United States
Environmental Protection
1=1 m m Agency
EPA/690/R-11/022F
Final
4-21-2011
Provisional Peer-Reviewed Toxicity Values for
Diisopropyl ether
(CASRN 108-20-3)
Superfund Health Risk Technical Support Center
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
CHEMICAL MANAGER
Nina Ching Y. Wang, PhD
National Center for Environmental Assessment, Cincinnati, OH
DRAFT DOCUMENT PREPARED BY
SRC, Inc.
7502 Round Pond Road
North Syracuse, NY 13212
PRIMARY INTERNAL REVIEWERS
Audrey Galizia, Dr. PH.
National Center for Environmental Assessment, Washington, DC
Dan D. Petersen, PhD, DABT
National Center for Environmental Assessment, Cincinnati, OH
This document was externally peer reviewed under contract to
Eastern Research Group, Inc.
110 Hartwell Avenue
Lexington, MA 02421-3136
Questions regarding the contents of this document may be directed to the U.S. EPA Office of
Research and Development's National Center for Environmental Assessment, Superfund Health
Risk Technical Support Center (513-569-7300).
-------
TABLE OF CONTENTS
COMMONLY USED ABBREVIATIONS ii
BACKGROUND 1
HISTORY 1
DISCLAIMERS 1
QUESTIONS REGARDING PPRTVS 2
INTRODUCTION 2
REVIEW 01 PERTINENT DATA 3
HUMAN STUDIES 3
ANIMAL STUDIES 3
Oral Exposure 3
Inhalation Exposure 3
Other Studies 9
Acute or Short-term Studies 9
Genotoxicity 9
FEASIBILITY OF DERIVING PROVISIONAL SUBCHRONIC AND CHRONIC
ORAL RID VALUES FOR DIISOPROPYL ETHER 10
DERIVATION OF PROVISIONAL SUBCHRONIC AND CHRONIC INHALATION
RfC VALUES I OR DIISOPROPYL ETHER 10
CHRONIC p-RfC 13
PROVISIONAL CARCINOGENICITY ASSESSMENT FOR DIISOPROPYL ETHER 14
REFERENCES 15
APPENDIX A. DETAILS OF BENCHMARK DOSE MODELING FOR
SUBCHRONIC RfC 18
l
Diisopropyl Ether
-------
COMMONLY USED ABBREVIATIONS
BMC
benchmark concentration
BMCL
benchmark concentration lower bound 95% confidence interval
BMD
benchmark dose
BMDL
benchmark dose lower bound 95% confidence interval
HEC
human equivalent concentration
HED
human equivalent dose
IUR
inhalation unit risk
LOAEL
lowest-observed-adverse-effect level
LOAELadj
LOAEL adjusted to continuous exposure duration
LOAELhec
LOAEL adjusted for dosimetric differences across species to a human
NOAEL
no-ob served-adverse-effect level
NOAELadj
NOAEL adjusted to continuous exposure duration
NOAELhec
NOAEL adjusted for dosimetric differences across species to a human
NOEL
no-ob served-effect level
OSF
oral slope factor
p-IUR
provisional inhalation unit risk
POD
point of departure
p-OSF
provisional oral slope factor
p-RfC
provisional reference concentration (inhalation)
p-RfD
provisional reference dose (oral)
RfC
reference concentration (inhalation)
RfD
reference dose (oral)
UF
uncertainty factor
UFa
animal-to-human uncertainty factor
UFC
composite uncertainty factor
UFd
incomplete-to-complete database uncertainty factor
UFh
interhuman uncertainty factor
UFl
LOAEL-to-NOAEL uncertainty factor
UFS
subchronic-to-chronic uncertainty factor
WOE
weight of evidence
11
Diisopropyl Ether
-------
FINAL
4-21-2011
PROVISIONAL PEER-REVIEWED TOXICITY VALUES FOR
DIISOPROPYL ETHER (CASRN 108-20-3)
BACKGROUND
HISTORY
On December 5, 2003, the U.S. Environmental Protection Agency's (EPA) Office of
Superfund Remediation and Technology Innovation (OSRTI) revised its hierarchy of human
health toxicity values for Superfund risk assessments, establishing the following three tiers as the
new hierarchy:
1) EPA's Integrated Risk Information System (IRIS)
2) Provisional Peer-Reviewed Toxicity Values (PPRTVs) used in EPA's Superfund
Program
3) Other (peer-reviewed) toxicity values, including
~ Minimal Risk Levels produced by the Agency for Toxic Substances and Disease
Registry (ATSDR);
~ California Environmental Protection Agency (CalEPA) values; and
~ EPA Health Effects Assessment Summary Table (HEAST) values.
A PPRTV is defined as a toxicity value derived for use in the Superfund Program when
such a value is not available in EPA's IRIS. PPRTVs are developed according to a Standard
Operating Procedure (SOP) and are derived after a review of the relevant scientific literature
using the same methods, sources of data, and Agency guidance for value derivation generally
used by the EPA IRIS Program. All provisional toxicity values receive internal review by a
panel of six EPA scientists and external peer review by three independently selected scientific
experts. PPRTVs differ from IRIS values in that PPRTVs do not receive the multiprogram
consensus review provided for IRIS values. This is because IRIS values are generally intended
to be used in all EPA programs, while PPRTVs are developed specifically for the Superfund
Program.
Because new information becomes available and scientific methods improve over time,
PPRTVs are reviewed on a 5-year basis and updated into the active database. Once an IRIS
value for a specific chemical becomes available for Agency review, the analogous PPRTV for
that same chemical is retired. It should also be noted that some PPRTV documents conclude that
a PPRTV cannot be derived based on inadequate data.
DISCLAIMERS
Users of this document should first check to see if any IRIS values exist for the chemical
of concern before proceeding to use a PPRTV. If no IRIS value is available, staff in the regional
Superfund and Resource Conservation and Recovery Act (RCRA) program offices are advised to
carefully review the information provided in this document to ensure that the PPRTVs used are
appropriate for the types of exposures and circumstances at the Superfund site or RCRA facility
in question. PPRTVs are periodically updated; therefore, users should ensure that the values
contained in the PPRTV are current at the time of use.
1
Diisopropyl Ether
-------
FINAL
4-21-2011
It is important to remember that a provisional value alone tells very little about the
adverse effects of a chemical or the quality of evidence on which the value is based. Therefore,
users are strongly encouraged to read the entire PPRTV document and understand the strengths
and limitations of the derived provisional values. PPRTVs are developed by the EPA Office of
Research and Development's National Center for Environmental Assessment, Superfund Health
Risk Technical Support Center for OSRTI. Other EPA programs or external parties who may
choose of their own initiative to use these PPRTVs are advised that Superfund resources will not
generally be used to respond to challenges of PPRTVs used in a context outside of the Superfund
Program.
QUESTIONS REGARDING PPRTVS
Questions regarding the contents of the PPRTVs and their appropriate use (e.g., on
chemicals not covered, or whether chemicals have pending IRIS toxicity values) may be directed
to the EPA Office of Research and Development's National Center for Environmental
Assessment, Superfund Health Risk Technical Support Center (513-569-7300), or OSRTI.
INTRODUCTION
No RfD, RfC, or cancer assessment for diisopropyl ether (DIPE; see Figure 1 for
chemical structure) is available on IRIS (U.S. EPA, 2009), in the HEAST (U.S. EPA, 1997), or
in the Drinking Water Standards and Health Advisories list (U. S. EPA, 2006). No relevant
documents were located in the Chemical Assessments and Related Activities (CARA) list
(U.S. EPA, 1991a, 1994a). The ATSDR (2008) has not published a Toxicological Profile for
DIPE, and no Environmental Health Criteria Document is available from the World Health
Organization (WHO, 2008). The carcinogenicity of DIPE has not been assessed by the
International Agency for Research on Cancer (IARC, 2008) or the National Toxicology Program
(NTP, 2005, 2008). The American Conference for Governmental Industrial Hygienists (ACGIH,
2007) has adopted a threshold limit value-time-weighted average (TLV-TWA) of 250 ppm
(1040 mg/m3) and a threshold limit value-short-term exposure limit (TLV-STEL; not to exceed
"3
15-minute exposure over an 8-hour work shift) of 310 ppm (1300 mg/m ) as protective against
irritation. The National Institute of Occupational Safety and Health-recommended exposure
"3
limit (REL) is 500 ppm (2090 mg/m ) based on irritation of eyes, skin, and respiratory system
and central nervous system effects (NIOSH, 2008). The Occupational Safety and Health
Administration permissible exposure limit (PEL) is 500 ppm (OSHA, 2008).
