EPA/690/R-22/004F | September 2022 | FINAL
United States
Environmental Protection
Agency
xvEPA
Provisional Peer-Reviewed Toxicity Values for
The Aromatic Medium Carbon Range Total Petroleum
Hydrocarbon (TPH) Fraction
(various CASRNs)
-------
A United Stiles
WtirV Protection
EPA/690/R-22/004F
September 2022
https://www.epa.gov/pprtv
Provisional Peer-Reviewed Toxicity Values for
The Aromatic Medium Carbon Range Total Petroleum
Hydrocarbon (TPH) Fraction
(various CASRNs)
Center for Public Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
CHEMICAL MANAGERS
Glenn E. Rice, ScD
Center for Public Health and Environmental Assessment, Cincinnati, OH
Allison L. Phillips, PhD
Center for Public Health and Environmental Assessment, Cincinnati, OH
CONTRIBUTORS
Elizabeth O. Owens, PhD
Center for Public Health and Environmental Assessment, Cincinnati, OH
Jeff Swartout, MS (deceased)
Center for Public Health and Environmental Assessment, Cincinnati, OH
Jacqueline Weinberger, Student Services Contractor
Oak Ridge Associated Universities
DRAFT DOCUMENT PREPARED BY
SRC, Inc.
7502 Round Pond Road
North Syracuse, NY 13212
PRIMARY INTERNAL REVIEWERS
Jeffry L. Dean II, PhD
Center for Public Health and Environmental Assessment, Cincinnati, OH
M. Margaret Pratt, PhD
Center for Public Health and Environmental Assessment, Washington, DC
PRIMARY EXTERNAL REVIEWERS
Eastern Research Group, Inc.
110 Hartwell Avenue
Lexington, MA 02421-3136
PPRTV PROGRAM MANAGEMENT
Teresa L. Shannon
Center for Public Health and Environmental Assessment, Cincinnati, OH
J. Phillip Kaiser, PhD, DABT
Center for Public Health and Environmental Assessment, Cincinnati, OH
Questions regarding the content of this PPRTV assessment should be directed to the U.S. EPA
Office of Research and Development (ORD) Center for Public Health and Environmental
Assessment (CPHEA) website at https://ecomments.epa.gov/pprtv.
iii Aromatic medium carbon range TPH fraction
-------
TABLE OF CONTENTS
COMMONLY USED ABBREVIATIONS AND ACRONYMS v
BACKGROUND 1
QUALITY ASSURANCE 1
DISCLAIMERS 2
QUESTIONS REGARDING PPRTVs 2
1. INTRODUCTION 3
1.1. DEFINITION OF THE AROMATIC MEDIUM CARBON RANGE FRACTION 3
1.2. OVERVIEW OF PHYSICOCHEMICAL PROPERTIES AND ENVIRONMENTAL
FATE 4
1.3. OVERVIEW OF MIXTURE ASSESSMENT METHODS 8
1.3.1. Indicator Chemical Approach 9
1.3.2. Hazard Index Approach 9
2. SUMMARY OF TOXICITY AND DOSE-RESPONSE ASSESSMENT APPROACH 11
2.1. IDENTIFICATION OF RELEVANT MIXTURES AND COMPOUNDS WITH
TOXICITY VALUES 13
2.2. IDENTIFICATION OF OTHER RELEVANT TOXICITY DATA 16
2.3. METHODS FOR INDICATOR CHEMICAL SELECTION 18
2.4. DEVELOPMENT OF EXPOSURE-RESPONSE ARRAYS 18
3. REVIEW OF POTENTIALLY RELEVANT DATA 20
3.1. NONCANCER EVIDENCE 20
3.2. CANCER EVIDENCE 22
3.2.1. Human Studies 22
3.2.2. Animal Studies—Oral 22
3.2.3. Animal Studies—Inhalation 23
3.2.4. Summary of Cancer Evidence 23
4. TOXICOKINETIC CONSIDERATIONS 24
5. MECHANISTIC CONSIDERATIONS AND GENOTOXICITY 26
6. DERIVATION 01 PROVISIONAL VALUES 27
6.1. DERIVATION OF ORAL REFERENCE DOSES 27
6.1.1. Oral Noncancer Assessment Using the Indicator Chemical Method for the
Aromatic Medium Carbon Range Fraction 32
6.1.2. Alternative Oral Noncancer Assessment Using the Hazard Index Method for
the Aromatic Medium Carbon Range Fraction 33
6.2. DERIVATION OF INHALATION REFERENCE CONCENTRATIONS 34
6.2.1. Inhalation Noncancer Assessment Using the Indicator Chemical Method for
the Aromatic Medium Carbon Range Fraction 38
6.2.2. Alternative Inhalation Noncancer Assessment Using the Hazard Index
Method for the Aromatic Medium Carbon Range Fraction 39
6.3. SUMMARY OF NONCANCER PROVISIONAL REFERENCE VALUES 39
6.4. CANCER WHIGII I-01 -EVIDENCE DESCRIPTOR 40
6.5. DERIVATION OF PROVISIONAL CANCER RISK ESTIMATES 41
APPENDIX A. LITERATURE SEARCH AND SCREENING 42
APPENDIX B. POTENTIALLY RELEVANT NONCANCER EVIDENCE 44
APPENDIX C. REFERENCES 72
iv Aromatic medium carbon range TPH fraction
-------
COMMONLY USED ABBREVIATIONS AND ACRONYMS
a2u-g
alpha 2u-globulin
IRIS
Integrated Risk Information System
ACGIH
American Conference of Governmental
IVF
in vitro fertilization
Industrial Hygienists
LC50
median lethal concentration
AIC
Akaike's information criterion
LD50
median lethal dose
ALD
approximate lethal dosage
LOAEL
lowest-observed-adverse-effect level
ALT
alanine aminotransferase
MN
micronuclei
AR
androgen receptor
MNPCE
micronucleated polychromatic
AST
aspartate aminotransferase
erythrocyte
atm
atmosphere
MOA
mode of action
ATSDR
Agency for Toxic Substances and
MTD
maximum tolerated dose
Disease Registry
NAG
7V-acetyl-P-D-glucosaminidase
BMC
benchmark concentration
NCI
National Cancer Institute
BMCL
benchmark concentration lower
NOAEL
no-observed-adverse-effect level
confidence limit
NTP
National Toxicology Program
BMD
benchmark dose
NZW
New Zealand White (rabbit breed)
BMDL
benchmark dose lower confidence limit
OCT
ornithine carbamoyl transferase
BMDS
Benchmark Dose Software
ORD
Office of Research and Development
BMR
benchmark response
PBPK
physiologically based pharmacokinetic
BUN
blood urea nitrogen
PCNA
proliferating cell nuclear antigen
BW
body weight
PND
postnatal day
C#
number of carbon atoms contained in a
POD
point of departure
molecule
PODadj
duration-adjusted POD
CA
chromosomal aberration
QSAR
quantitative structure-activity
CAS
Chemical Abstracts Service
relationship
CASRN
Chemical Abstracts Service registry
RBC
red blood cell
number
RDS
replicative DNA synthesis
CBI
covalent binding index
RfC
inhalation reference concentration
CHO
Chinese hamster ovary (cell line cells)
RfD
oral reference dose
CL
confidence limit
RGDR
regional gas dose ratio
CNS
central nervous system
RNA
ribonucleic acid
CPHEA
Center for Public Health and
SAR
structure-activity relationship
Environmental Assessment
SCE
sister chromatid exchange
CPN
chronic progressive nephropathy
SD
standard deviation
CYP450
cytochrome P450
SDH
sorbitol dehydrogenase
DAF
dosimetric adjustment factor
SE
standard error
DEN
diethylnitrosamine
SGOT
serum glutamic oxaloacetic
DMSO
dimethylsulfoxide
transaminase, also known as AST
DNA
deoxyribonucleic acid
SGPT
serum glutamic pyruvic transaminase,
EC
equivalent carbonEPA
also known as ALT
Environmental Protection Agency
SSD
systemic scleroderma
ER
estrogen receptor
TCA
trichloroacetic acid
FDA
Food and Drug Administration
TCE
trichloroethylene
FEVi
forced expiratory volume of 1 second
TWA
time-weighted average
GD
gestation day
UF
uncertainty factor
GDH
glutamate dehydrogenase
UFa
interspecies uncertainty factor
GGT
y-glutamyl transferase
UFc
composite uncertainty factor
GSH
glutathione
UFd
database uncertainty factor
GST
glutathiones-transferase
UFh
intraspecies uncertainty factor
Hb/g-A
animal blood-gas partition coefficient
UFl
LOAEL-to-NOAEL uncertainty factor
Hb/g-H
human blood-gas partition coefficient
UFS
subchronic-to-chronic uncertainty factor
HEC
human equivalent concentration
U.S.
United States of America
HED
human equivalent dose
WBC
white blood cell
i.p.
intraperitoneal
Abbreviations and acronyms not listed on this page are defined upon first use in the
PPRTV document.
v Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
PROVISIONAL PEER-REVIEWED TOXICITY VALUES FOR
THE AROMATIC MEDIUM CARBON RANGE TOTAL PETROLEUM
HYDROCARBON (TPH) FRACTION
BACKGROUND
A Provisional Peer-Reviewed Toxicity Value (PPRTV) is defined as a toxicity value
derived for use in the Superfund Program. PPRTVs are derived after a review of the relevant
scientific literature using established U.S. Environmental Protection Agency (U.S. EPA)
guidance on human health toxicity value derivations.
The purpose of this document is to provide support for the hazard and dose-response
assessment pertaining to chronic and subchronic exposures to substances of concern, to present
the major conclusions reached in the hazard identification and derivation of the PPRTVs, and to
characterize the overall confidence in these conclusions and toxicity values. It is not intended to
be a comprehensive treatise on the chemical or toxicological nature of this substance.
Currently available PPRTV assessments can be accessed on the U.S. EPA's PPRTV
website at https://www.epa.gov/pprtv. PPRTV assessments are eligible to be updated on a 5-year
cycle and revised as appropriate to incorporate new data or methodologies that might impact the
toxicity values or affect the characterization of the chemical's potential for causing adverse
human-health effects. Questions regarding nomination of chemicals for update can be sent to the
appropriate U.S. EPA's eComments Chemical Safety web page
(https ://ecomments epa. gov/chemical safety/).
QUALITY ASSURANCE
This work was conducted under the U.S. EPA Quality Assurance (QA) program to ensure
data are of known and acceptable quality to support their intended use. Surveillance of the work
by the assessment managers and programmatic scientific leads ensured adherence to QA
processes and criteria, as well as quick and effective resolution of any problems. The QA
manager, assessment managers, and programmatic scientific leads have determined under the
QA program that this work meets all U.S. EPA quality requirements. This PPRTV was written
with guidance from the CPHEA Program Quality Assurance Project Plan (PQAPP), the QAPP
titled Program Quality Assurance Project Plan (PQAPP) for the Provisional Peer-Reviewed
Toxicity Values (PPRTVs) and Related Assessments/Documents (L-CPAD-0032718-QP), and the
PPRTV development contractor QAPP titled Quality Assurance Project Plan—Preparation of
Provisional Toxicity Value (PTV) Documents (L-CPAD-0031971-QP). As part of the QA
system, a quality product review is done prior to management clearance. A Technical Systems
Audit may be performed at the discretion of the QA staff.
All PPRTV assessments receive internal peer review by at least two CPHEA scientists
and an independent external peer review by at least three scientific experts. The reviews focus on
whether all studies have been correctly selected, interpreted, and adequately described for the
purposes of deriving a provisional reference value. The reviews also cover quantitative and
qualitative aspects of the provisional value development and address whether uncertainties
associated with the assessment have been adequately characterized.
1 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
1 DISCLAIMERS
2 The PPRTV document provides toxicity values and information about the adverse effects
3 of the chemical and the evidence on which the value is based, including the strengths and
4 limitations of the data. All users are advised to review the information provided in this document
5 to ensure that the PPRTV used is appropriate for the types of exposures and circumstances at the
6 site in question and the risk management decision that would be supported by the risk
7 assessment.
8 Other U.S. EPA programs or external parties who may choose to use PPRTVs are
9 advised that Superfund resources will not generally be used to respond to challenges, if any, of
10 PPRTVs used in a context outside of the Superfund program.
11 This document has been reviewed in accordance with U.S. EPA policy and approved for
12 publication. Mention of trade names or commercial products does not constitute endorsement or
13 recommendation for use.
14 QUESTIONS REGARDING PPRTVS
15 Questions regarding the content of this PPRTV assessment should be directed to the
16 U.S. EPA ORD CPHEA website at https://ecomments.epa.gov/pprtv.
2 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
1. INTRODUCTION
This Provisional Peer-Reviewed Toxicity Value (PPRTV) assessment supports a
fraction-based approach to risk assessment for mixtures of petroleum hydrocarbons U.S. EPA
(2022). In this approach, total petroleum hydrocarbon (TPH) fractions are defined based on
expected transport in the environment and analytical methods used to quantify environmental
contamination by TPH mixtures. TPH components were first classified into aromatics and
aliphatics, and each of these two major fractions were further separated into low, medium, and
high carbon range fractions. This PPRTV assessment describes the derivation of both noncancer
and cancer toxicity values for the aromatic medium carbon range fraction of TPH. The toxicity
values described herein are used in the assessment of Complex Mixtures of Petroleum
Hydrocarbons that is intended to replace current approaches used at TPH-contaminated sites U.S.
HP A (2022).
In general, fraction-based approaches involve: (1) dividing a complex mixture into
groups based on similarities in their chemical structures or chemical properties; (2) measuring
the concentrations of these groups (or the components within the group) in environmental media
or estimating the rates of human exposure (e.g., mg/kg-day) to these groups; (3) selecting an
approach to characterize a dose-response relationship for the group; (4) combining the
dose-response approach and the exposure estimates for all members of the group to estimate
health risks from the group; and (5) estimating risks or hazards posed by exposure to the
complex mixture using the risk characterization information from the individual groups [adapted
from Atsdr (2018)1.
1.1. DEFINITION OF I II I AROMATIC MEDIUM CARBON RANGE FRACTION
The aromatic medium carbon range fraction includes aromatic hydrocarbons with a
carbon (C) range of C9-C10 (contains between 9 and 10 carbons, inclusive) and an equivalent
carbon (EC) number1 index range of EC9-EC < 112 that occur in, or co-occur with, petroleum
contamination. It should be noted that the aromatic high carbon range fraction also includes
C10 compounds, but unlike the aromatic high carbon range fraction, the aromatic medium
carbon range fraction is restricted to those with EC9-EC < 11; the aromatic high carbon range
fraction includes those compounds with an EC11-EC35. The EC index is equivalent to the
retention time of the compound on a boiling-point gas chromatography (GC) column (nonpolar
capillary column), normalized to the //-alkanes NJ PEP (2010). EC numbers are the physical
characteristic that underpin analytical separation of petroleum components. EC numbers are
useful because they are more closely related to environmental mobility than carbon number. For
instance, two chemicals with similar carbon numbers but different structures
(e.g., aliphatic vs. aromatic) could partition differently into environmental media and, ultimately,
have different environmental fates. Grouping based on EC numbers provides a consistent basis
for logically placing petroleum hydrocarbon compounds into fractions because EC measures
correlate with physicochemical properties such as water solubility, vapor pressure, Henry's law
constant, and soil adsorption coefficient (log Koc).
1 Based on an empirical relationship, the EC value can be estimated from a compound's boiling point (BP; °C) using
the following equation: EC = 4.12 + 0.02 (BP) + 6.5 x 10~5 (BP)2; see Gustafson et al. (1997a).
2This range reflects EC values rounded to the nearest whole number. For instance, isopropylbenzene (EC = 8.66) is
included in this fraction because its EC value rounds to 9.
3 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
Toxicological considerations also contributed to the definition of the aromatic medium
carbon range fraction. Substituted benzenes (C9-C10; contained within the aromatic medium
carbon fraction) were grouped separately from PAHs, naphthalenes, and 1,1-biphenyl (contained
within the aromatic high carbon fraction), which generally exhibit greater carcinogenicity and
noncancer toxicity. Example compounds in the aromatic medium carbon range fraction include
isopropylbenzene (a C9 aromatic compound with an EC of 8.66) and //-butylbenzene (a
CIO aromatic compound with an EC of 9.96). The selection of relevant compounds and mixtures
is described in Section 2 and Appendix A.
1.2. OVERVIEW OF PHYSICOCHEMICAL PROPERTIES AND ENVIRONMENTAL
FATE
The systematic chemical names, synonyms [following guidance in Nist (2020bVI,
CASRNs, chemical abbreviations, chemical structures, and molecular weights for chemicals in
this document are listed in Table 1 and in Appendix B of U.S. EPA (2022). The physicochemical
properties for chemicals of the aromatic medium carbon range fraction that have toxicity values
compiled from the CompTox Chemicals Dashboard U.S. EPA (2021) are provided in Table 2.
Section 2 details how fraction members with toxicity values were identified. As Table 2 shows,
the eight chemicals with toxicity values include both C9 and C10 compounds. The eight
chemicals selected to represent the components of the aromatic medium carbon range fraction
are all liquids at room temperature with moderate water solubility and high vapor pressure. All
fraction members contain an alkyl substituted aromatic ring; three of the members contain three
methyl groups on an aromatic ring (CASRNs 108-67-8, 95-63-6, and 526-73-8), three have one
branched alkyl group on an aromatic ring (CASRNs 98-82-8, 98-06-6, and 135-98-8), and two
have one linear alkyl group on an aromatic ring (CASRNs 103-65-1 and 104-51-8). Members of
this fraction are expected to have negligible to slow mobility in soil. Volatilization may occur
from water and moist soil (based upon measured and estimated Henry's law constant values) and
from dry soil surfaces (based upon measured vapor pressure data); however, adsorption to soil is
expected to attenuate volatilization of the fraction members from soil. Measured aerobic and
anaerobic biodegradation data are available for the representative compounds. Aromatic
hydrocarbons typically have slow biodegradation rates under aerobic conditions and slow to no
biodegradation under anaerobic conditions. However, more rapid biodegradation has been
reported for some of the members of this fraction. For example, in activated sludge from a
predominantly domestic wastewater treatment plant, .sfc-butylbenzene and //-butylbenzene had
56-67 and 72-80% removal, respectively, in 5 days at 25°C at test substance concentrations of
100 mg/L NCBI (2022a. b). Members of the aromatic medium carbon range fraction do not
contain hydrolysable functional groups; therefore, the rate of hydrolysis is expected to be
negligible for all fraction members. In the atmosphere, the rate of photooxidation is expected to
range from slow (fert-butylbenzene) to rapid for the vapor-phase forms of the fraction members.
The fraction members do not contain chromophores that absorb at wavelengths >290 nm and are
therefore not expected to be susceptible to direct photolysis by sunlight Mm (2013. 2008a. b, c,
2005. 2004a. b, 2003).
4 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 1. Synonyms and Abbreviations for Chemicals in This Document"
Chemical Name
(common synonyms)b
CASRN
Abbreviation
Structure
Molecular Weight
(g/mol)
1,2,3-T rimethylbenzene
(benzene, 1,2,3-trimethyl-)
526-73-8
1,2,3-TMB
H3C.
