EPA/63 5/R-16/079b
^^ LF^^ Public Comment Draft
www.epa.gov/iris
Toxicological Review of tert-Butyl Alcohol (tert-Butanol)
(CASRN 75-65-0)
Supplemental Information - tert-Butyl Alcohol
April 2016
NOTICE
This document is a Public Comment Draft. This information is distributed solely for the purpose of
pre-dissemination peer review under applicable information quality guidelines. It has not been
formally disseminated by EPA. It does not represent and should not be construed to represent any
Agency determination or policy. It is being circulated for review of its technical accuracy and
science policy implications.
Integrated Risk Information System
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
-------
Supplemental Information—tert-Butyl Alcohol
DISCLAIMER
This document is a preliminary draft for review purposes only. This information is
distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should
not be construed to represent any Agency determination or policy. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
This document is a draft for review purposes only and does not constitute Agency policy.
ii DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
CONTENTS
APPENDIX A. ASSESSMENTS BY OTHER NATIONAL AND INTERNATIONAL HEALTH AGENCIES A-l
APPENDIX B. INFORMATION IN SUPPORT OF HAZARD IDENTIFICATION AND DOSE-RESPONSE
ANALYSIS B-l
B.I. TOXICOKINETICS B-l
B.I.I. Absorption B-l
B.I.2. Distribution B-2
B.I.3. Metabolism B-2
B.1.4, Excretion B-5
B.I.5. Physiologically Based Pharmacokinetic Models B-6
B.2. PBPK MODEL EVALUATION SUMMARY B-9
B.2.1, Evaluation of Existing tert-Butanol Submodels B-9
B.2.2, Modification of Existing te/t-Butanol Submodels B-ll
B.2.3. Summary of the PBPK Model for te/t-Butanol B-16
B.2,4. te/t-Butanol Model Application B-16
B.2,5. PBPK Model Code B-16
B.3. OTHER PERTINENT TOXICITY INFORMATION B-17
B.3,1. Other Toxicological Effects B-17
B.3.2. Genotoxicity B-31
B.3,3. Summary B-35
APPENDIX C. DOSE-RESPONSE MODELING FOR THE DERIVATION OF REFERENCE VALUES FOR
EFFECTS OTHER THAN CANCER AND THE DERIVATION OF CANCER RISK
ESTIMATES C-l
C.I.I. Noncancer Endpoints C-l
C.I.2. Cancer Endpoints C-23
REFERENCES R-l
This document is a draft for review purposes only and does not constitute Agency policy.
iii DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
TABLES
Table A-l. Health assessments and regulatory limits by other national and international
health agencies A-l
Table B-l. PBPK model physiologic parameters and partition coefficients B-12
Table B-2. Rate constants for tert-butanol determined by optimization of the model with
experimental data B-15
Table B-3. Changes in kidney weight in animals following exposure to tert-butanol B-20
Table B-4. Changes in liver weight in animals following exposure to te/t-butanol B-23
Table B-5. Changes in liver histopathology in animals following exposure to tert-butanol B-25
Table B-6. Changes in urinary bladder histopathology in animals following oral exposure to
te/t-butanol B-27
Table B-7. Summary of genotoxicity (both in vitro and in vivo) studies of te/t-butanol B-34
Table C-l. Noncancer endpoints selected for dose-response modeling for te/t-butanol C-2
Table C-2. Summary of BMD modeling results for kidney transitional epithelial hyperplasia in
male F344 rats exposed to te/t-butanol in drinking water for 2 years (NTP,
1995); BMR = 10% extra risk C-3
Table C-3. Summary of BMD modeling results for kidney transitional epithelial hyperplasia in
female F344 rats exposed to te/t-butanol in drinking water for 2 years (NTP,
1995); BMR = 10% extra risk C-6
Table C-4. Summary of BMD modeling results for absolute kidney weight in male F344 rats
exposed to te/t-butanol in drinking water for 15 months (NTP, 1995); BMR =
10% rel. dev. from control mean C-9
Table C-5. Summary of BMD modeling results for absolute kidney weight in female F344 rats
exposed to te/t-butanol in drinking water for 15 months (NTP, 1995); BMR =
10% rel. dev. from control mean C-12
Table C-6. Summary of BMD modeling results for kidney inflammation in female rats
exposed to te/t-butanol in drinking water for 2 years (NTP, 1995); BMR = 10%
extra risk C-15
Table C-l. Summary of BMD modeling results for absolute kidney weight in male F344 rats
exposed to te/t-butanol via inhalation for 6 hr/d, 5d/wkfor 13 weeks (NTP,
1997); BMR = 10% relative deviation from the mean C-18
Table C-8. Summary of BMD modeling results for absolute kidney weight in female F344 rats
exposed to te/t-butanol via inhalation for 6 hr/d, 5d/wkfor 13 weeks (NTP,
1997); BMR = 10% relative deviation from the mean C-21
Table C-9. Cancer endpoints selected for dose-response modeling for te/t-butanol C-24
Table C-10. Summary of the oral slope factor derivations C-25
Table C-ll. Summary of BMD modeling results for thyroid follicular cell adenomas in female
B6C3F1 mice exposed to te/t-butanol in drinking water for 2 years (NTP, 1995);
BMR = 10% extra risk C-26
Table C-12. Summary of BMD modeling results for thyroid follicular cell adenomas or
carcinomas in male B6C3F1 mice exposed to te/t-butanol in drinking water for
2 years (NTP, 1995); BMR = 5% extra risk C-29
Table C-13. Summary of BMD modeling results for thyroid follicular cell adenomas or
carcinomas in male B6C3F1 mice exposed to te/t-butanol in drinking water for
2 years, high dose omitted (NTP, 1995); BMR = 5% extra risk C-32
This document is a draft for review purposes only and does not constitute Agency policy.
iv DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Table C-14. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with administered dose units and including all dose groups (NTP, 1995); BMR
= 10% extra risk C-35
Table C-15. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with administered dose units and excluding high-dose group (NTP, 1995); BMR
= 10% extra risk C-37
Table C-16. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with PBPK (tert-butanol, mg/L) dose units and including all dose groups (NTP,
1995); BMR = 10% extra risk C-39
Table C-17. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with PBPK (tert-butanol, mg/L) dose units and excluding high-dose group (NTP,
1995); BMR = 10% extra risk C-41
Table C-18. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with PBPK (metabolized, mg/hr) dose units and including all dose groups (NTP,
1995); BMR = 10% extra risk C-43
Table C-19. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with PBPK (metabolized, mg/hr) dose units and excluding high-dose group
(NTP, 1995); BMR = 10% extra risk C-45
Table C-20. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with administered dose units and including all dose groups; reanalyzed data
(Hard et al., 2011; NTP, 1995); BMR = 10% extra risk C-47
Table C-21. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with administered dose units and excluding high-dose group; re-analyzed data
(Hard et al., 2011; NTP, 1995); BMR = 10% extra risk C-47
Table C-22. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with PBPK (tert-butanol, mg/L) dose units and including all dose groups;
reanalyzed data (Hard et al., 2011; NTP, 1995); BMR = 10% extra risk C-50
Table C-23. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with PBPK (tert-butanol, mg/L) dose units and excluding high-dose group;
reanalyzed data (Hard et al., 2011; NTP, 1995); BMR = 10% extra risk C-50
Table C-24. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with PBPK (metabolized, mg/hr) dose units and including all dose groups;
reanalyzed data (Hard et al., 2011; NTP, 1995); BMR = 10% extra risk C-52
Table C-25. Summary of BMD modeling results for renal tubule adenoma or carcinoma in
male F344 rats exposed to tert-butanol in drinking water for 2 years modeled
with PBPK (metabolized, mg/hr) dose units and excluding high-dose group;
reanalyzed data (Hard et al., 2011; NTP, 1995); BMR = 10% extra risk C-52
This document is a draft for review purposes only and does not constitute Agency policy.
v DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Table C-26. Summary of the inhalation unit risk derivation C-56
FIGURES
Figure B-l. Biotransformation of tert-butanol in rats and humans B-4
Figure B-2. Comparison of the tert-butanol portions of existing MTBE models with tert-
butanol blood concentrations from i.v. exposure by Poet et al. (1997) B-10
Figure B-3. Schematic of the PBPK submodel for tert-butanol in rats B-12
Figure B-4. Comparison of the EPA model predictions with measured tert-butanol blood
concentrations for i.v., inhalation, and oral gavage exposure to tert-butanol B-15
Figure B-5. Comparison of the EPA model predictions with measured amounts of tert-
butanol in blood after repeated inhalation exposure to tert-butanol B-17
Figure B-6. Exposure-response array of other effects following oral exposure to tert-butanol B-29
Figure B-7. Exposure-response array of other effects following inhalation exposure to tert-
butanol B-30
Figure C-l. Plot of incidence by dose, with fitted curve for LogLogistic model for kidney
transitional epithelial hyperplasia in male F344 rats exposed to tert-butanol in
drinking water for 2 years (NTP, 1995); BMR = 10% extra risk; dose shown in
mg/kg-d C-4
Figure C-2. Plot of incidence by dose, with fitted curve for Multistage 3° model for kidney
transitional epithelial hyperplasia in female F344 rats exposed to tert-butanol
in drinking water for 2 years (NTP, 1995); BMR = 10% extra risk; dose shown in
mg/kg-d C-6
Figure C-3. Plot of mean response by dose, with fitted curve for Linear model with constant
variance for absolute kidney weight in male F344 rats exposed to tert-butanol
in drinking water for 15 months (NTP, 1995); BMR = 10% rel. dev. from control
mean; dose shown in mg/kg-d C-10
Figure C-4. Plot of mean response by dose, with fitted curve for Exponential (M4) model
with constant variance for absolute kidney weight in female F344 rats exposed
to te/t-butanol in drinking water for 15 months (NTP, 1995); BMR = 10% rel.
dev. from control mean; dose shown in mg/kg-d C-13
Figure C-5. Plot of incidence by dose, with fitted curve for Logprobit model for kidney
inflammation in female rats exposed to tert-butanol in drinking water for 2
years (NTP, 1995); BMR = 10% extra risk; dose shown in mg/kg-d C-15
Figure C-6. Plot of mean response by concentration, with fitted curve for Hill model for
absolute kidney weight in male F344 rats exposed to tert-butanol via inhalation
for 6 hr/d, 5d/wkfor 13 weeks (NTP, 1997); BMR = 10% relative deviation from
the mean; concentration shown in mg/m3 C-19
Figure C-7. Plot of mean response by concentration, with fitted curve for Hill model for
absolute kidney weight in female F344 rats exposed to tert-butanol via
inhalation for 6 hr/d, 5d/wk for 13 weeks (NTP, 1997); BMR = 10% relative
deviation from the mean; concentration shown in mg/m3 C-22
This document is a draft for review purposes only and does not constitute Agency policy.
vi DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Figure C-8. Plot of mean response by concentration, with fitted curve for Power model for
absolute kidney weight in female F344 rats exposed to tert-butanol via
inhalation for 6 hr/d, 5d/wkfor 13 weeks (NTP, 1997); BMR = 10% relative
deviation from the mean; concentration shown in mg/m3 C-22
Figure C-9. Plot of incidence by dose, with fitted curve for Multistage 3° model for thyroid
follicular cell adenomas in female B6C3F1 mice exposed to tert-butanol in
drinking water for 2 years (NTP, 1995); BMR = 10% extra risk; dose shown in
mg/kg-d C-26
Figure C-10. Plot of incidence by dose, with fitted curve for Multistage 1° model for thyroid
follicular cell adenomas or carcinomas in male B6C3F1 mice exposed to tert-
butanol in drinking water for 2 years (NTP, 1995); BMR = 5% extra risk; dose
shown in mg/kg-d C-29
Figure C-ll. Plot of incidence by dose, with fitted curve for Multistage 2° model for thyroid
follicular cell adenomas or carcinomas in male B6C3F1 mice exposed to tert-
butanol in drinking water for 2 years, high dose omitted (NTP, 1995); BMR = 5%
extra risk; dose shown in mg/kg-d C-32
Figure C-12. Plot of incidence by dose, with fitted curve for Multistage 2° model for renal
tubule adenoma or carcinoma in male F344 rats exposed to tert-butanol in
drinking water for 2 years modeled with administered dose units and including
all dose groups (NTP, 1995); BMR = 10% extra risk; dose shown in mg/kg-d C-35
Figure C-13. Plot of incidence by dose, with fitted curve for Multistage 1° model for renal
tubule adenoma or carcinoma in male F344 rats exposed to tert-butanol in
drinking water for 2 years modeled with administered dose units and excluding
high-dose group (NTP, 1995); BMR = 10% extra risk.; dose shown in mg/kg-d C-37
Figure C-14. Plot of incidence by dose, with fitted curve for Multistage 1° model for renal
tubule adenoma or carcinoma in male F344 rats exposed to tert-butanol in
drinking water for 2 years modeled with PBPK (tert-butanol, mg/L) dose units
and including all dose groups (NTP, 1995); BMR = 10% extra risk.; dose shown
in mg/L C-39
Figure C-15. Plot of incidence by dose, with fitted curve for Multistage 1° model for renal
tubule adenoma or carcinoma in male F344 rats exposed to tert-butanol in
drinking water for 2 years modeled with PBPK (tert-butanol, mg/L) dose units
and excluding high-dose group (NTP, 1995); BMR = 10% extra risk; dose shown
in mg/L C-41
Figure C-16. Plot of incidence by dose, with fitted curve for Multistage 1° model for renal
tubule adenoma or carcinoma in male F344 rats exposed to tert-butanol in
drinking water for 2 years modeled with PBPK (metabolized, mg/hr) dose units
and including all dose groups (NTP, 1995); BMR = 10% extra risk; dose shown in
mg/hr C-43
Figure C-17. Plot of incidence by dose, with fitted curve for Multistage 1° model for renal
tubule adenoma or carcinoma in male F344 rats exposed to tert-butanol in
drinking water for 2 years modeled with PBPK (metabolized, mg/hr) dose units
and excluding high-dose group (NTP, 1995); BMR = 10% extra risk; dose shown
in mg/hr C-45
Figure C-18. Plot of incidence by dose, with fitted curve for Multistage 1° model for renal
tubule adenoma or carcinoma in male F344 rats exposed to tert-butanol in
drinking water for 2 years modeled with administered dose units and excluding
This document is a draft for review purposes only and does not constitute Agency policy.
vii DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
high-dose group; re-analyzed data (Hard et al., 2011; NTP, 1995); BMR = 10%
extra risk; dose shown in mg/kg-d C-48
Figure C-19. Plot of incidence by dose, with fitted curve for Multistage 1° model for renal
tubule adenoma or carcinoma in male F344 rats exposed to tert-butanol in
drinking water for 2 years modeled with PBPK (tert-butanol, mg/L) dose units
and excluding high-dose group; reanalyzed data (Hard et al., 2011; NTP, 1995);
BMR = 10% extra risk; dose shown in mg/L C-51
Figure C-20. Plot of incidence by dose, with fitted curve for Multistage 1° model for renal
tubule adenoma or carcinoma in male F344 rats exposed to tert-butanol in
drinking water for 2 years modeled with PBPK (metabolized, mg/hr) dose units
and excluding high-dose group; reanalyzed data (Hard et al., 2011; NTP, 1995);
BMR = 10% extra risk.; dose shown in mg/hr C-53
This document is a draft for review purposes only and does not constitute Agency policy.
viii DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
ABBREVIATIONS
AIC Akaike's information criterion
ARCO ARCO Chemical Company
BMD benchmark dose
BMDL benchmark dose lower confidence limit
BMDS Benchmark Dose Software
BMDU benchmark dose upper confidence limit
BMR benchmark response
BW body weight
CFR Code of Federal Regulations
CHO Chinese hamster ovary
CYP450 cytochrome P450
DMSO dimethyl sulfoxide
DNA deoxyribonucleic acid
EDTA ethylenediaminetetraacetic acid
EPA U.S. Environmental Protection Agency
ETBE ethyl tert-butyl ether
HBA 2-hydroxyisobutyrate
HL human leukemia
ICso half-maximal inhibitory concentration
i.p. intraperitoneal
i.v. intravenous
MFO mixed function oxidase
MPD 2-methyl-l,2-propanediol
MTBE methyl tert-butyl ether
NADPH nicotinamide adenine dinucleotide
phosphate
NTP National Toxicology Program
•OH hydroxyl radical
PBPK physiologically based pharmacokinetic
POD point of departure
SD standard deviation
TWA time-weighted average
This document is a draft for review purposes only and does not constitute Agency policy.
ix DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
i APPENDIX A. ASSESSMENTS BY OTHER NATIONAL
2 AND INTERNATIONAL HEALTH AGENCIES
Table A-l. Health assessments and regulatory limits by other national and
international health agencies
Organization
National Institute of Occupational
Safety and Health (NIOSH, 2007)
Occupational Safety and Health
(OSHA, 2006)
Food and Drug Administration
(FDA, 2011a, b)
Toxicity value
Recommended Exposure Limit - 100 ppm (300 mg/m3) time-weighted
average (TWA) for up to a 10-hour workday and a 40-hour work week
Permissible Exposure Limit for general industry - 100 ppm (300 mg/m3) TWA
for an 8-hour workday
te/t-Butyl alcohol: Indirect food additive that may be safely used in surface
lubricants employed in the manufacture of metallic articles that contact
food, subject to the provisions of this section (21 Code of Federal
Regulations [CFR] 178.3910); substance may be used as a defoaming agent
(21 CFR 176.200).
This document is a draft for review purposes only and does not constitute Agency policy.
A-l DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
i APPENDIX B. INFORMATION IN SUPPORT OF
2 HAZARD IDENTIFICATION AND DOSE-RESPONSE
3 ANALYSIS
4 B.I. TOXICOKINETICS
5 Little information is available on the absorption, distribution, metabolism, or excretion of
6 tert-butyl alcohol (tert-butanol) in humans. The studies identified for this Toxicological Assessment
7 were conducted in conjunction with methyl tert-butyl ether (MTBE) or ethyl tert-butyl ether
8 (ETBE), as tert-butanol is a metabolite of both compounds. Several studies examining some aspect
9 of the toxicokinetic behavior of tert-butanol in animals have been identified. Many were carried out
10 in conjunction with other specific endpoints (e.g., developmental). ARCO [1983] determined no
11 differences in the pharmacokinetics of tert-butanol following either oral (i.e., gavage) or inhalation
12 exposure. Although some information is available for both oral and inhalation exposures, many
13 studies administered tert-butanol via intraperitoneal (i.p.) or intravenous (i.v.) injection. Although
14 these studies do not inform the absorption of tert-butanol, they can provide information on its
15 distribution, metabolism, and excretion.
16 B.I.I. Absorption
17 Toxicity data on tert-butanol submitted by industry to the U.S. Environmental Protection
18 Agency (EPA) under Section 8(e) of the Toxic Substances Control Act and other reporting
19 requirements indicate that tert-butanol is rapidly absorbed after oral administration. Very little of
20 the administered dose was excreted in the feces of rats, indicating 99% of the compound was
21 absorbed. Comparable blood levels of tert-butanol and its metabolites have been observed after
22 acute oral (350 mg/kg) or inhalation (6,060 mg/m3 for 6 hours) exposures in male Sprague-Dawley
23 rats (ARCO. 1983): the absorption rate after inhalation exposure could not be determined, however,
24 because the blood was saturated with radioactivity after 6 hours of exposure to 6,060 mg/m3. In
25 another study (Faulkner etal.. 1989). blood concentrations indicated that absorption was complete
26 at 1.5 hours following the last of six oral gavage doses of 10.5 mmoles tert-butanol/kg (twice daily)
27 in female C57BL/6J mice. There was an apparent zero-order decline in tert-butanol concentration
28 for most of the elimination phase, and no differences in absorption or elimination rates was
29 observed between mice on a repeated dosing regimen and control mice administered equivalent
30 volumes of tap water every 12 hours before administration of a single dose of 10.5 mmoles tert-
31 butanol/kg. The study therefore concluded that previous exposures did not affect the absorption or
32 elimination of tert-butanol (Faulkner etal.. 1989).
This document is a draft for review purposes only and does not constitute Agency policy.
B-l DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 B.1.2. Distribution
2 The available animal data suggest that tert-butanol is distributed throughout the body
3 following oral, inhalation, and i.v. exposures [Poetetal.. 1997: Faulkner et al.. 1989: ARCO.
4 1983). Nihlenetal. [1995] calculated partition coefficients for tert-butanol using blood from human
5 volunteers and available information about the relative content of water and fat in each tissue. The
6 calculated tissue:blood partition coefficients for tert-butanol were slightly above 1 (from 1.02 to
7 1.06) for most tissues, except for fatblood, which was 0.646. The same study evaluated the
8 partition coefficients of three oxygenated ethers, including MTBE and ETBE, which are metabolized
9 to tert-butanol (see Section B.I.4). The study concluded that, although tert-butanol preferentially
10 distributes in body water, the ethers distribute uniformly throughout the body with preference for
11 fatty tissues (Nihlenetal.. 1995).
12 In a study aimed at determining whether tert-butanol (or metabolites) can bind to
13 a?n-globulin, Williams and Borghoff (2001) exposed F-344 rats to a single gavage dose of 500
14 mg/kg 14C-tert-butanol. They found the radiolabel in three tissues (kidney, liver, and blood) in both
15 sexes, but male rats retained more of the tert-butanol equivalents than females (Williams and
16 Borghoff, 2001). Radioactivity was found in the low-molecular-weight protein fraction isolated
17 from the kidney cytosol in male rats but not in female rats, indicating that tert-butanol or one of its
18 metabolites was bound to a2u-globulin. Further analysis determined that tert-butanol, and not its
19 metabolite acetone, was bound. Most tert-butanol in the kidney cytosol was eluted as the free
20 compound in both males and females, but a small amount was associated with the high-molecular-
21 weight protein fraction in both males and females. In another study on (X2u-globulin
22 nephropathy, Borghoff etal. (2001) found similar results after F-344 rats were exposed to 0, 250,
23 450, or 1750 ppm tert-butanol by inhalation for 10 consecutive days. Male rat tert-butanol kidney-
24 to-blood ratios were significantly elevated over ratios in females at all dose levels and exposure
25 durations. Although the female tert-butanol kidney-to-blood ratio remained similar with both
26 duration and concentration, the male tert-butanol kidney-to-blood ratio increased with duration.
27 The liver-to-blood ratios were similar regardless of exposure duration, concentration, or sex. Both
28 of these studies indicate distribution to the liver and kidney with kidney retention of tert-butanol in
29 the male rat.
30 B.I.3. Metabolism
31 A general metabolic scheme for tert-butanol, illustrating the biotransformation in rats and
32 humans, is shown in Figure B-l. Urinary metabolites of tert-butanol in a human male volunteer who
33 ingested a gelatin capsule containing 5 mg/kg [13C]-tert-butanol were reported to be 2-methyl-l,2-
34 propanediol (MPD) and 2-hydroxyisobutyrate (Bernauer et al., 1998). Minor metabolites of
35 unconjugated tert-butanol, tert-butanol glucuronides, and traces of the sulfate conjugate also were
36 detected. The study was approved by an ethical review board, but no information regarding
37 informed consent was reported. In the same study, 2-hydroxyisobutyrate, MPD, and tert-butanol
38 sulfate were identified as major metabolites in rats, while acetone, tert-butanol, and tert-butanol
This document is a draft for review purposes only and does not constitute Agency policy.
B-2 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 glucuronides were identified as minor metabolites [Bernauer et al., 1998]. Baker etal. [1982] found
2 that tert-butanol was a source of acetone, but acetone production might have been stimulated from
3 other sources.