Figure 1. Chemical Structure of DIPE
Literature searches were conducted from the 1960s through March 2011 for studies
relevant to the derivation of provisional toxicity values for DIPE. Databases searched include
MEDLINE, TOXLINE (with NTIS), BIOSIS, TSCATS/TSCATS2, CCRIS, DART, GENETOX,
HSDB, RTECS, Chemical Abstracts, and Current Contents (last 6 months). An IUCLID Data
2
Diisopropyl Ether
-------
FINAL
4-21-2011
Set for DIPE submitted by ExxonMobil Biomedical Sciences (2005) under EPA's High
Production Volume (HPV) Challenge Program was also reviewed for relevant information.
An evaluation of the cancer literature indicates that a major study related to the
carcinogenicity of DIPE has been conducted by the Ramazzini Institute. Following a report from
the National Toxicology Program (NTP), EPA has placed the development of health
assessments, such as DIPE, that may rely on Ramazzini Institute cancer data on hold. The NTP
report, referred to in EPA's June 15, 2010 press release (U.S. EPA, 2010), recommended that
pathology reviews be carried out to resolve differences of opinion in the diagnoses of certain
tumors reported in a methanol research study completed by the Ramazzini Institute. As a result,
EPA and the National Institute of Environmental Health Sciences (NIEHS) are jointly
sponsoring an independent Pathology Working Group (PWG) review of select studies conducted
at the Institute. The cancer assessment for DIPE will remain on hold until the completion of the
PWG review.
REVIEW OF PERTINENT DATA
HUMAN STUDIES
Silverman et al. (1946) exposed a mixed-sex group of 12 human subjects to 300 ppm
"3
(1250 mg/m ) of DIPE vapor for 15 minutes. No irritation of the eyes, nose, or throat was
reported; although, about one-third of study subjects objected to the unpleasant odor of the
solvent at this concentration. No studies were located examining the effects of longer-term
inhalation exposure or oral exposure in humans.
ANIMAL STUDIES
Oral Exposure
No relevant noncancer studies on oral exposure to DIPE have been located.
Inhalation Exposure
Subchronic and developmental inhalation studies were conducted by Dalbey and
Feuston (1996). In the subchronic study, groups of Sprague-Dawley rats (14/sex) were
whole-body exposed to 0 (untreated), 0 (sham-exposed), 2000, 13,800, or 29,700 mg/m3 (0, 480,
3300, or 7100 ppm) of DIPE 6 hours/day, 5 days/week, for approximately 13 weeks. Because
commercial grade (92% pure) test material was used, test animals were also exposed to low
concentrations of a mixture containing more than 20 low molecular weight alkanes,
cycloalkanes, alkenes, alcohols, and ketones. The DIPE concentrations reported above represent
91-95% of the total chemical exposures in the treatment groups. Sham-exposed controls were
individually housed in the inhalation chambers, and untreated controls were observed in a
separate animal room. Food and water were provided ad libitum but not during exposures. Test
animals were monitored daily during the week (not on weekends) for clinical signs, and
individual body weights were recorded weekly. Blood samples collected prior to terminal
sacrifice were analyzed for serum chemistry (glucose, urea nitrogen [BUN], total protein,
albumin, globulin, A/G ratio, sorbitol dehyrogenase [SDH], aspartate aminotransferase [AST],
alanine aminotransferase [ALT], alkaline phosphatase [ALP], total bilirubin, creatinine,
cholesterol, triglycerides, uric acid, chloride, calcium, sodium, potassium, and phosphorus) and
hematology (white blood cells [WBCs], red blood cells [RBCs], hemoglobin [Hgb], hematocrit
[Hct], mean corpuscular volume [MCV], mean corpuscular hemoglobin [MCH], mean
3
Diisopropyl Ether
-------
FINAL
4-21-2011
corpuscular hemoglobin concentration [MCHC], platelets, and differential cell count).
Following sacrifice, all test animals were necropsied, and organ weights (adrenals, kidney,
spleen, brain, liver, testes, epididymides, ovaries, thymus, heart, prostate, uterus, and the right
middle lung lobe) were collected. Slides for histopathological examination were prepared from
over 40 tissues, and all gross lesions in the sham-exposed and high-dose groups; liver and kidney
in the mid-dose males; and lungs, tracheobronchial lymph nodes, and gross lesions in the
untreated control group. The left cauda epididymides from 10 male rats in each control group
and the high-dose group were used for evaluation of sperm morphology and number.
No mortality was reported, and there were no treatment-related clinical signs observed
over the course of the study (Dalbey and Feuston, 1996). Treated males tended to gain more
weight compared to controls during the first half of the study. A statistically significant
(p < 0.05) difference was seen in mid-dose males (see Table 1). Similar trends were not
observed in female rats. Serum chemistry and hematology analyses were generally
unremarkable, except for a statistically significant (p < 0.05) increase in serum cholesterol in
high-dose males (see Table 2). Absolute liver weights were statistically significant increased
(p < 0.05) in males and females in the mid- and high-dose groups in a dose-related manner (see
Table 1). Relative liver weights were not reported, but a comparison of the ratio of mean liver
weight to mean terminal body weight in the different study groups suggests that relative liver
weights were also increased in these groups in relation to dose. Microscopic examination
revealed mild hepatocellular hypertrophy only in the male rats from the high-dose group.
Absolute kidney weights of mid- and high-dose males were significantly increased (see Table 1).
Comparison of the ratio of mean kidney weight to mean terminal body weight in the different
study groups suggests that relative kidney weight was not increased in the mid-dose group and
only slightly increased in the high-dose group. Microscopic examination of the kidney showed a
mild increase in hyaline droplets in the proximal convoluted tubules of males of the high-dose
group only. No other organs had changes in weight or morphology attributed to exposure to
DIPE. There were no differences between treated males and controls (both untreated and
sham-exposed) in sperm or spermatid counts. The number of abnormal sperm was significantly
increased in high-dose males (5.3% versus 2.8% in control rats), but this increase was not
considered by the researchers to be of biological significance because no specific type of
abnormality was increased and because the prevalence of abnormal sperm in the high-dose group
was within the range of historical controls (2.85.6%).
4
Diisopropyl Ether
-------
FINAL
4-21-2011
Table 1. Absolute Body and Organ Weights of Sprague-Dawley Rats Exposed to DIPE
by Inhalation for 90 Days"'
Exposure Group (mg/m3)
Organ
0
(untreated)
0
(sham-exposed)
2000
13,800
29,700
Males
Body weight (g)
438 ±28
449 ± 34
466 ± 34
482 ± 32°
462 ± 36
Liver (g)
12.1 ± 1.4
12.3 ± 1.3
12.8 ± 1.5
15.4 ± l.ld
16.9 ± 2.2d
Kidneys (g)
2.92 ±0.26
2.86 ±0.36
2.96 ±0.33
3.24 ± 0.27d
3.26 ± 0.43d
Females
Body weight (g)
287 ± 20
276 ± 24
280 ± 19
276 ± 17
280 ± 17
Liver (g)
8.04 ±0.81
7.45 ± 1.09
7.64 ±0.68
8.23 ± 1.16s
9.11 ± 0.81d
Kidneys (g)
1.86 ±0.14
1.81 ±0.15
1.79 ± 0.11
1.90 ±0.15
1.94 ± 0.18e
aDalbey and Feuston (1996).
Values are presented as means ± SD.