CH3
^CH3
120.195
1,2,4-T rimethylbenzene
(benzene, 1,2,4-trimethyl-)
95-63-6
1,2,4-TMB
H3C\
H3CT
0
^ch3
120.195
1,3,5-T rimethylbenzene
(benzene, 1,3,5-trimethyl-)
108-67-8
1,3,5-TMB
ch3
•I
fH,
120.195
Isopropylbenzene
(cumene;
[propan-2-yl]benzene;
benzene, [1-methylethyl]-)
98-82-8
VI
H3C N '
120.195
HFAN
(light aromatic solvent naphtha
[petroleum];
solvent naphtha, petroleum, light
aromatic;
super high flash naphtha;
aromatic solvent;
solvent, aromatic petroleum;
solvent naphtha;
light aromatic solvent naphtha;
low boiling point naphtha-
unspecified;
solvent naphtha [petroleum], light
aromatic)
64742-95-6
Various
Various
w-Butylbenzene
(benzene, butyl-)
104-51-8
134.222
w-Propylbenzene
(propylbenzene;
benzene, propyl-)
103-65-1
120.195
5 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 1. Synonyms and Abbreviations for Chemicals in This Document"
Chemical Name
(common synonyms)b
CASRN
Abbreviation
Structure
Molecular Weight
(g/mol)
sec- Bu ty 1 ben ze n e
([butan-2-y ljbenzene;
benzene, [1-methylpropyl]-)
135-98-8
h3c
T
/CH3
134.222
tert- Bu ty 1 be n zen e
(benzene, [1,1-dimethylethyl]-)
98-06-6
H3(
ch3
: ch3
134.222
aOnly chemicals with toxicity values are listed.
' Synonyms are listed according to National Institute of Standards and Technology Nist (2020b') and include valid
synonyms from the U.S. EPA CompTox Chemicals Dashboard; https://comptox.epa.gov/dashboard: accessed
03-30-2020.
U.S. EPA = U.S. Environmental Protection Agency.
6 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 2. Physicochemical Properties of Aromatic Medium Carbon Range Chemicals with Toxicity Values"
Chemical
Isopropyl-
benzene
M-Propyl-
benzene
1,3,5-TMB
1,2,4-TMB
1,2,3-TMB
to*?-Butyl-
benzene
sec-Butyl-
benzene
rt-Butyl-
benzene
Structure
"VO
H;jC N '
A.
IIX,
ch3
H,C ,CH,
CH3
HSC CH3
6
2,
CASRN
98-82-8
103-65-1
108-67-8
95-63-6
526-73-8
98-06-6
135-98-8
104-51-8
Molecular formula
C9H12
C9H12
C9H12
C9H12
C9H12
C10H14
C10H14
C10H14
EC numberb
8.66
8.94
9.15
9.36
9.65
9.36
9.57
9.96
Molecular weight
(g/mol)
120.195
120.195
120.195
120.195
120.195
134.222
134.222
134.222
Melting point (°C)
-96
-99.6
-44.9
-46.0
-25.4
-58.3
-73.6
-88.0
Boiling point (°C)
152
159
164
169
176
169
174
183
Vapor pressure
(mm Hg at 25°C)
4.50
3.42
2.48
2.10
1.69
2.20
1.75
1.06
Henry's law constant
(atm-m3/mole at
25°C)
1.15 x 10-2
1.05x 10"2
8.77 x 10-3
6.16 x 10-3
4.36 x 10-3
7.89 x 10-3
8.03 x 10-3
8.05 x 10-3
Water solubility
(mol/L at 25°C)
5.24 x 10~4
4.65 x 10~4
3.99 x 10~4
4.78 x 10-4
6.30 x 10-4
2.19 x 10-4
1.30 x 10-4
8.73 x 10-5
Log Kow
3.66
3.71
3.50
3.63
3.63
4.11
4.57
4.38
Log Koa
3.98
4.09
4.54*
4.54*
4.54*
4.46*
4.31*
4.60*
Log Koc
3.00*
2.87
2.82
3.05*
2.80
3.34*
3.76*
3.39
'Data arc presented as experimental averages from the U.S. EPA CompTox Chemicals Dashboard unless otherw ise stated; https://eomptox.epa.gov/dashboard: updated
02-03-2021.
bEC number was developed by the TPHCWG and is proportional to the BP of a chemical. EC number is analogous to an //-paraffin retention time index and can be
estimated using: EC = 4.12 + 0.02 (BP) + 6.5 x 10~5 (BP)2 NIST (2020a: Edwards et al. (1997: Gustafson et at (1997b).
*Predicted value.
BP = boiling point; EC = equivalent carbon; Koa = octanol-air partition coefficient; Koc = soil adsorption coefficient; Kow = octanol-water partition coefficient;
TMB = trimethylbenzene; TPHCWG = Total Petroleum Hydrocarbon Criteria Working Group.
7
Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
1.3. OVERVIEW OF MIXTURE ASSESSMENT METHODS
A number of different approaches have been developed and used to estimate risks and
hazards posed by exposures to chemical mixtures encountered in the environment. Among the
simplest of these approaches to implement is the indicator chemical approach Atsdr (2018). The
indicator chemical approach estimates the risk or hazard of a mixture by evaluating the
dose-response assessment developed for a component of the mixture to the exposure rate of the
entire mixture. The indicator chemical approach is used when there are only measures of the
concentrations of this fraction (i.e., no information is available on the concentrations of
individual chemicals in this fraction) (see Section 1.3.1). The hazard index (HI) approach (the
other approach that will be addressed in this PPRTV assessment) can be used when there are
measured concentrations of specific chemical compounds. In the HI approach, the individual
chemical intake rates (or concentrations in the air) are divided by the reference dose (RfD) (or
reference concentration [RfC]) for the chemical to estimate a hazard quotient (HQ). The HQs are
summed to estimate the HI (see Section 1.3.2). The indicator chemical approach has greater
uncertainty than the HI approach (see Figure 1).
The U.S. Environmental Protection Agency's (U.S. EPA) Supplementary Guidance for
Conducting Health Risk Assessment of Chemical Mixtures U.S. EPA (2000, 1986) describes the
following two broad categories of approaches for assessing human health risks and health
hazards associated with environmental exposures to chemical mixtures: component methods and
whole mixture methods. Component-based approaches, which involve analyzing the toxicity of a
mixture's individual components, have more uncertainty and are recommended when toxicity
data on a complex mixture of concern, or on a sufficiently similar mixture (discussed below), are
unavailable U.S. EPA (2000, 1986). In this PPRTV, a component approach, the HI approach, is
described for assessing noncancer hazards posed by exposures to the aromatic medium carbon
range fraction.
Chemical mixture assessments are conducted most appropriately with quantitative
dose-response information resulting from comparable exposures to the mixture of concern. If the
dose-response data are insufficient to develop a health reference value for the specific mixture of
concern in the environment, the second option that the U.S. EPA's Supplementary Guidance for
Conducting Health Risk Assessment of Chemical Mixtures U.S. EPA (2000, 1986) recommended
is a "sufficient similarity" approach that uses a health reference value from a characterized
surrogate mixture to estimate the hazard or risk associated with exposures to the mixture of
concern. This method requires chemistry and toxicity data on both the potential surrogate
mixture and the mixture of concern (e.g., a key event that is related to the apical endpoint
observed in an epidemiological study or whole animal study), and a health reference value
(e.g., from an in vivo study) on the surrogate mixture. If the chemistry and toxicity data indicate
that the mixtures are "sufficiently similar" to one another, then the health reference value for the
surrogate mixture can be used as a proxy for the mixture of concern. No data were identified that
were suitable to implement a whole mixture approach.
The choice of a chemical mixtures risk assessment method is driven by the available data.
Starting with the method requiring the least information and then discussing the method
requiring more information, the following subsections summarize the indicator chemical
approach and the HI approach. Figure 1 summarizes the two approaches and the preference for
using each approach.
8 Aromatic medium carbon range TPH fraction
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
EPA 690 R-22 004F
Data driven approaches
Available Exposure Data
Fraction Measure
Aromatic medium carbon fraction
Approach
Indicator Chemical Approach
TMBs are indicator chemicals
Individual
Component Measures
Oral: n-propvlbenzene, n-
butylbenzene, sec-
butylbenzene, tert-
butylbenzene,
isopropylbenzene, or TMBs
Inhalation: n-propylbenzene,
isopropylbenzene, or TMBs
Hazard Index Approach
Component HQs; TMBs are a surrogate for
the remainder of the fraction mass HQ
Two approaches are available to estimate the noncancer hazards associated with exposure to the aromatic
medium range fraction. Approach selection should be driven by the available exposure data. Increased
analytical characterization of fraction components allows for more refined health hazard estimates with less
inherent uncertainty. Approach preference is inversely correlated with approach uncertainty.
HQ = hazard quotient; TMB = trimethylbenzene.
Figure 1. Provisional Peer-Reviewed Toxicity Approaches for the Aromatic Medium
Carbon Range TPH Fraction Noncancer Assessment
1.3.1. Indicator Chemical Approach
When the chemical composition of a mixture or a mixture fraction is not known, or
toxicity measures are not available for individual chemicals in a mixture, the toxicity of an
individual chemical can be used as an indicator for the toxicity of a mixture or a mixture fraction
Atsdr (2018). Atsdr (2018) describes an indicator chemical as "a chemical. . . selected to
represent the toxicity of a mixture because it is characteristic of other components in the mixture
and has adequate dose-response data." Indicator chemical approaches are typically implemented
to assess health risks in a health-protective manner; the chemical chosen as an indicator is among
the best characterized toxicologically and likely among the most potent components of the
mixture. The indicator chemical needs to have adequate dose-response data to indicate hazard
potential or a dose-response relationship for noncancer outcomes, depending on the purpose of
the assessment. The health risk value of the indicator chemical is integrated with exposure
estimates for the mixture or mixture fraction to estimate health hazard associated with the
fraction (i.e., calculate fraction-specific HI for a specific exposure pathway). This approach does
not scale for potency of individual constituents; instead, it assumes that the toxicity of all
measured members of the fraction can be adequately estimated, given the purpose of the risk
assessment, by the indicator chemical.
1.3.2. Hazard Index Approach
The HI approach combines estimated population exposures with toxicity information to
characterize the potential for adverse effects. The HI is not a risk estimate, in that it is not
expressed as a probability, nor is it an estimate of a toxicity measure (e.g., percentage decrement
9 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
1 in enzyme activity). Instead, the HI is an indicator of potential hazard. In the HI approach, a HQ
2 is calculated as the ratio of human exposure (E) to a health hazard reference value (RfV) for each
3 mixture component chemical (/) US t-'P-\ ()t)86). These HQs are summed to yield the HI for the
4 mixture. In health risk assessments, the U.S. EPA's preferred RfVs are the RfD for the oral
5 exposure route and the RfC for the inhalation exposure route.
A_J it—jfl/Vf
:=i ;=i
7 The HI is based on dose addition I' S T-P-V (2000; Svendsgaard and Hertzberu (1
8 the hazard is evaluated as the potency-weighted sum of the component exposures. The HI is
9 dimensionless, so E and the RfV must be in the same units.
10 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
2. SUMMARY OF TOXICITY AND DOSE-RESPONSE ASSESSMENT APPROACH
Toxicity and dose-response assessment for the aromatic medium carbon range fraction
entailed five basic steps, outlined here and described in more detail below. Mixtures and
compounds that met structural criteria (see Section 1.1) and had available toxicity values from
designated sources were identified. The dose-response assessment for the fraction can include
selection of a health reference value from an indicator chemical. Alternatively, if exposure
measurement data are available for component chemicals, those health reference values from
component chemicals with existing toxicity values can be used to apply the HI approach.
In Step 1 and Step 2 of the assessment, literature searches were performed for the
mixtures and compounds with toxicity values and for other mixtures and compounds that are
relevant to the fraction. Searches date-limited to assessments published from 2009 forward
(e.g., Agency for Toxic Substances and Disease Registry [ATSDR], Integrated Risk Information
System [IRIS]) were conducted in February 2018, and updated most recently in August 2021.
The start date of 2009 was selected to determine whether new information suggested that toxicity
values for mixtures or compounds relevant to the fraction should be updated from those
identified in the U.S. EPA (2009b) PPRTV for complex mixtures of aliphatic and aromatic
hydrocarbons. Step 3 in the assessment involved searching PubMed for new noncancer toxicity
data on compounds and mixtures lacking either Integrated Risk Information System (IRIS) oral
or inhalation toxicity values. These literature searches were conducted in February 2018, updated
in August 2021, and were date-limited to studies published from 2007 forward, in order to
capture studies that were published since the searches performed in U.S. EPA (2009b). Step 4 in
the assessment involved searching recent comprehensive reviews on the toxicity of petroleum
components or classes of compounds relevant to the fraction, as well as Organisation for
Economic Co-operation and Development (OECD) Screening Information Data Set (SIDS)
assessments3 and the Petroleum High Production Volume (HPV) Testing Group website, to
identify other mixtures or compounds within this carbon range with existing toxicity data that
may inform hazard identification for the fraction. Step 5 of the assessment involved informal
searches of PubMed to identify mixtures that met structural criteria of the fraction and had
toxicity data available. Toxicity data criteria included human studies of any duration by oral,
inhalation, and dermal exposure, and animal studies of oral or inhalation exposure lasting at least
28 days (or any duration of gestational exposure). Mixture toxicity data were considered relevant
only if the mixture composition under study was quantitatively defined to enable assessment of
relevance to the fraction. Figure 2 shows a schematic depiction of the process, and further detail
is provided below.
'The OECD Existing Chemicals Database (https://hpvchemicals.oecd.org') was reviewed for relevant categories, and
dossiers for the following categories were screened: diethylbenzenes (DEBs), C9 aromatic solvents, and C10-C13
aromatic solvents. In addition, the Oecd (1994) hazard characterization for 1.4-DEB was reviewed for relevant
toxicity data.
11 Aromatic medium carbon range TPH fraction
-------
EPA 690 R-22.004F
Step 4 Step 5 Step 2
Compounds and mixtures relevant to the aromatic medium carbon range fraction with available toxicity values or data were identified in a five-step
process. Table 3 lists individual compounds and mixtures and links them to their corresponding identification source during the literature search process.
Table 3 also indicates compounds with available toxicity values, or in the absence of toxicity values, available toxicity data.
ATSDR = Agency for Toxic Substances and Disease Registry: C = carbon; EC = equivalent carbon: HPV = High Production Volume: IRIS = Integrated
Risk Information System; OECD = Organisation for Economic Co-operation and Development; PPRTV = Provisional Peer-Reviewed Toxicity Value;
RfC = reference concentration; RID = reference dose; SIDS = Screening Information Data Set.
Figure 2. Selection of Compounds and Mixtures for Aromatic Medium Carbon Range Fraction Hazard Identification and
Dose-Response Assessment
12
Aromatic medium carbon range TPH fraction
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
EPA/690/R-22/004F
2.1. IDENTIFICATION OF RELEVANT MIXTURES AND COMPOUNDS WITH
TOXICITY VALUES
Step 1 (see Figure 2) in the assessment of the toxicity for the aromatic medium carbon
range fraction was to identify constituents of the fraction that have existing toxicity values from
any of the sources considered for the U.S. EPA (2009b) PPRTV assessment for complex
mixtures of aliphatic and aromatic hydrocarbons (these included IRIS, PPRTVs, ATSDR
Minimal Risk Levels [MRLs], Massachusetts Department of Environmental Protection
[MassDEP], Total Petroleum Hydrocarbon Criteria Working Group [TPHCWG], and Health
Effects Assessment Summary Tables [HEAST]). Of these sources, only IRIS, PPRTVs, and
ATSDR MRLs have been updated since 2009, so only these sources were consulted for toxicity
values. Based on the U.S. EPA's previous assessments and assessment activities as well as those
relevant chemicals reviewed by the MassDEP MassDep (2003) or TPHCWG Edwards et al.
0997), the U.S. EPA compiled an initial list of 12 chemicals and 1 mixture (i.e., high flash
aromatic naphtha [HFAN]). [See full list in Appendix A and description of approach and results
in Wang et al. (2012).! Table 3 lists the chemicals and mixtures for which Pub Med searches were
performed, published toxicity data were identified, and toxicity values were identified. At least
one updated subchronic or chronic oral or inhalation reference value was available for eight
chemicals (isopropylbenzene, //-propylbenzene, 1,3,5-trimethylbenzene [TMB], 1,2,4-TMB,
1,2,3-TMB, /m-butylbenzene, sec-butylbenzene, and //-butylbenzene), and one mixture
(HFAN).
13 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 3. Chemicals and Mixtures Identified in Literature Searches
CASRN
Chemical Name
Literature Search
Identification Source
PubMed
Searches
Performed
Toxicity
Values
Identified
Toxicity
Data
Identified
526-73-8
1,2,3-Trimethyl-
benzene
Initial list of
12 compounds + 1 mixture3, b,c
X
95-63-6
1,2,4-Trimethyl-
benzene
Initial list of
12 compounds + 1 mixture3, b,c
X
135-01-3
1,2-Diethylbenzene
Recent reviews of petroleum
toxicity'1'e
X
108-67-8
1,3,5-Trimethyl-
benzene
Initial list of
12 compounds + 1 mixture3, b,c
X
141-93-5
1,3 -Diethy lbenzene
Recent reviews of petroleum
toxicity'1'e
X
105-05-5
1,4-Diethylbenzene
Recent reviews of petroleum
toxicity'1'e
X
620-14-4
1 -Methy 1-3 -ethyl-
benzene
Initial list of
12 compounds + 1 mixture3
X
622-96-8
1 -Methy 1-4-ethyl-
benzene
Initial list of
12 compounds + 1 mixture'1 ' g
X
X
535-77-3
1-Methy 1-3-iso-
propy lbenzene
Initial list of
12 compounds + 1 mixture3
X
Various
Alkylbenzenes
(various)
PubMed searches'-h
X
X
25340-17-4
Diethy lbenzene
mixture
OECD SIDS1
X
64742-95-6,
88845-25-4, and
64742-94-5
High flash aromatic
naphtha
Initial list of
12 compounds + 1 mixture3'b
X
X
538-93-2
Isobuty lbenzene
Initial list of
12 compounds + 1 mixture3
X
98-82-8
Isopropy lbenzene
Initial list of
12 compounds + 1 mixture3, b,c
X
104-51-8
«-Buty lbenzene
Initial list of
12 compounds + 1 mixture3, b
X
X
103-65-1
«-Propy lbenzene
Initial list of
12 compounds + 1 mixture3, b
X
X
Various
Naphtha solvent
PubMed searchesf,J
X
X
99-87-6
/Msopropyltolucnc
Updated PPRTV, IRIS, and
ATSDR MRL databases'5
X1
135-98-8
sec-Buty lbenzene
Initial list of
12 compounds + 1 mixture3, b
X
X
14 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 3. Chemicals and Mixtures Identified in Literature Searches
CASRN
Chemical Name
Literature Search
Identification Source
PubMed
Searches
Performed
Toxicity
Values
Identified
Toxicity
Data
Identified
98-06-6
fcrt-Butylbcnzcnc
Initial list of
12 compounds + 1 mixturea b
X
X
aU.S. EPA developed the initial list of 12 chemicals and 1 mixture relevant to the aromatic medium carbon range.
The list included all individual hydrocarbons considered previously by the U.S. EPA's STSC in the evaluation of
hydrocarbons, as well as all those with toxicity data reviewed by the MassDEP MassDep (2003) or TPHCWG
Edwards et al. (19971.
bAt least one updated (since 2009) subchronic oral or inhalation reference value was available for these compounds
following searches of the IRIS, PPRTV, and ATSDR databases.
"Because these compounds had IRIS, PPRTV, or ATSDR toxicity values for both oral and inhalation routes, no
additional literature searches were performed.
•'These compounds/mixtures were identified in McK.ce et al. (2015). a recent review of petroleum toxicity.