4 No studies identified specific enzymes responsible for biotransforming tert-butanol. Using a
5 purified enzyme from Sprague-Dawley rats or whole-liver cytosol from Wistar rats, alcohol
6 dehydrogenase had negligible or no activity toward tert-butanol [Videlaetal.. 1982: Arslanianetal..
7 1971]. Other in vitro studies have implicated the liver microsomal mixed function oxidase (MFO]
8 system, namely cytochrome P450 (CYP450] [Cederbaumetal.. 1983: Cederbaum and Cohen. 1980].
9 In the 1983 study, incubation of tert-butanol at 35 mM with Sprague-Dawley rat liver microsomes
10 and a nicotinamide adenine dinucleotide phosphate- (NADPH] generating system resulted in
11 formaldehyde the production at a rate of approximately 25 nmoles/mg protein/30 min. According
12 to study authors, the amount of formaldehyde generated by tert-butanol was approximately 30% of
13 the amount of formaldehyde formed during the metabolism of 10 mM aminopyrene in a similar
14 microsomal system. The rate of formaldehyde generation from tert-butanol increased to about
15 90 nmol/mg protein/30 min upon addition of azide, which inhibits catalase and thereby prevents
16 the decomposition of hydrogen peroxide (HzOz). In other experiments in the same study,
17 formaldehyde formation was greatly reduced when H202 was included but NADPH was absent or
18 when the microsomes were boiled prior to incubation. Additionally, the rate of formaldehyde
19 formation in the microsomal oxidizing system depended on the concentration of tert-butanol, with
20 apparent Km and Vmax values of 30 mM and 5.5 nmol/min/mg protein, respectively. The study
21 authors concluded that tert-butanol is metabolized to formaldehyde by a mechanism involving
22 oxidation of NADPH, microsomal electron flow, and the generation of hydroxyl-radical (-OH] from
23 H202, possibly by a Fenton-type or a Haber-Weiss iron-catalyzed reaction involving CYP450, which
24 might serve as the iron chelate [Cederbaum and Cohen. 1980].
This document is a draft for review purposes only and does not constitute Agency policy.
B-3 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
glucuronide-O——CH3
CH3
t-butyl glucuronide
_
HO-
HO-
rats,
CH3
|
— — CH
1
CH3
humans
OH
CYP450
3 ^ |_| Q
rats,
humans Q|-|
t-butanol 0 ,. . , 0
-CH-,
[O]
CH-,
2-hydroxyisobutyric acid
2-methyl-1,2-propanediol
H2C=O
formaldehyde
O"
rats
v°
CH,
O-
acetone
-CH,
CH3
~ t-butyl sulfate
3 Source: NSF International (2003), ATSDR (1996), Bernauer et al. (1998), Amberg et al. (1999),
4 and Cederbaum and Cohen (1980).
5 Figure B-l. Biotransformation of tert-butanol in rats and humans.
6 In a follow-up study, tert-butanol was oxidized to formaldehyde and acetone by various
7 systems known to generate -OH radical, including rat liver microsomes or other nonmicrosomal
8 -OH-generating systems [Cederbaum etal.. 1983). The nonmicrosomal tests included two chemical
9 systems: (1) the iron-catalyzed oxidation of ascorbic acid (ascorbate-Fe-EDTA
10 [ethylenediaminetetraacetic acid]) and (2) the Fenton system of chelated ferrous iron and HzOz. In
11 both Fenton-type systems, HzOz served as a precursor for -OH. Additionally, a Haber-Weiss
12 enzymatic system involving xanthine oxidation by xanthine oxidase in the presence of Fe-EDTA was
13 used. In this system, -OH is thought to be produced by the interaction of H 20 2 and superoxide
14 (C>2-~). Further experiments demonstrated the involvement of -OH in either the ascorbate-Fe-EDTA
15 or the xanthine oxidation systems based on inhibition of formaldehyde and acetone production
16 from tert-butanol when -OH-scavenging agents (e.g., benzoate, mannitol) were added. Some
17 experiments in this study of the oxidation of tert-butanol by the microsomal metabolizing system of
18 the liver were similar to those in the previous study [Cederbaum and Cohen. 1980] except that
19 acetone formation, in addition to formaldehyde, also was measured. Again, these experiments
20 showed the dependence of the microsomal metabolizing system on an NADPH-generating system
21 and the ability of H202 to enhance, but not replace, the NADPH-generating system. Addition of
This document is a draft for review purposes only and does not constitute Agency policy.
B-4 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 chelated iron (Fe-EDTA) boosted the microsomal production of formaldehyde and acetone, while
2 -OH-scavenging agents inhibited their production. The study authors noted that neither Fe-EDTA
3 nor • OH-scavenging agents is known to affect the CYP450-catalyzed oxidation of typical MFC
4 substrates such as aminopyrene or aniline. The study also showed that known CYP45 0 inhibitors,
5 such as metyrapone or SKF-525A, inhibited the production of formaldehyde from aminopyrene but
6 not from tert-butanol. Finally, typical inducers of CYP450 and its MFC metabolizing activities, such
7 as phenobarbital or 3-methylcholanthrene, had no effect on microsomal metabolism of tert-butanol
8 to formaldehyde and acetone. According to the study authors, the oxidation of tert-butanol appears
9 to be mediated by -OH (possibly via H202), which can be produced by any of the tested systems by a
10 Fenton-type reaction as follows:
11 H202 + Fe2+-chelate -> -OH + OH- + Fe3+-chelate
12 According to this reaction, reduction of ferric iron (Fe3+) to ferrous iron (Fe2+) is required
13 for continuous activity. The study authors concluded that the nature of the iron and the pathway of
14 iron reduction within the microsomes remain to be elucidated even though an NADPH-dependent
15 electron transfer or 02-~ might be involved [Cederbaum etal.. 1983).
16 B.1.4. Excretion
17 Human data on the excretion of tert-butanol derives from studies of MTBE and ETBE
18 (Nihlenetal.. 1998a. b). Eight or ten male human volunteers were exposed to 5, 25, or 50 ppm
19 MTBE (18.0, 90.1, 757 mg/m3) or ETBE (20.9,104, and 210 mg/m3) by inhalation during 2 hours of
20 light exercise. The half-life of tert-butanol in urine following MTBE exposure was 8.1 ± 2.0 hours
21 (average of the 25- and 50-ppm MTBE doses); the half-life of tert-butanol in urine following ETBE
22 exposure was 7.9 ± 2.7 hours (average of 25- and 50-ppm ETBE doses). In both studies, the urinary
23 excretion of tert-butanol was less than 1% of the uptake or absorption of MTBE or ETBE. The renal
24 clearance rate of tert-butanol was 0.67 ±0.11 mL/hr-kg with MTBE exposure (average of 25- and
25 50-ppm MTBE doses); the renal clearance rate was 0.80 ± 0.34 mL/hr-kg with ETBE exposure
26 (average of 25- and 50-ppm ETBE doses).
27 Ambergetal. (2000) exposed six volunteers (three males and three females, 28 ± 2 years
28 old) to 18.8 and 170 mg/m3 ETBE. Each exposure lasted 4 hours, and the two concentrations were
29 administered to the same volunteers 4 weeks apart Urine was collected at 6-hour intervals for
30 72 hours following exposure. tert-Butanol and two metabolites of tert-butanol,
31 2-hydroxyisobutyrate (HBA) and MPD, also were identified in urine. At an ETBE level of 170
32 mg/m3, tert-butanol displayed a half-life of 9.8 ± 1.4 hours. At the low-exposure ETBE
33 concentration, the tert-butanol half-life was 8.2 ± 2.2 hours. The predominant urinary metabolite
34 identified was HBA, excreted in urine at 5-10 times the amount of MPD and 12-18 times the
35 amount of tert-butanol (note: urine samples had been treated with acid before analysis to cleave
36 conjugates). HBA in urine showed a broad maximum at 12-30 hours after exposure to both
This document is a draft for review purposes only and does not constitute Agency policy.
B-5 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 concentrations, with a slow decline thereafter. MPD in urine peaked at 12 and 18 hours after
2 exposure to 170 and 18.8 mg/m3 ETBE, respectively, while tert-butanol peaked at 6 hours after
3 exposure to both concentrations.
4 Ambergetal. [2000] exposed F344 NH rats to 18.8 and 170 mg/m3 ETBE. Urine was
5 collected for 72 hours following exposure. Similar to humans, rats excreted mostly HBA in urine,
6 followed by MPD and tert-butanol. The half-life for tert-butanol in rat urine was 4.6 ± 1.4 hours at
7 ETBE levels of 170 mg/m3, but half-life could not be calculated at the ETBE concentration of
8 18.8 mg/m3. Corresponding half-lives were 2.6 ± 0.5 and 4.0 ± 0.9 hours for MPD and 3.0 ± 1.0 and
9 4.7 ± 2.6 hours for HBA. In Sprague-Dawley rats treated with radiolabeled tert-butanol by gavage at
10 1, 30, or 500 mg/kg, a generally constant fraction of the administered radioactivity (23-33%) was
11 recovered in the urine at 24 hours postdosing. Only 9% of a 1500-mg/kg administered dose was
12 recovered in urine, however, suggesting that the urinary route of elimination is saturated following
13 this dose [ARCO. 1983]. Among all tested doses, most of the urinary radiolabel was attributed to a
14 polar fraction that was not characterized, while only 0.3-5.5% of the administered dose was
15 considered tert-butanol. The saturation in urinary elimination of radioactivity with the increased
16 dose was considered a manifestation of saturated metabolic capacity; however, no further
17 information was provided on the fate or balance of the administered radiolabel at any of the tested
18 tert-butanol doses [ARCO. 1983].
19 B.I.5. Physiologically Based Pharmacokinetic Models
20 No physiologically based pharmacokinetic (PBPK] models have been developed specifically
21 for administration of tert-butanol. Some models have been used to study tert-butanol as the
22 primary metabolite after oral or inhalation exposure to MTBE or ETBE. The most recent models for
23 MTBE oral and inhalation exposure include a component for the binding of tert-butanol to
24 a2u-globulin fBorghoffetal.. 2010: Leavens and Borghoff. 20091
25 Faulkner and Hussain [1989] used a one-compartment, open model with Michaelis-Menten
26 elimination kinetics to fit tert-butanol blood concentrations obtained from C57BL/6J mice given i.p.
27 injections of 5,10, or 20 mmol/kg tert-butanol. Elimination was indistinguishable from first-order
28 kinetics in the range of concentrations studied. An increase in Vmax and decrease in apparent
29 volume of distribution with dose are consistent with this model and suggest the existence of
30 parallel elimination processes.
31 Borghoff etal. [1996] developed a PBPK model for MTBE and its metabolite tert-butanol in
32 rats. Doses and blood levels were taken from several published studies. The initial model included a
33 tissue-specific, five-compartment model using blood, liver, kidney, muscle, and fat with liver
34 metabolism rate constants. The model predicted the accumulation of tert-butanol in blood, but not
35 its clearance. A two-compartment model was better at predicting tert-butanol blood levels, but the
36 volume of total body water had to be changed to obtain an adequate fit, suggesting dose-dependent
37 changes in the kinetics of tert-butanol. Overall, evaluation of the tert-butanol models suggests that
This document is a draft for review purposes only and does not constitute Agency policy.
B-6 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 the clearance of tert-butanol from the blood of rats after exposure to MTBE involves processes
2 beyond metabolic elimination.
3 Nihlen and Tohanson [1999] developed a PBPK model for evaluation of inhalation exposure
4 in humans to the gasoline additive ETBE. Model compartments for ETBE included lungs (with
5 arterial blood), liver, fat, rapidly perfused tissues, resting muscles, and working muscles. The same
6 set of compartments and an additional urinary excretion compartment were used for the
7 metabolite, tert-butanol. First-order metabolism was assumed in the model, and tissue/blood
8 partition coefficients were determined by in vitro methods [Nihlen etal.. 1995]. Estimates of
9 individual metabolite parameters of eight subjects were obtained by fitting the PBPK model to
10 experimental data from humans (5, 25, or 50 ppm ETBE; 2-hour exposure] [Nihlen etal.. 1998a].
11 This model primarily was applied to predict levels of the biomarkers ETBE and tert-butanol in
12 blood, urine, and exhaled air after various scenarios, such as prolonged exposure, fluctuating
13 exposure, and exposure during physical activity [Nihlen and Johanson. 1999].
14 Rao and Ginsberg [1997] developed a PBPK model for MTBE and its principal metabolite,
15 tert-butanol, based on the Borghoff etal. [1996] model. The modified model included a skin
16 compartment to simulate dermal absorption of MTBE during bathing or showering. A brain
17 compartment was added as a target organ for MTBE-induced neurological responses. MTBE
18 metabolism to tert-butanol was assumed to occur in the liver through two saturable pathways. The
19 tert-butanol portion of the model included further metabolism of tert-butanol in the liver,
20 exhalation in the lungs, and renal excretion (in the human model only]. The model was validated
21 against published human and rat data and was used to help determine the contribution of tert-
22 butanol to the acute central nervous system effects observed after MTBE dosing.
23 The Rao and Ginsberg [1997] model used peak concentrations of MTBE and tert-butanol in
24 the blood and brain for interspecies, route-to-route, and low-/high-dose extrapolations. The
25 MTBE/tert-butanol PBPK model was adapted to humans by adjusting physiology according to
26 literature values, incorporating the blood/air partition coefficient for humans reported by Tohanson
27 etal. [1995]. and allometrically scaling the metabolic rate based on body weight. A renal
28 elimination component was added to account for the small percentage of MTBE disposition that
29 occurs in humans via urinary excretion of tert-butanol. tert-Butanol concentrations in human blood
30 during and after MTBE exposure (25 or 50 ppm for 2 hours] were accurately predicted by the
31 human model [Tohanson etal.. 1995].
32 Kim etal. [2007] expanded the Borghoff etal. [1996] model to develop a multi-exposure
33 route model for MTBE and its primary metabolite, tert-butanol, in humans. The significant features
34 and advantages of the Kim etal. [2007] model are that parameters used for quantifying the
35 pharmacokinetic behavior of MTBE and tert-butanol are calibrated using time-series
36 measurements from controlled-exposure experiments in humans as reported by Prahetal. [2004].
37 MTBE partition coefficient values described in the Licataetal. [2001] model and skin compartment
38 parameters from the Rao and Ginsberg [1997] model were incorporated. The PBPK model for
This document is a draft for review purposes only and does not constitute Agency policy.
B-7 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 MTBE consists of nine primary compartments representing the lungs, skin, fat, kidney, stomach,
2 intestine, liver, rapidly perfused tissue, and slowly perfused tissue. The tert-butanol model consists
3 of three compartments representing blood, liver, and other tissue.
4 Leavens and Borghoff [2009] developed a PBPK model for inhalation exposures in male and
5 female rats that expanded on Borghoff etal. [1996] and Rao and Ginsberg [1997] to include the sex-
6 specific effects of MTBE binding to a2u-globulin, a protein unique to male rats, and to describe the
7 induction of tert-butanol metabolism after repeated exposures. Although the primary purpose of
8 the model was to estimate MTBE and tert-butanol tissue concentrations after MTBE exposure, the
9 model also was parameterized to include inhalation uptake of tert-butanol. The tert-butanol portion
10 of the model was calibrated using data from rat exposures to tert-butanol and to MTBE. Model
11 compartments included blood, brain, fat, gastrointestinal tissues, kidney, liver, poorly perfused
12 tissues (blood flow <100 mL/min/100 g of tissue: bone, muscle, skin, fat], and rapidly perfused
13 tissues.
14 Distribution of MTBE and tert-butanol was assumed perfusion (i.e., blood-flow] limited.
15 This model used the same assumptions as Borghoff etal. [1996] regarding MTBE metabolism and
16 kinetics and further assumed that tert-butanol was metabolized only in the liver through one low-
17 affinity pathway and excreted through urine. The model described binding of MTBE or tert-butanol
18 with a2u-globulin in the kidney, due to the high concentration of a2u-globulin in the kidney. As
19 chemicals bind to a2u-globulin, the rate of hydrolysis of the protein decreases and causes
20 accumulation in the kidney; however, there is no evidence that binding of a2u-globulin affects its
21 synthesis, secretion, or circulating concentrations [Borghoff etal. [1990] as cited in Leavens and
22 Borghoff [2009]]. Equations describing this phenomenon were included in the model for male rats
23 only to account for the effects of the binding with a2u-globulin on metabolism of MTBE and tert-
24 butanol. Partition coefficient values in the model that differed from those published in previous
25 PBPK models included poorly perfused tissues:blood and kidney:blood values. The kidney:blood
26 value was based on calculated kidney:blood concentrations in female rats only because of the lack
27 of a2u-globulin-associated effects in female rats. The deposition of tert-butanol during inhalation in
28 the nasal cavity and upper airways was reflected in the high blood:air partition coefficient for tert-
29 butanol, and the ability of tert-butanol to induce its own metabolism after chronic exposure also
30 was taken into account. No differences in the induction of metabolism were reported between
31 males and females. The model simulated concentrations of MTBE and tert-butanol in the brain,
32 liver, and kidney of male and female rats following inhalation exposure at concentrations of 100,
33 400,1,750, or 3,000 ppm MTBE, and compared them to measured concentrations of MTBE and tert-
34 butanol from rats exposed at those levels.
3 5 Concentrations of MTBE and tert-butanol in the brain and liver were similar in male and
36 female rats during exposure and postexposure, but the concentrations of the chemicals in the
37 kidney significantly differed between male rats and female rats. The additional parameter
38 accounting for a2u-globulin protein binding in this PBPK model more accurately reflects the
This document is a draft for review purposes only and does not constitute Agency policy.
B-8 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 metabolism of both MTBE and tert-butanol in male rat kidneys over time compared with other
2 PBPK models. The model highlights that binding can stimulate increased renal effects in male rats
3 after exposure to MTBE and tert-butanol. The assumptions made to reflect tert-butanol metabolism
4 induction and deposition in the nasal cavity and upper airways generally were supported by
5 measured data from rats exposed to 250-, 450-, or 1,750-ppm tert-butanol as evidenced by the fact
6 that the model was within one standard deviation of the mean concentrations for most data points.
7 The model overpredicted the concentration of tert-butanol in the brain, liver, and kidney of male
8 rats, however, after repeated exposures.
9 Borghoffetal. [2010] modified the PBPK model of Leavens and Borghoff [2009] by adding
10 oral gavage and drinking water exposure components to compare different dose metrics to the
11 toxicity observed across different studies. The Borghoffetal. [2010] model assumed first-order
12 uptake of MTBE absorption from the gut, with 100% of the MTBE dose absorbed for both drinking
13 water and oral gavage exposures. They conducted a series of pharmacokinetic studies comparing
14 the effects of different rat strains and different dosing vehicles on the blood concentration-time
15 profiles of MTBE and tert-butanol following MTBE exposure. The effects of exposure to MTBE via
16 drinking water, oral gavage, and inhalation routes over 7 and 91 days on male and female rats were
17 modeled and compared with measured data collected from F344 rats (exposed 28 days] and Wistar
18 Han rats (exposed 14 and 93 days].
19 The model predicted the blood concentrations of tert-butanol observed after 250 or 1,000
20 mg/kg-day administration of MTBE in males and females and the blood concentrations of MTBE
21 after 1,000 mg/kg-day. The model did not predict peak concentrations of MTBE, however, after 250
22 mg/kg-day in males or females using either olive oil or 2% Emulphor as vehicles. When comparing
23 strains, the blood concentrations were similar across strain and sex, except in female Sprague-
24 Dawley rats administered 1,000 mg/kg-day MTBE. Female Sprague-Dawley rats had a significantly
25 (p-value not specified] higher blood concentration of both MTBE and tert-butanol compared with
26 F344 and Wistar Han females. The study authors considered this an outlier, however, and
27 maintained the metabolic patterns were similar. The model overpredicted the amount of MTBE in
28 the male rat kidney but accurately predicted the level of tert-butanol in the male rat kidney at all
29 exposures tested. The model did not accurately predict the kidney concentrations of tert-butanol in
30 the female kidney after exposure to MTBE via drinking water, but the study authors attributed the
31 inaccuracies to the study design as opposed to the model formulation. All tert-butanol entering the
32 submodel comes from MTBE metabolism in the liver, and the model does not include a separate
33 oral intake of tert-butanol.
34 B.2. PBPK MODEL EVALUATION SUMMARY
35 B.2.1. Evaluation of Existing tert-Butanol Submodels
36 The Blancato et al. (2007) and Leavens and Borghoff [2009] PBPK models for MTBE were
37 evaluated by comparing predictions from the tert-butanol portions of the models with the tert-butanol
This document is a draft for review purposes only and does not constitute Agency policy.
B-9 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4
5
6
7
i.v. data of Poet et al. (1997) (see Figure B-2). Neither model adequately represented the tert-butanol
blood concentrations. Modifications of model assumptions for alveolar ventilation, explicit pulmonary
compartments, and induction of metabolism of tert-butanol did not significantly improve model fits
to the data. Attempts to reoptimize model parameters in the tert-butanol submodels of Blancato et al.
(2007) and Leavens and Borghoff [2009] to match blood concentrations from the i.v. dosing study
were unsuccessful.
(A) (B)
10000
1000
300mg/kg • male
— 150 mg/kg • male
75 mg/kg • male
- • -37.5 mg/kg » male
« female
o female
o female
* female
0 2 4 6 8 10 12 14 16 18 20 22 24
time (hours)
10000
1000
300 mg/kg » male o female
— ISO mg/kg • male n female
75 mg/kg • male o female
- • -37.5 mg/kg » male & female
0.1
0 2 4 6 8 10 12 14 16 18 20 22 24
time (hours)
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Neither the (A) Blancato etal. (2007) nor the (B) Leavens and Borghoff (2009) model adequately
represents the measured tert-butanol blood concentrations.
Figure B-2. Comparison of the tert-butanol portions of existing MTBE models
with tert-butanol blood concentrations from i.v. exposure by Poet et al.
f!9971.
The PBPK submodel for tert-butanol in rats was developed in acslX (Advanced Continuous
Simulation Language, Aegis, Inc., Huntsville, Alabama) by modifying information from the many PBPK
models developed in rats and humans for the structurally related substance, MTBE, and its metabolite
tert-butanol (Borghoff etal.. 2010; Leavens and Borghoff. 2009; Blancato et al.. 2007; Kim etal..
2007; Rao and Ginsberg. 1997; Borghoff et al.. 1996). A brief description comparing the Blancato et al.
(2007) and Leavens and Borghoff (2009) models is provided, followed by an evaluation of the MTBE
models and the assumptions adopted from MTBE models or modified in the tert-butanol model.
The Blancato et al. (2007) model is an update of the earlier Rao and Ginsberg (1997) model, and
the Leavens and Borghoff (2009) model is an update of the Borghoff et al. (1996) model. Both
the Blancato et al. (2007) and Leavens and Borghoff (2009) models are flow-limited models that
predict amounts and concentrations of MTBE and its metabolite tert-butanol in blood and six tissue
compartments: liver, kidney, fat, brain, and rapidly and slowly perfused tissues. These tissue
compartments are linked through blood flow, following an anatomically accurate, typical,
physiologically based description [Andersen. 1991]. The parent (MTBE) and metabolite
This document is a draft for review purposes only and does not constitute Agency policy.