"Reported to be significantly different from untreated controls by the researchers but />-values not shown. Following
methods reported by the researchers, ANOVA was performed for this review, followed by group comparisons using
Duncan's multiple range test. Based on this evaluation, it appears the study authors were evaluating statistical
significance at p< 0.05. However, not all of the statements about statistical significance made by the authors were
validated at this level. Discrepancies were that the re-analysis showed a statistically significant (p < 0.05) difference
from both control groups, rather than just untreated controls, for body weight in mid-dose males and no difference
from sham-exposed controls for liver weight in mid-dose females.
Significantly different from both control groups (/?-value not reported, see footnote c for further discussion).
"Significantly different from sham-exposed controls (/?-value not reported, see footnote c for further discussion).
Table 2. Serum Chemistry and Hematology Values in Sprague-Dawley Rats Exposed to
DIPE by Inhalation for 90 Days"'
Parameter
Exposure Group (mg/m3)
0
(untreated)
0
(sham-exposed)
2000
13,800
29,700
Males
Creatinine (mg/dL)
0.61 ±0.06
0.64 ± 0.04
0.64 ± 0.04
0.67 ± 0.06°
0.69 ± 0.03°
Cholesterol (mg/dL)
71 ± 10
74 ± 13
77 ± 17
77 ±9
95 ± 22d
SDH (IU/L)
11 ± 5
16 ±7
13 ±6
9 ±3'
9 ±3'
Lymphocytes0
92 ±3
92 ±4
90 ±6
90 ±4
87 ± 6°
Monocytes0
1 ± 2
1 ± 2
2 ± 2
2 ± 2
3 ± 2°
Females
Potassium (mmol/L)
4.96 ±0.35
4.68 ±0.24
4.58 ±0.41
4.51 ±0.37°
4.45 ± 0.4°
Lymphocytes0
92 ±3
88 ±5
86 ±6°
85 ± 7°
86 ± 3°
aDalbey and Feuston (1996).
Values are presented as means ± SD.
"Reported to be significantly different from untreated controls by the researchers, but /^-values not shown. Following
methods reported by the researchers, ANOVA was performed for this review, followed by group comparisons using
Tukey 's studentized range test. Based on this evaluation, it appears the study authors were evaluating statistical
significance at p< 0.05. However, not all of the statements about statistical significance made by the authors were
validated at this level. Discrepancies were that the re-analysis showed statistically significant (p < 0.05) differences
from both control groups, rather than just untreated controls, for creatinine in high-dose males and that levels of
lymphocytes and monocytes in high-dose males were not different than untreated controls.
Significantly different from both control groups (/?-value not reported, see footnote d for further discussion)
"Percent of total WBCs.
Significantly different from sham-exposed controls (/?-value not reported, see footnote d for further discussion).
5
Diisopropyl Ether
-------
FINAL
4-21-2011
Dalbey and Feuston (1996) identified the liver as the most sensitive target in male rats for
DIPE, with no effects at 2000 mg/m3, increases in liver weight at 13,800 mg/m3 and larger
increases in liver weight, hepatocellular hypertrophy, and increased serum cholesterol at
29,700 mg/m3. The changes observed in this study were increased liver weights, hepatocellular
hypertrophy, and elevated levels of serum cholesterol. The only other target identified in this
study was the kidney in male rats. Some of the experimental data suggest that development of
kidney toxicity in male rats following exposure to DIPE may involve an a2U-globulin-mediated
mode of action. Generally, kidney effects observed in animals are assumed to be relevant for
assessment of human toxicity. However, a number of chemicals have been shown to induce
accumulation of 2U-globulin in hyaline droplets in male rat kidney. The 2U-globulin accumulation
in hyaline droplets initiates a sequence of events that leads to renal nephropathy and, eventually,
to renal tubular tumor formation. The phenomenon is unique to the male rats since female rats
and other laboratory mammals administered the same chemicals do not accumulate 2u-globulin in
the kidney and do not develop renal tubule tumors (U.S. EPA, 1991b). However, there is a lack
of a2U-globulin immunohistochemical data for DIPE-induced nephrotoxicity. In the absence of
minimum information demonstrating the involvement of a2U-globulin processes, male rat renal
toxicity associated with exposure to DIPE is considered relevant for risk assessment purposes.
The study authors identified a no-observed-effect-level (NOEL) of 2000 mg/m3. For the purpose
"3
of this review, a LOAEL of 13,800 mg/m based on the increased liver weight (-28% in males;
6% in females) and aNOAEL of 2000 mg/m3 are identified.
In the developmental study, groups of 22 mated female Sprague-Dawley rats were
whole-body exposed to 0 (untreated), 0 (sham-exposed), 1800, 12,940, or 28,200 mg/m3 (0, 430,
3095, or 6745 ppm) DIPE vapor for 6 hours/day on Days 6-16 of gestation (Dalbey and Feuston,
1996). Because commercial grade (92% pure) test material was used, test animals were also
exposed to low concentrations of other chemicals, but the DIPE concentrations reported above
represent 92-95% of the total chemical exposures in the treatment groups. The sham-exposed
controls were housed in the study chambers without chemical treatment, and untreated controls
were observed in a separate animal room. Food and water were provided ad libitum, but not
during exposures. Dams were observed daily for clinical signs, and body-weight and food
consumption were recorded periodically throughout gestation. Dams were sacrificed on GD 20,
at which time blood samples were collected for serum chemistry analyses of the same parameters
as in the subchronic study. In addition, all organs were examined grossly, the ovaries were
inspected for corpora lutea, and the gravid uterus was weighed and examined for numbers of
implantation sites, early and late resorptions, and live and dead fetuses. All fetuses were
weighed, sexed, and examined for external anomalies. Half of the fetuses from each litter were
processed and examined for visceral anomalies, while the other half were processed and
examined for skeletal anomalies (due to overmaceration, roughly 23-32% of the litters processed
for skeletal evaluation could not be examined, spread evenly across the different treatment
groups).
Transient lacrimation and salivation were observed in some of the pregnant rats in the
high-dose group during exposure; in these cases, normal behaviors resumed shortly after the
daily exposure ended (Dalbey and Feuston, 1996). As shown in Table 3, there was a general
decrease in body-weight gain during the exposure period for all females housed in the inhalation
chambers relative to the untreated controls (including sham-exposed controls, indicating a
possible effect from handling and treatment). Compared to sham-exposed controls, the decrease
6
Diisopropyl Ether
-------
FINAL
4-21-2011
in average body-weight gain was statistically significant (p < 0.05) only for the high-dose group.
Food consumption was statistically significantly (p < 0.05) reduced in comparison to untreated
and sham-exposedcontrols in both the mid- and high-dose groups during the first week of
treatment (see Table 3). This indicates a possible food aversion during the first week of the
study that may have affected initial body-weight gains. No serum chemistry or gross pathology
changes were found in the treated dams. Reproductive parameters were not affected by
exposure, and there was no effect on fetal body weight. The only significant developmental
finding was a dose-related increase in the incidence of rudimentary ("small, discrete
ossification") or short ("less than one half the length of the preceding rib") 14th ribs in fetuses
from both the mid- and high-dose groups, both on the basis of number of fetuses affected and
number of litters affected (see Table 4). The study authors did not identify any effect levels. A
NOAEL of 1800 mg/m3 and a LOAEL of 12,940 mg/m3 for both maternal (reduced feed
consumption and body weight) and developmental (increased incidence of rudimentary 14th ribs
in fetuses) effects are identified for this review.
Table 3. Body-Weight Gain and Food Consumption in Dams Exposed to
DIPE by Inhalationa b
Parameter
Exposure Group (mg/m3)
0
(untreated)
0
(sham-exposed)
1800
12,940
28,200
Number of pregnant females
22
20
21
21
22
Body-weight gain (g) (GDs 6-16)
69± 11
50 ± 9°
58 ± 14°
42 ± 9°
33 ± 13d
Net weight gain (g)e
51.7 ± 13.8
37.3 ± 11.3°
41.5 ± 13.3°
31.9 ± 11.5°
29 ± 8.1°
Food Consumption (g/kg-d)
GDs 6-13
GDs 13-16
92.8 ±6.5
87.2 ±7.7
88.0 ±6.8
84.4 ±5.2
88.4 ±5.4
84.0 ±4.4
77.7 ± 8.4d
81.3 ± 7.4°
68.5 ± 6.7d
76.1 ± 7.0°
aDalbey and Feuston (1996).
bValues are presented as means ± SD.