"Toxicity data for these compounds were found in Gagnaire et al. (1990).
fThese compounds were identified in PubMed searches date-limited to studies published from 2007 forward,
conducted in February 2018.
'Toxicity data for this compound were found in Swiercz et al. (2000).
''Human olfaction data for these compounds were found in Cometto-Muniz and Abraham (2009).
'Toxicity data for this mixture were found in Oecd (2007).
J A human case report for this mixture was found in Magdalan et al. (2009).
kDuring review of the updated IRIS, PPRTV, and ATSDR MRL databases, these compounds were identified in the
PPRTV database as meeting structural criteria for inclusion and having toxicity assessments.
'This compound was not considered relevant to this assessment because this compound does not typically occur or
co-occur with petroleum contamination. The U.S. EPA (2011) PPRTV for isopropvltoluene identified no studies
investigating health effects following oral exposures for short-term, subchronic, or chronic durations; further, no
developmental or reproductive oral exposure studies were identified. The PPRTV reported data on
/?-isopropyltoluc lie-induced toxicity in animals exposed orally limited to a single-dose, acute toxicity study that
reported depression, coma, bloody lacrimation, diarrhea, irritability, and scrawny appearance in treated
Osbourne-Mendel rats. The PPRTV identified one subchronic inhalation study and two acute inhalation studies;
these reported mortality in mice, but not in rats or guinea pigs. A necropsy in the mice revealed hyperemic lungs,
mottled liver, and pale kidney.
ATSDR = Agency for Toxic Substances and Disease Registry; IRIS = Integrated Risk Information System;
MassDEP = Massachusetts Department of Environmental Protection; MRL = minimal risk level;
OECD = Organisation for Economic Co-operation and Development; PPRTV = provisional peer-reviewed toxicity
value; SIDS = Screening Information Data Set; STSC = Superfund Technical Support Center; TPHCWG = Total
Petroleum Hydrocarbon Criteria Working Group; U.S. EPA = U.S. Environmental Protection Agency.
1 In Step 2 (see Figure 2), all existing chemicals in the IRIS, PPRTV, and ATSDR MRL
2 databases were searched to determine whether they included any other compounds or mixtures
3 (not on the initial list) meeting the structural criteria for inclusion (C9-C10 and
4 EC9-EC <11 aromatics). Searches of the IRIS and ATSDR databases did not identify any
5 additional compounds, but review of the PPRTV database identified one additional compound
6 with a toxicity assessment (but no health risk values), />isopropyltoluene (p-cymene). Review of
7 the Petroleum Hydrocarbon Criteria Working Group's (1998) Selection of Representative TPH
8 Fractions Based on Fate and Transport Considerations Gustafson et al. (1997b) indicated that
9 this compound typically does not occur or co-occur with petroleum contamination, a finding
10 further confirmed by Ullmann's Encyclopedia of Industrial Chemistry Eggersdorfer (2012).
15 Aromatic medium carbon range TPH fraction
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
EPA/690/R-22/004F
Therefore, />-isopropyltoluene was not considered relevant to the assessment of the aromatic
medium carbon range fraction and is not discussed further. Table 4 shows the health risk values
available for the eight relevant compounds and the HFAN mixture.
Table 4. Summary of Available Toxicity Values for Mixtures and
Constituents of Aromatic Medium Carbon Range (C9-C10, EC9-EC <11)
Fraction"
CASRN
Name
C
EC
Oral Reference Dose
(mg/kg-d)
Inhalation Reference
Concentration (mg/m3)
Subchronic
Chronic
Subchronic
Chronic
98-82-8
Isopropylbenzene
9
8.66
-
0.1 (IRIS)
-
0.4 (IRIS)
103-65-1
//-Propylbcnzcnc
9
8.94
0.1b (PPRTV*)
0.1b (PPRTV*)
lb (PPRTV*)
lb (PPRTV*)
108-67-8
1,3,5-TMB
9
9.15
0.04 (IRIS)
0.01 (IRIS)
0.2 (IRIS)
0.06 (IRIS)
95-63-6
1,2,4-TMB
9
9.36
0.04 (IRIS)
0.01 (IRIS)
0.2 (IRIS)
0.06 (IRIS)
98-06-6
fer/-Butylbenzene
10
9.36
0.1c (PPRTV*)
0.1c (PPRTV*)
-
-
135-98-8
sec-Butylbenzene
10
9.57
0.1c (PPRTV*)
0.1c (PPRTV*)
-
-
526-73-8
1,2,3-TMB
9
9.65
0.04 (IRIS)
0.01 (IRIS)
0.2 (IRIS)
0.06 (IRIS)
104-51-8
//-Butylbcnzcnc
10
9.96
0.1 (PPRTV)
0.05 (PPRTV)
-
-
64742-95-6
HFAN
9-10
NA
0.3 (PPRTV*)
0.03 (PPRTV*)
1 (PPRTV)
0.1 (PPRTV)
"Toxicity values shown were selected from the following sources in order of preference: IRIS, PPRTV, ATSDR,
HEAST, MassDEP, or TPHCWG. None of the mixtures or constituents in this fraction has an OSF or IUR.
bBased on the identification of ethylbenzene (CASRN 100-41-4) as an appropriate analogue chemical.
°Based on the identification of isopropylbenzene (CASRN 98-82-8) as an appropriate analogue chemical.
* Screening provisional toxicity value. Screening values are not assigned confidence statements; however,
confidence in these values is presumed to be low. Screening values are derived when the data do not meet all
requirements for deriving a provisional toxicity value. Screening values are derived using the same methodologies
and undergo the same development and review processes (i.e., internal and external peer review, etc.) as
provisional values; however, there is generally more uncertainty associated with these values.
ATSDR = Agency for Toxic Substances and Disease Registry; C = carbon; EC = equivalent carbon;
HEAST = Health Effects Assessment Summary Tables; HFAN = high flash aromatic naphtha; IRIS = Integrated
Risk Information System; IUR = inhalation unit risk; MassDEP = Massachusetts Department of Environmental
Protection; NA = not applicable; OSF = oral slope factor; PPRTV = provisional peer-reviewed toxicity value;
TMB = trimethylbenzene; TPHCWG = Total Petroleum Hydrocarbon Criteria Working Group.
2.2. IDENTIFICATION OF OTHER RELEVANT TOXICITY DATA
Among the 12 compounds and 1 mixture on the initial list determined to be relevant to
the fraction, both oral and inhalation IRIS toxicity values were available for four compounds
(isopropylbenzene and the three TMB isomers). Therefore, these compounds were not included
in the comprehensive literature searches. In Step 3 (see Figure 2), literature searches were
conducted in PubMed to identify any new studies that could fill data gaps for the remaining eight
compounds and one mixture. The literature searches were conducted in February 2018, were
most recently updated in August 2021, and were date-limited to studies published from 2007
forward, in order to capture studies that were published since the searches performed for the
2009 PPRTV assessment for complex TPH mixtures. A summary of the literature search strategy
16 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
is provided in Appendix A. As detailed in the appendix, studies considered relevant to hazard
identification included animal studies using inhalation or oral exposure routes, in which
exposures continued for at least 28 days (or any duration of gestational exposure), at least one
health outcome was assessed, and an untreated or vehicle control group was included. Human
studies of any duration in which exposure was known or presumed to be through oral, inhalation,
or dermal routes and at least one health outcome was assessed were considered relevant.
The updated literature search identified two human studies: an acute human olfaction
study of alkylbenzenes Cometto-Muniz and Abraham (2009) and a case report of exposure to
naphtha solvent Magdatan et al. (2009). The only relevant animal study that was identified in the
updated literature search was a 4-week study of 1 -methyl-4-ethylbenzene Swiercz et al. (2000).
No relevant reviews or secondary sources were identified.
In Step 4 (see Figure 2), to determine whether additional relevant compounds or mixtures
had been tested for repeat-dose and/or reproductive/developmental toxicity since 2007, recent
reviews of petroleum toxicity McKee et al. (2015; Infante and Bingham (2012; OECD (2012a. b,
2007) and the Petroleum HPV Testing Group website were searched. Mixtures considered
relevant to the fraction met the following criteria (see Figure 2):
1. at least 90% of the mixture consisted of identified compounds within the C9-C10 and
EC9-EC <11 ranges.
2. 99% of the mixture consisted of aromatic compounds (<1% aliphatic).
3. the mixture had been tested in animals in at least one repeat-dose (>28 days) or
reproductive/developmental toxicity study using inhalation or oral exposure routes
and including an untreated or vehicle control group.
4. human mixture studies of any duration by oral, inhalation, and dermal exposure, and
animal studies of oral or inhalation exposure lasting at least 28 days (or any duration
of gestational exposure).
Using the same criteria, in Step 5, PubMed searches were conducted to identify mixtures
with relevant toxicity data.
None of the mixtures described on the Petroleum HPV Testing Group website met these
criteria. Oecd (2007) described toxicity data on a commercial diethylbenzene (DEB) mixture that
met these criteria. Thus, including HFAN, toxicity data for two mixtures were considered
potentially relevant to the assessment of the aromatic medium carbon fraction. HFAN consists of
TMB and ethyltoluene isomers; by definition, it must contain a combined total of 75% TMB and
ethyltoluene isomers (of which at least 22% is ethyltoluene and at least 15% is TMB) U.S. EPA
(2009c). Commercial DEB typically contains 60-65% 1,3-DEB, 27-30% 1,4-DEB, and 4-5%
1,2-DEB Oecd (2007). No further detail on the chemical compositions of these mixtures was
available in the sources reviewed.
Searching the review by McKee et al. (2015) resulted in the identification of three
additional compounds that met structural criteria and had toxicity data that met inclusion criteria
(see Figure 2): 1,2-, 1,3-, and 1,4-DEB. The three DEB isomers were tested for neurotoxicity in a
study of rats exposed orally for 10 weeks Gagnaire et al. (1990). No other mixtures or
compounds with toxicity data were identified. Human and animal studies that met criteria
17 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
outlined above were reviewed to support selection of surrogates for the aromatic medium carbon
range fraction toxicity values.
2.3. METHODS FOR INDICATOR CHEMICAL SELECTION
Only compounds or mixtures with at least one U.S. EPA or ATSDR toxicity value
(see Table 4) were to be considered for use as potential indicator chemicals (or indicator
mixtures) for derivation of the fraction-specific toxicity values; however, no ATSDR toxicity
values were identified. Toxicity data for other compounds that did not have toxicity values were
used for hazard identification and to assess consistency in effects and potency across the
components of the fraction. The method for selecting indicator chemicals was adapted from the
2009 complex TPH mixtures document U.S. EPA (2009b). First, mixtures consisting of fraction
component chemicals were preferred over individual compounds, provided that the mixture
study was adequate and the mixture exhibited in vivo toxic effects similar to those exhibited by
the individual fraction components. If suitable mixture data were lacking, a representative
compound exhibiting in vivo toxic effects and potency similar to those exhibited by other
compounds in the fraction was chosen. If the components of the fraction varied widely in toxic
effects or potency, the toxicity value for the most potent component (i.e., the component with
lowest reference value) was selected as an indicator chemical for the fraction. Finally, if toxicity
values were available for many or most of the individual compounds in a fraction, and these
compounds are typically monitored at sites of hydrocarbon contamination, then a component
approach would be considered.
2.4. DEVELOPMENT OF EXPOSURE-RESPONSE ARRAYS
In order to assess consistency in effects and potency across the components of the
fraction, experimental data from compound-specific IRIS and PPRTV documents and primary
data sources (identified from literature searches) were used to create exposure-response arrays
provided in Appendix B. Data were extracted only from certain studies (i.e., studies that
provided dose-response data enabling the identification of no-observed-adverse-effect levels
[NOAELs] and lowest-observed-adverse-effect levels [LOAELs]). Target-organ-specific
NOAELs and LOAELs were determined using the following methodology.
1. Whenever possible, NOAELs and LOAELs were identified from existing IRIS or
PPRTV assessments. For chemicals in which both types of assessments were
available, preference was given to IRIS (in accordance with U.S. EPA Office of
Superfund Remediation and Technology Innovation [OSRTI] hierarchy of human
health toxicity values for Superfund assessments). In general, these assessments
explicitly identified NOAEL and LOAEL values only for the most sensitive target of
toxicity, so characterization of additional adverse effect levels allowed for a
comprehensive comparison of toxic effects across additional endpoints and tissues.
2. All other target-organ-specific effect levels (i.e., for targets other than the most
sensitive target identified in IRIS or PPRTV assessments, and all targets evaluated in
newly identified studies) were determined using scientific judgment, taking into
consideration factors such as statistical significance (at ap-walue < 0.05), biological
significance (e.g., a greater than or equal to 10% increase in liver weight), magnitude
and direction of change, and study quality. In the case of chemicals with existing IRIS
or PPRTV assessments, NOAELs and LOAELs could often be identified from
existing study summaries.
18 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
1 Dose-response data were presented in exposure-response arrays by health outcome and
2 exposure route (see Appendix B). From left to right, compounds exhibiting an effect are shown
3 before those not exhibiting an effect, to facilitate identification of patterns. Within the group
4 exhibiting an effect, compounds are ordered from lowest LOAEL to highest. For compounds that
5 do not exhibit an effect, NOAELs in the arrays are ordered by EC number (low to high from left
6 to right), with mixtures shown last. Both administered doses and exposure concentrations
7 reported in the arrays and in text reflect time-weighted average (TWA) exposures, to facilitate
8 comparisons across studies and compounds. Consistency across the fraction was evaluated by
9 assessing if comparable outcomes were observed for members of the fraction, and if these effects
10 were observed at similar dose levels.
19 Aromatic medium carbon range TPH fraction
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
EPA/690/R-22/004F
3. REVIEW OF POTENTIALLY RELEVANT DATA
3.1. NONCANCER EVIDENCE
Compound-specific IRIS and PPRTV documents, supplemented by the literature search
findings and the review articles described above [particularly McKee et al. (2015)1 were assessed
to evaluate available noncancer data for the aromatic medium carbon range fraction compounds.
Critical effects identified with existing toxicity values include neurological, hepatic, renal,
body-weight, hematological, endocrine, respiratory, and developmental effects. Appendix B
summarizes the evidence provided by human and experimental animal studies of noncancer
health outcomes. Table 5 presents an overview of the human and animal data available to
evaluate the primary toxicological endpoints identified for the fraction (neurological, hepatic,
renal, body weight, hematological, endocrine, respiratory, and developmental). As Table 5
shows, both oral and inhalation data available to assess consistency in effects across members of
the fraction are limited. There are no dependable human or animal data for at least three
members of the fraction (//-propylbenzene, and tert- and .sec-butylbenzene). There are oral or
inhalation body-weight data for 11 members, and there are neurotoxicity endpoint data available
for 9 members. For all other primary toxicological endpoints, there are oral or inhalation data for
5-7 members of the fraction. Most of the animal data are from inhalation toxicity studies.
Comprehensive systemic toxicity was evaluated in rats and mice in subchronic and chronic
inhalation studies for one member of the fraction (isopropylbenzene). In general, studies ranged
in duration from 4 to 18 weeks; several of these studies (e.g., DEBs and TMBs) evaluated only
neurological endpoints. Developmental inhalation toxicity studies were available for four
members of the fraction (isopropylbenzene, 1,3,5- and 1,2,4-TMB, and HFAN). Finally, unless
otherwise specified, the term "significant," used throughout the document, refers to statistical
significance at ap-walue of < 0.05.
20 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 5. Overview of Human and Animal Data Availability for Evidence Integration3'b
CASRN
Name
C
EC
Neurological
Hepatic
Renal
Body
Weight
Hemato-
logical
Endocrine
Respiratory
Develop-
mental
98-82-8
Isopropylbenzene
9
8.66
I
I
1,0
I
I
I
I
I
103-65-1
M-Propylbenzenec
9
8.94
622-96-8
1 -Methy 1-4-ethy lbenzene
9
9.07
I
I
108-67-8
1,3,5-T rim ethylbenzene
9
9.15
H, I
O
O
0,1
H, 0
H
I
95-63-6
1,2,4-T rim ethylbenzene
9
9.36
H, I
I
I
I
H, I
I
H, I
I
98-06-6
tert- Bu ty 1 be n zen ec
10
9.36
135-98-8
sec- Bu ty 1 ben ze n e'
10
9.57
526-73-8
1,2,3-T rim ethylbenzene
9
9.65
H, I
I
I
I
H, I
I
H, I
141-93-5
1,3 -Diethy lbenzene
10
9.91
O
0
105-05-5
1,4-Diethy lbenzene
10
9.96
O
0
0
0
0
0
104-51-8
w-Butylbenzene
10
9.96
0
0
0
0
0
135-01-3
1,2-Diethy lbenzene
10
9.96
O
0
64742-95-6
HFAN
9-10
NA
I
0,1
0,1
0,1
0,1
0
0,1
NA
Diethylbenzenes (mixture)
10
NA
0,1
0,1
includes human and animal studies meeting inclusion criteria. Bolded compounds and mixtures have at least one oral or inhalation toxicity value available (see Table 4).
bCompounds are arranged by increasing EC number.
°In the absence of human or animal data, screening toxicity values were derived using appropriate analogue chemicals (ethylbenzene and isopropylbenzene) in the
PPRTV assessments of these compounds.
C = carbon; EC = equivalent carbon; H = human data; HFAN = high flash aromatic naphtha; I = animal inhalation studies; NA = not applicable; O = animal oral studies.
21
Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
Critical effects used to derive oral or inhalation toxicity values for the aromatic medium
carbon range fraction compounds and mixtures include neurological effects (decreased pain
sensitivity), hepatic toxicity (hepatocellular hypertrophy), renal toxicity (increased kidney weight
and histopathology), decreased body weight, hematological toxicity, endocrine system toxicity,
and developmental toxicity (decreased fetal/pup body weights and delayed skeletal ossification).
The available data for most of the aromatic medium carbon range compounds and mixtures are
limited for endpoints other than body weight and are altogether absent for three members of the
fraction (//-propylbenzene, and tert- and .sfc-butylbenzene). A majority of toxicity data are from
subchronic studies of the inhalation route of exposure, and few compounds have been tested for
toxicity following chronic oral or inhalation exposure.
Based on review of the available oral and inhalation toxicity data, there is evidence that
several members of the fraction, and especially TMBs and DEBs, can induce neurological
effects; however, most of the compounds in the fraction have not been evaluated for sensitive
measures of neurological function. Information across half of the compounds (oral and inhalation
exposure) composing the fraction suggests that aromatic medium carbon range fraction
compounds and mixtures can induce hepatic effects in the form of increased liver weight, often
accompanied by histological effects (most frequently hepatocellular hypertrophy via oral
exposure). Similarly, data show that several members of the aromatic medium carbon range
fraction induce significant increases in relative kidney weight after oral or inhalation exposure;
this effect was seen (infrequently) in conjunction with serum chemistry changes (i.e., increased
blood urea nitrogen [BUN] after oral exposure to 1,4-DEB) and in the absence of corresponding
histological changes (other than rat-specific male nephropathy). Data on body-weight effects
after oral and inhalation exposure to a variety of aromatic medium carbon range fraction
compounds and mixtures indicate that members of the fraction can be expected to induce
body-weight reductions generally at high doses.
The available data are not considered adequate to evaluate consistency in effects or
potencies across fraction members for hematological or endocrine effects. Because there are data
for only five compounds, data are insufficient to determine if respiratory effects are consistently
associated with inhalation exposure to members of the aromatic medium carbon range fraction
(there are no oral exposure data on respiratory effects). Finally, data from oral and inhalation
developmental toxicity studies consistently identify decreased fetal body weights and delays in
skeletal development for several members of the aromatic medium carbon range fraction.
