi staged Armitage-Doll model
estimation of upper bound potency for a number of chemicals (Gaylor and Kodell, 1980). It was
one of the first demonstrations of a nonparametric extrapolation method to assess risk at
environmental doses.
Most continuous response curves can be parametrically fit. The Agency fits the
experimental points with a polynomial function [P(d) = b0 + q,d + q2d2+ q3d3 q4d4...] employed by
Global86 software. In the low dose region linearity of dose and response is assumed and thus the
95% UCL of the q, coefficient is determined (the latter terms are so small that they can be ignored)
and is called the qx*. P(d) in the low range is estimated by qx* • (exposure). There is one
exception to being able to "fit' the curve, and that is when the bioassay has only one dose point
(Gaylor and Kodel, 1980). Gaylor and Kodel (1980) have stated: "In the special case where only
one dosage level of a chemical is administered to animals, obviously no mathematical model can be
obtained." It seems prudent, then, to seek a nonpresumptive method to estimate risk that differs
from the LMS method in methodology.
Considering the current CE one dose case (inhalation study at 7,070 mg/m3 and a
concurrent control), the degrees of freedom are n -1 = 2 - 1 = 1, a straight line. The more the
degrees of freedom for a data set, the more power or sensitivity it possesses. A two-point set,
concurrent control and one experimental point, has low power and low sensitivity to accurately
detect a specific response. Of course, if the one point of the bioassay is not duplicated or
7/12/99
71
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
replicated, the true variance is not known and the precision is unknown. In such a case where
there is just one de and one d0 (concurrent control), a low-dose interpolation method has been
suggested (Gaylor and Kodel, 1980). For example, at the 90% response point for CE, if one
assumes -100% as the 95% UCL on a 90% response rate, then cancer unit slope - 1.0/7,070
mg/m3 = 1.4E-4/mg/m3. It is notable this agrees with LMS and this is expected since the LMS is
essentially doing the same type of calculation.
Some time ago it was suggested that (1) not only may MDTs (maximum dose tested in a
bioassay) correlate between rats and mice, but also (2) that the MDTs for a given chemical seem to
correlate (in a 30-fold range) with the respective cancer potencies (Bernstein et al., 1985). The
cancer potency is abbreviated as TD50 and is used as a "midpoint" to characterize a cancer
dose-response curve. The TD50 is defined as the average daily dose (mg/kg bw/day) rate that is
estimated to halve the probability of remaining tumor free at a specified organ site in a 2 yr. study.
The TD50 varies over alO8 range of dose for the various chemicals bioassayed so far.
The first finding suggests that MDT ~ MTD = maximum dose tolerated in a 90-day study for both
rodent species. These rodent correlations suggest human parameters may also relate in a parallel
manner (Allen et al., 1988). The second finding suggests that knowing the MDT for a chemical
may allow an estimation of TD50 (cancer potency).
Further, for a given chemical, the TD50 seems to correlate with TD10, TD01, and even in the
low range the TD0 0001 (Gaylor, 1989). The latter is the dose at 10"4 % incidence or 1:106 which is
sometimes referred to as the virtually safe dose (VSD). The U.S. EPA makes no value judgement
at 10"6 incidence (risk) as being virtually safe or not. Here, we make use of that point as a reference
point on the continuous dose-response curve in the low-dose and linear range. It has been
suggested, additionally, that the q,* (cancer slope) varies inversely as the log (TD50) (Gaylor and
Gold, 1995). These relationships are summarized in Table 15 and developed further below.
Table 15. Chemical-specific dose parameters
MTD from 90-day study
MDT (chronic) « MTD
MDT - TD50
TD50 10~6 Risk Dose
MDT 10 6 Risk Dose
check: 10~6 Risk Dose
1/qj*
Note: These relationships are characterized in Gaylor and Gold, 1995; Krewski et al., 1993; Bernstein et al.,
1985. Specific constants relating these parameters may found in these references.
A log-normal distribution is
assumed.
In an expanded study including 69 tumor sites and 38 chemicals for rats and mice (138
cases) chosen for their varied chemical structures, empirical extrapolation from MDT to TD50 and
then to 10 6 Risk Dose correlated with the LMS model estimates of 10 6 Risk Dose at 10 6 risk
7/12/99
72
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
(Gaylor, 1989). The Gaylor empirical method found (MDT)/(10 6 Risk Dose) = k where the
empirical constant k was first estimated to be 340,000 (geometric mean of the 138 cases). Only 3
of 138 cases were extreme, i.e., over 10-fold different from the geometric mean of the ratio.
Because cancer potencies are known to vary over eight orders of magnitude, the relative constancy
of (MTD)/( 10 6 Risk Dose) ratio for a number of chemicals suggests that once the MTD is
estimated in a 90-day study, the 10~6 Risk Dose may be approximated also.