°Reported to be significantly different from untreated controls by the researchers but /^-values not shown. Following
methods reported by the researchers, ANOVA was performed for this review, followed by group comparisons using
Dunnett's test. Based on this evaluation, it appears the study authors were evaluating statistical significance at
p < 0.05. The critical effect of reduced maternal weight gain at 12,940 mg/m3 was found to be statistically
significantly (p < 0.05) different from untreated controls, as reported by the study authors. Minor discrepancies were
that the re-analysis showed statistically significant differences from both control groups, rather than just untreated
controls, for net weight gain and GDs 13-16 food consumption in the mid- and high-dose groups.
Significantly different from both control groups (p-value not reported, see footnote c for further discussion).
"Carcass weight minus GD 6 weight.
7
Diisopropyl Ether
-------
FINAL
4-21-2011
Table 4. Selected Developmental Anomalies in Fetuses from Dams Exposed to
DIPE by Inhalationa b
Rudimentary/
short 14th ribs
Exposure Group (mg/m3)
0
(untreated)
0
(sham-exposed)
1800
12,940
28,200
Number of viable
fetuses examined for
skeletal anomalies0
168
155
167
156
173
Number of fetuses
affected
4(3)
4 (3.5)
6(5)
20 (17)d
33 (28)d
Number of litters
examined for skeletal
anomalies6
17
14
15
15
15
Number of litters
affected
4(24)
1(7)
4(27)
7 (47)f
13 (87)d
aDalbey and Feuston (1996).
bValues are number affected (%).
Approximately one-half of fetuses examined for skeletal anomalies.
Significantly different from both control groups (p-value not reported, see footnote c for further discussion).
eDue to overmaceration, not all litters were evaluated for abnormal skeletal development.
fReported to be significantly different from sham-exposed controls by the researchers, but p-valuc not shown.
Following methods reported by the researchers, group comparisons were performed using Fisher's exact test. The
author's claims of statistical significance were validated at p < 0.05.
A subchronic neurotoxicity screening study was performed on groups of 10 male and
10 female Sprague-Dawley rats whole-body exposed 5 days/week, 6 hours/day, for 13 weeks to
0, 1900, 13,600, or 29,500 mg/m3 (0, 450, 3250, or 7060 ppm) of DIPE (Rodriguez and
Dalbey, 1997). Because commercial grade (92% pure) test material was used, test animals were
also exposed to low concentrations of other chemicals, but the DIPE concentrations reported
above represent 91-94% of the total chemical exposures in the treatment groups. The rats were
housed in the inhalation chambers for the study duration, except for scheduled behavioral testing
when the rat to be tested was removed from the chamber to another room overnight and
evaluated the following day. Rats were observed for clinical signs prior to the daily exposure,
and body weight was recorded weekly. Neurotoxicity potential was evaluated via a functional
observational battery (FOB), measurement of motor activity in a figure-8 maze, and
neuropathology. The FOB was conducted following Weeks 0, 2, 4, 8, and 13 of exposure, and
the motor activity was determined following Weeks 0, 4, 8, and 13 of exposure. At study
termination, the rats were anesthetized; intravascularly perfused; and the brain, spinal cord, and
peripheral nerves were removed and processed for microscopic examination.
Clinical signs and body weight were not affected by exposure to DIPE (Rodriguez and
Dalbey, 1997). The FOB identified no effects clearly related to treatment; although, a few
sporadic, statistically significant (p < 0.05) changes were observed (reduced pinna reflex in
low-dose males during Week 2, reduced general activity of low- and high-dose females during
Week 4, increased rectal temperature of low-dose males during Week 4). Motor activity in the
figure-8 maze decreased in all groups as the animals aged but decreased significantly faster in
high-dose females than in controls. No treatment-related effects on morphology were observed
in either the central or peripheral nervous system. A single low-dose female rat had a
8
Diisopropyl Ether
-------
FINAL
4-21-2011
hypoplastic condition of the cortex (the 2 x 4-mm cavity), but this incident was not considered to
be associated with the exposure to DIPE by the study authors. The study authors concluded that
"only minor neurological changes were observed" and considered the neurological effects of
DIPE in rats as "minimal" at concentrations up to 29,500 mg/m3. In addition, they did not
"3
identify any effect levels. A NOAEL of 29,500 mg/m is identified for this review.
Other Studies
Acute or Short-term Studies
Machle et al. (1939) exposed six rabbits to DIPE via gavage at doses ranging from
1620-8200 mg/kg. A rapid intense intoxication, including narcosis, was produced, and two
rabbits dosed at 7200 and 8200 mg/kg died from respiratory failure within the first hour
following dosing. Another rabbit dosed at 6000 mg/kg died within 15 hours from irritation of
the intestinal tract. The minimal lethal dose for rabbits was found to be between 5075 and
6525 mg/kg. Kimura et al. (1971) determined the acute oral LD50 for DIPE in 14-day old,
young adult, and adult rats as 4640, 11,963, and 11,600 mg/kg, respectively. DIPE was
significantly more toxic to immature rats than adult rats (p < 0.05) (Kimura et al., 1971).
Machle et al. (1939) also exposed test animals (monkey, rabbit, and guinea pig) to vapor
concentrations of DIPE at 0.1%, 0.3%, 1.0%, 3.0% and 6.0% by volume in air (approximately
1000, 3000, 10,000, 30,000, and 60,000 ppm). All animals exposed to 6.0% DIPE died from
respiratory failure. A monkey and two rabbits exposed to 3.0%> DIPE exhibited signs of
anesthesia, and the monkey showed signs of beginning respiratory failure. Overall,
concentration-dependent acute toxicity to DIPE treatment was observed.
DIPE applied to the clipped skin of rabbits for 1 hour produced no deleterious effects
(Machle et al., 1939). However, repeated dermal exposures of 1 hour each for 10 days caused
skin reddening and a well-developed dermatitis in rabbits (Machle et al., 1939). In addition, a
review by Mehlman (2000) indicated that DIPE produced minor injury and irritation to rabbit
eyes in an unpublished study conducted by the Union Carbide Chemical Company.
Genotoxicity
Limited genotoxicity testing of DIPE has produced negative results. Studies in
Salmonella typhimurium (strains TA98, TA100, TA 1535, TA1537, TA1538) and Escherichia
coli (strain WP2 uvr A pKMlOl) using a modified assay for volatile solvents found that DIPE is
not mutagenic in bacteria, with or without metabolic activation (Brooks et al., 1988). DIPE did
not induce mitotic gene conversion in Saccharomyces cerevisiae JD1 or chromosome damage in
the rat liver RL4 chromosome assay (Brooks et al., 1988), nor sister chromatid exchanges
(SCEs) in Chinese hamster ovary (CHO) cells (Brooks et al., 1988).
9
Diisopropyl Ether
-------
FINAL
4-21-2011
FEASIBILITY OF DERIVING PROVISIONAL SUBCHRONIC AND CHRONIC
ORAL RfD VALUES FOR DIISOPROPYL ETHER
Oral data are limited to acute studies in rats (Kimura et al., 1971) and rabbits
(Machle et al., 1939). The available data are not sufficient for derivation of a subchronic or
chronic p-RfD for DIPE.