3.2. CANCER EVIDENCE
3.2.1. Human Studies
No human studies were available to address the carcinogenic potential of the TMB
isomers or other members of the aromatic medium carbon range fraction by any route of
exposure.
3.2.2. Animal Studies—Oral
A single carcinogenicity study in rats orally exposed to 1,2,4-TMB for 104 weeks was
identified Clark et al.. 1989 as cited in U.S. EPA (2009c). The only noteworthy finding was a
nonsignificant increase in the incidence of neuroesthesioepitheliomas (3/100 treated animals
based on the combined sexes compared to 0/100 controls). Several study limitations were
apparent, including use of one rodent species, treatment at a single dose level (800 mg/kg-day),
and lack of quantitative mortality data ("slight" or "intermediate" reductions in survival were
22 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
reported). Therefore, the available oral data are not sufficient to adequately assess the
carcinogenic potential of 1,2,4-TMB (or other members of the aromatic medium carbon range
fraction).
3.2.3. Animal Studies—Inhalation
No neoplasms were reported in rats treated with HFAN for 12 months Clark et al., 1989
as cited in U.S. EPA (2009c).
New studies identified in the PubMed searches included a 105-week chronic toxicity/
carcinogenicity study of isopropylbenzene in rats and mice by the National Toxicology Program
Ntp (2009). Statistically significantly increased incidences of respiratory epithelial adenomas of
the nose in both sexes and renal adenoma or carcinoma (combined) in males were observed in
rats. Increased interstitial cell adenomas were also reported in the male testis; however, the NTP
report stated that these were possibly related to isopropylbenzene exposure. While the incidence
of interstitial cell adenomas reported in the highest dose group in the male rats was statistically
significantly increased compared to the control group and there was a positive trend in the
incidences reported among all exposed groups, the incidence in the high-dose group was within
the range for historical chamber controls when studies with all exposure routes were considered.
Interstitial cell hyperplasia and adenoma are common proliferative lesions in F344/N rats
(i.e., the test species) and reportedly will develop in nearly all male rats of this strain that are
allowed to complete their natural life span Ntp (2009). In mice, the incidences of alveolar/
bronchiolar adenomas were significantly increased in both sexes; increased incidences of
hemangiosarcomas and follicular cell adenomas in males4 (possibly related to exposure) and
hepatocellular adenomas or carcinomas in females also were reported. Based on these data, the
NTP concluded that there was clear evidence of carcinogenicity in male rats and male and
female mice, and some evidence of carcinogenic activity in female rats.
No studies evaluating carcinogenicity were available for other members of the aromatic
medium carbon range fraction.
3.2.4. Summary of Cancer Evidence
Few data are available to assess the carcinogenic potential of compounds and mixtures in
the aromatic medium carbon range fraction. No human data were identified. Animal data are
limited to 1,2,4-TMB and isopropylbenzene. Several limitations were identified in the
carcinogenicity study of 1,2,4-TMB. Only data from a newly identified study for
isopropylbenzene Ntp (2009) are considered adequate to assess carcinogenic potential of
individual fraction members. At this time, U.S. EPA has not formally evaluated the Ntp (2009)
study and has not estimated the cancer potency associated with the study results.
4NTP described the increased incidence of hemangiosarcomas in males as "equivocal;" these tumors were observed
only in the highest dose group. The incidences of follicular cell adenoma increased with a statistically significant
positive trend in male mice; however, the incidence in the highest dose group was at the upper end of the historical
ranges for chamber controls in inhalation studies and for historical controls (all exposure routes). NTP described
these increases to be "possibly related to cumene exposure."
23 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
4. TOXICOKINETIC CONSIDERATIONS
Reviews of toxicokinetic information on aromatic medium carbon chain length
hydrocarbons have been performed by McKee et al. (2015) and Infante and Bingham (2012). In
general, these chemicals are well absorbed by oral and inhalation exposure and distributed
widely throughout the body, with initial concentration in adipose tissue. Metabolism is efficient
and occurs primarily via oxidation and conjugation of the alkyl side chains off the aromatic ring.
These water-soluble metabolites are rapidly eliminated in the urine.
Compounds in the aromatic medium carbon range fraction are readily absorbed following
oral exposure. F344 rats exposed to radiolabeled isopropylbenzene (cumene) by gavage showed
maximum blood concentration levels 4 hours after dosing (the earliest time point sampled) at
33 mg/kg and 8-16 hours after dosing at 1,350 mg/kg U.S. EPA (1997). Based on recovery of
urinary metabolites (time point not specified), absorption exceeded 70%. Absorption of
/>isopropyltoluene (cymene) was at least 60-80% in rats and guinea pigs treated at 100 mg/kg
and 52-74%) in rats and various marsupial species treated at 50 or 200 mg/kg, based on urinary
metabolites excreted within 48 hours U.S. EPA (2011). Absorption of n- and /c/V-butylbenzene
was similarly found to exceed 66-81 % in rabbits U.S. EPA (2012b). More than 99%> of an orally
administered dose of 1,2,4-TMB was absorbed (and subsequently eliminated) within 24 hours in
rats McKee et al. (2015).
Absorption following inhalation exposure is rapid and extensive. Isopropylbenzene was
detected in the blood of F344 rats within 5 minutes of the start of inhalation exposure U.S. EPA
(1997). Based on recovery of urinary metabolites (time point not specified), absorption exceeded
70%o in rats exposed to 100, 500, or 1,500 ppm isopropylbenzene U.S. EPA (1997). Mean
respiratory tract retention was reported to be 50% (45-64%) in volunteers exposed to
isopropylbenzene at 240, 480, or 720 mg/m3 for 8-hour periods U.S. EPA (1997). Respiratory
retentions for the three TMB isomers were approximately 70% in human subjects exposed at
5-150 mg/m3 for 8 hours and 60% in subjects exposed to 2-25 ppm (10-125 mg/m3) for 2 hours
while performing light activity McKee et al. (2015; Infante and Bingham (2012). Consistent with
these results, measured blood-air partition coefficients for these compounds are high in humans
and laboratory animals. Human blood-air partition coefficients are 47 for //-propylbenzene and
37 for isopropylbenzene U.S. EPA (2009d). Blood-air partition coefficients for TMB isomers
(1,2,4-TMB, 1,2,3-TMB, and 1,3,5-TMB, respectively) are 59.1, 66.5, and 43.0 in humans and
57.7, 62.6, and 55.7 in rats U.S. EPA (2016b).
Compounds in the aromatic medium carbon range fraction are widely distributed in the
body after inhalation or oral exposure. Studies with isopropylbenzene, TMB isomers, and
/m-butylbenzene all showed elevated concentrations of the administered chemical in fat (also
stomach by oral route) » kidney > liver > brain >blood in rats after exposure, regardless of
route U.S. EPA (2016a, 2012b. 1997). Levels in fat were 10- to 100-fold higher than in other
tissues. Modeling of tissue-air partition coefficients suggests that tissue distribution in humans
would be similar to that seen in rats U.S. EPA (2016a; McKee et al. (2015). Due to the high
lipophilicity of these chemicals, it has been estimated that 85% of alkylbenzene in the blood is
bound to red blood cells (RBCs) U.S. EPA (2012b). TMBs and /c/v-butylbenzene have been
shown to cross the placenta U.S. EPA (2016a; Infante and Bingham (2012).
24 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
Similar to absorption, metabolism of aromatic medium carbon range compounds is rapid
and extensive. Metabolism proceeds primarily by oxidation of side chains on the aromatic ring to
form the corresponding alcohols and carboxylic acids, followed by conjugation with glycine
(hippuric acid), cysteine (mercapturic acid), glucuronic acid, or sulfate U.S. EPA (2016a; McKee
et al. (2015; U.S. EPA (2012b. 2011. 2009d. 1997). Example chemicals (and principal
metabolites) are isopropylbenzene (2-phenyl-2-propanol and its glucuronide and sulfate
conjugates) U.S. EPA (1997). //-propylbenzene (glucuronides of ethyl phenyl carbinol and benzyl
methyl carbinol) U.S. EPA (2009d). 1,2,4-TMB (3,4-dimethylhippuric acid) U.S. EPA (2016a).
1,3,5-TMB (3,5-dimethylhippuric acid) McKee et al. (2015), /c/7-butylbenzene
(2,2-dimethyl-2-phenylethyl glucuronide) U.S. EPA (2012b). and isopropyltoluene
(2-[4-methylphenyl]propan-l-ol, 2-[4-methylphenyl]propan-2-ol) U.S. EPA (2011). Oxidation of
the aromatic ring to form the corresponding phenol is a minor metabolic pathway for at least
some of these chemicals U.S. EPA (2016a; Infante and Bingham (2012; U.S. EPA (2011).
Where data in multiple species are available, metabolic profiles are similar in rats, rabbits, and
humans U.S. EPA (2016a; McKee et al. (2015). Metabolism of aromatic medium carbon range
compounds occurs in the liver, lung, and other tissues [e.g., kidney, adrenal, brain, and bone
marrow U.S. EPA (2012b. 2009d. 1997)1. Several of these compounds have been shown to
induce metabolic enzymes, and therefore, their own metabolism McKee et al. (2015; U.S. EPA
(2012b. 2009d). There is some experimental evidence for saturation of metabolism at high
exposure levels: blood concentrations of 4-ethyltoluene in rats were 10-fold higher after a single
6-hour exposure at 1,000 mg/m3 than at 250 mg/m3 [fourfold difference in exposure
concentration McKee et al. (2015)1.
Metabolites of aromatic medium carbon range compounds are water soluble and rapidly
excreted in the urine McKee et al. (2015; U.S. EPA (2012b). Less than 1% of the absorbed
fraction remained in the body 72 hours after inhalation exposure of isopropylbenzene at
1,200 ppm to rats U.S. EPA (1997). Similarly, >99% of an oral dose of 1,2,4-TMB was
eliminated as metabolites in the urine within 24 hours after dosing in rats McKee et al. (2015). In
studies of ^-isopropyltoluene, at least 60-80% of an oral dose in rats and guinea pigs and
52-74%) of an oral dose in rats and various marsupial species was excreted as metabolites in the
urine within 48 hours U.S. EPA (2011). Small amounts of unchanged parent compound may also
be found in the urine U.S. EPA (2016a. 2012b). Following inhalation exposure, unchanged
parent compound may be exhaled via the lungs. Human subjects who retained 60%> of inhaled
TMB in the lung subsequently exhaled approximately 30%> of the retained material Infante and
Bingham (2012). In one human study, elimination of 1,3,5-TMB via breath was biphasic, with an
initial half-life of 60 minutes and a terminal half-life of 600 minutes U.S. EPA (2016a; McKee et
al. (2015). Breath concentrations of 1,3,5-TMB in this study returned to pre-exposure levels
within 24 hours McKee et al. (2015).
25 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
5. MECHANISTIC CONSIDERATIONS AND GENOTOXICITY
Mechanistic information for health effects associated with exposure to compounds in the
aromatic medium carbon range is limited. There is evidence that renal histopathology induced by
isopropylbenzene in male rats reflects an alpha 2u-globulin (a2u-g)-specific nephropathy that is
specific to male rats, and is therefore not an appropriate endpoint for human health risk
assessment U.S. EPA (1997). Similar changes were noted in one study of an HFAN mixture U.S.
EPA (2009c). Renal effects in studies of other fraction members were limited primarily to
increases in kidney weight (see discussion of renal effects in Section 3.1). For isopropylbenzene,
increases in kidney weight were unrelated to a2u-g-specific nephropathy, as they occurred in
females as well as males [increased kidney weight in females was the critical effect for both the
RfD and RfC for isopropylbenzene U.S. EPA (1997)1. Therefore, increases in renal weight
associated with other fraction members does not necessarily result from a2u-g-specific
nephropathy. In fact, there is no evidence for a2u-g-specific nephropathy among specific fraction
member compounds other than isopropylbenzene (HFAN is a mixture of C8-C10 aromatics that
can include isopropylbenzene).
Genotoxicity data for aromatic medium carbon range compounds primarily indicate little
to no genotoxic potential. Almost all relevant mixtures or compounds were negative with or
without metabolic activation in in vitro tests for point mutations in bacteria or mammalian cells
U.S. EPA (2016a: OECD (2012a. b; U.S. EPA (2012b. 2011. 2009c. d; Oecd (2007; U.S. EPA
(1997; Oecd (1994). The only exception was 1,2,3-TMB, which produced reverse mutations in
Salmonella without, but not with, metabolic activation U.S. EPA (2016a). All three TMB
isomers were at least weakly positive for sister chromatid exchange (SCE) in mice tested in vivo
U.S. EPA (2016a). However, the larger C9 fraction to which TMB belongs was negative for SCE
in Chinese hamster ovary (CHO) cells in vitro Oecd (2012b). Other tests of fraction members
(including the TMB isomers) for clastogenicity (chromosomal aberrations [CAs] or
micronucleus [MN] formation) in rodents in vivo or in rodent cells in vitro were negative or
equivocal U.S. EPA (2016a; OECD (2012a. b; U.S. EPA (2012b; Oecd (2007; U.S. EPA (1997;
Oecd (1994). By contrast, clastogenicity was reported for a commercial HFAN mixture in human
lymphocytes with activation U.S. EPA (2009c). This HFAN mixture also was reported to
produce deoxyribonucleic acid (DNA) damage in Escherichia coli without activation U.S. EPA
(2009c). The only other finding relevant to DNA damage was equivocal evidence for
unscheduled DNA synthesis (UDS) in rat hepatocytes by isopropylbenzene U.S. EPA (1997).
There was also equivocal evidence for BALB/3T3 cell transformation by isopropylbenzene U.S.
EPA (1997).
26 Aromatic medium carbon range TPH fraction
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
EPA/690/R-22/004F
6. DERIVATION OF PROVISIONAL VALUES
6.1. DERIVATION OF ORAL REFERENCE DOSES
Subchronic provisional reference doses (p-RfDs) or RfDs are available for eight
constituents of the fraction. The critical effects for these subchronic toxicity values are liver and
kidney toxicity (//-propylbenzene, based on analogy to ethylbenzene), decreased pain sensitivity
(1,3,5-, 1,2,4-, and 1,2,3-TMB), increased kidney weight (tert- and sec-butylbenzene, based on
analogy to the chronic RfD for isopropylbenzene), liver histology (//-butylbenzene), and anemia
(HFAN). There are nine available chronic RfDs for constituent compounds (isopropylbenzene, in
addition to the compounds identified above). The chronic RfDs are based on the same studies
and the same points of departure (PODs) as the corresponding subchronic RfDs. The chronic
RfD for isopropylbenzene is based on increased kidney weights. Table 6 summarizes the
subchronic and chronic RfDs for constituent compounds and mixtures, with PODs, uncertainty
factors, critical effects, and confidence descriptors. As shown in Table 6 and discussed in
Appendix B, the data available to assess consistency in critical effects across members of the
fraction are limited for effects on endpoints other than body weight. The potencies with RfDs
being within one order of magnitude of one another are comparable.
27 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 6. Available RfD Values for Aromatic Medium Carbon Range Fraction (C9-C10, EC9-EC < ll)a
Indicator Chemical or
Components
POD
(mg/kg-d)
POD Type
UFc
UF
Components
p-RfD or
RfD
(mg/kg-d)
Confidence in
p-RfD or RID '
Critical Effect(s)
Species, Mode,
and Duration
Primary Reference
(source)
Subchronic
//-Propylbcnzcnc (C9,
EC8.94)
97.1
NOELadj
1,000
UFa, UFh,
UFS
0.1h
Low
Based on
ethylbenzene as an
analogue; increased
liver and kidney
weights (hepatic,
urinary);
histopathologic
changes in kidney
Rat, gavage,
5 d/wkfor 182 d
Wolf (1956s) as cited
in U.S. EPA (2009d)
1,3,5-T rimethylbenzene
(C9, EC9.15)
3.5
BMDL
(HED)
100
UFa, UFd,
UFh
0.04
Low
Decreased pain
sensitivity in male
Wistar ratsc
(nervous)
Rat, 6 h/d, 5 d/wk
for 13 wk
Korsak and
Rvd/vnski (1996) as
cited in U.S. EPA
(2016b)
1,2,4-T rimethylbenzene
(C9, EC9.36)
3.5
BMDL
(HED)
100
UFa, UFd,
UFh
0.04
Low
Decreased pain
sensitivity in male
Wistar ratsc
(nervous)
Rat, 6 h/d, 5 d/wk
for 13 wk
Korsak and
Rvd/vnski (1996) as
cited in U.S. EPA
(2016b)
fer/-Butylbenzene (CIO,
EC9.36)
110
NOAELadj
1,000
UFa, UFd,
UFh, UFs
0.1h
Low
Based on
isopropylbenzene as
an analogue;
increased kidney
weight (urinary)
Rat, 5 d/wk for
194 d
Wolf (1956) as cited
U.S. EPA (2012b)
sec-Butylbenzene (CIO,
EC9.57)
110
NOAELadj
1,000
UFa, UFd,
UFh, UFs
0.1b
Low
Based on
isopropylbenzene as
an analogue;
increased kidney
weight (urinary)
Rat, 5 d/wk for
194 d
Wolf (1956) as cited
in U.S. EPA (2012a)
1,2,3-T rimethylbenzene
(C9, EC9.65)
3.5
BMDL
(HED)
100
UFa, UFd,
UFh
0.04
Low
Decreased pain
sensitivity in male
Wistar ratsc
(nervous)
Rat, 6 h/d, 5 d/wk
for 13 wk
Korsak and
Rvd/vnski (1996) as
cited in U.S. EPA
(2016b)
28
Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 6. Available RfD Values for Aromatic Medium Carbon Range Fraction (C9-C10, EC9-EC < ll)a
Indicator Chemical or
Components
POD
(mg/kg-d)
POD Type
UFc
UF
Components
p-RfD or
RfD
(mg/kg-d)
Confidence in
p-RfD or RID '
Critical Effect(s)
Species, Mode,
and Duration
Primary Reference
(source)
«-Butylbenzene (CIO,
EC9.96)
137
BMDL
1,000
UFa, UFd,
UFh
0.1b
Low
Increased incidence
of hepatocellular
hypertrophy in F0 and
Fi parent male rats
(hepatic)
Rat, gavage,
two-generation
Izumi et al. (2005) as
cited in U.S. EPA
(2010)
HFAN (C9-C10)
85
BMDL
300
UFa, UFd,
UFh
0.3b
Low
Mild anemia,
evidenced by a
decrease in RBC
count (hematological)
Dog, gelatin
capsules, 13 wk
Biodvnamics
(1990b) as cited in
U.S. EPA (2009c)
Chronic
Isopropylbenzene (C9,
EC8.66)
110
NOAELadj
1,000
UFa, UFd,
UFh, UFs
0.1
Low-medium
Increased average
kidney weight in
female Wistar rats
(urinary)
Rat, 5 d/wk for
194 d
Wolf (1956) as cited
in U.S. EPA (1997)
//-Propylbcnzcnc (C9,
EC8.94)
97.1
NOELadj
1,000
UFa, UFh,
UFS
0.1h
Low
Based on
ethylbenzene as an
analogue; increased
liver and kidney
weights (hepatic,
urinary)
Rat; gavage;
5 d/wk for 182 d
Wolf (1956) as cited
in U.S. EPA (2009a)
1,3,5-T rimethylbenzene
(C9, EC9.15)
3.5
BMDL
(HED)
300
UFa, UFd,
UFh, UFs
0.01
Low
Decreased pain
sensitivity in male
Wistar ratsc
(nervous)
Rat, 6 h/d, 5 d/wk
for 13 wk
Korsak and
Rvd/vnski (1996) as
cited in U.S. EPA
(2016b)
1,2,4-T rimethylbenzene
(C9, EC9.36)
3.5
BMDL
(HED)
300
UFa, UFd,
UFh, UFs
0.01
Low
Decreased pain
sensitivity in male
Wistar ratsc
(nervous)
Rat, 6 h/d, 5 d/wk
for 13 wk
Korsak and
Rvd/vnski (1996) as
cited in U.S. EPA
(2016b)
29
Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 6. Available RfD Values for Aromatic Medium Carbon Range Fraction (C9-C10, EC9-EC < ll)a
Indicator Chemical or
Components
POD
(mg/kg-d)
POD Type
UFc
UF
Components
p-RfD or
RfD
(mg/kg-d)
Confidence in
p-RfD or RID '
Critical Effect(s)
Species, Mode,
and Duration
Primary Reference
(source)
fcrt-Butylbcnzcnc (CIO,
EC9.36)
110
NOAELadj
1,000
UFa, UFd,
UFh, UFs
0.1h
Low
Based on
isopropylbenzene as
an analogue;
increased kidney
weight (urinary)
Rat, 5 d/wk for
194 d
Wolf (1956s) as cited
U.S. EPA (2012b)
sec-Butylbenzene (CIO,
EC9.57)
110
NOAELadj
1,000
UFa, UFd,
UFh, UFs
0.1h
Low
Based on
isopropylbenzene as
an analogue;
increased kidney
weight (urinary)
Rat, 5 d/wk for
194 d
Wolf (1956s) as cited
in U.S. EPA (2012a)
1,2,3-T rimethylbenzene
(C9, EC9.65)
3.5
BMDL
(HED)
300
UFa, UFd,
UFh, UFs
0.01
Low
Decreased pain
sensitivity in male
Wistar ratsc
(nervous)
Rat, 6 h/d, 5 d/wk
for 13 wk
Korsak and
Rvdzvnski (1996)
as cited in U.S. EPA
(2016b)
«-Butylbenzene (CIO,
EC9.96)
137
BMDL
3,000
UFa, UFd,
UFh, UFs
0.05b
Low
Increased incidence
of hepatocellular
hypertrophy in F0 and
Fi parent male
Cij:CD (SD) IGS rats
(hepatic)
Rat, gavage,
two-generation
Izumi et al. (2005) as
cited in U.S. EPA
(2010)
30
Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 6. Available RfD Values for Aromatic Medium Carbon Range Fraction (C9-C10, EC9-EC < ll)a
Indicator Chemical or
Components
POD
(mg/kg-d)
POD Type
UFc
UF
Components
p-RfD or
RfD
(mg/kg-d)
Confidence in
p-RfD or RID '
Critical Effect(s)
Species, Mode,
and Duration
Primary Reference
(source)
HFAN (C9-C10)
85
BMDL
3,000
UFa, UFd,
UFh, UFs
0.0 3h
Low
Mild anemia,
evidenced by a
decrease in RBC
count (hematological)
Dog, gelatin
capsules, 13 wk
Biodvnamics
(1990b) as cited in
U.S. EPA (2009c)
11 Bo I (led row shows the compound and toxicity value selected as the indicator chemical for the fraction if analytical chemistry data do not identify concentrations of
individual chemicals composing this fraction.