B-10 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 (tert-butanol) models are linked by the metabolism of MTBE to tert-butanol in the liver. Routes of
2 exposure included in the models are oral and inhalation for MTBE; Leavens and Borghoff [2009]
3 included inhalation exposure to tert-butanol. Oral doses are assumed 100% bioavailable and 100%
4 absorbed from the gastrointestinal tract represented with a first-order rate constant. Following
5 inhalation of MTBE or tert-butanol, the chemical is assumed to enter the systemic blood supply
6 directly, and the respiratory tract is assumed to be at pseudo-steady state. Metabolism of MTBE by
7 CYP450s to formaldehyde and tert-butanol in the liver is described with two Michaelis-Menten
8 equations representing high- and low-affinity enzymes. tert-Butanol is either conjugated with
9 glucuronide or sulfate or further metabolized to acetone through MPD and HBA; both processes are
10 described by a single Michaelis-Menten equation in the models. All model assumptions are valid for
11 tert-butanol and were applied to the EPA-modified tert-butanol PBPK model, except for the
12 separate brain compartment The brain compartment was lumped with other richly perfused
13 tissues in the EPA-modified tert-butanol PBPK model.
14 In addition to differences in parameter values between the Blancato etal. [2007] and
15 the Leavens and Borghoff (2009) models, the model structure has three differences: (1) the alveolar
16 ventilation was reduced during exposure, (2) the rate of tert-butanol metabolism increased over time
17 due to induction of CYP enzymes, and (3) binding of MTBE and tert-butanol to a2u-globulin was
18 simulated in the kidney of male rats. The Blancato etal. [2007] model was configured through EPA's
19 PBPK modeling framework, ERDEM (Exposure-Related Dose Estimating Model], which includes
20 explicit pulmonary compartments. The modeling assumptions related to alveolar ventilation,
21 explicit pulmonary compartments, and induction of metabolism of tert-butanol are discussed in this
22 model evaluation section.
23 MTBE and tert-butanol binding to (X2u-globulin in the kidneys of male rats were incorporated in
24 the PBPK model of MTBE by Leavens and Borghoff (2009). Binding to a2u-globulin is one hypothesized
25 mode of action for the observed kidney effects in MTBE-exposed animals. For a detailed description of
26 the role of (X2u-globulin and other modes of action in kidney effects, see the kidney mode of action
27 section of the Toxicological Review (Section 1.2.1). In the Leavens and Borghoff (2009) model, binding of
28 MTBE to (X2u-globulin was applied to sex differences in kidney concentrations of MTBE and tert-
29 butanol, but acceptable estimates of MTBE and tert-butanol pharmacokinetics in the blood are
30 predicted in other models that did not consider (X2u-globulin binding. Given the uncertainty of tert-
31 butanol binding to (X2u-globulin, it was not included in the tert-butanol PBPK submodel.
32 B.2.2. Modification of Existing tert-Butanol Submodels
33 To account for the tert-butanol blood concentrations after i.v. tert-butanol exposure, the
34 model was modified by adding a pathway for reversible sequestration of tert-butanol in the blood (see
35 Figure B-3). The PBPK model represented the rate of change in the amount of tert-butanol in the
36 sequestered blood compartment (Abioodz) with the following equation, where KON is the binding rate
This document is a draft for review purposes only and does not constitute Agency policy.
B-ll DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 constant, CV is the free tert-butanol concentration in blood, KOFF is the unbinding rate constant, and
2 Cbiood2 is the concentration of tert-butanol bound in blood (equal to Abbod2/Vbiood).
4
5
6
7
8
9
10
11
dAblood2/dt=K0N*CV*-KoFF*Cbl,
Iood2
IV Dose
Inhalation Exhalation
A 1
Alveolar Air
Blood
Sbjndmg
.^dissociation
Sequestered
Rapidly Perfused
Slowly Perfused
Fat
Kidney
Liver
Oral
Dose
K
ELIM2
Urinary excretion
VMTBA, KMTBA
Metabolism
Exposure can be via multiple routes, including inhalation, oral, or i.v. dosing. Metabolism of tert-butanol,
which occurs in the liver, is described by Michaelis-Menten equations with one pathway for tert-butanol.
tert-Butanol is cleared via exhalation and via urinary excretion. See Table B-l for definitions of parameter
abbreviations.
Figure B-3. Schematic of the PBPK submodel for tert-butanol in rats.
Table B-l. PBPK model physiologic parameters and partition coefficients
Body weight and organ volumes as fraction of body weight
Body weight (kg)
Body fraction that is blood perfused (Fperf)
Liver
Kidney
Fat
0.25 (Brown et al., 1977)
0.8995 (Brown et al., 1977)
0.034 (Brown et al., 1977)
0.007 (Brown et al., 1977)
0.07 (Brown et al., 1977)
This document is a draft for review purposes only and does not constitute Agency policy.
B-12 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Rapidly perfused 0.04 (Brown et al., 1977)
Slowly perfused 0.7485 a
Blood 0.074 (Brown et al., 1977)
Cardiac output and organ blood flows as fraction of cardiac output
Cardiac output (L/hr) 5.38 (Brown et al., 1977)b
Alveolar ventilation (L/hr) 5.38 (Brown et al., 1977)c
Liver 0.174 (Brown etal., 1977)d
Kidney 0.141 (Brown et al., 1977)
Fat 0.07 (Brown etal., 1977)
Rapidly perfused 0.279 e
Slowly perfused 0.336 (Brown etal., 1977)
Partition coefficients for tert-butanol
Blood:air 481 (Borghoff etal., 1996)
Liverblood 0.83 (Borghoffetal., 1996)
Fat:blood 0.4 (Borghoffetal., 1996)
Rapidly perfused:blood 0.83 (Borghoff etal., 1996)
Slowly perfused:blood 1.0 (Borghoff etal., 1996)
Kidney:blood 0.83 (Borghoffetal., 1996)
a Fperf - Z(other compartments).
b 15.2*BW°-75 (BW = body weight).
c Alveolar ventilation is set equal to cardiac output.
d Sum of liver and gastrointestinal blood flows.
e 1 - Z(all other compartments).
1 The physiologic parameter values obtained from the literature are shown in Table B-l
2 [Brown et al.. 1977]. tert-Butanol partition coefficients, determined by the ratios of measured
3 tissue:air and blood:air partition coefficients [Borghoff etal.. 1996}. also were obtained from
4 literature. The parameters describing rate constants of metabolism and elimination of tert-butanol also
5 were obtained from the literature (Blancato et al.. 2007) and were kept fixed because they were
6 optimized to tert-butanol blood concentrations measured after MTBE exposure, which is also
7 metabolized to tert-butanol. The parameters describing tert-butanol absorption and tert-butanol
8 sequestration in blood were estimated by optimizing the model to the time-course data for blood tert-
9 butanol for rats exposed via i.v., inhalation, and oral routes (Leavens and Borghoff. 2009; Poet et al..
10 1997; ARCO. 1983). The model parameters were estimated with the acsIX optimization routine to
11 minimize the log-likelihood function of estimated and measured tert-butanol concentrations. The
12 Nedler-Mead algorithm was used with heteroscedasticity and allowed to vary between 0 and 2. The
13 predictions of the model with optimized parameters have a much-improved fit to the tert-butanol
14 blood concentrations after tert-butanol i.v. exposures, as shown in panel A of Figure B-4. Additionally,
This document is a draft for review purposes only and does not constitute Agency policy.
B-13 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 the model adequately estimated the tert-butanol blood concentrations after inhalation and oral gavage
2 exposures. The optimized parameter values are shown in Table B-2. The ARCO (1983) study measured
3 tert-butanol in plasma only, unlike the Poet et al. (1997) and Leavens and Borghoff (2009) studies,
4 which measured tert-butanol in whole blood. Based on the measurements of plasma and whole blood
5 by JPEC (2008), the concentration of tert-butanol in plasma is approximately 60% of the concentration
6 in whole blood. The tert-butanol plasma concentrations measured by ARCO were increased (divided by
7 60%) to the expected concentration in whole blood for comparison with the PBPK model.
8 Induction of tert-butanol-metabolizing enzymes was included in the Leavens and Borghoff
9 [2009] model of MTBE based on their study of rats exposed for 8 days to tert-butanol via inhalation.
10 The enzyme induction equation and parameters developed in the Leavens and Borghoff [2009]
11 model that were applied to the tert-butanol submodel are as follows.
12 Vmax tert-butanol IND = Vmax tert-butanol *INDMAX(l-exp(-KIND*t]]
13 Vmax tert-butanol IND is the maximum metabolic rate after accounting for enzyme induction,
14 Vmax tert-butanol is the metabolism rate constant from Table B-2 for both tert-butanol pathways,
15 andlNDMAXis the maximum percent increase in Vmax tert-butanol (124.9]. KIND is the rate
16 constant for enzyme induction (0.3977/day]. The increased tert-butanol metabolism better
17 estimates the measured tert-butanol blood concentrations as can be seen in the comparison of the
18 model predictions and experimental measurements shown in Figure B-5. The model better
19 predicted blood concentrations in female rats than male rats. The male rats had lower tert-butanol
20 blood concentrations after repeated exposures compared with female rats, and this difference could
21 indicate greater induction of tert-butanol metabolism or other physiologic changes such as
22 ventilation or urinary excretion in males. The current data for tert-butanol metabolism do not
23 provide sufficient information for resolving this difference between male and female rats.
24
This document is a draft for review purposes only and does not constitute Agency policy.
B-14 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
TBA inhalation exposure concentrations
400
0.1
024
8 10 12 14 16 18 20 22 24
time (hours)
3 10 12 14 16 IB 20 22 U
time (hours)
(C)
TB
u
c
o
'ro
dj
2
5
CO
t-
100
10
1
0.1
0.01
0.001
-
:
mr, 11
A gavage 500 mg/kg A
1 mg/kg •
A A
•
""" ?
•
0 3 6 9 12
time (hours)
5 Source: (A) i.v. data from Poet etal. (1997); (B) inhalation data from Leavens and Borghoff (2009); and (C)
6 oral gavage data from ARCO (1983) with the optimized parameter values as shown in Table B-2.
7 Figure B-4. Comparison of the EPA model predictions with measured tert-
8 butanol blood concentrations for i.v., inhalation, and oral gavage exposure to
9 tert-butanol.
10 Table B-2. Rate constants for tert-butanol determined by optimization of the model
11 with experimental data
Rate Constant
Value
Source or Reference
Metabolism (VMTBA; mg/kg-hr)a
Metabolism (KMTBA; mg/L)
Urinary elimination (KELIMZ; 1/hr)
TBA sequestration rate constant (KON; L/hr)
TBA unsequestration rate constant (KOFF; L/hr)
Absorption from gastrointestinal tract (KASZ; 1/hr)
8.0 Blancatoetal. (2007)
28.8 Blancatoetal. (2007)
0.5 Blancatoetal. (2007)
0.148 Optimized
0.0134 Optimized
0.5 Optimized
a Scaled by BW°-7 (0.250-7 = 0.379), BW = body weight.
This document is a draft for review purposes only and does not constitute Agency policy.
B-15 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 B.2.3. Summary of the PBPK Model for tert-Butanol
2 A PBPK model for tert-butanol was developed by modifying previous models for MTBE and
3 tert-butanol (Leavens and Borghoff. 2009; Blancato et al.. 2007). Published tert-butanol sub-models
4 do not adequately represent the tert-butanol blood concentrations measured in the i.v. study [Poet
5 etal., 1997]. The addition of a sequestered blood compartment for tert-butanol substantially
6 improved the model fit The alternative modification—changing to diffusion-limited distribution
7 between blood and tissues—also improved the model fit, but was considered less biologically
8 plausible. Physiological parameters and partition coefficients were obtained from published
9 measurements. The rate constants for tert-butanol metabolism and elimination were from a
10 published PBPK model of MTBE with a tert-butanol subcompartment (Blancato et al.. 2007).
11 Additional model parameters were estimated by calibrating to data sets for i.v., oral, and inhalation
12 exposures and repeated dosing studies for tert-butanol. Overall, the model produced acceptable fits
13 to multiple rat time-course datasets of tert-butanol blood levels following inhalation or oral gavage
14 exposures.
15 B.2.4. tert-Butanol Model Application
16 The PBPK model as described above was applied to toxicity studies to predict tert-butanol
17 blood concentrations (the preferred internal dose metric in the absence of evidence linking any
18 specific metabolite of tert-butanol to any toxic effect). For simulation studies where tert-butanol
19 was administered in drinking water, the consumption was modeled as episodic, based on the
20 pattern of drinking observed in rats [Spiteri, 1982].
21 B.2.5. PBPK Model Code
22 The PBPK acslX model code is available electronically through EPA's Health and
23 Environmental Research Online (HERO] database. All model files may be downloaded in a zipped
24 workspace from HERO (U.S. EPA, 201#, HEROID##].
25
This document is a draft for review purposes only and does not constitute Agency policy.
B-16 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
male rats without induction
male rats with induction
female rats without induction
female rats with induction
3 Male rats were exposed to 239, 444, or 1726 ppm and female rats were exposed to 256, 444, or
4 1914 ppm tert-butanol for up to 8 consecutive days (Borghoff et al., 2001). tert-Butanol blood
5 concentrations are better predicted by the model after 8 days of exposure with enzyme induction (right
6 panels) compared to without enzyme induction (left panels).
7 Figure B-5. Comparison of the EPA model predictions with measured amounts
8 of tert-butanol in blood after repeated inhalation exposure to tert-butanol.
9 B.3. OTHER PERTINENT TOXICITY INFORMATION
10 B.3.1. Other Toxicological Effects
11 B.3.1.1. Synthesis of Other Effects
12 Effects other than those related to kidney, thyroid, reproductive, developmental, and
13 neurodevelopmental effects were observed in some of the available rodent studies. These include
14 liver and urinary bladder effects. As previously mentioned in the Study Selection section of the
15 Toxicological Review, all studies discussed employed inhalation, oral gavage, or drinking water
16 exposures for >30 days. Studies are arranged in evidence tables by effect, species, duration, and
This document is a draft for review purposes only and does not constitute Agency policy.
B-17 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 design. The design, conduct, and reporting of each study was reviewed, and each study was
2 considered adequate to provide information pertinent to this assessment.
3 Central nervous system effects similar to those ethanol causes, in terms of animals
4 appearing intoxicated and having withdrawal symptoms after cessation of oral or inhalation
5 exposure, were observed with tert-butanol. Severity of central nervous system symptoms such as
6 withdrawal increased with dose and duration of exposure. Study quality and utility concerns (e.g.,
7 inappropriate exposure durations, lack of data reporting, small number of animals per treatment
8 group) associated with these studies [Grant and Samson. 1981: Snell. 1980: Thurmanetal..
9 1980: McComb and Goldstein, 1979a, bj Wood and Laverty, 1979], however, preclude an
10 understanding of potential neurotoxicity following tert-butanol exposure, and therefore, central
11 nervous system studies are not discussed further.
12 Exposure-response arrays of these effects on liver and urinary bladder are provided in
13 Figure B-6 and Figure B-7 for oral and inhalation studies, respectively.
14 Kidney effects
15 Absolute and relative kidney weight numerical data are presented in Table B-3.
16 Liver effects
17 Liver weight and body weight were demonstrated to be proportional and liver weight
18 normalized to body weight was concluded optimal for data analysis [Bailey etal.. 2004]: thus, only
19 relative liver weight is presented and considered in the determination of hazard. Although some
20 rodent studies observed liver effects (organ weight changes and histopathologic lesions], the effects
21 were not consistent across the database. Increases in relative liver weight with tert-butanol
22 exposure were observed, but the results pertaining to histopathologic changes were inconsistent
23 (Table B-4]. The NTP (1995] oral subchronic and chronic studies did not observe treatment-related
24 effects on liver histopathology in either sex of F344 rats. In a 10-week study in Wistar rats, several
25 liver lesions (including necrosis] and increased liver glycogen were observed in male rats (no
26 females were included in the study] with the only dose used (Acharyaetal.. 1997: Acharya etal..
27 1995]. The study provided no incidence or severity data. The dose used in this rat study was in the
28 range of the lower doses used in the NTP (1995] subchronic rat study. An increased incidence of
29 fatty liver was observed in the male mice of the highest dose group in the 2-year mouse bioassay,
30 but no histopathological changes were seen in the subchronic mouse study (NTP, 1995]. No
31 treatment-related effects in liver histopathology were observed in rats or mice of the NTP (1997]
32 subchronic inhalation study.
3 3 Urinary bladder effects
34 Subchronic studies reported effects in the urinary bladder (Table B-6], although the chronic
35 studies indicated little progression in incidence with increased exposure. Transitional epithelial
36 hyperplasia of the urinary bladder was observed in male rats and male mice after 13 weeks of
This document is a draft for review purposes only and does not constitute Agency policy.
B-18 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 exposure at doses of 3,610 mg/kg-day (male rats) and >3,940 mg/kg-day (male mice). In rats, the
2 increase in transitional epithelial hyperplasia of the urinary bladder was not observed in the 2-year
3 study. Male mice exposed at the high dose (2,070 mg/kg-day) for 2 years exhibited minimal
4 transitional epithelial hyperplasia of the urinary bladder. Neither female rats nor female mice
5 showed increased incidences of this lesion. Both sexes of mice demonstrated incidence of minimal
6 to mild inflammation in the urinary bladder after both subchronic and chronic exposures, with a
7 greater incidence in males compared to females.
8 B.3.1.2. Mechanistic Evidence
9 No mechanistic evidence is available for these effects.
10 B.3.1.3. Summary of Other Toxicity Data
11 Based on lack of consistency and lack of progression, the available evidence does not
12 support liver and urinary bladder effects, respectively, as potential human hazards of tert-butanol
13 exposure.
14
This document is a draft for review purposes only and does not constitute Agency policy.
B-19 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
Table B-3. Changes in kidney weight in animals following exposure to
tert-butanol
Reference and study design
Results
Kidney weight (percent change as compared to control)
Lvondell Chemical Co. (2004)
Sprague-Dawley rat;
12/sex/treatment
Gavage 0, 64, 160, 400, or
1,000 mg/kg-d
Males: 9 weeks beginning 4
weeks prior to mating
Females: = 10 weeks (4 weeks
prior to mating through PND21)
NTP (1995)
F344/N rat; 10/sex/treatment
40 mg/mL
M: 0, 230, 490, 840, 1,520,
3,610a mg/kg-d
F: 0, 290, 590, 850, 1,560,
3,620a mg/kg-d
13 weeks
Males
Dose
(mg/kg-d)
0
64
160
400
1,000
Females
Dose
(mg/kg-d)
0
64
160
400
1,000
Males
Dose
(mg/kg-d)
0
230
490
840
1,520
3,610
Left absolute Left relative Right absolute
weight
0
+6
+9
+12*
+18*
weight
0
+8
+14*
+14*
+28*
weight
0
+6
+6
+14*
+20*
Left absolute Left relative Right absolute
weight
0
-1
0
+3
+4
Absolute
weight
0
+12*
+17*
+16*
+26*
All dead
weight
0
-2
0
+2
0
Females
Relative Dose
weight (mg/kg-d)
0 0
+19* 290
+26* 590
+32* 850
+54* 1,560
All dead 3,620
weight
0
+2
+1
+4
+7
Absolute
weight
0
+19*
+16*
+29*
+39*
+36*
Right relative
weight
0
+8
+11*
+17*
+31*
Right relative
weight
0
0
0
+2
+2
Relative
weight
0
+17*
+15*
+28*
+40*
+81*
This document is a draft for review purposes only and does not constitute Agency policy.
B-20 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Reference and study design
NTP (1995)
B6C3Fi mouse; 10/sex/treatment
Drinking water (0, 2.5, 5, 10, 20,
40 mg/mL)
M: 0, 350, 640, 1,590, 3,940,
8,2 10a mg/kg-d
F: 0, 500, 820, 1,660, 6,430,
11,620 a mg/kg-d
13 weeks
NTP (1995)
F344/N rat; 60/sex/treatment
(10/sex/treatment evaluated at
15 months)
Drinking water (0, 1.25, 2.5, 5, or
10 mg/mL)
M: 0, 90, 200, or 420a mg/kg-d
F: 0, 180, 330, or 650a mg/kg-d
2 years
NTP (1997)
F344/N rat; 10/sex/treatment
Inhalation analytical
concentration: 0, 134, 272, 542,
1,080, or 2,101 ppm (0, 406, 824,
1,643, 3,273 or 6,368 mg/m3)
(dynamic whole-body chamber)
6 hr/d, 5 d/wk
13 weeks
Generation method (Sonimist
Ultrasonic spray nozzle
nebulizer), analytical
concentration and method were
reported
Right kidney weights measured
Results
Males
Dose Absolute
(mg/kg-d) weight
0 0
350 +1
640 +3
1,590 +2
3,940 +6
8,210 0
Males
Dose Absolute
(mg/kg-d) weight
0 0
90 +4
200 +11
420 +7
Females
Relative Dose Absolute
weight (mg/kg-d) weight
000
+1 500 0
+2 820 -3
+8 1,660 +1
+22* 6,430 +6
+48* 11,620 +12*
Females
Relative Dose Absolute
weight (mg/kg-d) weight
000
+8 180 +8*
+15* 330 +18*
+20* 650 +22*
Relative
weight
0
-3
-1
0
+15*
+35*
Relative
weight
0
+14*
+21*
+42*
Only rats sacrificed at 15 months were evaluated for organ weights.
Males
Concentration Absolute
(mg/m-) weight
0 0
406 +1
824 -2
1,643 +3
3,273 +11*
6,368 +9.8*
Females
Relative Absolute
weight weight
0 0
+1 -4
-1 0
+2 +4
+8* +2
+9* +4
Relative
weight
0
-1
+1
+4
+2
+9*
This document is a draft for review purposes only and does not constitute Agency policy.
B-21 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Reference and study design
NTP (1997)
B6C3Fi mouse; 10/sex/treatment
Inhalation analytical
concentration: 0, 134, 272, 542,
1,080, or 2,101 ppm (0, 406, 824,
1,643, 3,273 or 6,368 mg/m3)
(dynamic whole-body chamber)
6 hr/d, 5 d/wk
13 weeks
Generation method (Sonimist
Ultrasonic spray nozzle
nebulizer), analytical
concentration and method were
reported
Right kidney weights measured
Results
Concentration
(mg/m-)
0
406
824
1,643
3,273
6,368
Males
Absolute
weight
0
-6
-1
+4
-10
+3
Relative
weight
0
-4
+3
+3
-3
+6
Females
Absolute
weight
0
+1
+5
+1
0
+3
Relative
weight
0
-3
+9
-2
+7
+15*
1
2
3
4
5
a The high-dose group had an increase in mortality.
* Statistically significant p < 0.05 as determined by the study authors.
Percentage change compared to control = (treated value - control value) 4- control value x 100.
Conversions from drinking water concentrations to mg/kg-d performed by study authors.
Conversion from ppm to mg/m3 is 1 ppm = 3.031 mg/m3.
This document is a draft for review purposes only and does not constitute Agency policy.