DERIVATION OF PROVISIONAL SUBCHRONIC AND CHRONIC
INHALATION RfC VALUES FOR DIISOPROPYL ETHER
Noncancer inhalation effects are summarized in Table 5. The Dalbey and Feuston (1996)
study presents the most sensitive effects (maternal and developmental effects) in comparison to
other studies. It is selected as the principal study. The developmental study found both maternal
"3
and developmental effects at 12,940 and 28,200 mg/m (Dalbey and Feuston, 1996), with a
NOAEL of 1800 mg/m3 for both. Effects seen in the developmental toxicity study included a
statistically significant increased (p < 0.05) incidence of rudimentary and short 14th ribs among
rat fetuses and litters, and reduced body-weight gain and food consumption in the dams at
"3
>12,940 mg/m . Occasional lacrimation and salivation by the dams during exposure
were also observed at 28,200 mg/m3. Even though the body weight of female rats was not
affected at any concentration in the subchronic study (Dalby and Feuston, 1996), there was a
statistically significant decrease (p < 0.05) in the body-weight gain (GDs 6-16) for the pregnant
"3
female rats at >1800 mg/m concentration levels in the developmental study. This observation
suggests that pregnant female rats may be more sensitive to DIPE.
Dose-response modeling was performed for the maternal body-weight changes and fetal
skeletal variations in the developmental toxicity study (Dalbey and Feuston, 1996). For the
maternal body-weight change, both body-weight gain during GDs 6-16 and net weight gains
(carcass weight minus GD 6 weight) were modeled (see Table 3). Treated groups were
compared to the sham-exposed control group only. Body-weight gains were significantly
different in the sham-exposed controls in comparison to the untreated controls, suggesting that
handling and treatment procedures may have had some effect on the exposed rats, and the
untreated controls may not be an appropriate comparison group for this analysis.
10
Diisopropyl Ether
-------
FINAL
4-21-2011
Table 5. Summary of Inhalation Noncancer Dose-Response Information for DIPE
Species
Sex
Exposure
Concentration
(mg/m3)
Exposure
NOAEL
(mg/m3)
LOAEL
(mg/m3)
Responses
Comments
Reference
Subchronic Exposure
Rat,
14/sex
M/F
0, 0 (sham),
2000, 13,800,
29,700
Whole-body,
6 hours/day,
5 days/week,
13 weeks
2000
13,800
Increased liver weight.
Dalbey and
Feuston, 1996
Rat,
10/sex
M/F
0, 1900,
13,600, 29,500
Whole-body,
6 hours/day,
5 days/week,
13 weeks
29,500
Only minor, sporadic
neurological effects
were observed.
Rodriguez and
Dalbey, 1997
Developmental Toxicity
Rat,
22/dose
F
0, 0 (sham),
1800, 12,940,
28,200
Whole-body,
6 hours/day on
GDs 6-15
1800
12,940
Decreased body-weight
gains and food
consumption in dams;
increased incidence of
rudimentary/short 14th
ribs in fetuses.
Dalbey and
Feuston, 1996
11
Diisopropyl Ether
-------
FINAL
4-21-2011
For the skeletal variations, only the data for number of litters affected were modeled.
Incidence of affected litters, rather than individual fetuses, was modeled because fetuses or pups
within litters do not respond independently. The litter is generally considered the experimental
unit in most developmental toxicity studies, and statistical analyses are generally performed
based on incidence per litter (not reported for the skeletal variations in the Dalbey and
Feuston [1996] study) or number of litters affected with a particular endpoint (U.S. EPA, 1991c).
These data are shown in Table 4. There was some indication of an effect of handling on the
incidence of skeletal variations (4/17 for untreated vs. 1/14 for sham-exposed), the sham-exposed
controls were therefore used in the analysis rather than the untreated controls. The
sham-exposed controls represent the closest control condition to the treated rats and are
consistent with the analysis of the maternal data.
Appendix A contains details of the modeling. No model provided an adequate fit to the
maternal body-weight gain data from GDs 6-16. Based on net maternal weight gain, the BMC
"3
with a benchmark response (BMR) of 10% relative deviation is 10,141 mg/m , and the BMCL is
7261 mg/m3. Based on the incidence of rudimentary and short 14th ribs, the BMC5 is 652 mg/m3,
3 3
and the BMCL5 is 264 mg/m . The BMCL of 264 mg/m was selected as the point of departure
(POD) for the derivation of the subchronic and chronic p-RfCs
In order to derive the subchronic p-RfC, the rat BMCL was first converted to a human
equivalent concentration (HEC). According to the EPA guidance document, A Review of the
Reference Dose and Reference Concentration Processes (2002), an adjustment to continuous
exposure for inhalation developmental effects is typically made. In general, any chemical in
vapor form that leads to inhalation toxicity outside of the respiratory tract or results in systemic
toxicity would require use of the Category 3 gas equation for calculating a HEC. The BMCLhec
of 66 mg/m3 was calculated from the rat BMCL of 264 mg/m3 using EPA (1994b) methodology
for an extrarespiratory effect produced by a Category 3 gas, as follows:
BMCLadj = 264 mg/m3 x 6 hrs 24 hrs
= 66 mg/m3
BMCLhec = BMCLadj x (Hb/g)A ^ (Hb/g)H
66 mg/m x 1
66 mg/m3
where:
(Hb/g)A ^ (Hb/g)H = the ratio of the blood:gas (air) partition coefficient of the chemical
for the laboratory animal species to the human value. In the
absence of data for DIPE, a default value of 1 is used.
12
Diisopropyl Ether
-------
FINAL
4-21-2011
"3
The subchronic p-RfC for DIPE, based on the BMCLhec of 66 mg/m for rudimentary
and short 14th ribs in fetal rats exposed during gestation (Dalbey and Fueston, 1996), is derived
as follows:
Subchronic p-RfC = BMCLhec ^ UF
= 66 mg/m3 -MOO
= 0.7 mg/m3
The composite UF of 100 is composed of the following:
UFr: A factor of 10 is applied for extrapolation to a potentially susceptible human
subpopulation because data for evaluating susceptible human response are
insufficient.
UFa: A factor of 3 is applied to account for interspecies extrapolation
(toxicodynamic portion only) because a dosimetric adjustment was made.
UFd: A factor of 3 is applied for database deficiencies because data for a
inhalation multigeneration reproduction study are not available. The database
includes a subchronic study and a developmental study in rats.
UFS: A factor of 1 is applied for subchronic-to-chronic extrapolation because a rat
developmental study is chosen as the principal study. The effects associated with
this study represent a sensitive lifestage and is not considered to be
duration-dependent.
UFl: A factor of 1 is applied for LOAEL-to-NOAEL extrapolation because the
current approach is to address this factor as one of the considerations in selecting
a BMR for benchmark dose modeling. In this case, a BMR of 5% change in the
incidence of rudimentary and short 14th ribs in fetal rats (a developmental effect)
was selected under an assumption that it represents a minimal biologically
significant change.
Confidence in the principal study (Dalbey and Feuston, 1996) is high. This study
included an appropriate number of animals and exposure levels and investigated a suitable range
of endpoints. Confidence in the database is medium. Only one species has been evaluated (rat)
in a subchronic study, a neurotoxicity study, and a developmental study. A multigeneration
reproduction study is not available. Confidence in the subchronic p-RfC is medium.
CHRONIC p-RfC
The chronic p-RfC for DIPE, based on the BMCLhec of 66 mg/m3 for rudimentary and
short 14th ribs in fetal rats exposed during gestation (Dalbey and Fueston, 1996), is derived as
follows:
Chronic p-RfC = BMCLhec-UF
= 66 mg/m3 -M00
= 0.7 mg/m3
13
Diisopropyl Ether
-------
FINAL
4-21-2011
The composite UF of 100 is composed of the following:
UFr: A factor of 10 is applied for extrapolation to a potentially susceptible human
subpopulation because data for evaluating susceptible human response are
insufficient.
UFa: A factor of 3 is applied to account for interspecies extrapolation
(toxicodynamic portion only) because a dosimetric adjustment was made.
UFd: A factor of 3 is applied for database deficiencies because data for a
inhalation multigeneration reproduction study are not available. The database
includes a subchronic study and a developmental study in rats.