bToxicity values are provisional values obtained from an existing PPRTV assessment. Values in italics are screening provisional values obtained from an existing
PPRTV assessment. Screening provisional values are not assigned confidence statements; however, confidence in these values is presumed to be low. Screening
provisional values are derived when the available data do not meet the requirements for deriving a provisional toxicity value.
Toxicity values based on route-to-route extrapolation (inhalation to oral) using a modified PBPK model.
ADJ = adjusted; BMDL = benchmark dose lower confidence limit; C = carbon; EC = equivalent carbon; HED = human equivalent dose; HFAN = high flash aromatic
naphtha; NOAEL = no-observed-adverse-effect level; NOEL = no-observed-effect level; PBPK = physiologically based pharmacokinetic; POD = point of departure;
PPRTV = provisional peer-reviewed toxicity value; RBC = red blood cell; p-RfD = provisional reference dose; RfD = reference dose; SD = standard deviation;
UF = uncertainty factor; UFA = interspecies uncertainty factor; UFC = composite uncertainty factor; UFD = database uncertainty factor; UFH = intraspecies uncertainty
factor; UFS = subchronic-to-chronic uncertainty factor.
31
Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
6.1.1. Oral Noncancer Assessment Using the Indicator Chemical Method for the Aromatic
Medium Carbon Range Fraction
If available analytical chemistry data do not identify concentrations of individual
chemicals composing this fraction, the subchronic and chronic p-RfDs (0.04 and
0.01 mg/kg-day, respectively) for TMBs are recommended as the indicator chemical for the
aromatic medium carbon range fraction U.S. EPA (2016a). Given the limited data, the
compounds that resulted in the lowest RfDs for these effects and target tissues were considered
as the basis for indicator chemical selection. The RfDs for the TMBs, derived by route-to-route
extrapolation (from inhalation to oral) using a modified physiologically based pharmacokinetic
(PBPK) model, are based on neurological effects (decreased pain sensitivity; see Section 2.3).
Available data generally support the nervous system as a target of the aromatic medium carbon
compounds. Evaluation of available data as discussed in Appendix B suggests that use of the
TMBs' RfD values is reasonably anticipated to be protective for effects associated with
exposures to other constituents of the fraction. Users of the indicator chemical method should
understand that there could be more uncertainty associated with the application of this toxicity
value to the aromatic medium carbon range fraction than for its derivation in U.S. EPA (2016b).
The IRIS review of TMBs cited Korsak and Rydzynski (1996) as cited in U.S. EPA
(2016b) as the principal study for the subchronic and chronic RfDs. A study summary was not
provided in the IRIS assessment; however, the U.S. EPA (2007) PPRTV for 1,2,4-TMB (the
compound that served as the driver for the toxicity value) provided the following summary:
In the subchronic experiment, rats were exposed to 1,2,4-trimethylbenzene
at concentrations of 0, 25, 100 or 250ppm (0, 123, 491 or 1,227 mg/m3),
6 hours/day, 5 days/week for 3 months and observedfor exposure-related clinical
signs and body weight effects (Korsak and Rydzynski, 1996). Rotarod
performance and hot-plate behavior were measured as indices of the
neurotoxicity of trimethylbenzene isomers. Rotarod performance was tested prior
to start of the study, weekly during exposure, and 2 weeks after the termination of
the exposure. Hot-plate behavior was tested immediately after termination of the
exposure. Fisher's exact test was usedfor analysis of rotarod performance and
the Kruskall-Wallis test usedfor changes in pain sensitivity (hot plate behavior).
Exposures to 1,2,4-trimethylbenzene did not result in any apparent body weight
effects or clinical signs of toxicity. However, exposure-related indicators of
neurotoxicity were noted. Rotarod performance failure increased in a
concentration-related manner in the groups exposed to 1,2,4-trimethylbenzene,
but reached the level of statistical significance (40% failure; p < 0.05) only in the
highest (1,227 mg/m3) exposure group following 8 or 13 weeks of exposure. The
incidence of rotarod performance failure in control rats was 0% throughout the
study period. Although the mean rotarod performance failure rate in the highest
exposure group remained at 30% after a 2-week recovery period, the rate was not
significantly different from controls. Pain-sensitivity was also decreased in a
concentration dependent manner (evidenced by increased latency of the paw-lick
response). As shown in Table 2, the increased latency reached the level of
statistical significance in the 491- and 1,227-mg/m3 groups. After a 2-week
recovery period, the highest (1,227 mg/m3) exposure group no longer exhibited a
significant difference in pain sensitivity, relative to controls. This study identified
32 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
a NOAEL of 123 mg/m3 and a LOAEL of 491 mg/m3 (6 hours/day, 5 days/week)
for significantly decreased pain sensitivity.
U.S. EPA (2016a) used PBPK model estimates of internal blood dose metrics for
1.2.4-TMB coupled with benchmark dose (BMD) modeling to generate a POD. First, BMD
modeling of the data (for decreased pain sensitivity following 1,2,4-TMB exposure) identified a
benchmark concentration lower confidence limit with one standard deviation (BMCLisd) of
140.54 mg/m3 (based on external air concentrations and subsequently adjusted for continuous
exposure). Using the available PBPK model, the BMCLisd was converted to a duration-adjusted
POD (PODadj) of 0.099 mg/L. The PODadj value represents the internal blood dose metric of
average weekly venous blood concentration of 1,2,4-TMB, which is considered by the U.S. EPA
to be the most relevant internal dose metric. To derive an oral toxicity value based on these
(inhalation) data, an oral exposure component was added to the PBPK model by the U.S. EPA.
The model assumed 100% absorption of ingested 1,2,4-TMB. The human PBPK model was run
to estimate a human BMDL (HED) that would result from the same weekly average venous
blood concentration observed in the PODadj in animals (0.099 mg/L). The resultant BMDL
(HED) of 3.5 mg/kg-day was used to derive the subchronic and chronic RfDs for 1,2,4-TMB,
which was applied to all TMBs (see Table 6).
Confidence in the principal study was low to medium. Although the study was well-
conducted, peer-reviewed, and amenable to dose-response analyses (i.e., used an appropriate
number of exposure groups), there was uncertainty with respect to the actual concentrations
achieved (only target concentrations were reported), and reported measures of variance (type of
measures [e.g., standard deviation] not explicitly specified). Confidence in the oral database (for
TMBs) was low, because only acute neurotoxicity data and one subchronic toxicity study (for
1.3.5-TMB) were available. Owing to low confidence in the oral database, low to medium
confidence in the principal study, and uncertainty associated with the applicability of the PBPK
model for route-to-route extrapolation, confidence in the subchronic and chronic RfDs was low.
While toxicological data from mixtures such as HFAN might be preferred in some cases, the
p-RfD for HFAN is based on a screening value, and the Agency has more confidence in EPA's
IRIS TMB oral assessments as the indicator chemical.
6.1.2. Alternative Oral Noncancer Assessment Using the Hazard Index Method for the
Aromatic Medium Carbon Range Fraction
If the available analytical chemistry data quantify the concentrations of TMBs,
//-propylbenzene, //-butylbenzene, sec-butylbenzene, fert-butylbenzene, or isopropylbenzene
separately from the remainder of the aromatic medium carbon range fraction, it is recommended
that HQs for the individual chemicals with analytical data be calculated and an HI for the
mixture be developed using the calculated HQs.
For subchronic oral exposures, the following subchronic RfDs or p-RfDs can be used as
the denominator in the HQ equations: TMBs (0.04 mg/kg-day), //-propylbenzene
(0.1 mg/kg-day), //-butylbenzene (0.1 mg/kg-day), sec-butylbenzene (0.1 mg/kg-day), and
/^/'/-butylbenzene (0.1 mg/kg-day). In this alternative approach, the subchronic RfD for TMBs
(0.04 mg/kg-day) is recommended for use with the remainder of the fraction, including any other
fraction members analyzed individually (see Table 6).
33 Aromatic medium carbon range TPH fraction
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
EPA/690/R-22/004F
For chronic oral exposures, the following chronic RfDs or p-RfDs can be used as the
denominator in the HQ equations: TMBs (0.01 mg/kg-day), isopropylbenzene (0.1 mg/kg-day),
//-propylbenzene (0.1 mg/kg-day), //-butylbenzene (0.05 mg/kg-day), .sf c-butylbenzene
(0.1 mg/kg-day), and /^/'/-butylbenzene (0.1 mg/kg-day). In this alternative approach, the chronic
RfD for TMBs (0.01 mg/kg-day) is recommended for use with the remainder of the fraction,
including any other fraction members analyzed individually (see Table 6).
In some cases, toxicological data from mixtures such as HFAN might be preferred;
however, the p-RfD for HFAN is based on a screening value. The Agency has more confidence
in an HI approach as an alternative to the indicator chemical approach than for the surrogate
mixture approach for this fraction.
6.2. DERIVATION OF INHALATION REFERENCE CONCENTRATIONS
The available subchronic and chronic RfC values, with PODs, uncertainty factors, critical
effects, and confidence ratings are presented in Table 7. As shown in the table, there are
subchronic and chronic RfCs or provisional reference concentrations (p-RfCs) for one mixture
(HFAN) and four individual compounds (w-propylbenzene and 1,3,5-, 1,2,4-, and 1,2,3-TMB)
relevant to the aromatic medium carbon range fraction. In addition, there is a chronic RfC for
isopropylbenzene. Critical effects for the RfCs included maternal toxicity (decreased body
weight), developmental toxicity, increased adrenal and kidney weights, and decreased pain
sensitivity.
34 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 7. Available RfC Values for Aromatic Medium Carbon Range Fraction (C9-C10, EC9-EC < ll)a
Indicator Chemical or
Components
POD
POD
Type
(all are
HECs)
UFc
UF
Components
RfC or
p-RfC
(mg/m3)
Confidence in
p-RfC or RfCa
Critical Effect(s)
Species, Mode, and
Duration
Primary
Reference
(source)
Subchronic
//-Propylbcnzcnc (C9,
EC8.94)
434
NOAEL
300
UFa, UFh,
UFd
lh
Low
Based on ethylbenzene
as an analogue;
developmental toxicity
(developmental)
Rat, 6-7 h/d, 7 d/wk for
3 wk prior to mating and
GDs 1-19; rabbits,
6-7 h/d, 7 d/wk on
GDs1-24
Andrews (1981)
and Hardin (1981)
as cited in U.S.
EPA (2009a)
1,3,5-T rimethylbenzene
(C9, EC9.15)
18.15
BMCL
100
UFa, UFh,
UFd
0.2
Low-medium
Decreased pain
sensitivity in male
Wistar rats (nervous)
Rat, 6 h/d, 5 d/wk for
13 wk
Korsak and
Rvd/vnski (1996)
as cited in U.S.
EPA (2016b)
1,2,4-T rimethylbenzene
(C9, EC9.36)
18.15
BMCL
100
UFa, UFh,
UFd
0.2
Low-medium
Decreased pain
sensitivity in male
Wistar rats (nervous)
Rat, 6 h/d, 5 d/wk for
13 wk
Korsak and
Rvd/vnski (1996)
as cited in U.S.
EPA (2016b)
1,2,3-T rimethylbenzene
(C9, EC9.65)
18.15
BMCL
100
UFa, UFh,
UFd
0.2
Low-medium
Decreased pain
sensitivity in male
Wistar rats (nervous)
Rat, 6 h/d, 5 d/wk for
13 wk
Korsak and
Rvd/vnski (1996)
as cited in U.S.
EPA (2016b)
HFAN (C9-C10)
125
LOAEL
100
UFa, UFh,
UFl
lb
Moderate
Decreased maternal
body weight vs. controls
(reproductive) in
CD-I mice
Mouse, 6 h/d, 7 d/wk on
GDs 6-15
McK.ee et al.
(1990) as cited in
U.S. EPA (2009c)
35
Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 7. Available RfC Values for Aromatic Medium Carbon Range Fraction (C9-C10, EC9-EC < ll)a
Indicator Chemical or
Components
POD
POD
Type
(all are
HECs)
UFc
UF
Components
RfC or
p-RfC
(mg/m3)
Confidence in
p-RfC or RfC
Critical Effect(s)
Species, Mode, and
Duration
Primary
Reference
(source)
Chronic
Isopropylbenzene (C9,
EC8.66)
435
NOAEL
1,000
UFa, UFd,
UFh, UFs
0.4
Medium
Increased kidney
weights in female rats
and adrenal weights in
male and female
F344 rats (endocrine,
urinary)
Rat, 6 h/d, 5 d/wk for
13 wk
Cushman (1995)
as cited in U.S.
EPA (1997)
//-Propylbcnzcnc (C9,
EC8.94)
434
NOAEL
300
UFa, UFd,
UFh
lh
Low
Based on ethylbenzene
as an analogue;
developmental toxicity
(developmental)
Rat, 6-7 h/d, 7 d/wk for
3 wk prior to mating and
GDs 1-19; rabbit, 6-7 h/d,
7 d/wk on GDs 1-24
Andrews (1981)
and Hardin (1981)
as cited in U.S.
EPA (2009a)
1,3,5-T rimethylbenzene
(C9, EC9.15)
18.15
BMCL
300
UFa, UFd,
UFh, UFs
0.06
Low-medium
Decreased pain
sensitivity in male
Wistar rats (nervous)
Rat, 6 h/d, 5 d/wk for
13 wk
Korsak and
Rvd/vnski (1996)
as cited in U.S.
EPA (2016b)
1,2,4-T rimethylbenzene
(C9, EC9.36)
18.15
BMCL
300
UFa, UFd,
UFh, UFs
0.06
Low-medium
Decreased pain
sensitivity in male
Wistar rats (nervous)
Rat, 6 h/d, 5 d/wk for
13 wk
Korsak and
Rvd/vnski (1996)
as cited in U.S.
EPA (2016b)
1,2,3-T rimethylbenzene
(C9, EC9.65)
18.15
BMCL
300
UFa, UFd,
UFh, UFs
0.06
Low-medium
Decreased pain
sensitivity in male
Wistar rats (nervous)
Rat, 6 h/d, 5 d/wk for
13 wk
Korsak and
Rvd/vnski (1996)
as cited in U.S.
EPA (2016b)
36
Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 7. Available RfC Values for Aromatic Medium Carbon Range Fraction (C9-C10, EC9-EC < ll)a
Indicator Chemical or
Components
POD
POD
Type
(all are
HECs)
UFc
UF
Components
RfC or
p-RfC
(mg/m3)
Confidence in
p-RfC or RfC
Critical Effect(s)
Species, Mode, and
Duration
Primary
Reference
(source)
HFAN (C9-C10)
125
LOAEL
1,000
UFa, UFh,
UFl, UFs
0.1b
Moderate
Decreased maternal
body weight vs. controls
(reproductive) on GD 15
in CD-I mice
Mouse, 6 h/d, 7 d/wk on
GDs 6-15
McK.ee et al.
(1990) as cited in
U.S. EPA (2009c)
aBolded row shows the compounds and toxicity value selected as the indicator chemical for the fraction if analytical chemistry data do not identify concentrations of
individual chemicals composing this fraction.
bToxicity values are provisional values obtained from an existing PPRTV assessment. Values in italics are screening provisional values obtained from an existing
PPRTV assessment. Screening provisional values are not assigned confidence statements; however, confidence in these values is presumed to be low. Screening
provisional values are derived when the available data do not meet the requirements for deriving a provisional toxicity value.
BMCL = benchmark concentration lower confidence limit; C = carbon; EC = equivalent carbon; GD = gestation day; HEC = human equivalent concentration;
HFAN = high flash aromatic naphtha; LOAEL = lowest-observed-adverse-effect level; NOAEL = no-observed-adverse-effect level; POD = point of departure;
PPRTV = Provisional Peer-Reviewed Toxicity Value; p-RfC = provisional reference concentration; RfC = reference concentration; UF = uncertainty factor;
UFa = interspecies uncertainty factor; UFC = composite uncertainty factor; UFD = database uncertainty factor UFH = intraspecies uncertainty factor;
UFS = subchronic-to-chronic uncertainty factor.
37
Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
As shown in Table 7, the values of RfCs for fraction members show consistency across
the fraction with respect to the toxicological effects exerted (most frequently neurological and
developmental effects). The data show that an indicator chemical identifying effects on these
targets would be reasonably anticipated to be representative of the effects of the fraction as a
whole. Therefore, the compounds that resulted in the lowest RfCs for these effects were
considered as the basis for surrogate selection (see Section 2.3).