B-22 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
Table B-4. Changes in liver weight in animals following exposure to
tert-butanol
Reference and study design
Acharvaetal.(1995)
Wistar rat; 5-6 males/treatment
Drinking water (0 or 0.5%), 0 or 575 mg/kg-
d
10 weeks
Lvondell Chemical Co. (2004)
Sprague-Dawley rat; 12/sex/treatment
Gavage 0, 64, 160, 400, or 1,000 mg/kg-d
Males: 9 weeks beginning 4 weeks prior to
mating
Females: 4 weeks prior to mating through
PND21
NTP (1995)
F344/N rat; 10/sex/treatment
Drinking water (0, 2.5, 5, 10, 20, or 40
mg/mL)
M: 0, 230, 490, 840, 1,520, 3,610a mg/kg-d
F: 0, 290, 590, 850, 1,560, 3,620a mg/kg-d
13 weeks
NTP (1995)
B6C3Fi mouse; 10/sex/treatment
Drinking water (0, 2.5, 5, 10, 20, or 40
mg/mL)
M: 0, 350, 640, 1,590, 3,940,
8,210a mg/kg-d
F: 0, 500, 820, 1,660, 6,430,
ll,620a mg/kg-d
13 weeks
NTP (1995)
Results
No significant treatment-related effects (results were only provided in a figure)
Percent change compared to control:
Males
Dose
(mg/kg-d)
0
64
160
400
1,000
Absolute
weight
0
-1
-3
-2
+8
Relative
weight
0
0
+1
-1
+16*
Females
Dose
(mg/kg-d)
0
64
160
400
1,000
Absolute
weight
0
-4
-7
+2
+8
Relative weight
0
-4
-5
+1
+3
Percent change compared to control:
Males
Dose
f mp/kp-rH
V's/^s u/
0
230
490
840
1,520
3,610
Absolute
wsisht
0
-2
+1
+5
+8
All dead
Relative
wsisht
0
+4
+8*
+20*
+31*
All dead
Females
Dose
f mp/kp-rH
V's/^s u/
0
290
590
850
1,560
3,620
Absolute
wsisht
0
+11*
+10*
+12*
+15*
+9*
Relative
wsisht
0
+9*
+9*
+11*
+16*
+41*
Percent change compared to control:
Males
Dose
(mg/kg-d)
0
350
640
1,590
3,940
8,210
Absolute
weight
0
+2
-1
-1
0
-16
Relative
weight
0
+3
-2
+5
+14*
+22*
Females
Dose
(mg/kg-d)
0
500
820
1,660
6,430
11,620
Absolute
weight
0
-1
-5
-8
-2
-6
Relative
weight
0
-4
-3
_9*
+6
+13*
Percent change compared to control:
Males
Females
This document is a draft for review purposes only and does not constitute Agency policy.
B-23 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Reference and study design
F344/N rat; 60/sex/treatment
(10/sex/treatment evaluated at 15 months)
Drinking water (0, 1.25, 2.5, 5 or 10 mg/mL)
M: 0, 90, 200, or 420a mg/kg-d
F: 0, 180, 330, or 650a mg/kg-d
2 years
NTP (1997)
F344/N rat; 10/sex/treatment
Inhalation analytical concentration: 0, 134,
272, 542, 1,080, or 2,101 ppm (0, 406, 824,
1,643, 3,273 or 6,368 mg/m3) (dynamic
whole body chamber)
6 hr/d, 5 d/wk
13 weeks
Generation method (Sonimist Ultrasonic
spray nozzle nebulizer), analytical
concentration and method were reported
NTP (1997)
B6C3Fi mouse; 10/sex/treatment
Inhalation analytical concentration: 0, 134,
272, 542, 1,080, or 2,101 ppm (0, 406, 824,
1,643, 3,273 or 6,368 mg/m3) (dynamic
whole body chamber)
6 hr/d, 5 d/wk
13 weeks
Generation method (Sonimist Ultrasonic
spray nozzle nebulizer), analytical
concentration and method were reported
Results
Dose Absolute Relative Dose
(mg/kg-d) weight weight (mg/kg-d
0 000
90 +2 +7 180
200 +8 +11 330
420 +1 +14* 650
Absolute
weight
0
-14*
-3
-6
Relative
weight
0
-8
-1
+9*
Only animals sacrificed at 15 months were evaluated for organ weights. Organ
weights were not measured in the 2-year mouse study
Percent change compared to control:
Males
Concentration Absolute Relative
(mg/m3) weight weight
0 00
406 -8 -8
824 -2 -1
1,643 +1 -1
3,273 +10 +7
6,368 +5 +5
Percent change compared to control:
Males
Concentration Absolute Relative
(mg/m3) weight weight
0 00
406 -1 0
824 +4 +9
1,643 +7 +5
3,273 -8 -2
6,368 +5 +7
Females
Absolute
weight
0
0
0
+3
+9
+4
Females
Absolute
weight
0
+1
+1
+5
+2
+8
Relative
weight
0
+3
0
+2
+9*
+8*
Relative
weight
0
-4
+5
+1
+9*
+21*
1
2
3
4
5
6
aThe high-dose group had an increase in mortality.
* Statistically significant p < 0.05 as determined by study authors.
Conversions from drinking water concentrations to mg/kg-d performed by study authors.
Conversion from ppm to mg/m3 is 1 ppm = 3.031 mg/m3.
Percentage change compared to control = (treated value - control value) 4- control value x 100.
This document is a draft for review purposes only and does not constitute Agency policy.
B-24 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
Table B-5. Changes in liver histopathology in animals following exposure to
tert-butanol
Reference and study design
Results
Acharyaetal. (1997)
Acharvaetal. (1995)
Wistar rat; 5-6 males/treatment
Drinking water (0, 0.5%), 0, 575 mg/kg-d
10 weeks
T" liver glycogen (~ 7 fold)*
^incidence of centrilobular necrosis, vacuolation of hepatocytes, loss of
hepatocyte architecture, peripheral proliferation, and lymphocyte
infiltration (incidences and results of statistical tests not reported)
NTP (1995)
F344/N rat; 10/sex/treatment
Drinking water (0, 2.5, 5,10, 20, or 40 mg/mL)
M: 0, 230, 490, 840, 1,520, 3,610a mg/kg-d
F: 0, 290, 590, 850, 1,560, 3,620a mg/kg-d
13 weeks
No treatment-related effects observed.
NTP (1995)
B6C3Fi mouse; 10/sex/treatment
Drinking water (0, 2.5, 5,10, 20, 40 mg/mL)
M: 0, 350, 640, 1,590, 3,940, 8,210a mg/kg-d
F: 0, 500, 820, 1,660, 6,430, ll,620a mg/kg-d
13 weeks
No treatment-related effects observed.
NTP (1995)
F344/N rat; 60/sex/treatment
(10/sex/treatment evaluated at 15 months)
Drinking water (0,1.25, 2.5, 5,10 mg/mL)
M: 0, 90, 200, or 420a mg/kg-d
F: 0,180, 330, or 650a mg/kg-d
2 years
No treatment-related effects observed.
NTP (1995)
B6C3Fi mouse; 60/sex/treatment
Drinking water (0, 5,10, 20 mg/mL)
M: 0, 540,1,040, or 2,070a mg/kg-d
F: 0, 510,1,020, or 2,110 mg/kg-d
2 years
Males
Dose
(mg/kg-d)
0
540
1,040
2,070
Incidence of fatty
change
Females
Dose
(mg/kg-d)
0
510
1,020
2,110
Incidence of fatty
change
NTP (1997)
F344/N rat; 10/sex/treatment
Inhalation analytical concentration: 0,134,
272, 542, 1,080, or 2,101 ppm (0, 406, 824,
1,643, 3,273 or 6,368 mg/m3) (dynamic whole
body chamber)
6 hr/d, 5 d/wk
13 weeks
Generation method (Sonimist Ultrasonic spray
nozzle nebulizer), analytical concentration and
method were reported
No treatment-related effects observed in the high dose group (only
treatment group with liver endpoints evaluated).
This document is a draft for review purposes only and does not constitute Agency policy.
B-25 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Reference and study design
Results
NTP (1997)
B6C3Fi mouse; 10/sex/treatment
Inhalation analytical concentration: 0,134,
272, 542, 1,080, or 2,101 ppm (0, 406, 824,
1,643, 3,273 or 6,368 mg/m3) (dynamic whole
body chamber)
6 hr/d, 5 d/wk
13 weeks
Generation method (Sonimist Ultrasonic spray
nozzle nebulizer), analytical concentration and
method were reported
Authors stated that there were no treatment-related microscopic changes,
but data were not provided.
4
aThe high-dose group had an increase in mortality.
* Statistically significant p < 0.05 as determined by study authors.
Conversions from drinking water concentrations to mg/kg-d performed by study authors.
Conversion from ppm to mg/m3 is 1 ppm = 3.031 mg/m3.
This document is a draft for review purposes only and does not constitute Agency policy.
B-26 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
Table B-6. Changes in urinary bladder histopathology in animals following
oral exposure to tert-butanol
Reference and study design
NTP (1995)
F344/N rat; 10/sex/treatment
Drinking water (0, 2.5, 5, 10, 20,
40 mg/mL)
M: 0, 230, 490, 840, 1,520, 3,610a
mg/kg-d
F: 0, 290, 590, 850, 1,560, 3,620a
mg/kg-d
13 weeks
NTP (1995)
B6C3Fi mouse; 10/sex/treatment
Drinking water (0, 2.5, 5, 10, 20,
40 mg/mL)
M: 0, 350, 640, 1,590, 3,940,
8,210a mg/kg-d
F: 0, 500, 820, 1,660, 6,430,
ll,620a mg/kg-d
13 weeks
NTP (1995)
F344/N rat; 60/sex/treatment
(10/sex/treatment evaluated at
15 months)
Drinking water (0, 1.25, 2.5, 5, or
10 mg/mL)
M: 0, 90, 200, 420a mg/kg-d
F: 0, 180, 330, 650a mg/kg-d
2 years
Results
Incidence (severity):
Males
Transitional
epithelial
Dose (mg/kg-d) hyperplasia
0 0/10
230 not evaluated
490 not evaluated
840 0/10
1,520 1/10 (3.0)
3,610 7/10* (2.9)
Severity: 1 = minimal, 2 = mild, 3 = moderate,
Incidence (severity):
Males
Transitional
Dose epithelial Inflam-
( mg/kg-d) hyperplasia mation
0 0/10 0/10
350 not evaluated
640 not evaluated
1,590 0/10 0/10
3,940 6/10* (1.3) 6/10* (1.3)
8,210 10/10* (2.0) 10/10*
(2.3)
Severity: 1 = minimal, 2 = mild, 3 = moderate,
Females
Dose (mg/kg-
d)
0
290
590
850
1,560
3,620
4 = marked
Females
Transitional epithelial
hyperplasia
0/10
not evaluated
not evaluated
not evaluated
0/10
3/10 (2.0)
Transitional
Dose
(mg/kg-d)
0
500
820
1,660
6,430
11,620
4 = marked
epithelial Inflam-
hyperplasia mation
0/10 0/10
0/10 0/10
not evaluated
not evaluated
0/10 0/10
3/9(2.0) 6/9* (1.2)
No treatment-related effects observed
This document is a draft for review purposes only and does not constitute Agency policy.
B-27 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Reference and
NTP (1995)
study design
B6C3Fi mouse; 60/sex/treatment
Drinking water (0, 5, 10, or 20
mg/mL)
M: 0, 540, 1,040, 2,070a mg/kg-d
F: 0, 510, 1,020, 2,
2 years
110 mg/kg-d
Results
Incidence (severity):
Males
Dose
(mg/kg-d)
0
540
1,040
2,070
Severity: 1 =
Transitional
epithelial
hyperplasia
1/59 (2.0)
3/59 (1.7)
1/58 (1.0)
17/59*
(1.8)
minimal, 2 = mild,
Inflam-
mation
0/59
3/59 (1.7)
1/58 (1.0)
37/59* (2.0)
3 = moderate, 4
Females
Dose
(mg/kg-d)
0
510
1,020
2,110
= marked
Transitional
epithelial
hyperplasia
0/59
0/60
0/59
3/57(1.0)
Inflam-
mation
0/59
0/60
0/59
4/57*
(2.0)
1 aThe high-dose group had an increase in mortality.
2 * Statistically significant p < 0.05 as determined by study authors.
3 Conversions from drinking water concentrations to mg/kg-d performed by study authors.
This document is a draft for review purposes only and does not constitute Agency policy.
B-28 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
• = exposures at which the endpoint was reported statistically significant by study authors
D = exposures at which the endpoint was reported not statistically significant by study authors
x = exposures at which all animals died and were unable to be examined for the endpoint
URINARY Transitionalepitheliuin hyperplasia; M rat(C)
BLADDER Transitional epithelium hyperplasia; Frat(C) •
EFFECTS
Subchronic
Transitional epithelium hyperplasia; M mouse (Cj
Inflammation; M mouse (C)
Chronic Inflammation; Fmouse(C)
Transitional epithelium liyperplasia; F mouse (C)
LIVER Increased glycogen; M rat(A)
EFFECTS Relative weight; M rat(A)
frt
Subchronic
Relative weight M rat(C)
Relative weight Frat(C)
Absolute weight, M rat(C)
Absolute weight Frat(C) -
t-'atty tissue; M mouse (C)
Chronic pat|y t(gs(|e. p mouse ^
0
G
D
B— E
D— E
D— E
D— E
•
D
Dm •
• M M
1 — B— •
• D D
1 — B — D
ODD
Q-E
D— £
-B •
D — 0
DM H
3— •
3— •
3— •
3— a
• V
BI—I
Bm m
3— •
l-j
10
100
3
4
1,000 10,000 100,000
Dose(mg/kg-day)
Sources: (A) (Acharya etal. (1997); Acharya et al. (1995)); (B) Lyondell Chemical Co. (2004); (C) NTP (1995)
Figure B-6. Exposure-response array of other effects following oral exposure
to tert-butanol.
This document is a draft for review purposes only and does not constitute Agency policy.
B-29 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
• = exposures at which the endpoint was reported statistically significant by study authors
D = exposures at which the endpoint was reported not statistically significant by study authors
2
3
4
LIVER EFFECTS
Absolute liver weight; M rat (A)
Relative liver weight; M rat (A)
Absolute liver weight; F rat (A)
Relative liver weight; F rat (A)
Absolute liver weight; M mouse (A)
Relative liver weight; M mouse (A)
Absolute liver weight; F mouse (A)
Relative liver weight; F mouse (A)
Liver histopathology; M rat (A)
Liver histopathology; Frat(A)
Liver histopathology; M mouse (A]
Liver histopathology; F mouse (A)
Source: (A) NTP (1997)
100 1,000
Concentration (mg/m;i)
10,000
Figure B-7. Exposure-response array of other effects following inhalation
exposure to tert-butanol.
This document is a draft for review purposes only and does not constitute Agency policy.
B-30 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 B.3.2. Genotoxicity
2 The genotoxic potential of tert-butanol has been studied using a variety of genotoxicity
3 assays, including bacterial reverse mutation assays, gene mutation assays, chromosomal
4 aberrations, sister chromatid exchanges, micronucleus formation, and deoxyribonucleic acid (DNA)
5 strand breaks and adducts. The available genotoxicity data for tert-butanol are discussed below,
6 and the data summary is provided in Table B-7.
7 B.3.2.1. Bacterial Systems
8 The mutagenic potential of tert-butanol has been tested by Zeiger etal. [1987] using
9 different Salmonella typhimurium strains both in the presence and absence of S9 metabolic
10 activation. The preincubation assay protocol was followed. Salmonella strains TA98, TA100,
11 TA1535, TA1537, andTA1538 were exposed to five concentrations (100, 333,1,000, 3,333, or
12 10,000 ug/plate] and tested in triplicate. No mutations were observed in any of the strains tested,
13 in either the presence or absence of S9 metabolic activation.
14 Conflicting results have been obtained with tert-butanol-induced mutagenicity in strain
15 Salmonella strain TA102, a strain that is sensitive to damage at A-T sites inducible by oxidants and
16 other mutagens and is excision-repair proficient In a study by Williams-Hill et al. [1999],
17 tert-butanol induced an increase in the number of revertants in the first three concentrations with
18 S9 activation in a dose-response manner. The number of revertants decreased in the last two
19 concentrations. No discussion was provided on why the revertants decreased at higher
20 concentrations. The results of this study indicated that test strain TA102 might be a more sensitive
21 strain for monitoring tert-butanol levels [Williams-Hill etal.. 1999]. In another study by Mcgregor
22 etal. [2005]. however, experiments were conducted on TA102 in two different laboratories using
23 similar protocols. tert-Butanol was dissolved in dimethyl sulfoxide [DMSO] or distilled water and
24 tested in both the presence and absence of S9 metabolic activation. No statistically significant
25 increase in mutants was observed in either solvent medium. In one experiment where tert-butanol
26 was dissolved in water, a significant, dose-related increase in the number of revertants occurred,
27 reaching almost twice the control value at a concentration of 2,250 ug/plate. Of note is that DMSO is
28 known to be a free radical scavenger, and its presence at high concentrations might mask a
29 mutagenic response caused by oxidative damage.
30 Mutagenicity of tert-butanol has been studied in other systems including Neurospora crassa
31 and Saccharomyces cerevisiae. Yeast strain Neurospora crassa atthe ad-3Alocus (allele 38701] was
32 used to test the mutagenic activity of tert-butanol at a concentration of 1.75 mol/L for 30 minutes.
33 tert-Butanol did not induce reverse mutations in the tested strain at the exposed concentration
34 [Dickey etal., 1949]. tert-butanol without exogenous metabolic activation, however, significantly
35 increased the frequency of petite mutations (the mitochondrial DNA deletion rho-] in
36 Saccharomyces cerevisiae laboratory strains K5-A5, MMY1, D517-4B, and DS8 [Timenez etal.. 1988].
37 This effect on mitochondrial DNA, also observed with ethanol and other solvents, was attributed by
3 8 the study authors to the alteration in the lipid composition of mitochondrial membranes, and
This document is a draft for review purposes only and does not constitute Agency policy.
B-31 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 mitochondrial DNA's close association could be affected by membrane composition (Jimenez etal.,
2 1988).
3 B.3.2.2. In Vitro Mammalian Studies
4 To understand the role of tert-butanol-induced genotoxicity in mammalian systems, in vitro
5 studies have been conducted in different test systems and assays. tert-Butanol was tested to
6 evaluate its ability to induce forward mutations at the thymidine kinase locus (tk) in the L5178Y
7 tk+/- mouse lymphoma cells using forward mutation assay. Experiments were conducted in both
8 the presence and absence of S9 metabolic activation. The mutant frequency was calculated using
9 the ratio of mutant clones per plate/total clones per plate x 200. tert-Butanol did not reliably
10 increase the frequency of forward mutations in L5178Y tk+/- mouse lymphoma cells with or
11 without metabolic activation, although one experiment without addition of S9 yielded a small
12 increase in mutant fraction at the highest tested concentration (5,000 |ig/mL] [McGregor etal.,
13 19881.
14 To further determine potential DNA or chromosomal damage induced by tert-butanol in in
15 vitro systems, NTP [1995] studied sister chromatid exchanges and chromosomal aberrations.
16 Chinese hamster ovary (CHO) cells were exposed to tert-butanol in both the presence and absence
17 of S9 activation at concentrations of 160-5,000 [ig/mL for 26 hours. tert-Butanol did not induce
18 sister chromatid exchanges in any concentration tested, although in one experiment, percent
19 relative change of sister chromatid exchanges per chromosome scored slightly increased. The same
20 authors also studied the effect of tert-butanol on chromosomal aberration formation. CHO cells
21 were exposed to four concentrations (160, 500,1,600, or 5,000 [ig/mL] of tert-butanol in both the
22 presence and absence of S9. No significant increase in chromosomal aberration was observed in
23 any concentration tested. Of note is that, due to severe toxicity at the highest concentration
24 (5,000 |ig/mL], only 13 metaphase cells were scored instead of 100 in the chromosomal aberration
25 assay.
26 Sgambato etal. [2009] examined the effects of tert-butanol on DNA damage using a normal
27 diploid rat fibroblast cell line. Cells were treated with 0- to 100-mM tert-butanol for 48 hours to
28 determine the half-maximal inhibitory concentration (ICso; 0.44 ± 0.2 mM]. The 48-hour ICso
29 concentration then was used to determine DNA content, cell number, and phases of the cell cycle
30 after 24 and 48 hours of exposure. Total protein and DNA oxidative damage also were measured. A
31 comet assay was used to evaluate DNA fragmentation at time 0 and after 30 minutes, 4 hours, or 12
32 hours of exposure to the ICso concentration. tert-Butanol inhibited cell division in a dose-dependent
33 manner as measured by the number of cells after 24 and 48 hours of exposure at ICso
34 concentrations, and with concentrations at l/10th the ICso. Cell death did not increase, suggesting a
35 reduction in cell number due to reduced replication rather than to cytotoxicity. tert-Butanol caused
36 an accumulation in the G0/Gi phase of replication, related to different effects on the expression of
37 the cyclin Dl, p27Kipl, and p53 genes. An initial increase in DNA damage as measured by nuclear
3 8 fragmentation was observed at 3 0 minutes. The DNA damage declined drastically after 4 hours and
This document is a draft for review purposes only and does not constitute Agency policy.
B-32 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 disappeared almost entirely after 12 hours of exposure to tert-butanol. This reduction in the extent
2 of DNA fragmentation after the initial increase is likely the result of an efficient DNA repair
3 mechanism activated by cells following DNA damage induced by tert-butanol.
4 DNA damage caused by tert-butanol was determined by single-cell gel electrophoresis
5 (comet assay) in human promyelocytic leukemia (HL-60) cells. The cells were exposed to
6 concentrations ranging from 1 to 30 mmol/L for 1 hour, and 100 cells were evaluated for DNA
7 fragmentation. A dose-dependent increase in DNA damage was observed between 1 and
8 30 mmol/L. No cytotoxicity was observed at the concentrations tested [Tangetal.. 1997].
9 B.3.2.3. In Vivo Mammalian Studies
10 Few in vivo studies are available to understand the role of tert-butanol on genotoxicity. The
11 National Toxicology Program studied the effect of tert-butanol in a 13-week toxicity study [NTP,
12 1995]. Peripheral blood samples were obtained from male and female B6CF1 mice exposed to tert-
13 butanol in drinking water at doses of 3,000-40,000 ppm. Slides were prepared to determine the
14 frequency of micronuclei in 10,000 normochromatic erythrocytes. In addition, the percentage of
15 polychromatic erythrocytes among the total erythrocyte population was determined. No increase in
16 micronucleus formation in peripheral blood lymphocytes was observed either in male or female
17 B6C3Fi mice exposed for 13 weeks to tert-butanol in drinking water at concentrations as high as
18 40,000 ppm (2,110 mg/kg-day] (NTP. 1995].
19 Male Kumming mice (8 per treatment] were administered 0, 0.099, 0.99,10,101, or
20 997 [ig/kg BW14C-tert-butanol in saline via gavage with specific activity ranging from 1.60 to
21 0.00978 mCi/mol (Yuan etal.. 2007]. Animals were sacrificed 6 hours after exposure, and liver,
22 kidney, and lung were collected. Tissues were prepared for DNA isolation with samples from the
23 same organs from every two mice combined. DNA adducts were measured using accelerated mass
24 spectrometry. The results of this study showed a dose-response increase in DNA adducts in all
25 three organs measured, although the methodology used to detect DNA adducts is considered
26 sensitive but could be nonspecific. The authors stated that tert-butanol was found, for the first time,
27 to form DNA adducts in mouse liver, lung, and kidney. Because this is a single and first-time study,
28 further validation of this study will provide certainty in understanding the mechanism of tert-
29 butanol-induced DNA adducts.
This document is a draft for review purposes only and does not constitute Agency policy.
B-33 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
Table B-7. Summary of genotoxicity (both in vitro and in vivo) studies oftert-
butanol
Test system
Dose/ Cone.