UFS: A factor of 1 is applied for subchronic-to-chronic extrapolation because a rat
developmental study is chosen as the principal study. The effects associated with
this study represent a sensitive lifestage and is not considered to be
duration-dependent.
UFl: A factor of 1 is applied for LOAEL-to-NOAEL extrapolation because the
current approach is to address this factor as one of the considerations in selecting
a BMR for benchmark dose modeling. In this case, a BMR of 5% change in the
incidence of rudimentary and short 14th ribs in fetal rats (a developmental effect)
was selected under an assumption that it represents a minimal biologically
significant change.
As stated in the derivation of a subchronic p-RfC, confidence in the principal study
(Dalbey and Feuston, 1996) is high. Confidence in the database is medium because there are no
multigenerational reproductive toxicity studies. Confidence in the chronic p-RfC is medium.
PROVISIONAL CARCINOGENICITY ASSESSMENT
FOR DIISOPROPYL ETHER
As stated in Introduction on page 4, an evaluation of the cancer literature indicates that a
major study related to the carcinogenicity of DIPE has been conducted by the Ramazzini
Institute. As specified earlier, the cancer assessment for DIPE will remain on hold until the
completion of the PWG review.
14
Diisopropyl Ether
-------
FINAL
4-21-2011
REFERENCES
ACGIH (American Conference of Governmental Industrial Hygienists). 2007. 2007 Threshold
Limit Values for Chemical Substances and Physical Agents and Biological Exposure Indices.
ACGIH, Cincinnati, OH.
ATSDR (Agency for Toxic Substances and Disease Registry). 2008. Toxicological Profile
Information Sheet. U.S. Department of Health and Human Services, Public Health Service.
Online, http://www.atsdr.cdc.gov/toxpro2.html.
Belpoggi, F., M. Soffritti, F. Minardi, L. Bua, E. Cattin and C. Maltoni. 2002. Results of
long-term carcinogenicity bioassays on tert-amyl-methyl-ether (TAME) and di-isopropyl-ether
(DIPE) in rats. Ann. NY Acad. Sci. 982:70-86.
Brooks, T.M., A.L. Meyer and D.H. Hutson. 1988. The genetic toxicology of some
hydrocarbon and oxygenated solvents. Mutagenesis. 3(3):227-232.
Cruzan, G. 2009. Assessment of the cancer potential of methanol. Crit Rev Toxicol, 39:
347-363.
Dalbey, W. and M. Feuston. 1996. Subchronic and developmental toxicity studies of vaporized
diisopropyl ether in rats. J. Toxicol. Environ. Health. 49:29-43.
ExxonMobil Biomedical Sciences. 2005. IUCLID Data Set. Diisopropyl ether. Submitted to
U.S. Environmental Protection Agency, High Production Volume Challenge Program. Online.
http://www.epa.gov/chemrtk/pubs/summaries/diisoeth/cl6164tc.htm.
Greim H, A. Hartwig, U. Reuter, H.B. Richter-Reichhelm and H.W. Thielmann. 2009.
Chemically induced pheochromocytomas in rats: mechanisms and relevance for human risk
assessment. Crit. Rev. Toxicol. 39:695-718.
Hailey, J.R. 2004. Lifetime study in rats conducted by the Ramazzini Foundation Pathology
Working Group Chairperson's report. NIEHS. RTP, NC.
IARC (International Agency for Research on Cancer). 2008. Search IARC Monographs.
Online. http://monographs.iarc.fr/ENG/Monographs/allmonos90.php.
Kimura, E.T., D.M. Ebert and P.W. Doge. 1971. Acute toxicity and limits of solvent residues
for sixteen organic solvents. Toxicol. Appl. Pharmacol. 19:699-704.
Machle, W., E.W. Scott and J. Treon. 1939. The physiological response to isopropyl ether and
to a mixture of isopropyl ether and gasoline. J. Ind. Hyg. Toxicol. 21:72-96.
McConnell, E.E., H.A. Solleveld, J.A. Swenberg; G.A. Boorman. 1986. Guidelines for
combining neoplasms for evaluation of rodent carcinogenesis studies. J. Natl. Cancer Inst.
76:283-289.
15
Diisopropyl Ether
-------
FINAL
4-21-2011
Mehlman, M.A. 2000. Ethers. In: Patty's Toxicology. E. Bingham, B. Cohrssen and C.H.
Powell, (Ed.). John Wiley & Sons, Inc. DOI: 10.1002/0471435139.tox072. Online.
http://mrw.interscience.wiley.com/emrw/9780471125471/pattys/tox/article/tox072/current/html?
hd%253DAll%252CEthers.
NIOSH (National Institute for Occupational Safety and Health). 2008. NIOSH Pocket Guide to
Chemical Hazards. Index by CASRN. Online, http://www2.cdc.gov/nioshtic-2/nioshtic2.htm.
NTP (National Toxicology Program). 2005. 11th Report on Carcinogens. U.S. Department of
Health and Human Services, Public Health Service, National Institutes of Health, Research
Triangle Park, NC. Online, http://ntp-server.niehs.nih.gov.
NTP (National Toxicology Program). 2008. Management Status Report. Online.
http://ntp.niehs.nih.gov/index.cfm?objectid=78CC7E4C-FlF6-975E-72940974DE301C3F.
OSHA (Occupational Safety and Health Administration). 2008. OSHA Standard 1910.1000
Table Z-l. Part Z, Toxic and Hazardous Substances. Online, http://www.osha-slc.gov/
OshStd_data/l 910 1000 TABLEZ-1 .html.
Rodriguez, S.C. and W.E. Dalbey. 1997. Subchronic neurotoxicity of vaporized diisopropyl
ether in rats. Int. J. Toxicol. 16:599-610.
Silverman, L., H.F. Schulte and M.W. First. 1946. Further studies on sensory response to
certain industrial solvent vapors. J. Ind. Hyg. Toxicol. 28:262-266.
Soffritti, M., R. Belpoggi, D. Cevolani, M. Guarino, M.P. Ani and C. Maltoni. 2002. Results of
long-term experimental studies on the carcinogenicity of methyl alcohol and ethyl alcohol in rats.
Ann. NY Acad. Sci. 982:46-69.
Soffritti, M., F. Belpoggi, D.D. Esposti, and L. Lambertini. 2005. Aspartame induces
lymphomas and leukaemias in rats. Eur. J. Oncol., 10(2): 107-116.
Soffritti, M., F. Belpoggi, E. Tibaldi, D.D. Esposti, and M. Lauriola. 2007. Life-span exposure
to low doses of aspartame beginning during prenatal life increases cancer effects in rats.
Environ. Hlth. Perspect. 115(9): 1293-1297.
Spector, W.S. (ed.). 1956. Handbook of Toxicology, Vol. 1, Saunders, Philadelphia, PA.
U.S. EPA. 1988. Recommendations for and Documentation of Biological Values for Risk
Assessment. Prepared by the Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Cincinnati, OH. EPA/600/6-87/008.
U.S. EPA. 1991a. Chemical Assessments and Related Activities (CARA). Office of Health and
Environmental Assessment, Washington, DC. April.
U.S. EPA. 1991b. Alpha2u-Globulin: Association with Chemically Induced Renal Toxicity and
Neoplasia in the Male Rat. Risk Assessment Forum, U.S. Environmental Protection Agency,
Washington, D.C. EPA/625/3-91/019F. September 1991.
16
Diisopropyl Ether
-------
FINAL
4-21-2011
U.S. EPA. 1991c. Guidelines for Developmental Toxicity Risk Assessment. Risk Assessment
Forum, U.S. Environmental Protection Agency, Washington, DC. EPA/600/FR-91/001.
December 1991.
U.S. EPA. 1994a. Chemical Assessments and Related Activities (CARA). Office of Health and
Environmental Assessment, Washington, DC. December.
U.S. EPA. 1994b. Methods of Derivation of Inhalation Reference Concentrations and
Application of Inhalation Dosimetry. Office of Research and Development, National Center for
Environmental Assessment, Washington, DC. October 1994. EPA/600/8-90/066F.