6.2.1. Inhalation Noncancer Assessment Using the Indicator Chemical Method for the
Aromatic Medium Carbon Range Fraction
If available analytical chemistry data do not identify concentrations of individual
chemicals in this fraction, the lowest subchronic and chronic p-RfCs (0.2 mg/m3 and 0.06 mg/m3,
respectively) for TMBs (0.2 and 0.06 mg/m3, respectively; see Table 7); these values are
recommended as indicator chemicals for the aromatic medium carbon range fraction U.S. EPA
(2016a). The RfCs for TMBs are based on neurological effects (decreased pain sensitivity), and
available data generally support the nervous system as a target of the aromatic medium carbon
compounds. Use of these values is anticipated to be protective for exposure to other constituents
based on the available toxicological information (see Appendix B). However, users of the
indicator chemical method should understand that there could be more uncertainty associated
with the application of this toxicity value to the aromatic medium carbon range fraction than for
its derivation in the IRIS assessment U.S. EPA (2016b) and application to individual TMBs in
the environment.
The IRIS review of TMBs cited Korsak and Rvdzvnski (1996) as cited in U.S. EPA
(2016b) as the principal study for the subchronic and chronic RfCs. A summary of this study was
provided in the preceding section (see Section 6.1). The PODadj of 0.099 mg/L (described in the
preceding section) was converted to a BMCL (HEC) of 18.15 mg/m3 based on the available
PBPK model; this was used to derive subchronic and chronic RfCs for 1,2,4-TMB, which were
applied to all TMBs (see Table 7). As indicated in the preceding section, confidence in the
principal study was low to medium. Confidence in the inhalation database was also low to
medium. Although acute, short-term, subchronic, and developmental inhalation toxicity studies
in rats and mice are available, there were no chronic or developmental neurotoxicity studies. In
addition, supporting studies (with respect to the critical effect) were primarily from the same
research group. Taken together, confidence in the subchronic and chronic RfCs was also low to
medium.
Previously, in the PPRTV TPH Mixtures document U.S. EPA (2009b), the HFAN
subchronic and chronic p-RfCs were recommended for assessing noncancer hazards associated
with inhalation route exposures to this fraction, based on a 2009 PPRTV assessment U.S. EPA
(2009c). In 2016, the U.S. EPA IRIS Program published TMB subchronic and chronic p-RfCs of
0.2 and 0.06 mg/m3, respectively U.S. EPA (2016a). which are lower than the respective HFAN
values of 1 and 0.1 mg/m3, respectively U.S. EPA (2009c) (see Table 7). Because these are IRIS
values rather than PPRTVs, these IRIS single chemical values should be used in the indicator
chemical approach rather than HFAN-based surrogate mixture approach. The 2009 TPH mixture
assessment indicates that HFAN toxicity values are similar to values for other individual
compounds in the fraction, which supports using HFAN as a surrogate for the fraction; however,
the 2016 TMB values are much lower than the HFAN values and that logic is not applicable.
38 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
6.2.2. Alternative Inhalation Noncancer Assessment Using the Hazard Index Method for
the Aromatic Medium Carbon Range Fraction
If the available analytical chemistry data quantify the concentrations of TMBs,
//-propylbenzene, or isopropylbenzene separately from the remainder of the aromatic medium
carbon range fraction, it is recommended that HQs for the individual chemicals with analytical
data be calculated and an HI for the mixture be developed using the calculated HQs.
For subchronic inhalation exposures, the subchronic RfCs or p-RfCs for TMBs
(0.2 mg/m3) or ^-propylbenzene (1.0 mg/m3) can be used as the denominator in the HQ
equations (see Table 7). In this alternative approach, the subchronic RfC for TMBs (0.2 mg/m3)
is recommended for use with the remainder of the fraction, including any other fraction members
analyzed individually.
For chronic inhalation exposures, the following chronic RfCs or p-RfCs can be used in
the denominator of the HQ equations: TMBs (0.06 mg/m3), isopropylbenzene (0.4 mg/m3), and
//-propylbenzene (1.0 mg/m3) (see Table 7). In this alternative approach, the chronic RfC for
TMBs (0.06 mg/m3) is recommended for use with the remainder of the fraction, including any
other fraction members analyzed individually. As stated in Section 6.2.1, in the U.S. EPA's
PPRTV TPH Mixtures document U.S. EPA (2009b). the HFAN subchronic and chronic p-RfCs
were recommended for assessing noncancer hazards associated with inhalation route exposures
to this fraction, based on a 2009 PPRTV assessment U.S. EPA (2009c). In 2016, the U.S. EPA
IRIS Program published TMB subchronic and chronic p-RfCs of 0.2 and 0.06 mg/m3,
respectively (U.S. EPA, 2016a), which are lower than the HFAN values of 1 and 0.1 mg/m3 U.S.
EPA (2009c) (see Table 7). Because these are IRIS values rather than PPRTVs, the U.S. EPA
has more confidence in using these IRIS single chemical values in an HI approach rather than the
HFAN values in surrogate mixture approach.
6.3. SUMMARY OF NONCANCER PROVISIONAL REFERENCE VALUES
Table 8 summarizes the noncancer health reference values for indicator chemicals used
when available analytical data and exposure estimates are limited to either air concentrations of
or oral exposure rates associated with the whole fraction. When analytical results, air
concentrations, or exposure rate measures for individual compounds with reference values are
available, then the hazards associated with these compounds can be assessed separately, using
the HI approach and reference values reported in Tables 6 and 7.
39 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 8. Summary of Noncancer Reference Estimates for Indicator
Chemicals for Aromatic Medium Carbon Range (C9-C10; EC9-EC <11)
Fraction of Total Petroleum Hydrocarbons
Toxicity Type
(units); Indicator
Chemical
Species/
Sex
Critical Effect
p-Reference
Value
POD
Method
POD
(HED/HEC)
UFc
Reference
Subchronic p-RfD
(mg/kg-d);
trimethylbenzenes
Rat/M
Neurotoxicity
(decreased pain
sensitivity)
0.04 mg/kg-d
BMDL
(HED)a
3.5
100
Korsak and
Rvdzvnski (1996)
as cited in U.S.
EPA (2016b)
Chronic p-RfD
(mg/kg-d);
trimethylbenzenes
Rat/M
Neurotoxicity
(decreased pain
sensitivity)
0.01 mg/kg-d
BMDL
(HED)a
3.5
300
Korsak and
Rvdzvnski (1996)
as cited in U.S.
EPA (2016b)
Subchronic p-RfC
(mg/m3);
trimethylbenzenes
Rat/M
Neurotoxicity
(decreased pain
sensitivity)
0.2 mg/m3
BMDL
(HEC)
18.15
100
Korsak and
Rvdzvnski (1996)
as cited in U.S.
EPA (2016b)
Chronic p-RfC
(mg/m3);
trimethylbenzenes
Rat/M
Neurotoxicity
(decreased pain
sensitivity)
0.06 mg/m3
BMDL
(HEC)
18.15
300
Korsak and
Rvdzvnski (1996)
as cited in U.S.
EPA (2016b)
aBased on route-to-route extrapolation (inhalation to oral) using a modified PBPK model.
BMDL = benchmark dose lower confidence limit; HEC = human equivalent concentration; HED = human
equivalent dose; M = male(s); PBPK = physiologically based pharmacokinetic; POD = point of departure;
p-RfC = provisional reference concentration; p-RfD = provisional reference dose; UFc = composite uncertainty
factor.
1 6.4. CANCER WEIGHT-OF-EVIDENCE DESCRIPTOR
2 Carcinogenicity assessments for mixtures and individual components of the aromatic
3 medium carbon range fraction that have assessments are shown below in Table 9. For all
4 components of the fraction, there are either inadequate data to assess carcinogenic potential (via
5 the oral or inhalation routes of exposure), or the available studies have not been formally
6 evaluated by the U.S. EPA, and the U.S. EPA has not estimated the cancer potency associated
7 with the study results.
40 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table 9. Available Cancer Weight-of-Evidence Evaluations for Aromatic
Medium Carbon Range Fraction (C9-C10, EC9-EC < 11)
Compound or Mixture
Cancer WOE
Source
Isopropylbenzene (C9, EC8.66)
"Inadequate Information to Assess Carcinogenic Potential"
U.S. EPA (1997)
//-Propylbcnzcnc (C9, EC8.94)
"Inadequate Information to Assess Carcinogenic Potential"
U.S. EPA (2009d)
1,3,5-Trimethylbenzene (C9,
EC9.15)
"Inadequate Information to Assess Carcinogenic Potential"
U.S. EPA (2016a)
1,2,4-Trimethylbenzene (C9,
EC9.36)
"Inadequate Information to Assess Carcinogenic Potential"
U.S. EPA (2016a)
fer/-Butylbenzene (CIO,
EC9.36)
"Inadequate Information to Assess Carcinogenic Potential"
U.S. EPA (2012b)
sec-Butylbenzene (CIO,
EC9.57)
"Inadequate Information to Assess Carcinogenic Potential"
U.S. EPA (2012a)
1,2,3-Trimethylbenzene (C9,
EC9.65)
"Inadequate Information to Assess Carcinogenic Potential"
U.S. EPA (2016a)
//-Butylbcnzcnc (CIO, EC9.96)
"Inadequate Information to Assess Carcinogenic Potential"
U.S. EPA (2010)
High flash aromatic naphtha
(C9-C10)
"Inadequate Information to Assess Carcinogenic Potential"
U.S. EPA (2009c)
C = carbon; EC = equivalent carbon; WOE = weight of evidence.
1 While data on genotoxicity testing of compounds and mixtures in the aromatic medium
2 carbon range fraction are limited, available information suggests little to no genotoxic potential
3 (see Section 5).
4 6.5. DERIVATION OF PROVISIONAL CANCER RISK ESTIMATES
5 None of the mixtures or constituents in this fraction had an oral slope factor (OSF) or
6 inhalation unit risk (IUR) from IRIS, PPRTVs, HEAST, MassDEP, or TPHCWG. Thus, a
7 provisional oral slope factor (p-OSF) or provisional inhalation unit risk (p-IUR) was not derived
8 for the fraction (see Table 10).
Table 10. Summary of Cancer Risk Estimates for Aromatic Medium
Carbon Range (C9-C10; EC9-EC < 11) Fraction of TPHs
Toxicity Type (units)
Species/Sex
Tumor Type
Cancer Risk Estimate
Principal Study
p-OSF (mg/kg-d) 1
NDr
p-IUR (lng/in3) 1
NDr
C = carbon; EC = equivalent carbon; NDr = not determined; p-IUR = provisional inhalation unit risk;
p-OSF = provisional oral slope factor; TPH = total petroleum hydrocarbon.
41 Aromatic medium carbon range TPH fraction
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
EPA/690/R-22/004F
APPENDIX A. LITERATURE SEARCH AND SCREENING
Literature searches were conducted in February 2018 and updated in August 2021 for
studies relevant to the derivation of provisional toxicity values for the aromatic medium carbon
range fraction of total petroleum hydrocarbons (TPHs). The following substances (CASRNs),
12 chemicals and 1 mixture, were included in the initial list for the revised aromatic medium
carbon range fraction: //-butylbenzene (104-51-8), isobutylbenzene (538-93-2), /t7-/-butylbenzene
(98-06-6), sec-butylbenzene (135-98-8), isopropylbenzene (535-77-3), «-propylbenzene
(103-65-1), l-methyl-4-ethylbenzene (622-96-8), l-methyl-3-isopropylbenzene (535-77-3),
l-methyl-3-ethylbenzene (620-14-4), 1,2,3-trimethylbenzene (TMB; 526-73-8), 1,2,4-TMB
(95-63-6), 1,3,5-TMB (108-67-8), and high flash aromatic naphtha (HFAN; 64742-95-6,
88845-25-4, and 64742-94-5). Because Integrated Risk Information System (IRIS) assessments
were available for isopropylbenzene and the three TMB isomers, these compounds were not
included in the literature searches. Literature searches were conducted for studies relevant to the
derivation of provisional toxicity values for the remaining eight chemicals and one mixture listed
above. Initial searches were date limited from 2007 to 2018 and were conducted using the
U.S. Environmental Protection Agency (U.S. EPA) Health and Environmental Research Online
(HERO) database of scientific literature. The PubMed database was searched using the HERO
interface. The updated search was conducted similarly using the same search strings in PubMed
and Web of Science from February 2018 through August 2021.
The results of the PubMed searches (title and abstract) were screened for relevance using
the Population, Exposure, Comparator, and Outcome (PECO) criteria outline in Table A-l.
Full-text screening for relevance to hazard identification was performed using the refined PECO
criteria shown in Table A-2.
Table A-l. PECO Criteria for Screening of Total Petroleum Hydrocarbon
Constituent Literature Search Results
PECO Element
Inclusion Criteria
Population
Humans (any population) or laboratory mammals (any life stage).
Exposure
Human: Exposure to the subject material alone or as the primary component of a mixture, known
or presumed to occur by oral, inhalation, and/or dermal routes.
Animal: In vivo, exposure to the subject material alone, by oral or inhalation (including
instillation) routes, for all durations of exposures (durations <28 d will be captured as supporting
information), including any duration during gestation. Other routes of exposure will be captured
as supporting information.
Comparator
Human: Includes any comparison/referent group (no exposure, lower exposure).
Animal: Includes concurrent negative (untreated, sham-treated, or vehicle) control.
Outcome
Assesses any cancer or noncancer endpoint in any tissue, organ, or physiological system.
PECO = Population, Exposure, Comparator, and Outcome.
42 Aromatic medium carbon range TPH fraction
-------
EPA/690/R-22/004F
Table A-2. PECO Criteria for Relevance to Hazard Identification
PECO Element
Inclusion Criteria
Population
Humans (any population) or laboratory mammals (any life stage).
Exposure
Human: Exposure to the subject material alone or as the primary component of a mixture, known
or presumed to occur by oral or inhalation routes.
Animal: In vivo, exposure to the subject material alone, by oral or inhalation routes, for durations
>28 d or any duration during gestation.
Comparator
Human: Includes any comparison/referent group (no exposure, lower exposure).
Animal: Includes concurrent negative (untreated, sham-treated, or vehicle) control.
Outcome
Assesses any cancer or noncancer health outcome in any tissue, organ, or physiological system.
PECO = Population, Exposure, Comparator, and Outcome.
43 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
APPENDIX B. POTENTIALLY RELEVANT NONCANCER EVIDENCE
DEVELOPMENT OF EXPOSURE-RESPONSE ARRAYS
As described in the main document, dose-response data were presented in
exposure-response arrays by health outcome and exposure route. In order to assess consistency in
effects and potency across the components of the fraction, experimental data from compound
specific Integrated Risk Information System (IRIS) and Provisional Peer Reviewed Toxicity
Value (PPRTV) documents and primary data sources (identified from literature searches) were
used to create exposure-response arrays. Exposure-response arrays present dose-response data by
health outcome and exposure route. From left to right, compounds exhibiting an effect are shown
before those not exhibiting an effect, to enable identification of patterns. Within the group
exhibiting an effect, compounds are ordered from lowest lowest-observed-adverse-effect level
(LOAEL) to highest. For compounds that do not exhibit an effect, no-observed-adverse-effect
levels (NOAELs) in the arrays are ordered by equivalent carbon (EC) number index (low to high
from left to right), with mixtures shown last. Both administered doses and exposure
concentrations reported in the arrays and in text reflect time-weighted average (TWA) exposures,
to facilitate comparisons across studies and compounds. Consistency across the fraction was
evaluated by assessing if comparable outcomes were observed for members of the fraction, and
if these effects were observed at similar dose levels. Unless otherwise specified, the term
"significant," used throughout this appendix, refers to statistical significance at ap-value < 0.05.
NEUROLOGICAL EFFECTS
A nervous system endpoint (decreased pain sensitivity) is the critical effect for the
subchronic and chronic reference concentration (RfC) values for trimethylbenzene (TMB)
isomers, with data for 1,2,4-TMB being the driver for these values U.S. EPA (2016b). Oral
toxicity values (i.e., subchronic and chronic reference doses [RfDs]) for TMBs were derived
based on route-to-route extrapolation (using a modified physiologically based pharmacokinetic
[PBPK] model) from the inhalation values. Neurological effects (including effects on motor
coordination, cognitive function, vision, and the inner ear) have been reported in humans
occupationally exposed to solvents including TMBs; however, the effects cannot be attributed to
specific compounds. Neurotoxicity data in humans are limited to TMBs, and there are no human
data for other members of the aromatic medium carbon range fraction. Animal studies examining
neurological endpoints are available for most of the compounds or mixtures with toxicity data;
however, the studies varied widely with respect to the nature of the neurological endpoints
evaluated.
Human Studies
U.S. EPA (2016b) reviewed the evidence for neurotoxicity in humans exposed to TMBs
alone or in complex mixtures. Much of the epidemiological evidence is from occupational
exposures to complex mixtures including TMBs, and the relative contribution of TMBs
compared with other constituents is not known. Associations between exposure of dockyard and
shipyard painters to solvent mixtures possibly containing TMBs and impaired performance in a
battery of neurological tests, including short-term memory (symbol digit substitution), motor
speed/coordination (finger tapping), and peripheral nerve function tests, were reported in several
studies U.S. EPA (2016b). Other neuropsychological symptoms (mood changes, equilibrium
complaints, and sleep disturbances) were also observed among shipyard painters. Similarly, paint
factory workers exposed to multiple unspecified solvents had detrimental neuropsychological
44 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
effects (memory problems, dizziness, hand tremble), and construction workers exposed to
solvent mixtures had impaired performance in memory tasks U.S. EPA (2016b).
Other researchers reported damage or dysfunction of the inner ear and increased
incidence of vertigo following workplace exposure (in paint and varnish factories and histology
laboratories) to TMBs and other organic solvents U.S. EPA (2016b). Suggestive evidence of
visual impairment (altered color vision and contrast) among furniture factory workers was
reported following exposure to complex solvent mixtures. Increased latencies for visual evoked
potentials (VEPs) were found in gasoline-exposed workers. There is suggestive evidence from
multiple studies of human exposure to solvent mixtures containing TMB isomers that exposure
results in toxicological effects on neuromuscular function and balance in humans, including
increased reaction time, increased hand tremble, decreased hand-eye coordination, and vertigo
U.S. EPA (2016b). Finally, symptoms associated with central nervous system (CNS) depression
(e.g., lightheadedness, fatigue) have been reported in workers occupationally exposed to solvent
mixtures containing TMBs U.S. EPA (2016b).
A significant, positive association between exposure and neurological symptoms (such as
abnormal fatigue) was reported among asphalt workers exposed to 1,2,4-TMB; however, the
association was not evident among asphalt workers with exposure to lower levels of 1,2,3-TMB
or 1,3,5-TMB U.S. EPA (2016b). Paint shop workers exposed to 49-295 mg/m3 of a solvent
mixture containing 50% 1,2,4-TMB, 30% 1,3,5-TMB, and unspecified amounts of 1,2,3-TMB
(listed as possibly present) exhibited a variety of neurological effects including nervousness,
tension, headaches, vertigo, and anxiety U.S. EPA (2016b).
Studies of adult volunteers who were acutely exposed to mixtures containing 1,2,4-TMB
reported significant and consistent increases in reaction time, although it is unclear whether
1,2,4-TMB or other constituents in the mixtures were responsible for the observed effects.
Neurobehavioral impairment was either weakly or inconsistently associated with exposure in a
volunteer study in which participants were exposed to aromatic or dearomatized white spirit
(white spirit contains a mixture of 1,2,4- and 1,3,5-TMB) for 4 hours U.S. EPA (2016b).