Results3
Comments
Reference
Bacterial Systems
Reverse Mutation Assay
Salmonella typhimuhum
(TA98, TA100, TA1535,
TA1537, TA1538)
Reverse Mutation Assay
Salmonella typhimurium
(TA102)
Reverse Mutation Assay
Salmonella typhimurium
(TA98, TA100, TA102,
TA1535, TA1537)
Reverse mutation
Neurospora crassa, ad-3A
locus (allele 38701)
Mitochondrial mutation
Saccharomyces cerevisiae
(K5-5A, MMY1, D517-4B,
and DS8)
100, 333, 1,000,
3,333, 10,000
u.g/plate
1,000-4,000
Mg/plate
5, 15, 50, 100, 150,
200, 500, 1,000,
1,500, 2,500, 5,000
u.g/plate
1.75mol/L
4.0% (vol/vol)
-S9
ND
+b
+S9
+
ND
Preincubation procedure was
followed. This study was part of
the NTP 1995 testing results.
Only tested with S9 activation
Experiments conducted in two
different laboratories, two
vehicles - distilled water and
DMSO were used, different
concentrations were used in
experiments from different
laboratories
Eighty four percent cell death
was observed; note it is a 1949
study
Mitochondrial mutations,
membrane solvent
Zeiger et al.
(1987);NTP
(1995)
Williams-Hill et
al. (1999)
Mcgregor et al.
(2005)
Dickey et al.
(1949)
Jimenez et al.
(1988)
In vitro Systems
Gene Mutation Assay,
Mouse lymphoma cells
L5178YTK+/-
Sister-chromatid exchange,
Chinese Hamster Ovary cells
Chromosomal Aberrations,
Chinese Hamster Ovary cells
DNA damage (comet assay),
Rat fibroblasts
625, 1,000, 1,250,
2,000, 3,000,
4,000, 5,000 u.g/mL
160, 500, 1,600,
2,000, 3,000,
4,000, 5,000 u.g/mL
160, 500, 1,600,
2,000, 3,000,
4,000, 5,000 u.g/mL
0.44mmol/L(IC5o)
-
+c
-
ND
Cultures were exposed for 4 h,
then cultured for 2 days before
plating in soft agar with or
without trifluorothymidine,
3 u.g/mL; this study was part of
the NTP 1995 testing results
This study was part of the NTP
1995 testing results
This study was part of the NTP
1995 testing results
Exposure duration - 30 min,
4 h, 12 h; this study provides
other information on effect of
cell cycle control genes and
mechanism of action for TBA
McGregor et al.
(1988);NTP
(1995)
Galloway,
1987; NTP
(1995)
Galloway,
1987 NTP
(1995)
Sgambato et al.
(2009)
This document is a draft for review purposes only and does not constitute Agency policy.
B-34 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Test system
DNA damage, (comet
assay), HL-60 leukemia cells
Dose/ Cone.
1, 5, 10, 30 mmol/L
Results3
+
ND
Comments
Exposure duration - Ih
Reference
Tang et al.
(1997)
In vivo Animal Studies
Micronucleus formation,
B6C3F1 mouse peripheral
blood cells
DNAadducts, male
Kunming mouse liver,
kidney and lung cells
3,000, 5,000,
10,000, 20,000,
40,000 ppm
0. 1-1,000 ug/kg
body weight
+
13-week, subchronic, drinking
water study
Gavage, 6-h exposure, DNA
adduct determined by
accelerator mass spectrometry
NTP (1995)
Yuan et al.
(2007)
1 a+ = positive; - = negative; ND = not determined.
2 bEffect is predicted to be due to mitochondrial membrane composition.
3 CDNA damage was completely reversed with increased exposure time.
4 B.3.3. Summary
5 tert-Butanol has been tested for its genotoxic potential using a variety of genotoxicity
6 assays. Bacterial assays that detect reverse mutations have been thought to predict carcinogenicity
7 with accuracy up to 80%. tert-Butanol did not induce mutations in most bacterial strains; however,
8 when tested in TA102, a strain that is sensitive to damage at A-T sites inducible by oxidants, an
9 increase in mutants was observed at low concentrations, although conflicting results were reported
10 in another study. Furthermore, the solvent (e.g., distilled water or DMSO) used in the genotoxicity
11 assay could influence results. In one experiment where tert-butanol was dissolved in distilled
12 water, a significant, dose-related increase in the number of mutants was observed, with the
13 maximum value reaching almost twice the control value. DMSO is known to be a radical scavenger,
14 and its presence in high concentrations might mask a mutagenic response modulated by oxidative
15 damage. Other species such as Neurospora crassa did not produce reverse mutations due to
16 exposure to tert-butanol.
17 tert-Butanol was tested in several human and animal in vitro mammalian systems for
18 genotoxicity (gene mutation, sister chromatid exchanges, chromosomal aberrations, and DNA
19 damage). No increase in gene mutations was observed in mouse lymphoma cells (L5178Y TK+/-).
20 These specific locus mutations in mammalian cells are used to demonstrate and quantify genetic
21 damage, thereby confirming or extending the data obtained in the more widely used bacterial cell
22 tests. Sister chromatid exchanges or chromosomal aberrations were not observed in CHO cells in
23 response to tert-butanol treatment. DNA damage was detected using comet assay, however, in both
24 rat fibroblasts and HL-60 leukemia cells, with either an increase in DNA fragmentation at the
25 beginning of the exposure or dose-dependent increase in DNA damage observed. An initial increase
26 in DNA damage was observed at 30 minutes that declined drastically following 4 hours of exposure
27 and disappeared almost entirely after 12 hours of exposure to tert-butanol. This reduction in the
28 extent of DNA fragmentation after an initial increase is likely the result of an efficient DNA repair
29 mechanism activated by cells following DNA damage induced by tert-butanol. A dose-dependent
This document is a draft for review purposes only and does not constitute Agency policy.
B-35 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 increase in DNA damage was observed in human cells tested; however, because the exposure
2 occurred for only 1 hour in this study, whether DNA-repair mechanisms would occur after a longer
3 period of observation cannot be discerned.
4 Limited in vivo animal studies have been conducted on DNA adduct and micro nucleus
5 formation. A dose-response increase in DNA adducts was observed in mouse liver, kidney, and lung
6 cells. The authors used accelerated mass spectrometry to detect DNA adducts, but the identity of
7 these adducts was not determined. The method uses 14C-labeled chemical for dosing, isolated DNA
8 is oxidized to carbon dioxide and reduced to filamentous graphite, and the ratios of 14C/12C are
9 measured. The ratio then is converted to DNA adducts based on nucleotide content of the DNA.
10 Confirmation of these data will further the understanding of the mechanism of tert-butanol-induced
11 DNA adducts. No increase in micronucleus formation was observed in mouse peripheral blood cells
12 in a 13-week drinking water study conducted by the National Toxicology Program.
13 Overall, a limited database is available for understanding the role of tert-butanol-induced
14 genotoxicity for mode of action and carcinogenicity. The database is limited in terms of either the
15 array of genotoxicity tests conducted or the number of studies within the same type of test In
16 addition, the results are either conflicting or inconsistent The test strains, solvents, or control for
17 volatility used in certain studies are variable and could influence results. Furthermore, in some
18 studies, the specificity of the methodology used has been challenged. Given the inconsistencies and
19 limitations of the database in terms of the methodology used, number of studies in the overall
20 database, coverage of studies across the genotoxicity battery, and the quality of the studies, the
21 weight of evidence analysis is inconclusive. The available data do not inform a definitive conclusion
22 on the genotoxicty of tert-butanol and thus the potential genotoxic effects of tert-butanol cannot be
23 discounted.
This document is a draft for review purposes only and does not constitute Agency policy.
B-36 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
i APPENDIX C. DOSE-RESPONSE MODELING FOR
2 THE DERIVATION OF REFERENCE VALUES FOR
3 EFFECTS OTHER THAN CANCER AND THE
4 DERIVATION OF CANCER RISK ESTIMATES
5 This appendix provides technical detail on dose-response evaluation and determination of
6 points of departure (PODs) for relevant endpoints. The endpoints were modeled using EPA's
7 Benchmark Dose Software (BMDS), version 2.1.2. The preambles for the cancer and noncancer
8 parts below describe the common practices used in evaluating the model fit and selecting the
9 appropriate model for determining the POD as outlined in the Benchmark Dose Technical Guidance
10 Document [U.S. EPA. 2000). In some cases, using alternative methods based on statistical judgment
11 might be appropriate; exceptions are noted as necessary in the summary of the modeling results.
12 C.I.I. Noncancer Endpoints
13 C.l.1.1. Data Sets
14 Data sets selected for dose-response modeling are provided in Table C-l. In all cases,
15 administered exposure was used in modeling the response data.
16 C.l.1.2. Model Fit
17 All models were fit to the data using the maximum likelihood method. The following
18 procedures were used, depending on whether data were dichotomous or continuous:
19 • For dichotomous models, the following parameter restrictions were applied: for log-logistic
20 model, restrict slope >1; for gamma and Weibull models, restrict power >1; for multistage
21 models, restrict beta values >0. Each model was tested for goodness-of-fit using a chi-
22 square goodness-of-fit test (x2 p-value < 0.10 indicates lack of fit). Other factors also were
23 used to assess model fit, such as scaled residuals, visual fit, and adequacy of fit in the low-
24 dose region and near the benchmark response (BMR).
25 • For continuous models, the following parameter restrictions were applied: for polynomial
26 models, restrict beta values >0; for Hill, power, and exponential models, restrict power >1.
27 Model fit was assessed by a series of tests. For each model, first the homogeneity of the
28 variances was tested using a likelihood ratio test (BMDS Test 2). If Test 2 was not rejected
29 (x2 p-value > 0.10), the model was fit to the data assuming constant variance. If Test 2 was
30 rejected (x2 p-value < 0.10), the variance was modeled as a power function of the mean, and
31 the variance model was tested for adequacy of fit using a likelihood ratio test (BMDS
32 Test 3). For fitting models using either constant variance or modeled variance, models for
This document is a draft for review purposes only and does not constitute Agency policy.
C-l DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 the mean response were tested for adequacy of fit using a likelihood ratio test (BMDS Test
2 4, with x2 p-value < 0.10 indicating inadequate fit). Other factors also were used to assess
3 the model fit, such as scaled residuals, visual fit, and adequacy of fit in the low-dose region
4 and near the BMR.
5 C.l.1.3. Model Selection
6 For each endpoint, the BMDL estimate (95% lower confidence limit on the BMD, as
7 estimated by the profile likelihood method) and the Akaike's information criterion (AIC) value were
8 used to select a best-fit model from among the models exhibiting adequate fit If the BMDL
9 estimates were "sufficiently close," that is, differed by no more than three-fold, the model selected
10 was the one that yielded the lowest AIC value. If the BMDL estimates were not sufficiently close, the
11 lowest BMDL was selected as the POD.
12 Table C-l. Noncancer endpoints selected for dose-response modeling for
13 tert-butanol
Endpoint/Study
Kidney transitional
epithelial hyperplasia
NTP (1995)
Kidney transitional
epithelial hyperplasia
NTP (1995)
Increased absolute
kidney weight
NTP (1995)
Increased absolute
kidney weight
NTP (1995)
Kidney inflammation
NTP (1995)
Increased absolute
kidney weight
NTP (1997)
Increased absolute
kidney weight
Species/
Sex
Rat (F344)/Male
Rat
(F344)/Female
Rat (F344)/Male
Rat
(F344)/Female
Rat
(F344)/Female
Rat (F344)/Male
Rat
(F344)/Female
Doses and effect data
Dose (mg/kg-d)
Incidence/Total
Dose (mg/kg-d)
Incidence/Total
Dose (mg/kg-d)
Mean±SD(n)
Dose (mg/kg-d)
Mean ±SD (n)
Dose (mg/kg-d)
Incidence/Total
Concentration
(mg/m3)
Mean ±SD (n)
Concentration
(mg/m3)
0
25/50
0
0/50
0
1.78 ±0.18
(10)
0
1.07 ± 0.09
(10)
0
2/50
90
32/50
180
0/50
90
1.85 ±0.17
(10)
180
1.16 ±0.10
(10)
180
3/50
0 406 825
1.21 ± 1.21 ± 1.18 ±
0.082 0.096 0.079
(10) (9) (10)
0 406 825
200
36/50
330
3/50
200
1.99 ±0.18
(10)
330
1.27 ±0.08
(10)
330
13/50
1,643
1.25 ±
0.111
(10)
1,643
420
40/50
650
17/50
420
1.9 ±0.23
(10)
650
1.31 ±0.09
(10)
650
17/50
3,274 6,369
1.34 ± 1.32 ±
0.054 0.089
(10) (10)
3,274 6,369
This document is a draft for review purposes only and does not constitute Agency policy.
C-2 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Endpoint/Study
NTP (1997)
Species/
Sex
Doses and effect data
Mean ±SD (n)
0.817 ±
0.136
(10)
0.782 ±
0.063
(10)
0.821 ±
0.061
(10)
0.853 ±
0.045
(10)
0.831 ±
0.054
(10)
0.849 ±
0.038
(10)
4
5
C.l.1.4. Modeling Results
Below are tables summarizing the modeling results for the noncancer endpoints modeled.
Table C-2. Summary of BMD modeling results for kidney transitional epithelial
hyperplasia in male F344 rats exposed to tert-butanol in drinking water for 2
years fNTP. 19951: BMR = 10% extra risk
Model"
Log-logistic
Gamma
Logistic
Log-probit
Multistage, 3°
Probit
Weibull
Dichotomous-Hill
Goodness of fit
p-value
0.976
0.784
0.661
0.539
0.784
0.633
0.784
0.968
AIC
248.0
248.5
248.8
249.2
248.5
248.9
248.5
250.0
BMDio
(mg/kg-d)
30
46
58
84
46
60
46
25
BMDLio
(mg/kg-d)
16
29
41
53
29
43
29
15
Basis for model selection
Log-logistic model selected as best-
fitting model based on lowest AIC
with all BMDL values sufficiently
close (BMDLs differed by slightly
more than 3-fold).
6
7
a Scaled residuals for selected model for doses 0, 90, 200, and 420 mg/kg-d were -0.076, 0.147, 0.046, and -0.137,
respectively.
This document is a draft for review purposes only and does not constitute Agency policy.
C-3 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
0.9
0.8
0.7
0.6
0.5
0.4
BMDL 3MD
Log-Logistic Model with 0.95 Confidence Level
Log-Logistic
0
50
100
150
17:1605/132011
200 250
dose
300
350
400
2
3
4
5
6
1
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
Figure C-l. Plot of incidence by dose, with fitted curve for LogLogistic model
for kidney transitional epithelial hyperplasia in male F344 rats exposed to
tert-butanol in drinking water for 2 years (NTP. 1995); BMR = 10% extra risk;
dose shown in mg/kg-d
Logistic Model. (Version: 2.13; Date: 10/28/2009)
Input Data File: M:\NCEA t-Butanol\BMD modeling\BMDS Output\17 NTP 1995b_Kidney
transitional epithelial hyperplasia, male rats_LogLogistic_10.(d)
Gnuplot Plotting File: M:\NCEA~t-Butanol\BMD modeling\BMDS Output\17 NTP
1995b Kidney transitional epithelial hyperplasia, male rats LogLogistic 10.pit
Fri May 13 17:16:25 2011
This document is a draft for review purposes only and does not constitute Agency policy.
C-4 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4
5
6
7
8 and do not appear in the correlation matrix )
10 background intercept
12 background 1 -0.71
14 intercept -0.71 1
16
17
18
19
20 95.0% Wald Confidence Interval
21 Variable Estimate Std. Err. Lower Conf. Limit Upper Conf. Limit
22 background
23 intercept
24 slope
25
26 * - Indicates that this value is not calculated.
27
28
29
30
31
32
33
34 Fitted model -122.02 2
35 Reduced model -127.533 1
36
37 AIC: 248.04
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52 Benchmark Dose Computation
53
54 Specified effect = 0.1
56 Risk Type = Extra risk
57
58 Confidence level = 0.95
59
60 BMD = 29.6967
61
62 BMDL = 15. 6252
63
This document is a draft for review purposes only and does not constitute Agency policy.
C-5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Supplemental Information—tert-Butyl Alcohol
Table C-3. Summary of BMD modeling results for kidney transitional epithelial
hyperplasia in female F344 rats exposed to tert-butanol in drinking water for
2 years (NTP. 1995): BMR = 10% extra risk
Model"
Gamma
Logistic
LogLogistic
LogProbit
Multistage 3°
Probit
Weibull
Dichotomous-Hill
Goodness of fit
p-value
0.83
0.50
0.79
0.89
0.92
0.62
0.76
N/Ab
AIC
91.41
92.81
91.57
91.19
89.73
92.20
91.67
117.89
BMDio
(mg/kg-d)
409
461
414
400
412
439
421
Error0
BMDLio
(mg/kg-d)
334
393
333
327
339
372
337
Error0
Basis for model selection
Multistage 3rd-order model
selected as best-fitting model
based on lowest AIC with all BMDL
values sufficiently close (BMDLs
differed by less than 3-fold).
aScaled residuals for selected model for doses 0,180, 330, and 650 mg/m3 were 0.0, -0.664, 0.230, and 0.016,
respectively.
bNo available degrees of freedom to estimate a p-value.
°BMD and BMDL computation failed for the Dichotomous-Hill model.
Multistage Model with O.95 Confidence Level
5
6
7
9
10
17:18 OS/13 2O11
Multistage
3OO
dose
Figure C-2. Plot of incidence by dose, with fitted curve for Multistage 3° model
for kidney transitional epithelial hyperplasia in female F344 rats exposed to
tert-butanol in drinking water for 2 years (NTP. 1995): BMR = 10% extra risk;
dose shown in mg/kg-d
This document is a draft for review purposes only and does not constitute Agency policy.
C-6 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 Input Data File: M:\NCEA te/"t-blltanol\BMD modeling\BMDS Output\20 NTP
2 1995b_Kidney transitional epithelial hyperplasia, female rats_Multi3_10.(d)
3 Gnuplot Plotting File: M:\NCEA te/"t-blltanol\BMD modeling\BMDS Output\20 NTP
4 1995b_Kidney transitional epithelial hyperplasia, female rats_Multi3_10.pit
5 ~ Mon May ^9 18:31:33 2011
6 ====================================================================
7
9
10
11 The form of the probability function is:
12
13
14
15
16 The parameter betas are restricted to be positive
17
18
19
20
21
22 Total number of observations = 4
23 Total number of records with missing values = 0
24 Total number of parameters in model = 4
25 Total number of specified parameters = 0
26 Degree of polynomial = 3
27
28
29
30
31
32
33
34
35 Default Initial Parameter Values
36 Background = 0
37 Beta(l) = 0
38 Beta(2) = 1.51408e-007
39 Beta(3) = 1.29813e-009
40
41
42
43
44
45
46
47
48 Beta (3)
49
50 Beta (3) 1
51
52
53
54 Parameter Estimates
55
56 95.0% Wald Confidence Interval
57 Variable Estimate Std. Err. Lower Conf. Limit Upper Conf. Limit
59
60
61
62
63 * - Indicates that this value is not calculated.
64
65
66
67
68
69 Model Log(likelihood) # Param's Deviance Test d.f. P-value
70 Full model -43.4002 4
This document is a draft for review purposes only and does not constitute Agency policy.
C-7 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 Fitted model -43.8652 1
2 Reduced model -65.0166 1
3
4 AIC: 89.7304
5
6
7
9 Dose Est. Prob. Expected Observed Size Residual
10
11
12
13
14
15
16 ChiA2 = 0.49 d.f. = 3
17
18
19 Benchmark Dose Computation
20
21 Specified effect = 0.1
22
23 Risk Type = Extra risk
24
25 Confidence level = 0.95
26
27 BMD = 411. 95
28
29 BMDL = 338. 618
30
31 BMDU = 469.73
32
33 Taken together, (338.618, 469.73 ) is a 90 % two-sided confidence
34
35
36
This document is a draft for review purposes only and does not constitute Agency policy.
C-8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Supplemental Information—tert-Butyl Alcohol
Table C-4. Summary of BMD modeling results for absolute kidney weight in
male F344 rats exposed to tert-butanol in drinking water for 15 months (NTP.
1995): BMR = 10% rel. dev. from control mean
Model"
Exponential (M2)b
Exponential (M3)c
Exponential (M4)
Exponential (M5)
Hill
Power'
Polynomial 3°B
Polynomial 2°h
Linear
Goodness of fit
p-value
0.123
0.123
0.167
N/Ae
0.301
0.126
AIC
-86.757
-86.757
-87.041
-85.880
-87.880
-86.804
BMDioRD
(mg/kg-d)
661
661
errord
errord
errord
657
BMDLioRD
(mg/kg-d)
307
307
0
0
errord
296
Basis for model selection
Of the models that provided an
adequate fit and a valid BMDL
estimate, the linear model was
selected based on lowest AIC.
a Constant variance case presented (BMDS Test 2 p-value = 0.777), selected model in bold; scaled residuals for
selected model for doses 0, 90, 200, and 420 mg/kg-d were -0.78, -0.11,1.65, -0.76, respectively.
b The Exponential (M2) model can appear equivalent to the Exponential (M3) model, however differences exist in
digits not displayed in the table.
c The Exponential (M3) model can appear equivalent to the Exponential (M2) model, however differences exist in
digits not displayed in the table.
d BMD or BMDL computation failed for this model.
e No available degrees of freedom to calculate a goodness-of-fit value.
f For the Power model, the power parameter estimate was 1. The models in this row reduced to the Linear
model.
8 For the Polynomial 3° model, the b3 coefficient estimate was 0 (boundary of parameters space). The models in
this row reduced to the Polynomial 2° model. For the Polynomial 3° model, the b3 and b2 coefficient estimates
were 0 (boundary of parameters space). The models in this row reduced to the Linear model.
h For the Polynomial 2° model, the b2 coefficient estimate was 0 (boundary of parameters space). The models in
this row reduced to the Linear model.
This document is a draft for review purposes only and does not constitute Agency policy.
C-9 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Linear Model, with BMP of 0.1 Pel. Dev. for the BMD and 0.95 Lower Confidence Limit for the BMDL
11:4605/262015
4
5
6
1
Figure C-3. Plot of mean response by dose, with fitted curve for Linear model
with constant variance for absolute kidney weight in male F344 rats exposed
to tert-butanol in drinking water for 15 months (NTP. 1995): BMR = 10% rel.
dev. from control mean; dose shown in mg/kg-d
Polynomial Model. (Version: 2.20; Date: 10/22/2014)
The form of the response function is: Y[dose] = beta_0 + beta_l*dose.
A constant variance model is fit
9
10
11
12
Benchmark Dose Computation.
BMR = 10% Relative deviation
BMD = 656.583
BMDL at the 95% confidence level = 295.826
13
14
Parameter Estimates
Variable
alpha
rho
beta_0
beta_l
Estimate
0.0361494
n/a
1.83173
0.000278979
Default Initial
Parameter Values
0.0362125
0
1.83173
0.000278979
This document is a draft for review purposes only and does not constitute Agency policy.
C-10 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Table of Data and Estimated Values of Interest
Dose
0
90
200
420
N
10
10
10
10
Obs Mean
1.78
1.85
1.99
1.9
Est Mean
1.83
1.86
1.89
1.95
Obs Std Dev
0.18
0.17
0.18
0.23
Est Std Dev
0.19
0.19
0.19
0.19
Scaled Resid
-0.777
-0.114
1.65
-0.763
Likelihoods of Interest
Model
Al
A2
A3
fitted
R
Log(likelihood)
48.474229
49.025188
48.474229
46.401914
45.368971
# Param's
5
8
5
3
2
AIC
-86.948457
-82.050377
-86.948457
-86.803828
-86.737942
Tests of Interest
Test
Testl
Test 2
Tests
Test 4
-2*log(Likelihood
Ratio)
7.31243
1.10192
1.10192
4.14463
Test df
6
3
3
2
p-value
0.2929
0.7766
0.7766
0.1259
This document is a draft for review purposes only and does not constitute Agency policy.