U.S. EPA. 1997. Health Effects Assessment Summary Tables. FY-1997 Update. Prepared by
the Office of Research and Development, National Center for Environmental Assessment,
Cincinnati OH for the Office of Emergency and Remedial Response, Washington, DC. July.
EPA/540/R-97/036. NTIS PB97-921199.
U.S. EPA. 2000. Benchmark Dose Technical Guidance Document [External Review Draft],
Risk Assessment Forum. Washington, DC. October 2000.
U.S. EPA. 2002. A Review of the Reference Dose and Reference Concentration Processes.
Risk Assessment Forum, U.S. Environmental Protection Agency, Washington, DC.
EPA/630/P-02/002F. December 2002.
U.S. EPA. 2005. Guidelines for Carcinogen Risk Assessment. U.S. Environmental Protection
Agency, Risk Assessment Forum, Washington, DC. EPA/630/P-03/001F. March 2005.
U.S. EPA. 2006. 2006 Edition of the Drinking Water Standards and Health Advisories. Office
of Water, Washington, DC. EPA 822-R-06-013. Online, http://www.epa.gov/waterscience/
drinking/standards/dwstandards.pdf.
U.S. EPA. 2009. Integrated Risk Information System (IRIS). Online. Office of Research and
Development, National Center for Environmental Assessment, Washington, DC.
http://www.epa.gov/iris/. Accessed July, 2009.
U.S. EPA 2010. EPA places four IRIS assessment on hold pending review. News releases
issued by the Office of Research and Development, Office of Science & Technology,
Washington, DC. Release date: 06/15/2010. Available online at
http://yosemite.epa.gov/opa/admpress.nsf/03dd877d6fl726c28525735900404443/b64d44f06a56
d5b285257742007c5002!OpenDocument
WHO (World Health Organization). 2008. Online catalogs for the Environmental Health
Criteria Series. Online, http://www.who.int/ipcs/publications/ehc/ehc_alphabetical/en/
index.html.
17
Diisopropyl Ether
-------
FINAL
4-21-2011
APPENDIX A. DETAILS OF BENCHMARK DOSE MODELING
FOR SUBCHRONIC RfC
MODEL-FITTING PROCEDURE FOR CONTINUOUS DATA:
The model-fitting procedure for continuous data using the EPA benchmark dose software
(BMDS) is as follows. The simplest model (linear) is first applied to the data while assuming
constant variance. If the data are consistent with the assumption of constant variance (p> 0.1),
then the fit of the linear model to the means is evaluated, and the polynomial, power, and Hill
models are fit to the data while assuming constant variance. An adequate model fit is judged by
three criteria: goodness-of-fit/>-value (p > 0.1), visual inspection of the dose-response curve, and
scaled residual at the data point (except the control) closest to the predefined benchmark
response (BMR). Among all the models providing adequate fits to the data, the lowest BMD
(BMDL) is selected as the point of departure (POD) when the difference between the BMDLs
estimated from these models is more than three-fold (unless it appears to be an outlier);
otherwise, the BMDL from the model with the lowest Akaike Information Criterion (AIC) is
chosen. If the test for constant variance is negative, the linear model is run again while applying
the power model integrated into the BMDS to account for nonhomogenous variance. If the
nonhomogenous variance model provides an adequate fit (p > 0.1) to the variance data, then the
fit of the linear model to the means is evaluated, and the polynomial, power, and Hill models are
also fit to the data and evaluated while the variance model is applied. Model fit and POD
selection proceed as described earlier. If the test for constant variance is negative and the
nonhomogenous variance model does not provide an adequate fit to the variance data, then the
data set is considered unsuitable for modeling.
MODEL-FITTING RESULTS FOR MATERNAL BODY-WEIGHT GAINS IN RATS
(DALBEY AND FEUSTON, 1996):
Following the above procedure, the continuous models in the EPA BMDS (version 2.1)
were fit to the data shown in Table 5 for maternal body-weight gains in rats using the
sham-exposed group as controls for both body-weight gain during GDs 6-16 and net weight gain
(carcass weight minus GD 6 weight). The models were run with a BMR of 1 standard deviation
(SD) from the control mean, as generally recommended by EPA (2000), and also with a relative
deviation of 10% from the control mean (10% change is generally considered to be biologically
significant for body weight). The results are shown in Table A-l. For the GDs 6-16
body-weight gain data, the assumption of constant variance did not hold, and the
nonhomogenous variance model did not provide an adequate fit. For the net weight gain data,
the constant variance model provided an adequate fit to the variance data, and the linear model
provided an adequate fit to the means. The power and higher-degree polynomial models all
defaulted back to the linear model. There were insufficient data points to fit the Hill model. The
fit of the linear model to the data is shown in Figure A-l. Benchmark concentration (BMC) and
the lowest bound of the BMC (BMCL) values were considerably higher using the BMR of 1 SD
(28,820 and 19,552 mg/m3) than using the BMR of 10% relative deviation (10,141 and
"3
7261 mg/m ). The lower BMCL values based on the 10% relative deviation were chosen to
represent the modeling results for this endpoint.
18
Diisopropyl Ether
-------
FINAL
4-21-2011
Table A-l. Model Predictions for Maternal Weight Gain in Pregnant Rats"
Model
Variance
p-V alueb
Means
p-V alueb
AIC
BMC
(mg/m3)
BMCL
(mg/m3)
GDs 6-16 BWgain (BMR = 1 SD)
Linear (constant variance)0
0.078
0.02
504.3
15346
12038
Linear (modeled variance)0
0.033
0.02
506.3
15179
11488
Net BW gain (BMR =1 SD)
Linear (constant variance)0
0.1599
0.207
494.5
28820
19552
Polynomial (constant variance)o d
0.1599
0.207
494.5
28820
19552
Power (constant variance)0
0.1599
0.207
494.5
28820
19552
Hill (constant variance)0
0.1599
NA
496.9
NA
NA
Net BW gain (BMR = 10%)
Linear (constant variance)0
0.1599
0.207
494.5
10,141
7261
aDalbey and Feuston, 1996.
bValues <0.10 fail to meet conventional goodness-of-fit criteria.
Coefficients restricted to be negative.
dOne degree polynomial shown. Higher degree polynomials default back to one degree.
"Power restricted to >1.
AIC = Akaike Information Criterion; BMC = maximum likelihood estimate of the concentration associated
with the selected benchmark response; BMCL = 95% lower confidence limit on the BMC; NA = Not
applicable; SD = standard deviation.
19
Diisopropyl Ether
-------
FINAL
4-21-2011
Linear Model wth 0.95 Confidence Level
on
ro
-------
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
alpha = 124.834
rho = 0 Specified
beta_0 = 39.0576
beta 1 = -0.000384966
FINAL
4-21-2011
the user,
alpha
beta_0
beta 1
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -rho
have been estimated at a boundary point, or have been specified by
and do not appear in the correlation matrix )
alpha beta_0 beta_l
1 -2.6e-010 1.4e-010
-2.6e-010 1 -0.7
1.4e-010 -0.7 1
Interval
Variable
Limit
alpha
160.762
beta_0
42.4156
beta_l
0.00017582
Estimate
123.433
39.094
-0.000385502
Parameter Estimates
Std. Err.
19.0461
1.69475
0.000106983
95.0% Wald Confidence
Lower Conf. Limit Upper Conf.
86.1029
35.7723
-0.000595185
Table of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Scaled Res.