The few available controlled human exposure studies of TMBs alone have not shown
neurological effects U.S. EPA (2016b). No neurological abnormalities were reported in routine
clinical examinations in two studies investigating the toxicokinetics of TMBs following
controlled human exposures to 5-150 mg/m3 of 1,2,3-, 1,2,4-, or 1,3,5-TMB; however, neither
results data nor details regarding the specific neurological tests performed were provided. In
another controlled toxicokinetics study, no overt CNS depression (measured as heart rate and
respiration) or increase in subjective CNS symptoms (headache, fatigue, nausea, dizziness,
intoxication) was observed in volunteers exposed to concentrations <123 mg/m3 of TMB isomers
U.S. EPA (2016b).
45 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
Animal Studies
Animals exposed orally to diethylbenzenes (DEBs) (individual isomers, as well as a
mixture containing 7% 1,2-DEB, 58% 1,3-DEB, and 35% 1,4-DEB) have been evaluated for
peripheral nervous system and CNS effects. No oral data on the neurotoxicity of the other
members of the aromatic medium carbon range fraction were located. Figure B-l is an
exposure-response array containing studies for which neurotoxicity effects levels could be
reliably determined. Reductions in sensory conduction velocity (SCV), motor nerve conduction
velocity (MCV), and amplitude of the sensory action potential (SAP) (of the tail nerve) occurred
after exposure to 57 mg/kg-day 1,2-DEB and >357 mg/kg-day of the DEB mixture (lowest doses
tested) Gagnaire et al. (1990). The same parameters were unaffected in rats treated with 1,3- or
1,4-DEB at 357 mg/kg-day. In a follow-up study, rats administered 1,2-DEB at >43 mg/kg-day
also showed significantly increased latencies with respect to parameters of brainstem auditory
evoked potentials (BAEPs) Gagnaire et al. (1992a).
Neurological effects evaluated after inhalation exposure to aromatic medium carbon
range compounds include clinical signs of neurotoxicity, neurobehavioral changes, peripheral
nervous system and CNS function, and brain histopathology. Figure B-2 is an exposure-response
array containing studies for which neurotoxicity effect levels could be reliably determined. Data
were available for six members of the fraction. Clinical signs of neurotoxicity (side-to-side
movement and head tilt) were observed in rats treated with isopropylbenzene at >92 mg/m3 for
4 weeks Monsanto Company. 1986 as cited in U.S. EPA (1997). Significant neurobehavioral
changes (impairments in active and passive avoidance, increased motor activity, and/or reduced
pain sensitivity) were observed in rats treated with 1,2,4-, 1,3,5-, or 1,2,3-TMB at >88 mg/m3
Wiaderna et al., 2002, Gralewicz and Wiadema, 2001, Wiadema et al., 1998, Gralewicz et al.,
1997a, and Korsak and Rvdzvnski, 1996, Lutz, 2010, all as cited in U.S. EPA (2016b).
Decreased pain sensitivity in rats treated with 1,2,4-TMB at >88 mg/m3 serves as the basis for
oral and inhalation subchronic and chronic RfDs and RfCs for all three TMB isomers Korsak and
Rvdzvnski, 1996 as cited in U.S. EPA (2016b). Pain sensitivity was not significantly affected in
rats exposed to HFAN at up to 1,157 mg/m3 for 90 days Douglas et al., 1993 as cited in U.S.
EPA (2009c). There was no effect on nervous system histology in this study or on brain weight
or histology in rats or mice following exposure to isopropylbenzene for 14 or 105 weeks at
concentrations up to 907 mg/m3 Ntp (2009). Male rats exposed to a DEB mixture for 18 weeks
showed decreased SCV, MCV, and amplitude of SAP (>486 mg/m3) and increased latency of
BAEP parameters (at >633 mg/m3) Gagnaire et al. (1992b). In a developmental study, no
neurobehavioral effects were reported in rats exposed to HFAN at up to 500 mg/m3 on gestation
days (GDs) 7-15; however, it is not clear if only pups or pups and dams were evaluated for
effects Lehotskv, 1989 as cited in U.S. EPA (2009c).
46 Aromatic medium carbon range TPH fraction
-------
EPA 690 R-22 004F
1000
100
I
>¦
—
'E
E
"O
<
_>>
fO
Q
LOAEL
¦LOAEL
ANOAEL
• Doses < NOAEL
o £
a *
i—i T3
— CD
OJ
> tI
U
LO >
&Q
ru
>
u
00
ID
ro
rg
cn
QJ
ro
c
cxo
ro
O
t/>
QJ
U
C -I—~
a> ro
*-> cn
fD , ^
Q_
<
in
<-
3
00
1,2-Diethylbenzene
o %
$ -
rH T3
a> ~
0) CO
cxo
03
>
u
-V:
$
00
Diethylbenzene Mix
a_
<
LO
o
cn >
cn u
i-H LO
03
4—*
a»
a»
i_
OJ
c
>
u
OJ
cc
00
— u
QD m
m I.
1,3-Diethylbenzene
Q-
<
LO
O
cn >
cn o
t—I LO
¦ *¦ -M
03 ^ ^
, u cc
a;
0)
03
<= tJ
Q0 a)
03 £
e? %
o
£ jc
c 3
° 00
1,4-Diethylbenzene
Figure B-l. Neurological Effects in Animals after Oral Exposure to Aromatics Medium Carbon Range Compounds
47
Aromatic medium carbon range TPH fraction
-------
EPA 690 R-22.004F
10000
1000
• Cone > LOAEL
¦ LOAEL
ANOAEL
° Cone < NOAEL
£
c
o
100
c
0>
u
c
o
u
0>
t-
D
t/)
o
Q-
X
LU
a;
no
H3
i—
0)
>
<
"D
QJ
.SP
'3
s
I
ai
E
10
21
it ^
3 *-
==
So
_o
o
2 -i?
_o
o
o 5
tU -it
£ 5
Is op ropy I benzene
^ 5
;>
r~ k—
o3 -5
i| ui 5
u .E >
w 5 nj
« (V
1-?
Q J0
^ -t Q,
¦o 1
n>
c oc
g o t3
-* E to
"D
05
U
I
ai
to
LD
~_ "O
v- m
TO
1,3,5-
Tri methyl benzene
Q_ rH
l/l ~
41
Z> TO
C QJ
U
X5 C
B "5 "S
u o K
M. S J?
n> "
™ i» 5
n! ^
fK. .dL *S
s ?s
•H Qm
-£ -»
< 5
a. —
i±i t>
— TO
l ! 2
" ir¥
™ > 5
8 ™ S
<-J J- c
„ O
? o !2
leg
13 <- "
3 ir
U <||
3 "
2 2
UJ Q
c/i £
ZD "a
Q- E
P TO
1,2,4-Tri met hyI be nze n e
< B
Cl O
^ E
^ -o
=> Pi -
3 ?
to "a
E £
ai to
-a Cl
.2 E
±; d>
O c
!5 E
o B
o -c
PM QJ
.. a.
li e
3 TO
J <-
1,2,3-
Trimethyl-
benzene
x» CL
fN ^
tTi
a, >
± >"00
TOO1-1
c s
Ofl
TO _s.
O -7
&d <-
TO
e?
Diethyl benzene
Mix
3 S
« E
< 2
IB (g Q S
£ £
-e ? -
m qj
00 n
cr. °
T> O _
irt >-"O
:-|s
SI
" rsp
O QJ
a 2:
HFAN
Figure B-2. Neurological Effects in Animals after Inhalation Exposure to Aromatic Medium Carbon Range Compounds
48
Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
Summary of Potentially Relevant Evidence
Available data indicate that neurological effects are associated with oral or inhalation
exposure to some fraction members. In oral and inhalation toxicity studies using individual DEB
isomers or a mixture of DEB isomers, effects on peripheral nervous system and/or CNS (SCV,
MCV, amplitude of SAP, and BAEPs) were consistently observed for 1,2-DEB and the DEB
mixture (but not 1,3-DEB or 1,4-DEB). Studies in animals show that exposure to TMBs via
inhalation induces neurobehavioral effects (specifically, decreased pain sensitivity and increased
motor activity and passive and active avoidance); data from human studies are available but are
insufficient to establish causal relationships (owing to coexposures with other compounds).
Limited data are available for the other members of the fraction. CNS effects (consisting of
clinical signs of neurotoxicity) were observed in at least one other inhalation study (in rats
exposed to isopropylbenzene). However, additional studies using this compound were not
dedicated to neurotoxicity studies; only brain histopathology was evaluated (no effects were
observed in subchronic and chronic studies in rats and mice). The most sensitive chemical- or
mixture-specific LOAELs for neurological endpoints ranged between 43 and 357 mg/kg-day in
subchronic oral toxicity studies in rats (see Figure B-l) and between 88 and 633 mg/m3 in
subchronic inhalation toxicity studies in rats (see Figure B-2).
Taken together, the available data indicate that some members of the aromatic medium
carbon range fraction can induce neurological effects. However, there are a number of
compounds comprising the aromatic medium carbon range fraction that have not been evaluated
for sensitive measures of neurological function.
HEPATIC EFFECTS
A hepatic effect (hepatocellular hypertrophy) is the critical effect for the subchronic and
chronic provisional reference doses (p-RfDs) for //-butylbenzene U.S. EPA (2010). In addition,
effects on liver histopathology serve as a cocritical effects (with effects on kidney
histopathology) for the screening-level subchronic and chronic p-RfDs for //-propylbenzene
[based on the use of ethylbenzene (CASRN 100-41-4) as an analogue chemical U.S. EPA
(2009a). No human data pertaining to the hepatotoxicity of aromatic medium carbon range
fraction members were identified. As shown in Table 5, oral and/or inhalation data on hepatic
effects in animals were located for seven members of the fraction. In general, the hepatic
endpoints evaluated in the studies were clinical chemistry parameters, liver weight, and
histology.
Human Studies
No human studies were available to address the potential for hepatic toxicity of the TMB
isomers or other members of the aromatic medium carbon range fraction by any route of
exposure.
49 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
Animal Studies
Animals orally administered four materials (three individual compounds and HFAN) in
the aromatic medium carbon range have been evaluated for hepatotoxicity. Figure B-3 is an
exposure-response array containing studies for which hepatic effects levels could be reliably
determined. Increased absolute and/or liver weight was observed in rats treated with 1,3,5-TMB
at 428 mg/kg-day for 90 days Adenuga et al.. 2014 as cited in U.S. EPA (2016b). 1,4-DEB at
750 mg/kg-day for 6 weeks MHW, 1993b as cited in Oecd (1994). //-butylbenzene at
300 mg/kg-day for two generations Izumi et al.. 2005 as cited in U.S. EPA (2010). and HFAN at
>357 mg/kg-day for 13 weeks Biodynamics, 1990a and Mobil Oil Company, 1994 as cited in
U.S. EPA (2009c). Relative liver weight was likewise increased in HFAN-treated dogs at
500 mg/kg-day; however, this effect may have been influenced by decreased terminal body
weights (20% lower than controls) Biodynamics. 1990b as cited in U.S. EPA (2009c). With the
exception of rats treated with 1,3,5-TMB, rats that exhibited increased liver weights also showed
increased incidences of hepatocellular hypertrophy. A significantly increased incidence of
hepatocellular hypertrophy (in Fo and Fi parental males) was the basis for subchronic and
chronic RfDs for //-butylbenzene Izumi et al.. 2005 as cited in U.S. EPA (2010). In addition,
liver and kidney toxicity were designated as cocritical effects for screening-level subchronic and
chronic RfDs for //-propylbenzene, based on analogy to ethylbenzene U.S. EPA (2009d).
Data evaluating hepatotoxicity via inhalation exposure were available for four members
of the aromatic medium carbon range fraction. Figure B-4 is an exposure-response array
containing studies for which hepatic effects levels could be reliably determined. Relative liver
weight was biologically and statistically significantly increased at concentrations of >227 mg/m3
in rats and >454 mg/m3 in mice treated with isopropylbenzene for 14 weeks [NTP (2009);
Cushman et al.. 1995 as cited in U.S. EPA (1997) and at 2,823 mg/m3 in rabbits treated with
isopropylbenzene on GDs 6-18 Parmer et al. (1997). Increased organ weight was accompanied
by histopathological evidence of liver damage (e.g., eosinophilic foci and necrosis) after
105 weeks in mice Ntp (2009). but not rats Ntp (2009). No significant, treatment-related effects
on liver weights and/or histopathology were observed in rats exposed to 1,2,4- or 1,2,3-TMB at
concentrations up to 220 mg/m3 for 3 months Korsak et al. 2000a. b as cited in U.S. EPA
(2016b). in rats exposed to HFAN at up to 327 mg/m3 for 3 months Clark et al.. 1985 as cited in
U.S. EPA (2009c). or in rabbits exposed to HFAN at up to 1,000 mg/m3 Clark et al.. 1989 and
Ungvarv and Tartal. 1985 as cited in U.S. EPA (2009c).
50 Aromatic medium carbon range TPH fraction
-------
EPA 690 R-22 004F
i
• Doses > LOAEL
¦ LOAEL
ANOAEL
o Doses < NOAEL
o.
o
gj ^
Q. ro
>- ca
c
CM i- fa
cq — a.
-Q
(0
<-
n-Butyl-
benzene
>¦
CiD
O
_D
KD
O
fM fD
oO
Q.
O
-a
cu
o
fM
QD
D
C .
V \
"D
<
03
— DC
8
£ 5
33 ™
o rH
<-
1,3,5-
Trimethyl-
benzene
_D
g 2
5-! fO
K Q-
Q a>
u -c
LU *
O u
c « >¦
— _C 4-j
"O 'q CO
Qj QJ O C£l
.t: a. h • ?
"J ^
_q — -c
ro aj
G) i/>
rH _Q
- ru
5 r
X £
1,4-Diethyl benzene
cn
o
o
CM
<
CL
Q.
o
I—
a>
CL
u
cr>
o
o
.
If
J2
"O — m
$ U QC
u o -itf
in +-• >
as as >
Q. nr>
a; i-i
CD _C
CTl H
3 |
2 1
t/">
Z>
c
"D J-
JC
o =
TO *->
. o
ro ~
o ^
cn
a>
a> ol
CL
Qj
1 S!
!<-
CD
HFAN
_Q
o
Ch
E
CD
C
>
Q
O
CO
4->
5
GO
CD
O
O
>
13
5
no
<-
T—1
LT
0)
o
5
o
CQ
rsl
o
<
4—'
Q_
T3
LU
0)
¦M
on
.2
Z)
a>
c
_>•
T3
_Q
CD
on
¦M
on
a
O
l/l
Q.
rc
Figure B-3. Hepatic Effects in Animals after Oral Exposure to Aromatic Medium Carbon Range Compounds
51
Aromatic medium carbon range TPH fraction
-------
EPA 690 R-22 004F
10000
1000
60
E
c
o
•J3 100
to
/>
O
a.
x
• Cone > LOAEL
¦ LOAEL
a NOAEL
o Cone < NOAEL
10
0)
60
ru
i_
a;
"D
aj
,60
*
E
<-
= 3
WJ S_ 1
cn x-
-------
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
EPA/690/R-22/004F
Summary of Potentially Relevant Evidence
Oral studies examining liver effects were limited to three compounds and one mixture
(HFAN) in studies of 6 weeks to about 3 months in duration. All oral studies showed increases in
liver weight; this effect was typically accompanied by changes in liver histopathology (namely,
hepatocellular hypertrophy). Hepatic effects, predominantly consisting of increased relative
weights, were also seen in inhalation studies in laboratory animals exposed to members of the
aromatic medium carbon range fraction. Histological changes observed in the livers of animals
exposed to members of the fraction included hepatocellular hypertrophy in multiple subchronic
oral toxicity studies and eosinophilic foci in singular chronic inhalation toxicity studies (with
several studies reporting no histological effects). Lowest LOAELs for hepatic endpoints ranged
from 300 to 750 mg/kg-day in oral studies in rats (see Figure B-3; a LOAEL of 500 mg/kg-day
was also identified in dogs), and from 227 to 454 mg/m3 in subchronic inhalation studies in rats
and mice (see Figure B-4; excluding a LOAEL of 2,823 mg/m3 for increased liver weight in a
developmental toxicity study in rabbits). In aggregate, the data suggest that many aromatic
medium carbon range fraction compounds and mixtures can promote increases in rodent liver
weight, sometimes accompanied by histological changes.
RENAL EFFECTS
A renal endpoint (increased kidney weights in female F344 rats) is the critical effect for
the chronic RfD and the cocritical effect (with increased adrenal weights) for the chronic RfC for
isopropylbenzene U.S. EPA (1997). In addition, increased kidney weight serves as the critical
effects for the screening-level subchronic and chronic p-RfDs for tert- and .scc -butylbenzene U.S.
EPA (2012a. b); these toxicity values are based on the use of isopropylbenzene as an analogue
chemical. Effects on kidney histopathology were also identified as the critical effect in the study
used to derive screening-level subchronic and chronic p-RfDs for //-propylbenzene U.S. EPA
(2009d), based on the use of ethylbenzene as an analogue chemical. No human data pertaining to
the renal toxicity of aromatic medium carbon range fraction members were identified. As shown
in Table 5, data on renal effects in animals were located for seven members of the fraction. In
general, the renal endpoints evaluated in the studies were kidney weight and histology; many
also measured clinical chemistry parameters.
Human Studies
No human studies were available to address the potential renal effects of the TMB
isomers or other members of the aromatic medium carbon range fraction by any route of
exposure.
Animal Studies
Reliable data evaluating renal toxicity in animals after oral exposure to aromatic medium
carbon range compounds were available for five members of the fraction. Figure B-5 is an
exposure-response array containing studies for which renal effects levels could be reliably
determined. These studies identified increased relative kidney weight as the most sensitive renal
effect. Increased organ weight was observed in rats treated at >331 mg/kg-day with
isopropylbenzene over 192 days (females only evaluated) Wolf. 1956 as cited in U.S. EPA
(1997). 428 mg/kg-day with 1,3,5-TMB for 90 days Adenuga et at.. 2014 as cited in U.S. EPA
(2016b). 150 mg/kg-day with 1,4-DEB for about 6 weeks MHW. 1993b as cited in Oecd (1994).
300 mg/kg-day with //-butylbenzene for two generations (both generations) Izumi et al.. 2005 as
cited in U.S. EPA (2010). and >357 mg/kg-day with HFAN for about 13 weeks Biodvnamics.
1990a and Mobil Oil Company, 1994 as cited in U.S. EPA (2009c). Kidney weight was likewise
53 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
increased in HFAN-treated dogs at 500 mg/kg-day; however, this effect may have been
influenced by decreased terminal body weights (20% lower than controls) Biodvnamics. 1990b
as cited in U.S. EPA (2009c). Increased serum blood urea nitrogen (BUN) was observed at the
same dose as increased kidney weights in rats treated with 1,4-DEB MHW, 1993b as cited in
Oecd (1994). Although changes in kidney histopathology were noted in at least one study of
HFAN Mobil Oil Company. 1994 as cited in U.S. EPA (2009c). effects were seen in males only
and the "male kidney sections had changes that may be consistent with nephropathy typical of
male rats (dose-related hyaline droplet deposition, and nondose-related cortical tubular
degeneration, consisting primarily of epithelial swelling)" U.S. EPA (2009c). These toxicities
were consistent with rat-specific nephropathy, and U.S. EPA (1991) reported that these
pathologies may not be considered relevant to humans. As such, the results of this are not
presented in Figure B-5.