C-ll DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Supplemental Information—tert-Butyl Alcohol
Table C-5. Summary of BMD modeling results for absolute kidney weight in
female F344 rats exposed to tert-butanol in drinking water for 15 months
(NTP. 1995): BMR = 10% rel. dev. from control mean
Model"
Exponential (M2)
Exponential (M3)b
Exponential (M4)
Exponential (M5)
Hill
Powerd
Polynomial 3°e
Polynomial 2°f
Linear
Goodness of fit
p-value
0.0594
0.176
N/AC
N/AC
0.0842
AIC
-144.00
-145.81
-145.65
-145.65
-144.70
BMDioRD
(mg/kg-d)
318
164
207
202
294
BMDLioRD
(mg/kg-d)
249
91.4
117
119
224
Basis for model selection
The Exponential (M4) model was
selected as the only model with
adequate fit.
a Constant variance case presented (BMDS Test 2 p-value = 0.852), selected model in bold; scaled residuals for
selected model for doses 0,180, 330, and 650 mg/kg-d were 0.21, -0.9, 0.94, -0.25, respectively.
b For the Exponential (M3) model, the estimate of d was 1 (boundary). The models in this row reduced to the
Exponential (M2) model.
c No available degrees of freedom to calculate a goodness-of-fit value.
d For the Power model, the power parameter estimate was 1. The models in this row reduced to the Linear
model.
e For the Polynomial 3° model, the b3 coefficient estimate was 0 (boundary of parameters space). The models in
this row reduced to the Polynomial 2° model. For the Polynomial 3° model, the b3 and b2 coefficient estimates
were 0 (boundary of parameters space). The models in this row reduced to the Linear model.
f For the Polynomial 2° model, the b2 coefficient estimate was 0 (boundary of parameters space). The models in
this row reduced to the Linear model.
This document is a draft for review purposes only and does not constitute Agency policy.
C-12 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Exponential 4 Model, with BMR of 0.1 Rel. Dev. for the BMD and 0.95 Lower Confidence Limit for the BMDL
Exponential 4
11:32 05/26 2015
2 Figure C-4. Plot of mean response by dose, with fitted curve for Exponential
3 (M4) model with constant variance for absolute kidney weight in female F344
4 rats exposed to tert-butanol in drinking water for 15 months (NTP. 1995);
5 BMR = 10% rel. dev. from control mean; dose shown in mg/kg-d
6 Exponential Model. (Version: 1.10; Date: 01/12/2015)
7 The form of the response function is: Y[dose] = a * [c-(c-l) * exp(-b * dose)].
8 A constant variance model is fit.
9 Benchmark Dose Computation.
10 BMR = 10% Relative deviation
11 BMD = 163.803
12 BMDL at the 95% confidence level = 91.3614
13
Parameter Estimates
Variable
Inalpha
rho
a
b
c
d
Estimate
-4.84526
n/a
1.06808
0.00258011
1.29013
n/a
Default Initial
Parameter Values
-4.89115
0
1.0203
0.00282085
1.35122
1
14
This document is a draft for review purposes only and does not constitute Agency policy.
C-13 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Table of Data and Estimated Values of Interest
Dose
0
180
330
650
N
10
10
10
10
Obs Mean
1.07
1.16
1.27
1.31
Est Mean
1.07
1.18
1.25
1.32
Obs Std Dev
0.09
0.1
0.08
0.09
Est Std Dev
0.09
0.09
0.09
0.09
Scaled Resid
0.2112
-0.8984
0.9379
-0.2507
Likelihoods of Interest
Model
Al
A2
A3
R
4
Log(likelihood)
77.82307
78.21688
77.82307
62.21809
76.90527
# Param's
5
8
5
2
4
AIC
-145.6461
-140.4338
-145.6461
-120.4362
-145.8105
Tests of Interest
Test
Testl
Test 2
Tests
Test 6a
-2*log(Likelihood
Ratio)
32
0.7876
0.7876
1.836
Test df
6
3
3
1
p-value
<0.0001
0.8524
0.8524
0.1755
This document is a draft for review purposes only and does not constitute Agency policy.
C-14 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
Table C-6. Summary of BMD modeling results for kidney inflammation in
female rats exposed to tert-butanol in drinking water for 2 years (NTP. 1995);
BMR = 10% extra risk
Model"
Gamma
Logistic
LogLogistic
LogProbit
Multistage 3°
Probit
Weibull
Dichotomous-Hill
Goodness of fit
p-value
0.084
0.082
0.092
0.243
0.072
0.108
0.081
N/Ab
AIC
169.9
169.7
169.8
167.6
170.3
169.2
170.0
169.5
BMD 10%
(mg/kg-d)
231
305
228
254
216
285
226
229
BMDLior.
(mg/kg-d)
135
252
124
200
132
235
134
186
Basis for model selection
LogProbit was selected on the
basis of the lowest AIC with all
BMDL values for fitting models
being sufficiently close (BMDLs
differed by less than 3-fold).
5
6
7
aSelected model in bold; scaled residuals for selected model for doses 0,180, 330, and 650 mg/kg-d were -0.067,
-0.700,1.347, and -0.724, respectively.
bNo available degrees of freedom to estimate a p-value.
LogProbit Model with O.95 Confidence Level
0.5
1
0.2
LogProbit
BMD
17:17 OS/13 2O1 1
3OO
dose
Figure C-5. Plot of incidence by dose, with fitted curve for Logprobit model for
kidney inflammation in female rats exposed to tert-butanol in drinking water
for 2 years (NTP. 1995): BMR = 10% extra risk; dose shown in mg/kg-d
This document is a draft for review purposes only and does not constitute Agency policy.
C-15 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
2 Probit Model. (Version: 3.2; Date: 10/28/2009)
3 Input Data File: M:/NCEA te/"t-blltanol/BMD modeling/BMDS Output/19 NTP
4 1995b_Kidney inflammation, female rats_LogProbit_10.(d)
5 Gnuplot Plotting File: M:/NCEA te/"t-blltanol/BMD modeling/BMDS Output/19 NTP
6 1995b_Kidney inflammation, female rats_LogProbit_10.pit
7 Fri May 13 17:17:59 2011
^
10 [notes]
12
13 The form of the probability function is:
14
15 P[response] = Background
16 + (1-Background) * CumNorm(Intercept+Slope*Log(Dose)) ,
17
18 where CumNorm(.) is the cumulative normal distribution function
19
20
21
22
23
24
25 Total number of observations = 4
26 Total number of records with missing values = 0
27 Maximum number of iterations = 250
28 Relative Function Convergence has been set to: le-008
29 Parameter Convergence has been set to: le-008
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48 background intercept
49
50 background 1 -0.51
51
52 intercept -0.51 1
53
54
55
56 Parameter Estimates
57
58 95.0% Wald Confidence Interval
59 Variable Estimate Std. Err. Lower Conf. Limit Upper Conf. Limit
60 background 0.0381743 0.0246892 -0.0102155 0.0865642
61 intercept -6.82025 0.161407 -7.1366 -6.5039
62 slope 1 NA
63
64
65
66
67
68
69
70
This document is a draft for review purposes only and does not constitute Agency policy.
C-16 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2 Model Log(likelihood) # Param's Deviance Test d.f. P-value
3 Full model
4 Fitted model
5 Reduced model
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
22 Benchmark Dose Computation
23
24 Specified effect = 0.1
25
26 Risk Type = Extra risk
27
28 Confidence level = 0.95
29
30 BMD = 254.347
31
32 BMDL = 199.789
33
34
35
This document is a draft for review purposes only and does not constitute Agency policy.
C-17 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Supplemental Information—tert-Butyl Alcohol
Table C-7. Summary of BMD modeling results for absolute kidney weight in
male F344 rats exposed to tert-butanol via inhalation for 6 hr/d, 5d/wk for 13
weeks (NTP. 1997); BMR = 10% relative deviation from the mean
Model"
Exponential (M2)
Exponential (M3)
Exponential (M4)
Exponential (M5)
Hill
Power0
Linear
Polynomial 5°d
Polynomial 4°e
Polynomial 3°
Polynomial 2°
Goodness of fit
p-value
<0.0001
<0.0001
<0.0001
<0.0001
0.763
0.0607
1.44E-04
1.44E-04
AIC
-205.06
-203.06
-203.06
-201.06
-226.82
-220.97
-207.06
-207.06
BMClORD
(mg/m3)
errorb
9.2E+07
errorb
errorb
1931
5364
-9999
-9999
BMCLioRD
(mg/m3)
errorb
7094
0
0
1705
3800
errorf
18436
Basis for model selection
Although the Hill model was the
only adequately fitting model
(p>0.1), the resulting fit was
essentially a step-function that
does not support interpolation
between the well-fit
observations.
a Constant variance case presented (BMDS Test 2 p-value = 0.390), selected model in bold; scaled residuals for
selected model for doses 0, 406, 825,1,643, 3,274, and 6,369 mg/m3 were 0.395, 0.374, -0.75, -1.96e-006,
0.381, and -0.381, respectively.
b BMC or BMCL computation failed for this model.
c For the Power model, the power parameter estimate was 1. The models in this row reduced to the Linear
model.
d For the Polynomial 5° model, the b5 and b4 coefficient estimates were 0 (boundary of parameters space). The
models in this row reduced to the Polynomial 3° model.
e For the Polynomial 4° model, the b4 coefficient estimate was 0 (boundary of parameters space). The models in
this row reduced to the Polynomial 3° model.
f BMC or BMCL computation failed for this model.
8-
1.2
1.15
Hill Model with O.95 Confidence Level
BMDL
BMD
1O:15 O4/3O 2O14
3OOO
dose
This document is a draft for review purposes only and does not constitute Agency policy.
C-18 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 Figure C-6. Plot of mean response by concentration, with fitted curve for Hill
2 model for absolute kidney weight in male F344 rats exposed to tert-butanol
3 via inhalation for 6 hr/d, 5d/wk for 13 weeks (NTP. 1997): BMR = 10%
4 relative deviation from the mean; concentration shown in mg/m3
5 Hill Model. (Version: 2.15; Date: 10/28/2009)
6 The form of the response function is: Y[dose] = intercept + v*doseAn/(kAn + doseAn).
7 A constant variance model is fit
8 Benchmark Dose Computation.
9 BMR = 10% Relative risk
10 BMD = 1931.35
11 BMDL at the 95% confidence level = 1704.82
12 Parameter Estimates
13
Variable
alpha
rho
intercept
V
n
k
Estimate
0.00687349
n/a
1.19966
0.130345
18
1685.82
Default Initial
Parameter Values
0.00750263
0
1.21
0.13
18
4451.94
This document is a draft for review purposes only and does not constitute Agency policy.
C-19 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Table of Data and Estimated Values of Interest
Dose
0
406
825
1643
3274
6369
N
10
9
10
10
10
10
Obs Mean
1.21
1.21
1.18
1.25
1.34
1.32
Est Mean
1.2
1.2
1.2
1.25
1.33
1.33
Obs Std Dev
0.0822
0.096
0.0791
0.111
0.0538
0.0885
Est Std Dev
0.0829
0.0829
0.0829
0.0829
0.0829
0.0829
Scaled Resid
0.395
0.374
-0.75
-0.00000196
0.381
-0.381
Likelihoods of Interest
Model
Al
A2
A3
fitted
R
Log(likelihood)
117.992549
120.600135
117.992549
117.41244
105.528775
# Param's
7
12
7
4
2
AIC
-221.985098
-217.20027
-221.985098
-226.82488
-207.05755
Tests of Interest
Test
Testl
Test 2
Tests
Test 4
-2*log(Likelihood
Ratio)
30.1427
5.21517
5.21517
1.16022
Test df
10
5
5
3
p-value
0.0008118
0.3902
0.3902
0.7626
This document is a draft for review purposes only and does not constitute Agency policy.
C-20 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Supplemental Information—tert-Butyl Alcohol
Table C-8. Summary of BMD modeling results for absolute kidney weight in
female F344 rats exposed to tert-butanol via inhalation for 6 hr/d, 5d/wk for
13 weeks (NTP. 1997); BMR = 10% relative deviation from the mean
Model"
Exponential (M2)
Exponential (M3)b
Exponential (M4)
Exponential (M5)
Hill
Power
Polynomial 3°d
Polynomial 2°e
Linear
Polynomial 5°
Polynomial 4°
Goodness of fit
p-value
0.0378
0.533
0.374
0.227
0.0392
0.0274
0.0274
0.0274
AIC
-261.52
-267.48
-265.71
-265.57
-261.61
-261.61
-261.61
-261.61
BMClORD
(mg/m3)
14500
error0
error0
error0
14673
14673
14673
14673
BMCLioRD
(mg/m3)
7713
0
0
error0
7678
7678
7569
7674
Basis for model selection
No model adequately fit the data.
a Modeled variance case presented (BMDS Test 2 p-value = 1.90E-04, BMDS Test 3 p-value = 0.374), no model
was selected as a best-fitting model.
b For the Exponential (M3) model, the estimate of d was 1 (boundary). The models in this row reduced to the
Exponential (M2) model.
0 BMC or BMCL computation failed for this model.
d For the Polynomial 3° model, the b3 coefficient estimate was 0 (boundary of parameters space). The models in
this row reduced to the Polynomial 2° model. For the Polynomial 3° model, the b3 and b2 coefficient estimates
were 0 (boundary of parameters space). The models in this row reduced to the Linear model.
e For the Polynomial 2° model, the b2 coefficient estimate was 0 (boundary of parameters space). The models in
this row reduced to the Linear model.
Note: Graphs of the better fitting models are provided for illustration.
This document is a draft for review purposes only and does not constitute Agency policy.
C-21 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
0.8
Hill Model
Hill
1O:32 O4/3O 2O14
3OOO
dose
2
3
4
5
Figure C-7. Plot of mean response by concentration, with fitted curve for Hill
model for absolute kidney weight in female F344 rats exposed to tert-butanol
via inhalation for 6 hr/d, 5d/wk for 13 weeks (NTP. 1997): BMR = 10%
relative deviation from the mean; concentration shown in mg/m3
0.8
Power Model with O.95 Confidence Level
Power
BMDL
BM13
1O:32 O4/3O 2O14
6OOO 8OOO 1OOOO 12OOO 14OOO
dose
7
8
9
10
11
Figure C-8. Plot of mean response by concentration, with fitted curve for
Power model for absolute kidney weight in female F344 rats exposed to tert-
butanol via inhalation for 6 hr/d, 5d/wk for 13 weeks (NTP. 1997): BMR =
10% relative deviation from the mean; concentration shown in mg/m3
This document is a draft for review purposes only and does not constitute Agency policy.
C-22 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 C.I.2. Cancer Endpoints
2 C.l.2.1. Data Sets
3 The cancer data sets selected for dose-response modeling are summarized in Table C-9. In
4 all cases, administered exposure was used in modeling the response data. Due to the significant
5 difference in survival in the high-dose male mice compared with the concurrent control, the Poly-3
6 procedure [Bailer and Portier. 1988] for adjusting tumor incidence rates for intercurrent mortality
7 was used. The procedure is based on the observation that the cumulative incidence of tumors tends
8 to increase with time raised to the second through the fourth powers for a large proportion of
9 cases. In the Poly-3 procedure, for a study of T weeks' duration, an animal that is removed from the
10 study after t weeks (t < T) without a specified type of tumor of interest is given a weight of (t/T)3.
11 An animal that survives until the terminal sacrifice at T weeks is assigned a weight of (T/T)3 = 1. An
12 animal that develops the specific type of tumor of interest obviously lived long enough to develop
13 the tumor, and is assigned a weight of 1. The Poly-3 tumor incidence, adjusted for intercurrent
14 mortality up to time T, is the number of animals in a dose group with the specified type of tumor
15 divided by the sum of the weights (the effective number of animals at risk). The tumor incidences,
16 adjusted using this procedure, also are provided in Table C-9.
17 C.l.2.2. Model Fit
18 The multistage model was fit to the cancer data sets. Model coefficients were restricted to
19 be non-negative (beta values > 0), to estimate a monotonically increasing function. Each model was
20 fit to the data using the maximum likelihood method, and was tested for goodness-of-fit using a chi-
21 square goodness-of-fit test (x2p-value < 0.05 * indicates lack of fit). Other factors were used to
22 assess model fit, such as scaled residuals, visual fit, and adequacy of fit in the low dose region and
23 near the BMR.
24 For each endpoint, the BMDL estimate (95% lower confidence limit on the BMD, as
25 estimated by the profile likelihood method) and AIC value were used to select a best-fit model from
26 among the models exhibiting adequate fit For the NTP (1995) and Hard etal. (2011) data, models
27 were run with all doses included, as well as with the high dose dropped. Dropping the high dose
28 resulted in a better fit to the data. Including the high dose caused the model to overestimate the
29 control.
A significance level of 0.05 is used for selecting cancer models because the model family (multistage) is
selected a priori (U.S. EPA. 2000).
This document is a draft for review purposes only and does not constitute Agency policy.
C-23 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
Table C-9. Cancer endpoints selected for dose-response modeling for tert-
butanol
Endpoint/Study
Species/Sex
Doses and Effect Data
Thyroid
Thyroid follicular cell
adenoma
NTP (1995)
Thyroid follicular cell
adenoma
NTP (1995)
B6C3Fi
mice/female
B6C3Fi
mice/male
Dose (mg/kg-d)
Incidence/Total
Dose (mg/kg-d)
Incidence/Total
lncidence/Poly-3
adjusted Total
0
2/58
0
1/60
1/50
510
3/60
540
0/59
0/50
1,020
2/59
1,040
4/59
4/51
2,110
9/59
2,070
2/60
2/35
Kidney3
Renal tubule adenoma or
carcinoma
NTP (1995)
Renal tubule adenoma or
carcinoma
NTP (1995)
Renal tubule adenoma or
carcinoma
NTP (1995)
Renal tubule adenoma or
carcinoma; Hard
reanalysis
NTP(1995);Hardetal.
(2011)
Renal tubule adenoma or
carcinoma; Hard
reanalysis
NTP(1995);Hardetal.
(2011)
Renal tubule adenoma or
carcinoma; Hard
reanalysis
NTP(1995);Hardetal.
(2011)
Rat (F344) /
Male
Rat (F344) /
Male
Rat (F344) /
Male
Rat (F344) /
Male
Rat (F344) /
Male
Rat (F344) /
Male
Dose (mg/kg-d)
Incidence /Total
Dose (PBPK, mg/L)
Incidence /Total
Dose (PBPK, mg/hr)
Incidence /Total
Dose (mg/kg-d)
Incidence /Total
Dose (PBPK, mg/L)
Incidence /Total
Dose (PBPK, mg/hr)
Incidence /Total
0
8/50
0
8/50
0
8/50
0
4/50
0
4/50
0
4/50
90
13/50
4.6945
13/50
0.7992
13/50
90
13/50
4.6945
13/50
0.7992
13/50
200
19/50
12.6177
19/50
1.7462
19/50
200
18/50
12.6177
18/50
1.7462
18/50
420
13/50
40.7135
13/50
3.4712
13/50
420
12/50
40.7135
12/50
3.4712
12/50
3
4
Endpoint presented if kidney tumors are acceptable for quantitation
This document is a draft for review purposes only and does not constitute Agency policy.
C-24 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Table C-10. Summary of the oral slope factor derivations
Tumor
Species/Sex
Selected Model
BMR
BMD
(mg/kg-
d)
POD=
BMDL
(mg/kg-d)
BMDLHED3
(mg/kg-d)
Slope factorb
(mg/kg-day)-1
Thyroid
Thyroid follicular cell
adenoma
Female
B6C3F1
mouse
3° Multistage
10%
2002
1437
201
5 x 10'4
Kidneyc
Renal tubule
adenoma or
carcinoma
Renal tubule
adenoma or
carcinoma [Hard et
al. (2011) reanalysis]
Male F344
rat; dose as
administered
Male F344
rat; dose as
administered
1° Multistage
(high dose
dropped)
1° Multistage
(high dose
dropped)
10%
10%
70
54
42
36
10.1
8.88
1 x 10'2
1 x ID'2
2
3
4
5
aHED PODs were calculated using BW3/4scaling (U.S. EPA, 2011).
bHuman equivalent slope factor = 0.1/BMDLioHED
Alternative endpoint if kidney tumors are acceptable for quantitation.
This document is a draft for review purposes only and does not constitute Agency policy.
C-25 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 C.l.2.3. Modeling Results
2
3
4
Table C-ll. Summary of BMD modeling results for thyroid follicular cell
adenomas in female B6C3F1 mice exposed to tert-butanol in drinking water
for 2 years fNTP. 19951: BMR = 10% extra risk
Model"
Three
Two
One
Goodness of fit
p-value
0.75
0.36
0.63
AICb
113.665
115.402
114.115
BMDio%c
(mg/kg-d)
2002
2186
1987
BMDLio%c
(mg/kg-d)
1437
1217
1378
Basis for model selection
Multistage 3° was selected on the basis of
the lowest AIC with all BMDL values for
fitting models being sufficiently close
(BMDLs differed by less than 3-fold).
a Selected (best-fitting) model shown in boldface type.
bAIC = Akaike Information Criterion.
c Confidence level = 0.95.
Multistage Cancer Model with 0.95 Confidence Level
0.3
0.25
0.2
0.15
0.1
0.05
Multistage Cancer
Linear extrapolation
BMDL
BMD
500
15:2205/132011
1000
dose
1500
2000
6
7
10
11
12
13
14
15
16
17
18
Figure C-9. Plot of incidence by dose, with fitted curve for Multistage 3° model
for thyroid follicular cell adenomas in female B6C3F1 mice exposed to tert-
butanol in drinking water for 2 years (NTP. 1995): BMR = 10% extra risk; dose
shown in mg/kg-d
Multistage Cancer Model. (Version: 1.9; Date: 05/26/2010)
Input Data File: M:\NCEA t-Butanol\BMD modeling\BMDS Output\29 NTP 1995b_Thyroid
folluclar cell andenoma, female mice MultiCancS 10.(d)
Gnuplot Plotting File: M:\NCEA t-Butanol\BMD modeling\BMDS Output\29 NTP
1995b_Thyroid folluclar cell andenoma, female mice_MultiCanc3_10.pit
Fri May 13 15:22:18 2011
This document is a draft for review purposes only and does not constitute Agency policy.
C-26 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4 The form of the probability function is:
6 P[response] = background + (1-background)*[1-EXP(
7 -betal*doseAl-beta2*doseA2-beta3*doseA3)
8
9
10
11
12
13
14
15 Total number of observations = 4
16 Total number of records with missing values = 0
17 Total number of parameters in model = 4
18 Total number of specified parameters = 0
19 Degree of polynomial = 3
20
21
22
23
24
25
26
27
28 Default Initial Parameter Values
29 Background = 0.0347373
30 Beta(l) = 0
31 Beta(2) = 0
32 Beta(3) = 1.36917e-011
33
34
35 Asymptotic Correlation Matrix of Parameter Estimates
36
37
38
39
40 and do not appear in the correlation matrix )
41
42 Background Beta (3)
43
44 Background 1 -0.53
45
46 Beta(3) -0.53 1
47
48
49
50
51
52 95.0% Wald Confidence Interval
53 Variable Estimate Std. Err. Lower Conf. Limit Upper Conf.
54 Limit
55
56
57
58
59
60 * - Indicates that this value is not calculated.