0 20
1800 21
1.294e+004
2.82e+004
37.3
41.5
21
22
31.9
29
39.1
38.4
34.1
28.2
11.3
13.3
11.5
8.1
11.1
11.1
11.1
11.1
-0.722
1.28
-0. 91
0.328
Model Descriptions for likelihoods calculated
Model A1: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma^2
Model A2:
Yij = Mu (i) + e (i j )
21
Diisopropyl Ether
-------
FINAL
4-21-2011
Var{e(ij)} = Sigma(i)^2
Model A3: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma^2
Model A3 uses any fixed variance parameters that
were specified by the user
Model R: Yi = Mu + e(i)
Var{e(i)} = Sigma^2
Likelihoods of Interest
Model
A1
A2
A3
fitted
R
Log(likelihood)
-242.684177
-240.100150
-242.684177
-244.259175
-250.296021
# Param's
5
8
5
3
2
AIC
495.368354
496.200300
495.368354
494.518350
504.592043
Test 1:
Test
Test
Test
Explanation of Tests
Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Are Variances Homogeneous? (A1 vs A2)
Are variances adeguately modeled? (A2 vs. A3)
Does the Model for the Mean Fit? (A3 vs. fitted)
(Note: When rho=0 the results of Test 3 and Test 2 will be the same.)
Tests of Interest
Test
-2*log(Likelihood Ratio) Test df
p-value
Test
Test
Test
Test
20.3917
5.16805
5.16805
3.15
0.002358
0.1599
0.1599
0.207
The p-value for Test 1 is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
The p-value for Test 2 is greater than .1. A homogeneous variance
model appears to be appropriate here
The p-value for Test 3 is greater than .1. The modeled variance appears
to be appropriate here
The p-value for Test 4 is greater than .1. The model chosen seems
to adeguately describe the data
Benchmark Dose Computation
Specified effect = 0.1
Risk Type = Relative risk
Confidence level = 0.95
22
Diisopropyl Ether
-------
BMD = 10141.1
BMDL = 72 61.4
FINAL
4-21-2011
23
Diisopropyl Ether
-------
FINAL
4-21-2011
MODEL-FITTING PROCEDURE FOR QUANTAL NONCANCER DATA:
The model-fitting procedure for dichotomous noncancer data is as follows. All available
dichotomous models in the EPA BMDS are fit to the incidence data using the extra risk option.
The multistage model is run for all polynomial degrees up to n-1 (where n is the number of dose
groups including control). An adequate model fit is judged by three criteria: goodness-of-fit
p-walue (p > 0.1), visual inspection of the dose-response curve, and scaled residual at the data
point (except the control) closest to the predefined BMR. Among all the models providing
adequate fit to the data, the lowest BMDL is selected as the POD when the difference between
the BMDLs estimated from these models is more than three-fold (unless it appears to be an
outlier); otherwise, the BMDL from the model with the lowest AIC is chosen. In accordance
with EPA (2000) guidance, BMDs and BMDLs associated with an extra risk of 10% are
calculated for all models. Although a 10% BMR is the default, in this case a 5% BMR was used
because the developmental effect (i.e., short ribs) was observed during a potentially sensitive
lifestage.
MODEL-FITTING RESULTS FOR INCIDENCE OF RUDIMENTARY/SHORT 14th
RIBS IN FETAL RATS (DALBEY AND FEUSTON, 1996):
Following the above procedure, the dichotomous models in the EPA BMDS (version 2.1)
were fit to the data shown in Table 4 for incidence of rudimentary and short 14th ribs in the
number of litters affected from the pregnant rats treated with DIPE during gestation. The
incidence of affected litters, rather than individual fetuses, was modeled because fetuses or pups
within litters do not respond independently. The sham-exposed group was used as the controls.
The results are shown in Table A-2. All models fit the data adequately. The BMCLs from the
models providing adequate fit differed by more than 3-fold. In accordance with EPA (2000)
guidance, the lowest BMCL was selected from among the models providing adequate fit. The
resulting benchmark concentration (BMC5) and associated 95% lower confidence limit (BMCL5)
"3
were 652 and 264 mg/m , respectively, based on the log-logistic model. The fit of the log-
logistic model to the data is shown in Figure A-2.
24
Diisopropyl Ether
-------
FINAL
4-21-2011
Table A-2. Model Predictions for Incidence of Rudimentary/Short 14th Ribs
in Fetal Rats"
Model
Degrees of
Freedom
x2
2
X
Goodness-of-Fit
/>-Valuc
AIC
BMC10
(mg/m3)
BMCL10
(mg/m3)
Scaled
Residual
of
Interest
Gamma (power >1)
2
1.6
0.4486
62.69
901.32
589.84
0.76
Logistic
2
1.42
0.4921
62.59
2356.76
1606.75
0.857
Log-Logistic (slope > 1)
1
2.26
0.1326
65.43
652.399
264.336
-0.202
Log Probit (slope >1)
1
1.93
0.1643
65.17
6221.59
1667.01
0.967
Multistage (degree = 1,
betas > 0)
2
1.6
0.4486
62.69
901.319
589.84
0.76
Multistage (degree = 2,
betas > 0)
1
1.37
0.2411
64.46
1379.01
599.884
0.875
Multistage (degree = 3,
betas > 0)
1
1.16
0.2815
64.25
1304.77
609.616
0.845
Probit
2
1.38
0.5017
62.55
2213.23
1576.94
0.85
Weibull (power > 1)
2
1.6
0.4486
62.69
901.319
589.84
0.76
Quantal-Linear
2
1.6
0.4486
62.69
901.319
589.84
0.76
aDalbey and Feuston (1996).
bValues <0.10 fail to meet conventional goodness-of-fit criteria.
AIC = Akaike Information Criterion; BMC = maximum likelihood estimate of the concentration
associated with the selected benchmark response; BMCL = 95% lower confidence limit on the BMC.
25
Diisopropyl Ether
-------
FINAL
4-21-2011
Log-Logistic Model with 0.95 Confidence Level
dose
14:10 03/29 2011
Figure A-2. Fit of Log-Logistic Model to Data on Incidence of Rudimentary/Short 14th
Ribs in Fetal Rats (Dalbey and Feuston, 1996)
BMC and BMCLs indicated are associated with an extra risk of 5% and are in units of mg/m3.
Dichotomous Hill Model. (Version: 1.0; Date: 09/24/2006)
Input Data File:
C:\USEPA\BMDS21\Data\dhl DIPE inh dich dev litters Dhl-BMR05-Restrict.(d)
Gnuplot Plotting File:
C:\USEPA\BMDS21\Data\dhl DIPE inh dich dev litters Dhl-BMR05-Restrict.pit
Tue Mar 29 15:10:23 2011
BMDS Model Run
The form of the probability function is:
P[response] = v*g +(v-v*g)/[1+EXP(-intercept-slope*Log(dose))]
where: 0 <= g < 1, 0 < v <= 1
v is the maximum probability of response predicted by the
model,
and v*g is the background estimate of that probability.
Dependent variable = Incidence
Independent variable = Dose
Slope parameter is restricted as slope >= 1
26
Diisopropyl Ether
-------
FINAL
4-21-2011
Total number of observations = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
v = -9999
g = -9999
intercept = -9.11383
slope = 1.0107 9
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -v -slope
have been estimated at a boundary point, or have been
specified by the user,
and do not appear in the correlation matrix )
g intercept
g 1 -0.4
intercept -0.4 1
Parameter Estimates
Confidence Interval
Variable Estimate
Upper Conf. Limit
v 1
g 0.0625
0. 175327
intercept -9.13496
-8 . 33721
slope 1
Std. Err.
NA
0. 0575656
0.407024
NA
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
95.0% Wald
Lower Conf. Limit
-0.0503265
-9. 93271
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced model
Log(likelihood)
-28.5552
-29.7575
-40.2066
Deviance Test d.f.
2.40467
23.3028
P-value
0.3005
<.0001
AIC:
63.515
27
Diisopropyl Ether
-------
FINAL
4-21-2011
Goodness of Fit
Scaled
Dose Est. Prob. Expected Observed Size Residual
0.0000 0.0625 0.875 1 14 0.138
1800.0000 0.2149 3.223 4 15 0.4883
12940.0000 0.6086 9.129 7 15 -1.126
28200.0000 0.7680 11.520 13 15 0.9054
ChiA2 = 2.345887 d.f. = 2 P-value = 0.3095
Benchmark Dose Computation
Specified effect = 0.05
Risk Type = Extra risk
Confidence level = 0.95
BMD = 4 8 8.101
BMDL = 162.338
28
Diisopropyl Ether
------- |