Renal effects seen after inhalation exposure to aromatic medium carbon range
compounds are almost strictly limited to significantly increased relative kidney weight.
Figure B-6 is an exposure-response array containing studies for which renal effects levels could
be reliably determined. Organ weight was increased in rats exposed to isopropylbenzene at
>454 mg/m3 for 14 weeks Ntp (2009), 1,055 mg/m3 for 13 weeks Cushman et at., 1995 as cited
in U.S. EPA (1997). and 526 mg/m3 for 4 weeks Monsanto Company. 1986 as cited in U.S. EPA
(1997); kidney weights were unaffected in mice treated at up to 907 mg/m3 for 14 or 105 weeks
Ntp (2009). Although histopathological effects (e.g., granular casts, mineralization of renal
papilla) were noted in isopropylbenzene-exposed rats, effects were seen in males only and were
considered in the assessment to be consistent with rat-specific nephropathy (i.e., their relevance
to humans is considered questionable) U.S. EPA (1997). In rats exposed to 1,2,4- or 1,2,3-TMB
at up to 220 mg/m3 for 3 months, there was no evidence of renal toxicity based on kidney
weights or histopathology Korsak et al. 2000a. b as cited in U.S. EPA (2016b). Similarly, rats
exposed to HFAN at up to 327 mg/m3 for 12 months showed no consistent signs of kidney
damage based on evaluations of clinical chemistry parameters and kidney histopathology Clark
et al.. 1989 as cited in U.S. EPA (2009c).
54 Aromatic medium carbon range TPH fraction
-------
EPA 690 R-22 004F
:
6
i
• Doses > LOAEL
¦ LOAEL
ANOAEL
o Doses < NOAEL
-D
01
CT> tj
U) QJ
i-H Q.
s. l/>
~ -M uo
U o 0)
^ c ns
O a) E
"O
0)
|z5
CO
Jii "o >
fD OJ UD
O C to
£ c 03
(*> t- 0)
LO 0)
(D
a>
o
•h 5 .E
>*
a>
<-
1,4-Diethylbenzene
I
S 03
< i
CL c
lu ro
? I
- (Q 4_,
"D Q- to
c C£L
O 4—• c
ro ^ 5b
LtT £* fN
O
O ^
(N
(L» _Q
U
n-Butyl-
benzene
>•
CL>
C
"n
<>-' 4-»
ai ns
o "S
= - *
c 0) fN
£ _Q
-- ro
kO
u">
a*
Isopropylbenzene
_D
s |
eg
< £
Q- ro
LU Q_
CO S
= Z
= o
"O c ¦*->
2 if £
^ p "i
ro E >
2 - ^
*17
O >
fN >
• >*
Z S
QJ
ro -*
§>-5
<
1,3,5-
Trimethyl-
benzene
- ro
•
a
o
CO
HFAN
>•
CD
C
"O
_Q ^
s ~
O)
Ch
bo
o
Q
$
• . CD
5. 4/1 "53
.9 => t
" .E 3
"O t
-------
EPA 690 R-22 004F
10000
1000
00
E.
c
•2 ioo
fC
LOAEL
¦ LOAEL
& NOAEL
o Cone < NOAEL
10
op
* 15 ro
CTi r-l
a>
o
Isopropylbenzene
t/> .12
D ¦*=
1,2,4-
Trimethyl-
benzene
£ o
TP % ™
q;
•— >* ...
u 4l ^
w c 5
ra TJ ¦="
-Q *H
O C
o o
V <11
"ra ^
S O
J*: *¦
ra
1,2,3-
Trimethyl-
benzene
D 5?
-- 2 m
¦o jc a
m ...
>* o
" £ E
_tu
ifc
HFAN
Figure B-6. Renal Effects in Animals after Inhalation Exposure to Aromatic Medium Carbon Range Compounds
56
Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
Summary of Potentially Relevant Evidence
Oral studies examining kidney effects were limited to four compounds and one mixture
(HFAN) in studies of 4-27 weeks in duration. All studies showed increases in kidney weight;
one study also reported a change in serum chemistry consistent with kidney damage
(i.e., increased BUN). Changes in kidney histology were reported only in male rats for the four
compounds and one mixture; these effects were consistent with rat-specific nephropathy. Kidney
effects (increased relative kidney weights) were also seen in inhalation studies of animals
exposed to one member of the aromatic medium carbon range fraction (isopropylbenzene). The
lowest LOAELs (by compound or mixture) for increased kidney weights ranged from 150 to
500 mg/kg-day in oral studies in rats and dogs (see Figure B-5). For inhalation exposure, the
lowest LOAEL was 454 mg/m3 for subchronic exposure to isopropylbenzene in rats
(see Figure B-6). Inhalation studies of other compounds comprising the medium carbon fraction
(including TMB isomers) did not indicate significant, treatment-related changes in kidney
weights or histology. Taken together, the data indicate that several members of the aromatic
medium carbon range fraction compounds and mixtures can produce increases in rodent kidney
weight, sometimes accompanied by serum chemistry and histological changes.
BODY-WEIGHT EFFECTS
Decreased maternal body weight on GD 15 is the critical effect in the study used to
derive the subchronic and chronic provisional reference concentrations (p-RfCs) for HFAN U.S.
EPA (2009c). No human studies examining body-weight effects of aromatic medium carbon
range compounds were identified in the sources reviewed. As Table 5 shows, animal studies
(oral or inhalation) that examined body weight as an endpoint are available for nearly all of the
compounds and mixtures with toxicity data. Exceptions are //-propylbenzene and tert- and
.sfc-butylbenzene. In this section, body-weight changes of at least 10% relative to controls in
adult animals are considered LOAELs, and smaller changes are not. For studies that reported
body-weight gain but did not report absolute body weights, and for studies of maternal weight
gain during gestation, statistically significant changes from control are described.
Human Studies
No human studies were available to address the potential for impacts on body weight of
the TMB isomers or other members of the aromatic medium carbon range fraction by any route
of exposure.
Animal Studies
Figure B-7 shows the effects of orally-administered aromatic medium carbon range
compounds and mixtures on body weight; data are available for seven materials. In studies of
1,3,5-TMB, 1,3-DEB, and //-butylbenzene, no treatment-related effects on body weight or
body-weight gain were observed Adenuga et al.. 2014 as cited in U.S. EPA (2016b; Izurni. 2005
as cited in U.S. EPA (2010; Gagnaire et al. (1990). However, decreased body
weight/body-weight gain was seen after exposure to rats to 1,4-DEB at 750 mg/kg-day for
6 weeks MHW, 1993b as cited in Oecd (1994), 1,2-DEB at >43 mg/kg-day for 8 weeks Gagnaire
et al. (1992a; Gagnaire et al. (1990). HFAN at >625 mg/kg-day on GDs 6-15 Biodvnamics.
1990c as cited in U.S. EPA (2009c). and a mixture of DEB isomers at >357 mg/kg-day for
10 weeks Gagnaire et al. (1990). Dogs administered HFAN at 500 mg/kg-day (highest dose
tested) for up to 90 days showed a 20% reduction in terminal body weights [albeit not
statistically significant Biodvnamics. 1990c as cited in U.S. EPA (2009c)l.
57 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
Body-weight effects evaluated in subchronic and chronic inhalation toxicity studies are
shown in Figure B-8; effects evaluated in developmental inhalation toxicity studies are shown in
Figure B-9. Data via the inhalation route were available for six members of the aromatic medium
carbon range fraction and a mixture of members of this fraction. As shown in Figure B-8,
significant effects in subchronic or chronic studies are limited to rats exposed to mixtures
comprising the aromatic medium range carbon fraction. HFAN-treated rats showed decreased
body weights after treatment at 1,157 mg/m3 for 90 days Douglas et al.. 1993 as cited in U.S.
EPA (2009c). In addition, body-weight gain was significantly reduced in rats exposed to mixed
DEB isomers at >486 mg/m3 for 18 weeks Gagnaire et al. (1992b). In contrast, no significant
effects on body weight were observed following exposure of rats and mice to isopropylbenzene
at up to 907 mg/m3 for 14 or 105 weeks Ntp (2009) or in rats exposed to isopropylbenzene at up
to 1,055 mg/m3 for 13 weeks Cushman et al.. 1995 as cited in U.S. EPA (1997). In addition, rats
exposed to 1 -methyl-4-ethylbenzene at up to 417 mg/m3 for 4 weeks Swiercz et al. (2000). or
any of the three individual TMB isomers at up to 220 mg/m3 for 3 months showed no significant,
treatment-related effects on body weight or body-weight gain Wiaderna et al.. 2002. Korsak et
al.. 2000a. b. Gralewicz and Wiaderna. 2000. Wiaderna et al.. 1998. Gralewicz et al.. 1997a. and
Korsak and Rvdzvnski. 1996. 1997. all as cited in U.S. EPA (2016b).
Significant effects on body weight/body-weight gain were observed in dams exposed to
members of the aromatic medium carbon range fraction during the gestational period
(see Figure B-9). Rats and rabbits exposed to high concentrations of isopropylbenzene during
gestation (at >1,488 and 2,823 mg/m3, respectively) showed significantly reduced body-weight
gains Parmer et al. (1997). After gestational exposure to 1,3,5-TMB (at >369 mg/m3) or
1,2,4-TMB (at >738 mg/m3), body-weight gain was likewise significantly decreased Saillenfait
et al.. 2014 as cited in U.S. EPA (2016b). Rabbits exposed to HFAN at up to 1,000 mg/m3 on
GDs 7-20 showed no significant, treatment-related effects on body weight Ungvarv and Tatrai.
1985 as cited in U.S. EPA (2009c). but mice exposed during gestation were affected at
concentrations as low as 125 mg/m3 McKee et al.. 1990 as cited in U.S. EPA (2009c).
58 Aromatic medium carbon range TPH fraction
-------
EPA 690 R-22 004F
• Doses > LOAEL
¦ LOAEL
A NOAEL
o Doses < NOAEL
i
A
O
ra
o
o
fN
CTi
Ch
Q>
Ob
(Tl
o>
1
tH
H
4—1
•
f0
75
(0
15
m
oc
4—1
cc
¦*->
¦*->
1,2-Diethylbenzene
5
Di ethyl-
benzene
Mixture
<3-
CTi
a\
_Q
m
o
-t-1
Ch
u
cn»
C£
<—I
o
if
§
c
~=
T5
'vD
cr>
5
I .£ 2
E
>¦ 0>
Q *-*
° 'S
CO ro
c
>>
O
o
cn
O
<0
o
O
Ol
CTi
o>
§
V-J
cr.
CTi
§
«?r
8
Psl
Cv]
cn
rsj
<
Gl
-t-J
CO
oc
i-H
U
<
CL
+-j
<0
CT1
*—i
<
CL
+-»
03
LLl
LA
c
LU
cc
o
LLl
DC
i/5
*H
1/)
to
if
o
CO
if
Z)
o
3
5
o
3
5
c
oTl
E
c
c
ro
¦—
Q
ru
1—i
¦—
H
T3
e?
c
TJ
-Q
TJ
CD
>¦
p
2
¦4—>
O
"o
'o
00
o
c
T3 ra
S ^
° c
on cu
(o 00
in ^
o
o
C"J
n-Butyl-
benzene
o
G\
OV
o>
o>
"ro
c
00
(TJ
e>
1,3-Diethyl-
benzene
_Q
O ^
rsl
- Q_ f0
15 ^ a:
S ^
M r ro
D •- r-l
F
-------
EPA 690 R-22.004F
10000
"So1000
E,
c
_o
2
4-»
c
cu
u
c
o
u
0) 100
3
CO
O
a
x
LU
0)
to
ro
k.
(U
-O
cu
'55
g
i
LOAEL
¦ LOAEL
A NOAEL
o Cone < NOAEL
10
3
5
in ro
I 3
Isopropyl benzene
1-Methyi-
4-ethyi
benzene
3 <
u 3
U o
1,3,5-Tri
methyl
benzene
ro
CD
ro
ro
ro
cc
cc
cc
..
rH
U0
ai
H
O
g.
tH
o
T3
O
(N
O
fM
o
"ro
rM
ro
cm
»
CC
<
Cl
LU
Cl
T3
C
Q.
LU
ai
CL
UJ
"i
"c
D
§
"i
ro
ro
«i
D
HFAN
Figure B-8. Body-Weight Effects in Animals after Subchronic and Chronic Inhalation Exposure to Aromatic Medium Carbon Range
Compounds
60
Aromatic medium carbon range TPH fraction
-------
EPA 690 R-22 004F
10000
1000
no
E,
c
o
ra
c
a;
u
c
~
u
0)
1—
D
I/)
o
a.
x
a;
txo
ra
l_
a)
>
<
"C
a>
.2?
'5
3
ai
E
100
10
~ Cone > LOAEL
¦ LOAEL
A NOAEL
° Cone < NOAEL
HFAN
Isopropylbenzene
Figure B-9. Body-Weight Effects in Animals after Gestational Inhalation Exposure to Aromatic Medium Carbon Range Compounds
61
Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
Summary of Potentially Relevant Evidence
Compounds and mixtures in the aromatic medium carbon range fraction have been shown
to reduce body weights of rats, mice, rabbits, and dogs after oral and inhalation exposure.
Members that induced body-weight changes in oral studies in rats included HFAN and DEBs
(both individual isomers and mixed isomers); LOAELs for these effects ranged from
43 mg/kg-day (for 1,2-DEB) to 893 mg/kg-day (for HFAN) (see Figure B-7). In subchronic
inhalation toxicity studies, rats exposed to HFAN and DEB mixtures (at 1,157 and >486 mg/m3,
respectively) showed reductions in body weight/body-weight gain, whereas rats and mice
exposed to individual compounds from the fraction (in the case of many compounds, at lower
concentrations and/or for shorter time periods) did not (see Figure B-8). In general, rats, mice,
and rabbits exposed to aromatic medium carbon range fraction compounds or mixtures during
gestation showed treatment-related body-weight deficits; LOAELs for this effect were
125 mg/m3 for mice exposed to HFAN, between 369 and 1,488 mg/m3 for rats exposed to
1,3,5-TMB, 1,2,4-TMB, or isopropylbenzene, and 2,823 mg/m3 in rabbits exposed to
isopropylbenzene (see Figure B-9). Taken together, the inhalation and oral animal data indicate
that compounds in the aromatic medium carbon range fraction can be expected to induce
body-weight reductions at sufficiently high doses (generally >500 mg/kg-day or
duration-adjusted concentrations >300 mg/m3 for most compounds or mixtures).
HEMATOLOGICAL EFFECTS
A hematological endpoint (anemia in dogs, with males being affected more than females)
is the critical effect for the screening-level subchronic and chronic p-RfDs for HFAN U.S. EPA
(2009c). Hematological effects (including anemia and effects on blood clotting) have been
reported for humans occupationally exposed to solvent mixtures including TMBs;
hematotoxicity cannot be attributed to specific isomers. Hematological data in humans are
limited to TMBs; there are animal data (inhalation or oral) for six members of the aromatic
medium carbon range fraction.
Human Studies
Hematological effects have been reported in workers exposed by inhalation to mixtures
containing TMB isomers. Workers exposed to paint solvent containing 50% 1,2,4-TMB, 30%
1,3,5-TMB, and unspecified amounts of 1,2,3-TMB (listed as possibly present) exhibited
alterations in blood clotting and anemia at 295 mg/m3 U.S. EPA (2016b). Because occupational
exposure studies involve solvent mixtures, hematological and clinical chemistry effects cannot
be attributed to TMB isomers and may be due to other agents in the mixture. For example, the
hematological effects may be attributed to trace amounts of benzene present in the solvent
mixture U.S. EPA (2016b).
62 Aromatic medium carbon range TPH fraction
-------
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
EPA/690/R-22/004F
Animal Studies
Reliable data regarding the hematotoxicity of aromatic medium carbon range compounds
following oral exposure are limited to three members of the fraction. Figure B-10 is an
exposure-response array containing studies for which hematological effects levels could be
reliably determined. Increased monocytes were observed in rats treated with 1,3,5-TMB at
>143 mg/kg-day for 90 days Adenuga et al.. 2014 as cited in U.S. EPA (2016b). but no
significant effects were reported after treatment with 1,4-DEB at up to 750 mg/kg-day for
6 weeks MHW, 1993b as cited in Oecd (1994). Although one 13-week study reported no
significant, treatment-related effects on hematological parameters in rats treated with HFAN at
up to 893 mg/kg-day Mobil Oil Company. 1994 as cited in U.S. EPA (2009c). two other studies
of similar duration identified effects consistent with anemia (i.e., reductions in red blood cells
[RBCs], hematocrit [Hct], and/or hemoglobin [Hb] levels) in rats (females only) treated at
1,250 mg/kg-day for up to 96 days Biodvnamics. 1990a as cited in U.S. EPA (2009c) and in
dogs treated at >250 mg/kg-day for 90 days Biodvnamics. 1990b as cited in U.S. EPA (2009c).
There are limited data for hematological effects following inhalation exposure to
members of the aromatic medium carbon range fraction. Figure B-l 1 is an exposure-response
array containing studies for which hematological effects levels could be reliably determined. No
significant, treatment-related hematological effects were observed in rats or mice exposed to
isopropylbenzene at up to 907 mg/m3 for 14 weeks Ntp (2009). Rats exposed to TMB isomers
(1,2,4- and 1,2,3-TMB) at >88 mg/m3 showed decreased clotting time, decreased segmented
neutrophils, decreased RBCs, and/or changes in differential white blood cell (WBC) counts
Korsak et al. 2000a. b as cited in U.S. EPA (2016b). Effects consistent with anemia (decreased
Hct and mean corpuscular volume [MCV]) were reported in mice exposed to HFAN at
1,858 mg/m3 (highest concentration tested) on GDs 6-15 McKee et al.. 1990 as cited in U.S.
EPA (2009c); in contrast, a different study in rats reported no hematological effects following
12 months of exposure to HFAN at concentrations up to approximately 300 mg/m3 Clark et al..
1989 as cited in U.S. EPA (2009c).
63 Aromatic medium carbon range TPH fraction
-------
EPA 690 R-22 004F
10000
1000
>
TO
-a
i
HQ
tuD
100
0
G
"O
cu
k-
0)
*-»
.S2
'E
E
"O
<
*<5
Q
01
cxo
03
u_
0)
£
10
1
• Doses > LOAEL
¦ LOAEL
ANOAEL
° Doses < NOAEL
_Q
<—I
O
fM
<
Q_
,E In
"S §¦ |
u f-1-
~ o -
w X >
(n O >
,f E "5
O <-
(N
00
3
c
CI
¦D
<
1,3,5-
Trimethyl-
benzene
CT>
O
O
rNi
<
CL
LU
CO
Z>
.£ Dfi
£ ?
» %
ra ro
S 10
2 cc
J
c
O
CD
O")
o
o
rsi
<
CL ^
LU _£>
UO fO
c «
c ^
Ji:
u oc
1
CO
HFAN
"O ro ro
2i £ cc
7i 00 •-
° o -*
!8 o ^
J H
a £
u u
LU
-------
EPA 690 R-22 004F
10000
• Cone > LOAEL
¦ LOAEL
a NOAEL
o Conc< NOAEL
1000
60
E,
c
o
£ 100
c
0)
u
c
o
u
ai
w
3
in
£ 10
a>
60
m
L_
0J
Ǥ
"D
a>
bp
'a;
3
O)
E
"> ra
=> E
II
o o
rN u
¦sf ->
1,2,4-
Trimethyl-
benzene
o
N irt
>
flj 41
. c m
S "
Q w
8
oi ai
£
aJ ^
1,2,3-Trimethylbenzene
D
no • -
f °
o c
|