61
62
63
64
65
66 Model Log(likelihood) # Param's Deviance Test d.f. P-value
67 Full model
68 Fitted model
69 Reduced model
70
71
This document is a draft for review purposes only and does not constitute Agency policy.
C-27 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3 Goodness of Fit
4
5 Dose Est._Prob. Expected Observed
6
7
9
10
11
12
13
14
15
16
17 Specified effect =
18
19 Risk Type
20
21 Confidence level =
22
23 BMD =
24
25 BMDL =
26
27 BMDU =
28
29 Taken together, (1436.69, 3802.47) is a 90 % two-sided confidence
30
31
32
33
34
This document is a draft for review purposes only and does not constitute Agency policy.
C-28 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Supplemental Information—tert-Butyl Alcohol
Table C-12. Summary of BMD modeling results for thyroid follicular cell
adenomas or carcinomas in male B6C3F1 mice exposed to tert-butanol in
drinking water for 2 years (NTP. 1995): BMR = 5% extra risk
Model"
One, Two,
Three
Goodness of fit
p-value
0.202
AICb
61.6
BMDsr.
(mg/kg-d)
1788
BMDL5%C
(mg/kg-d)
787
Basis for model selection
Multistage 1° was selected. Only form of
multistage that resulted; fit adequate.
a Selected (best-fitting) model shown in boldface type.
bAIC = Akaike Information Criterion.
c Confidence level = 0.95.
Multistage Cancer Model, with BMR of 5% Extra Risk for the BMD and 0.95 Lower Confidence Limit for the BMDL
11:0206/052015
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Figure C-10. Plot of incidence by dose, with fitted curve for Multistage 1°
model for thyroid follicular cell adenomas or carcinomas in male B6C3F1 mice
exposed to tert-butanol in drinking water for 2 years (NTP. 1995); BMR = 5%
extra risk; dose shown in mg/kg-d
Multistage Model. (Version: 3.4; Date: 05/02/2014)
Input Data File: C:/Users/KHOGAN/BMDS/BMDS260/Data/msc_TBA NTP1995 MMthyroid tumors
poly3_Mscl-BMR05.(d)
Gnuplot Plotting File: C:/Users/KHOGAN/BMDS/BMDS260/Data/msc_TBA NTP1995 MMthyroid
tumors poly3_Mscl-BMR05.plt
Fri Jun 05 11:02:14 2015
This document is a draft for review purposes only and does not constitute Agency policy.
C-29 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
Dependent variable = Effect
Independent variable = Dose
Total number of observations = 4
Total number of records with missing values
Total number of parameters in model = 2
Total number of specified parameters = 0
Degree of polynomial = 1
Default Initial Parameter Values
Background = 0.0164855
Beta(l) = 2.58163e-005
Asymptotic Correlation Matrix of Parameter Estimates
Background Beta(l)
Background 1 -0.56
Beta(l) -0.56 1
Parameter Estimates
Variable
Background
Beta(1)
Model
Full model
Fitted model
Reduced model
AIC:
# Param's
4
Test d.f.
P-value
Est. Prob.
Benchmark Dose Computation
Specified effect = 0.05
Risk Type = Extra risk
This document is a draft for review purposes only and does not constitute Agency policy.
C-30 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2 Confidence level =
3
4 BMD =
5
6 BMDL =
7
8
9 BMDU did not converge for BMR = 0.050000
10 BMDU calculation failed
11 BMDU = Inf
This document is a draft for review purposes only and does not constitute Agency policy.
C-31 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4
Table C-13. Summary of BMD modeling results for thyroid follicular cell
adenomas or carcinomas in male B6C3F1 mice exposed to tert-butanol in
drinking water for 2 years, high dose omitted (NTP. 1995); BMR = 5% extra
risk
Model3
One stage
Two stage
Goodness of fit
p- value
0.105
0.174
AICb
46.0
44.9
BMD5%
(mg/kg-d)
1341
1028
BMDL5%C
(mg/kg-d)
538
644
Basis for model selection
Multistage 2° was selected based on lowest AIC.
a Selected (best-fitting) model shown in boldface type.
b AIC = Akaike Information Criterion.
Confidence level = 0.95.
Multistage Cancer Model, with BMR of 5% Extra Risk for the BMD and 0.95 Lower Confidence Limit for the BMDL
11:1806/052015
6
7
10
11
12
13
14
15
16
17
18
19
20
21
Figure C-ll. Plot of incidence by dose, with fitted curve for Multistage 2°
model for thyroid follicular cell adenomas or carcinomas in male B6C3F1 mice
exposed to tert-butanol in drinking water for 2 years, high dose omitted (NTP.
1995); BMR = 5% extra risk; dose shown in mg/kg-d
Multistage Model. (Version: 3.4; Date: 05/02/2014)
Input Data File: C:/Users/KHOGAN/BMDS/BMDS260/Data/msc_TBA NTP1995 MMthyroid tumors
poly3 -h_Msc2-BMR05.(d)
Gnuplot Plotting File: C:/Users/KHOGAN/BMDS/BMDS260/Data/msc_TBA NTP1995 MMthyroid
tumors poly3 -h_Msc2-BMR05.pit
Fri Jun 05 11:18:05 2015
This document is a draft for review purposes only and does not constitute Agency policy.
C-32 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
The form of the probability function is:
P[response] = background + (1-background)*[1-EXP(
-betal*doseAl-beta2*doseA2)]
Total number of observations = 3
Total number of records with missing values = 0
Total number of parameters in model = 3
Total number of specified parameters = 0
Degree of polynomial = 2
Variable
Background
Beta(1)
Beta(2)
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
Model
Full model
Fitted model
Reduced model
AIC:
P-value
This document is a draft for review purposes only and does not constitute Agency policy.
C-33 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2 Dose Est. Prob.
3 -
4
5
6
7
8 ChiA2 = 1.85 d.f. = 1
9
10
11 Benchmark Dose Computation
12
13 Specified effect = 0.05
14
15 Risk Type = Extra risk
16
17 Confidence level = 0.95
18
19 BMD = 1028.79
20
21 BMDL = 644.475
22
23
24 BMDU did not converge for BMR = 0.050000
25 BMDU calculation failed
26 BMDU = 14661. 6
27
28
This document is a draft for review purposes only and does not constitute Agency policy.
C-34 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information— tert- Butyl Alcohol
1
2
3
4
Table C-14. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with administered dose units and including all dose groups
fNTP. 19951: BMR = 10% extra risk.
Model"
Three
Two
One
Goodness of fit
P-
value
0.0806
0.0806
Scaled residuals
-0.989, 0.288, 1.719,
and -1.010
-0.989, 0.288, 1.719,
and -1.010
AIC
233.94
233.94
BMDiopct (mg/kg-d)
294
294
BMDLiopct (mg/kg-
d)
118
errorb
Basis for model
selection
Multistage 2° is
selected as the most
parsimonious model
of adequate fit.
a Selected model in bold.
b BMD or BMDL computation failed for this model.
0.1
1O:57 O4/3O 2O14
Multistage Cancer Model with O.95 Confidence Level
Multistage Cancer
Linear extrapolation
9
10
11
12
13
14
Figure C-12. Plot of incidence by dose, with fitted curve for Multistage 2°
model for renal tubule adenoma or carcinoma in male F344 rats exposed to
tert-butanol in drinking water for 2 years modeled with administered dose
units and including all dose groups (NTP. 1995); BMR = 10% extra risk; dose
shown in mg/kg-d.
This document is a draft for review purposes only and does not constitute Agency policy.
C-35 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Multistage Cancer Model. (Version: 1.9; Date: 05/26/2010)
The form of the probability function is: P[response] = background + (l-background)*[l-EXP(-
betal*doseAl-beta2*doseA2...)]
The parameter betas are restricted to be positive
Benchmark Dose Computation.
BMR = 10% Extra risk
BMD = 293.978
BMDL atthe 95% confidence level = 117.584
BMDU atthe 95% confidence level = 543384000
Taken together, (117.584, 543384000) is a 90% two-sided confidence interval for the BMD
Multistage Cancer Slope Factor = 0.000850453
Parameter Estimates
Variable
Background
Beta(l)
Beta(2)
Estimate
0.217704
0.000358397
0
Default Initial
Parameter Values
0.2335
0.000268894
0
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced
model
Log( likelihood)
-112.492
-114.97
-115.644
# Param's
4
2
1
Deviance
4.95502
6.30404
Test d.f.
2
3
p- value
0.08395
0.09772
AIC: = 233.94
Goodness of Fit Table
Dose
0
90
200
420
Est. Prob.
0.2177
0.2425
0.2718
0.327
Expected
10.885
12.127
13.591
16.351
Observed
8
13
19
13
Size
50
50
50
50
Scaled Resid
-0.989
0.288
1.719
-1.01
ChiA2 = 5.04 d.f=2 P-value = 0.0806
This document is a draft for review purposes only and does not constitute Agency policy.
C-36 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information— tert- Butyl Alcohol
1
2
3
4
Table C-15. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with administered dose units and excluding high-dose group
fNTP. 19951: BMR = 10% extra risk.
Model"
Two
One
Goodness of fit
P-
value
N/Ab
0.924
Scaled residuals
0.000, -0.000, and -
0.000
0.031, -0.078, and
0.045
AIC
173.68
171.69
BMDiopct (mg/kg-d)
75.6
70.1
BMDLiopct
(mg/kg-d)
41.6
41.6
Basis for model
selection
Multistage 1° was
selected as the only
adequately-fitting
model available
a Selected model in bold.
b No available degrees of freedom to calculate a goodness of fit value.
0.5
0.4
0.3 - _
0.2
Multistage Cancer Model with O.95 Confidence Level
Multistage Cancer
Linear extrapolation
200
11:02 04/30 2O14
9
10
11
12
13
14
Figure C-13. Plot of incidence by dose, with fitted curve for Multistage 1°
model for renal tubule adenoma or carcinoma in male F344 rats exposed to
tert-butanol in drinking water for 2 years modeled with administered dose
units and excluding high-dose group (NTP. 1995); BMR = 10% extra risk.; dose
shown in mg/kg-d.
This document is a draft for review purposes only and does not constitute Agency policy.
C-37 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Multistage Cancer Model. (Version: 1.9; Date: 05/26/2010)
The form of the probability function is: P[response] = background + (l-background)*[l-EXP(-
betal*doseAl-beta2*doseA2...)]
The parameter betas are restricted to be positive
Benchmark Dose Computation.
BMR = 10% Extra risk
BMD = 70.1068
BMDL at the 95% confidence level = 41.5902
BMDU at the 95% confidence level = 203.311
Taken together, (41.5902, 203.311) is a 90% two-sided confidence interval for the BMD
Multistage Cancer Slope Factor = 0.00240441
Parameter Estimates
Variable
Background
Beta(l)
Estimate
0.158399
0.00150286
Default Initial
Parameter Values
0.156954
0.0015217
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced
model
Log( likelihood
)
-83.8395
-83.8441
-86.9873
# Param's
3
2
1
Deviance
0.00913685
6.29546
Test d.f.
1
2
p- value
0.9238
0.04295
AIC: = 171.688
Goodness of Fit Table
ChiA2 = 0.01 d.f=l P-value = 0.9239
Dose
0
90
200
Est. Prob.
0.1584
0.2649
0.3769
Expected
7.92
13.243
18.844
Observed
8
13
19
Size
50
50
50
Scaled Resid
0.031
-0.078
0.045
This document is a draft for review purposes only and does not constitute Agency policy.
C-38 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information— tert- Butyl Alcohol
1
2
3
4
Table C-16. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with PBPK (tert-butanol, mg/L) dose units and including all
dose groups (NTP. 1995); BMR = 10% extra risk.
Model"
Three
Two
One
Goodness of fit
P-
value
0.0518
Scaled residuals
-1.373, 0.155, 1.889,
and -0.668
AIC
234.83
BMDiopct (mg/L)
51.8
BMDLioPct (mg/L)
13.9
Basis for model
selection
Multistage 1° was
selected as the
most parsimonious
model of adequate
fit.
a Selected model in bold.
Multistage Cancer Model with O.95 Confidence Level
1
0.5
0.4
O.3
Multistage Cancer
Linear extrapolation
BMI3
50
11:19 O4/3O 2O14
9
10
11
12
13
14
Figure C-14. Plot of incidence by dose, with fitted curve for Multistage 1°
model for renal tubule adenoma or carcinoma in male F344 rats exposed to
tert-butanol in drinking water for 2 years modeled with PBPK (tert-butanol,
mg/L) dose units and including all dose groups (NTP. 1995); BMR = 10% extra
risk.; dose shown in mg/L.
This document is a draft for review purposes only and does not constitute Agency policy.
C-39 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Supplemental Information—tert-Butyl Alcohol
Multistage Cancer Model. (Version: 1.9; Date: 05/26/2010)
The form of the probability function is: P[response] = background + (l-background)*[l-EXP(-
betal*doseAl-beta2*doseA2...)]
The parameter betas are restricted to be positive
Benchmark Dose Computation.
BMR= 10% Extra risk
BMD = 51.8357
BMDL at the 95% confidence level = 13.9404
BMDU at the 95% confidence level = error
Taken together, (13.9404, error) is a 90% two-sided confidence interval for the BMD
Multistage Cancer Slope Factor = error
Parameter Estimates
Variable
Background
Beta(l)
Estimate
0.243327
0.00203259
Default Initial
Parameter Values
0.253053
0.00150893
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced
model
Log( likelihood)
-112.492
-115.417
-115.644
# Pa ram's
4
2
1
Deviance
5.84883
6.30404
Test d.f.
2
3
p- value
0.0537
0.09772
AIC: = 234.834
Goodness of Fit Table
Dose
0
4.6945
12.6177
40.7135
Est. Prob.
0.2433
0.2505
0.2625
0.3034
Expected
12.166
12.526
13.124
15.171
Observed
8
13
19
13
Size
50
50
50
50
Scaled Resid
-1.373
0.155
1.889
-0.668
ChiA2 = 5.92 d.f=2 P-value = 0.0518
This document is a draft for review purposes only and does not constitute Agency policy.
C-40 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
Supplemental Information—tert-Butyl Alcohol
Table C-17. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with PBPK (tert-butanol, mg/L) dose units and excluding high-
dose group (NTP. 1995); BMR = 10% extra risk.
Model"
Two
One
Goodness of fit
P-
value
0.891
Scaled residuals
-0.054, 0.113, and -
0.057
AIC
171.70
BMDiopct (mg/L)
4.33
BMDLioPct (mg/L)
2.54
Basis for model
selection
Multistage 1° was
selected as the most
parsimonious model
of adequate fit.
a Selected model in bold.
Multistage Cancer Model with O.95 Confidence Level
6
1
9
10
11
12
13
0.3
0.2
0.1
Multistage Cancer
Linear extrapolation
BMDL
BMD
11 :2O O4/3O 2O14
6
dose
Figure C-15. Plot of incidence by dose, with fitted curve for Multistage 1°
model for renal tubule adenoma or carcinoma in male F344 rats exposed to
tert-butanol in drinking water for 2 years modeled with PBPK (tert-butanol,
mg/L) dose units and excluding high-dose group (NTP. 1995); BMR = 10%
extra risk; dose shown in mg/L.
This document is a draft for review purposes only and does not constitute Agency policy.
C-41 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Multistage Cancer Model. (Version: 1.9; Date: 05/26/2010)
The form of the probability function is: P[response] = background + (l-background)*[l-EXP(-
betal*doseAl-beta2*doseA2...)]
The parameter betas are restricted to be positive
Benchmark Dose Computation.
BMR = 10% Extra risk
BMD = 4.33496
BMDL atthe 95% confidence level = 2.53714
BMDU atthe 95% confidence level = 12.8097
Taken together, (2.53714,12.8097) is a 90% two-sided confidence interval for the BMD
Multistage Cancer Slope Factor = 0.0394144
Parameter Estimates
Variable
Background
Beta(l)
Estimate
0.162798
0.0243048
Default Initial
Parameter Values
0.164724
0.0238858
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced
model
Log( likelihood)
-83.8395
-83.8489
-86.9873
# Pa ram's
3
2
1
Deviance
0.0187339
6.29546
Test d.f.
1
2
p- value
0.8911
0.04295
AIC: = 171.698
Goodness of Fit Table
ChiA2 = 0.02 d.f=l P-value = 0.891
Dose
0
4.6945
12.6177
Est. Prob.
0.1628
0.2531
0.3839
Expected
8.14
12.654
19.195
Observed
8
13
19
Size
50
50
50
Scaled Resid
-0.054
0.113
-0.057
This document is a draft for review purposes only and does not constitute Agency policy.
C-42 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information— tert- Butyl Alcohol
1
2
3
4
Table C-18. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with PBPK (metabolized, mg/hr) dose units and including all
dose groups (NTP. 1995); BMR = 10% extra risk.
Model"
Three
Two
One
Goodness of fit
P-
value
0.0885
Scaled residuals
-0.920, 0.301, 1.677,
and -1.049
AIC
233.76
BMDiopct (mg/hr)
2.28
BMDLiopct (mg/hr)
0.954
Basis for model
selection
Multistage 1° was
selected as the most
parsimonious model
of adequate fit.
a Selected model in bold.
Multistage Cancer Model with O.95 Confidence Level
O.3 - -
0.1
11 .22 O4/3O 2O14
Multistage Cancer
Linear extrapolation
9
10
11
12
13
14
Figure C-16. Plot of incidence by dose, with fitted curve for Multistage 1°
model for renal tubule adenoma or carcinoma in male F344 rats exposed to
tert-butanol in drinking water for 2 years modeled with PBPK (metabolized,
mg/hr) dose units and including all dose groups (NTP. 1995); BMR = 10%
extra risk; dose shown in mg/hr.
This document is a draft for review purposes only and does not constitute Agency policy.
C-43 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Multistage Cancer Model. (Version: 1.9; Date: 05/26/2010)
The form of the probability function is: P[response] = background + (l-background)*[l-EXP(-
betal*doseAl-beta2*doseA2...)]
The parameter betas are restricted to be positive
Benchmark Dose Computation.
BMR= 10% Extra risk
BMD = 2.28299
BMDL at the 95% confidence level = 0.95436
BMDU at the 95% confidence level = error
Taken together, (0.95436, error) is a 90% two-sided confidence interval for the BMD
Multistage Cancer Slope Factor = error
Parameter Estimates
Variable
Background
Beta(l)
Estimate
0.21328
0.0461502
Default Initial
Parameter Values
0.229822
0.0349139
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced
model
Log( likelihood)
-112.492
-114.879
-115.644
# Pa ram's
4
2
1
Deviance
4.77309
6.30404
Test d.f.
2
3
p- value
0.09195
0.09772
AIC: = 233.758
Goodness of Fit Table
ChiA2=4.85 d.f=2 P-value = 0.0885
Dose
0
0.7992
1.7462
3.4712
Est. Prob.
0.2133
0.2418
0.2742
0.3297
Expected
10.664
12.088
13.71
16.487
Observed
8
13
19
13
Size
50
50
50
50
Scaled Resid
-0.92
0.301
1.677
-1.049
This document is a draft for review purposes only and does not constitute Agency policy.
C-44 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information— tert- Butyl Alcohol
1
2
3
4
Table C-19. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with PBPK (metabolized, mg/hr) dose units and excluding
high-dose group (NTP. 1995); BMR = 10% extra risk.
Model"
Two
One
Goodness of fit
P-
value
N/Ab
0.906
Scaled residuals
-0.000, -0.000, and -
0.000
0.037, -0.096, and
0.057
AIC
173.68
171.69
BMDiopct (mg/hr)
0.673
0.614
BMDLiopct (mg/hr)
0.365
0.364
Basis for model
selection
Multistage 1° was
selected on the
basis of lowest AIC.
a Selected model in bold.
b No available degrees of freedom to calculate a goodness of fit value.
Data from NTP1995
0.5
0.4
0.3 - _
0.2
Multistage Cancer Model with O.95 Confidence Level
Multistage Cancer
Linear extrapolation
1.6
1.8
11:24 04/30 2O14
9
10
11
12
13
14
Figure C-17. Plot of incidence by dose, with fitted curve for Multistage 1°
model for renal tubule adenoma or carcinoma in male F344 rats exposed to
tert-butanol in drinking water for 2 years modeled with PBPK (metabolized,
mg/hr) dose units and excluding high-dose group (NTP. 1995); BMR = 10%
extra risk; dose shown in mg/hr.
This document is a draft for review purposes only and does not constitute Agency policy.
C-45 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Multistage Cancer Model. (Version: 1.9; Date: 05/26/2010)
The form of the probability function is: P[response] = background + (l-background)*[l-EXP(-
betal*doseAl-beta2*doseA2...)]
The parameter betas are restricted to be positive
Benchmark Dose Computation.
BMR = 10% Extra risk
BMD = 0.613798
BMDL at the 95% confidence level = 0.364494
BMDU at the 95% confidence level = 1.77845
Taken together, (0.364494,1.77845) is a 90% two-sided confidence interval for the BMD
Multistage Cancer Slope Factor = 0.274353
Parameter Estimates
Variable
Background
Beta(l)
Estimate
0.158068
0.171653
Default Initial
Parameter Values
0.156284
0.174305
Analysis of Deviance Table
Model
Full model
Fitted model
Reduced
model
Log( likelihood)
-83.8395
-83.8465
-86.9873
# Pa ram's
3
2
1
Deviance
0.0138544
6.29546
Test d.f.
1
2
p- value
0.9063
0.04295
AIC: = 171.693
Goodness of Fit Table
ChiA2 = 0.01 d.f=l P-value = 0.9064
Dose
0
0.7992
1.7462
Est. Prob.
0.1581
0.266
0.3761
Expected
7.903
13.3
18.806
Observed
8
13
19
Size
50
50
50
Scaled Resid
0.037
-0.096
0.057
This document is a draft for review purposes only and does not constitute Agency policy.
C-46 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information— tert- Butyl Alcohol
1
2
3
4
Table C-20. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with administered dose units and including all dose groups;
reanalyzed data (Hard etal.. 2011; NTP. 1995); BMR = 10% extra risk.
Model"
Three
Two
One
Goodness of fit
P-
value
0.0117
Scaled residuals
-1.476, 1.100, 1.855,
and -1.435
AIC
218.68
BMDiopct (mg/kg-d)
184
BMDLiopct
(mg/kg-d)
94.8
Basis for model
selection
No model fit the
data.
6
7
8
9
10
No model was selected as a best-fitting model.
Table C-21. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with administered dose units and excluding high-dose group;
re-analyzed data fHard etal.. 2011: NTP. 1995): BMR = 10% extra risk.
Model"
Two
One
Goodness of fit
P-
value
0.572
Scaled residuals
-0.141, 0.461, and -
0.296
AIC
154.84
BMDiopct (mg/kg-d)
54.2
BMDLiopct
(mg/kg-d)
36.3
Basis for model
selection
Multistage 1° was
selected as the most
parsimonious model
of adequate fit.
a Selected model in bold.
This document is a draft for review purposes only and does not constitute Agency policy.
C-47 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Multistage Cancer Model with O.95 Confidence Level
1
2
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
0.5
0.3
Multistage Cancer
Linear extrapolation
BMDL
BMD
11:05 04/30 2O14
1OO
dose
Figure C-18. Plot of incidence by dose, with fitted curve for Multistage 1°
model for renal tubule adenoma or carcinoma in male F344 rats exposed to
tert-butanol in drinking water for 2 years modeled with administered dose
units and excluding high-dose group; re-analyzed data (Hard etal.. 2011: NTP.