Further analytical work examined and confirmed the properties of the empirical inverse
function of the upper bound on the low dose slope (b) and the MDT (Krewski et al., 1993). The
cancer slope "b" varies inversely with MTD: log (b) = (0.01 ± 0.05) - (1.05 ± 0.03) log (MDT).
These authors also showed the TD50 is also related to the MDT: log (TD50) = (1.04 ± 0.02) log
(MDT) - (0.10 ± 0.04). It is clear, then, because TD50 and b are related to MDT, that TD50 and b
are related to each other: log (MDT) = 0.961 log (TD50) + 0.0673. A comprehensive study
compiling 139 carcinogens from the NCI database showed the geometric mean of the MDT/TD50
ratio is 0.919 for mutagens and 0.764 for nonmutagens (Gaylor and Gold, 1995). That is, the
TD50 values tend to be greater than the MDTs. Variance for 78% of the 139 chemicals is within a
factor of 4-fold and for 98% of them variance is within a factor of 10-fold. From the open
literature, representing a more diverse set of chemicals, the MDT/TD50 ratios are 1.46 and 0.951
(Gaylor and Gold, 1995). These ratios suggest, in these approximate measures, that MDT/TD50
ratios are constant for most dose-response curves. Also suggested is that mutagens and
nonmutagens do not differ significantly. The variation of the MDT/TD50 ratio is similar to the
variation in cancer potency (1/TD50), thus the MTD (90-day study) is a reasonable surrogate for
the TD50 (chronic study) (Gaylor and Gold, 1995). It follows that at 10 6 risk with a
corresponding dose of 10 6 Risk Dose, "k" may be estimated by the ratio (MDT)/(10 6 Risk Dose)
for a given chemical. When a larger number of chemicals (317) are considered, the geometric
mean constant "k" is 740,000, which happens to be larger (in fact two fold) than the above
empirical constant. The change in "k" is likely because more chemicals were considered in this
study, therefore 740,000 probably better represents the "geometric average k" (Gaylor and Gold,
1995). The geometric range of 10~6 Risk Doses may be estimated in the range of MDT/7,400,000
- MDT/74,000 from the most potent carcinogens ever assayed to the least potent carcinogens, i.e.,
the 10 6 Risk Dose range is a geometric variation of 10-fold around the mean value. Obviously,
division by 7,400,000 yields the more conservative 10 6 Risk Dose estimate (< 10 6 risk) and
would be at the low end of 10 6 Risk Doses of chemicals previously tested. Human exposures <
MTD/107 have been proposed as negligible risk because it would be assumed the carcinogen is
similar to the most potent measured (Gaylor, 1989).
The LMS method uses a polynomial fit to cancer dose-response data, and the coefficients
and power depend on the data set. Because the NTP study has one experimental point (15,000
7/12/99
73
DRAFT-DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
ppm), there are not enough degrees of freedom to find a proper fit (NTP, 1989a). Gaylor and
Kodel (1980) have stated: "In the special case where only one dosage level of a chemical is
administered to animals, obviously no mathematical model can be obtained." It seems prudent to
use a another type of method to estimate risk; at least this approach backs up the default LMS
method. It is estimated (see section 5.3.2.) by nonparametric method 2 that the 10 6 Risk Dose =
1E-2 mg/m3 or 4 ppb (above) and the cancer slope estimated as a check is 1.05E-4/mg/m3
(= 10 6/ lE-2 mg/m3) for a lifetime of CE inhalation exposure. In Table 15 (last cell) it is indicated
that 10 6 Risk Dose 1/qi*, which indicates that the nonparametric method provides estimates
similar to those made by the LMS method: 1.05E-4/mg/m3 ( nonparametric #2 ) ~4E-4/mg/m3
(LMS). Therefore the Gaylor method is as conservative as the LMS method estimation of chemical
carcinogen risk. For further discussion see Gaylor and Gold (1995).
Doses calculated by the cancer slopes in Table 14 (LMS, nonparametric methods 1 and 2)
at 10 6 risk (a arbitrary low level in the linear range) arel, 3, and 4 ppb CE. It is not implied that at
lower environmental doses there is no risk from CE because a linear, no-threshold presumption is
made for genotoxic carcinogens. The hazard of the 1-4 ppb range of CE does imply that the
additive risk may be no more, and perhaps less, than 10"6 risk, and thus exposures up to 1-4 ppb
are less than rodent uterine spontaneous cancer, which occurs at a rate of 2.9%o incidence rate, by
a factor of 29,000 (0.0029/10"6). A human approximation of the margin of safety (MOS) may be
made by dividing the 1997 frequency of U.S. uterine incidence of 15.01/100,000 North American
females by 10 6, which indicates a MOS of 150 in the U.S. Thus, continuous CE inhalation
exposures < 1-4 ppb likely do not add a significant risk to ongoing human uterine cancer from all
causes.
***
7/12/99
74
DRAFT—DO NOT CITE OR QUOTE
------- |