1995); BMR = 10% extra risk; dose shown in mg/kg-d.
Multistage Cancer Model. (Version: 1.9; Date: 05/26/2010)
The form of the probability function is: P[response] = background + (l-background)*[l-EXP(-
betal*doseAl-beta2*doseA2...)]
The parameter betas are restricted to be positive
Benchmark Dose Computation.
BMR = 10% Extra risk
BMD = 54.1642
BMDL atthe 95% confidence level = 36.3321
BMDU atthe 95% confidence level = 101.125
Taken together, (36.3321,101.125) is a 90% two-sided confidence interval for the BMD
Multistage Cancer Slope Factor = 0.00275239
Parameter Estimates
Variable
Background
Beta(l)
Estimate
0.0855815
0.00194521
Default Initial
Parameter Values
0.0981146
0.00179645
Analysis of Deviance Table
Model
Full model
Log( likelihood)
-75.2622
# Param's
3
Deviance
Test d.f.
p- value
This document is a draft for review purposes only and does not constitute Agency policy.
C-48 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Fitted model
Reduced
model
-75.4201
-81.4909
2
1
0.315716
12.4574
1
2
0.5742
0.001972
1
2
3
4
AIC: = 154.84
Goodness of Fit Table
Dose
0
90
200
Est. Prob.
0.0856
0.2324
0.3803
Expected
4.279
11.622
19.015
Observed
4
13
18
Size
50
50
50
Scaled Resid
-0.141
0.461
-0.296
5
6
7
ChiA2 = 0.32 d.f=l P-value = 0.5715
This document is a draft for review purposes only and does not constitute Agency policy.
C-49 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
Supplemental Information—tert-Butyl Alcohol
Table C-22. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with PBPK (tert-butanol, mg/L) dose units and including all
dose groups; reanalyzed data (Hard etal.. 2011; NTP. 1995); BMR = 10% extra
risk.
Model"
Three
Two
One
Goodness of fit
P-
value
0.0048
Scaled residuals
-2.089,0.864,2.165,
and -0.929
AIC
220.82
BMDiopct (mg/L)
31.4
BMDLioPct (mg/L)
11.7
Basis for model
selection
No model fit the
data.
6
7
8
9
10
No model was selected as a best-fitting model.
Table C-23. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with PBPK (tert-butanol, mg/L) dose units and excluding high-
dose group; reanalyzed data (Hard etal.. 2011; NTP. 1995); BMR = 10% extra
risk.
Model"
Two
One
Goodness of fit
P-
value
0.364
Scaled residuals
-0.285, 0.750, and -
0.424
AIC
155.33
BMDiopct (mg/L)
3.35
BMDLioPct (mg/L)
2.21
Basis for model
selection
Multistage 1° was
selected as the most
parsimonious model
of adequate fit.
a Selected model in bold.
This document is a draft for review purposes only and does not constitute Agency policy.
C-50 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Multistage Cancer Model with O.95 Confidence Level
1
2
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
0.5
0.3
Multistage Cancer
Linear extrapolation
BMDL
BMD
11:30 04/30 2O14
6
dose
Figure C-19. Plot of incidence by dose, with fitted curve for Multistage 1°
model for renal tubule adenoma or carcinoma in male F344 rats exposed to
tert-butanol in drinking water for 2 years modeled with PBPK (tert-butanol,
mg/L) dose units and excluding high-dose group; reanalyzed data (Hard etal..
2011; NTP. 1995); BMR = 10% extra risk; dose shown in mg/L.
Multistage Cancer Model. (Version: 1.9; Date: 05/26/2010)
The form of the probability function is: P[response] = background + (l-background)*[l-EXP(-
betal*doseAl-beta2*doseA2...)]
The parameter betas are restricted to be positive
Benchmark Dose Computation.
BMR = 10% Extra risk
BMD = 3.34903
BMDL at the 95% confidence level = 2.20865
BMDU at the 95% confidence level = 6.49702
Taken together, (2.20865, 6.49702) is a 90% two-sided confidence interval for the BMD
Multistage Cancer Slope Factor = 0.0452765
Parameter Estimates
Variable
Background
Beta(l)
Estimate
0.0916116
0.03146
Default Initial
Parameter Values
0.110649
0.0276674
Analysis of Deviance Table
Model
Full model
Log( likelihood)
-75.2622
# Param's
3
Deviance
Test d.f.
p- value
This document is a draft for review purposes only and does not constitute Agency policy.
C-51 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Fitted model
Reduced
model
-75.664
-81.4909
2
1
0.803466
12.4574
1
2
0.3701
0.001972
1
2 AIC: = 155.328
3
4 Goodness of Fit Table
Dose
0
4.6945
12.6177
Est. Prob.
0.0916
0.2163
0.3892
Expected
4.581
10.817
19.462
Observed
4
13
18
Size
50
50
50
Scaled Resid
-0.285
0.75
-0.424
5
6
7
ChiA2 = 0.82 d.f=l P-value = 0.3643
9
10
11
12
13
Table C-24. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with PBPK (metabolized, mg/hr) dose units and including all
dose groups; reanalyzed data (Hard etal.. 2011: NTP. 1995): BMR = 10% extra
risk.
Model"
Three
Two
One
Goodness of fit
P-
value
0.0142
Scaled residuals
-1.367, 1.119, 1.783,
and -1.484
AIC
218.26
BMDiopct (mg/hr)
1.44
BMDLiopct (mg/hr)
0.770
Basis for model
selection
No model fit the
data.
a No model was selected as a best-fitting model.
14
15
16
17
18
Table C-25. Summary of BMD modeling results for renal tubule adenoma or
carcinoma in male F344 rats exposed to tert-butanol in drinking water for 2
years modeled with PBPK (metabolized, mg/hr) dose units and excluding
high-dose group; reanalyzed data (Hard etal.. 2011: NTP. 1995): BMR = 10%
extra risk.
Model"
Two
One
Goodness of fit
p-value
0.593
Scaled residuals
-0.130, 0.435, and -
0.281
AIC
154.81
BMDiopct (mg/hr)
0.474
BMDLiopct
(mg/hr)
0.319
Basis for model
selection
Multistage 1° was
selected as the most
parsimonious model
of adequate fit.
This document is a draft for review purposes only and does not constitute Agency policy.
C-52 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 Selected model in bold.
Multistage Cancer Model with O.95 Confidence Level
2
O
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
K/taltistage 'Cancer '-
Linear extrapolation
1 1 .33 O4/3O 2O1 4
O.8 1
dose
Figure C-20. Plot of incidence by dose, with fitted curve for Multistage 1°
model for renal tubule adenoma or carcinoma in male F344 rats exposed to
tert-butanol in drinking water for 2 years modeled with PBPK (metabolized,
mg/hr) dose units and excluding high-dose group; reanalyzed data (Hard et
al..2011: NTP. 19951: BMR = 10% extra risk.; dose shown in mg/hr.
Multistage Cancer Model. (Version: 1.9; Date: 05/26/2010)
The form of the probability function is: P[response] = background + (l-background)*[l-EXP(-
betal*doseAl-beta2*doseA2...)]
The parameter betas are restricted to be positive
Benchmark Dose Computation.
BMR= 10% Extra risk
BMD = 0.474241
BMDL at the 95% confidence level = 0.318504
BMDU at the 95% confidence level = 0.882859
Taken together, (0.318504, 0.882859) is a 90% two-sided confidence interval for the BMD
Multistage Cancer Slope Factor = 0.313968
Parameter Estimates
Variable
Background
Beta(l)
Estimate
0.0851364
0.222167
Default Initial
Parameter Values
0.0969736
0.206161
Analysis of Deviance Table
This document is a draft for review purposes only and does not constitute Agency policy.
C-53 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
Model
Full model
Fitted model
Reduced model
Log(likelihood)
-75.2622
-75.4029
-81.4909
# Pa ram's
3
2
1
Deviance
0.281435
12.4574
Test d.f.
1
2
p- value
0.5958
0.001972
1
2
3
4
AIC: = 154.806
Goodness of Fit Table
Dose
0
0.7992
1.7462
Est. Prob.
0.0851
0.234
0.3793
Expected
4.257
11.699
18.966
Observed
4
13
18
Size
50
50
50
Scaled Resid
-0.13
0.435
-0.281
5
6
7
ChiA2 = 0.29 d.f = 1 P-value = 0.5933
This document is a draft for review purposes only and does not constitute Agency policy.
C-54 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 C.l.2.4. Inhalation Unit Risk for Cancer
2 An inhalation unit risk was not derived because the relative contribution of the ot2u-globulin
3 and other, unknown, processes to renal tumor formation cannot be determined [U.S. EPA, 1991],
4 and therefore the male rat renal tumors are not considered suitable for quantitative analysis.
5 However, if renal tumors are considered suitable for analysis then route-to-route extrapolation
6 could be performed.
7 Dose Response Analysis -Adjustments and Extrapolation Methods
8 A PBPK model for tert-butanol in rats has been developed, as described in Appendix B.
9 Using this model, route-to-route extrapolation of the oral BMDL to derive an inhalation POD was
10 performed as follows. First, the internal dose in the rat at the oral BMDL (assuming continuous
11 exposure) was estimated using the PBPK model, to derive an "internal dose BMDL." Then, the
12 inhalation air concentration (again assuming continuous exposure) that led to the same internal
13 dose in the rat was estimated using the PBPK model, resulting in a route-to-route extrapolated
14 BMCL.
15 A critical decision in the route-to-route extrapolation is the selection of the internal dose
16 metric to use that established "equivalent" oral and inhalation exposures. For tert-butanol-induced
17 kidney effects, the two options are the concentration of tert-butanol in blood and rate of tert-
18 butanol metabolism. Note that using the kidney concentration of tert-butanol will lead to the same
19 route-to-route extrapolation relationship as tert-butanol in blood, since the distribution from blood
20 to kidney is independent of route. There are no data that suggest metabolites of tert-butanol
21 mediate its renal toxicity. In the absence of evidence that would suggest otherwise, it is assumed
22 that tert-butanol itself is the active toxicological agent. Therefore, the concentration of tert-butanol
23 in blood was selected as the dose metric to derive the BMCL.
24 The RfC methodology provides a mechanism for deriving a HEC from the BMCL determined
25 from the animal data. The approach takes into account the extra-respiratory nature of the
26 toxicological responses and accommodates species differences by considering blood:air partition
27 coefficients for tert-butanol in the laboratory animal (rat or mouse) and humans. According to the
28 RfC guidelines (U.S. EPA. 1994). tert-butanol is a Category 3 gas because extra-respiratory effects
29 were observed. Kaneko etal. (2000) measured a blood:gas partition coefficient of 531 ± 102 for
30 tert-butanol in the male Wistar rat, while Borghoff etal. (1996) measured a value of 481 ± 29 in
31 male F344 rats. A blood:gas partition coefficient of 462 was reported for tert-butanol in humans
32 (Nihlenetal.. 1995). The calculation (Hb/g)A +• (Hb/g)H was used to calculate a blood:gas partition
33 coefficient ratio to apply to the delivered concentration. Because F344 rats were used in the study,
34 the blood:gas partition coefficient for F344 rats was used. Thus, the calculation was: 481 4- 462 =
35 1.04. Therefore, a ratio of 1.04 was used to calculate the HEC. This allowed a BMCLnEc to be derived
36 as follows:
37
This document is a draft for review purposes only and does not constitute Agency policy.
C-55 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
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
BMCL.HEC = BMCL.ADJ (mg/m3) x (interspecies conversion)
= BMCLADj (mg/m3) x (481 + 462)
= BMCLADj (mg/m3) x (1.04)
The U.S. EPA Guidelines for Carcinogen Risk Assessment (U.S. EPA. 2005) recommend that the
method used to characterize and quantify cancer risk from a chemical is determined by what is
known about the MOA of the carcinogen and the shape of the cancer dose-response curve. The
linear approach is recommended if the MOA of carcinogenicity has not been established (U.S. EPA.
2005). In the case of tert-butanol, the mode of carcinogenic action for renal tubule tumors is not
fully understood (see Section 1.2.1). Therefore, a linear low-dose extrapolation approach was used
to estimate human carcinogenic risk associated with tert-butanol exposure.
Inhalation Unit Risk Derivation
The results from route-to-route extrapolation of the male rat renal tubule tumor data are
summarized in Table C-26. The lifetime inhalation unit risk for humans is defined as the slope of
the line from the lower 95% bound on the exposure at the POD to the control response (inhalation
unit risk = 0.1/BMCLio). This slope, a 95% upper confidence limit represents a plausible upper
bound on the true risk. Using linear extrapolation from the BMCLio, a human equivalent inhalation
unit risk was derived, as listed in Table C-26.
Two inhalation unit risks were derived from the NTP (1995) bioassay: one based on the
original reported incidences and one based on the Hardetal. (2011) reanalysis. The two estimates
differ by less than 20%, but the Hardetal. (2011) reanalysis is considered preferable, as it is based
on a PWG analysis. Therefore, the recommended inhalation unit risk for providing a sense of the
magnitude of potential carcinogenic risk associated with lifetime inhalation exposure to
tert-butanol is 1 x 1Q-3 per mg/m3, or 2 x 1Q-3 per ng/m3, based on the renal tubule tumor
response in male F344 rats.
Table C-26. Summary of the inhalation unit risk derivation
Tumor
Renal tubule adenoma or
carcinoma
Renal tubule adenoma or
carcinoma [Hard et al.
(2011) reanalysis]
Species/Sex
Male F344
rat
Male F344
rat
BMR
10%
10%
BMDL
(mg/kg-d)
41.6
36.3
Internal Dose3
(mg/L)
2.01
1.74
POD=
BMCLHEcb
(mg/m3)
68.7
59.8
Unit Riskc
(mg/m3)-1
1 x 10'3
2 x 10'3
27
28
29
30
31
a Average blood concentration of tert-butanol under continuous oral exposure at the BMDL
b Continuous inhalation human equivalent concentration that leads to the same average blood concentration of
tert-butanol as continuous oral exposure at the BMDL
cHuman equivalent inhalation unit risk = 0.1/BMCI.HEc.
This document is a draft for review purposes only and does not constitute Agency policy.
C-56 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
REFERENCES
2 Acharya, S; Mehta, K; Rodrigues, S; Pereira, J; Krishnan, S; Rao, CV. (1995). Administration of
3 subtoxic doses of t-butyl alcohol and trichloroacetic acid to male Wistar rats to study the
4 interactive toxicity. Toxicol Lett 80: 97-104. http://dx.doi.org/10.1016/0378-
5 4274(95)03340-Q.
6 Acharya, S; Mehta, K; Rodriguez, S; Pereira, J; Krishnan, S; Rao, CV. (1997). A histopathological
7 study of liver and kidney in male Wistar rats treated with subtoxic doses of t-butyl
8 alcohol and trichloroacetic acid. Exp Toxicol Pathol 49: 369-373.
9 Amberg, A; Rosner, E; Dekant, W. (1999). Biotransformation and kinetics of excretion of methyl-
10 tert-butyl ether in rats and humans. Toxicol Sci 51: 1-8.
11 Amberg, A; Rosner, E; Dekant, W. (2000). Biotransformation and kinetics of excretion of ethyl
12 tert-butyl ether in rats and humans. Toxicol Sci 53: 194-201.
13 http://dx.doi.0rg/10.1093/toxsci/53.2.194.
14 Andersen, ME. (1991). Physiological modelling of organic compounds. Ann Occup Hyg 35: 309-
15 321.
16 ARCO (ARCO Chemical Company). (1983). Toxicologist's report on metabolism and
17 pharmacokinetics of radiolabeled TBA 534 tertiary butyl alcohol with cover letter dated
18 03/24/1994. (8EHQ86940000263). Newton Square, PA.
19 Arslanian, MJ; Pascoe, E; Reinhold, JG. (1971). Rat liver alcohol dehydrogenase. Purification and
20 properties. Biochem J 125: 1039-1047.
21 ATS PR (Agency for Toxic Substances and Disease Registry). (1996). Toxicological profile for
22 methyl-tert-butyl ether [ATSDR Tox Profile]. Atlanta, GA: U.S. Department of Health and
23 Human Services, Public Health Service. http://www.atsdr.cdc.gov/ToxProfiles/tp91.pdf.
24 Bailer, AJ; Portier, G. (1988). Effects of treatment-induced mortality and tumor-induced
25 mortality on tests for carcinogenicity in small samples. Biometrics 44: 417-431.
26 Bailey, SA; Zidell, RH; Perry, RW. (2004). Relationships between organ weight and body/brain
27 weight in the rat: What is the best analytical endpoint? Toxicol Pathol 32: 448-466.
28 http://dx.doi.org/10.1080/01926230490465874.
29 Baker, RC; Sorensen, SM; Deitrich, RA. (1982). The in vivo metabolism of tertiary butanol by
30 adult rats. Alcohol Clin Exp Res 6: 247-251. http://dx.doi.Org/10.llll/i.1530-
31 0277.1982.tb04970.x.
32 Bernauer, U; Amberg, A; Scheutzow, D; Dekant, W. (1998). Biotransformation of 12C- and 2-
33 13C-labeled methyl tert-butyl ether, ethyl tert-butyl ether, and tert-butyl alcohol in rats:
34 identification of metabolites in urine by 13C nuclear magnetic resonance and gas
35 chromatography/mass spectrometry. Chem Res Toxicol 11: 651-658.
36 http://dx.doi.org/10.1021/tx970215v.
37 Blancato. JIM: Evans. MV: Power. FW: Caldwell. JC. (2007). Development and use of PBPK
38 modeling and the impact of metabolism on variability in dose metrics for the risk
39 assessment of methyl tertiary butyl ether (MTBE). J Environ Prot Sci 1: 29-51.
40 Borghoff, S: Murphy, J: Medinskv, M. (1996). Development of physiologically based
41 pharmacokinetic model for methyl tertiary-butyl ether and tertiary-butanol in male
This document is a draft for review purposes only and does not constitute Agency policy.
R-l DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 Fisher-344 rats. Fundam Appl Toxicol 30: 264-275.
2 http://dx.doi.org/10.1006/faat.1996.0064.
3 Borghoff, S; Parkinson, H; Leavens, T. (2010). Physiologically based pharmacokinetic rat model
4 for methyl tertiary-butyl ether; comparison of selected dose metrics following various
5 MTBE exposure scenarios used for toxicity and carcinogenicity evaluation. Toxicology
6 275: 79-91. http://dx.doi.Org/10.1016/i.tox.2010.06.003.
7 Borghoff, SJ; Short, BG; Swenberg, JA. (1990). Biochemical mechanisms and pathobiology of
8 alpha 2u-globulin nephropathy [Review]. Annu Rev Pharmacol Toxicol 30: 349-367.
9 http://dx.doi.org/10.1146/annurev.pa.30.040190.002025.
10 Borghoff. SJ: Prescott. JS: Janszen. DB: Wong. BA: Everitt. Jl. (2001). alpha2u-Globulin
11 nephropathy, renal cell proliferation, and dosimetry of inhaled tert-butyl alcohol in male
12 and female F-344 rats. Toxicol Sci 61:176-186.
13 Brown. MA: Cornell. BA: Davenport. JB. (1977). PERTURBATION OF BIOLOGICAL-MEMBRANES
14 WITH TERT-BUTANOL. Seikagaku 49: 1-1.
15 Cederbaum, Al; Cohen, G. (1980). Oxidative demethylation of t-butyl alcohol by rat liver
16 microsomes. Biochem Biophys Res Commun 97: 730-736.
17 Cederbaum, Al; Qureshi, A; Cohen, G. (1983). Production of formaldehyde and acetone by
18 hydroxyl-radical generating systems during the metabolism of tertiary butyl alcohol.
19 Biochem Pharmacol 32: 3517-3524. http://dx.doi.org/10.1016/0006-2952(83)90297-6.
20 Dickey, FH; Cleland, GH; Lotz, C. (1949). The role of organic peroxides in the induction of
21 mutations. PNAS 35: 581-586.
22 Faulkner, T; Hussain, A. (1989). The pharmacokinetics of tertiary butanol in C57BL/6J mice. Res
23 Comm Chem Pathol Pharmacol 64: 31-39.
24 Faulkner, TP; Wiechart, JD; Hartman, DM; Hussain, AS. (1989). The effects of prenatal tertiary
25 butanol administration in CBA/J and C57BL/6J mice. Life Sci 45:1989-1995.
26 FDA (U.S. Food and Drug Administration). (2011a). Indirect food additives: Adjuvants,
27 production aids, and sanitizers. Surface lubricants used in the manufacture of metallic
28 articles. 21 CFR 178.3910.
29 http://www.accessdata.fda.gov/scripts/cd rh/cfdocs/cfcfr/CFRSearch.cfm?fr=178.3910.
30 FDA (U.S. Food and Drug Administration). (2011b). Indirect food additives: Paper and
31 paperboard components. Defoaming agents used in coatings. 21 CFR 176.200.
32 http://www.accessdata.fda.gov/scripts/cd rh/cfdocs/cfcfr/CFRSearch.cfm?fr=176.200.
33 Grant, KA: Samson, HH. (1981). Development of physical dependence on t-butanol in rats: An
34 examination using schedule-induced drinking. Pharmacol Biochem Behav 14: 633-637.
35 http://dx.doi.org/10.1016/0091-3057(81)90124-6.
36 Hard. GC: Bruner. RH: Cohen. SM: Pletcher. JM: Regan. KS. (2011). Renal histopathology in
37 toxicity and carcinogenicity studies with tert-butyl alcohol administered in drinking
38 water to F344 rats: A pathology working group review and re-evaluation. Regul Toxicol
39 Pharmacol 59: 430-436. http://dx.doi.Org/10.1016/i.yrtph.2011.01.007.
40 Jimenez, J; Longo, E; Benitez, T. (1988). Induction of petite yeast mutants by membrane-active
41 agents. Appl Environ Microbiol 54: 3126-3132.
42 Johanson, G; Nihlen, A; Lof, A. (1995). Toxicokinetics and acute effects of MTBE and ETBE in
43 male volunteers. Toxicol Lett 82/83: 713-718. http://dx.doi.org/10.1016/0378-
44 4274(95)03589-3.
This document is a draft for review purposes only and does not constitute Agency policy.
R-2 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 Kaneko, T; Wang, PY; Sato, A. (2000). Partition coefficients for gasoline additives and their
2 metabolites. J Occup Health 42: 86-87. http://dx.doi.org/10.1539/ioh.42.86.
3 Kim. D: Andersen. ME: Pleil. JD: Nylander-French. LA: Prah. JD. (2007). Refined PBPK model of
4 aggregate exposure to methyl tertiary-butyl ether. Toxicol Lett 169: 222-235.
5 http://dx.doi.0rg/10.1016/i.toxlet.2007.01.008.
6 Leavens, T; Borghoff, S. (2009). Physiologically based pharmacokinetic model of methyl tertiary
7 butyl ether and tertiary butyl alcohol dosimetry in male rats based on binding to
8 alpha2u-globulin. Toxicol Sci 109: 321-335. http://dx.doi.org/10.1093/toxsci/kfp049.
9 Licata, AC; Dekant, W; Smith, CE; Borghoff, SJ. (2001). A physiologically based pharmacokinetic
10 model for methyl tert-butyl ether in humans: Implementing sensitivity and variability
11 analyses. Toxicol Sci 62:191-204. http://dx.doi.Org/10.1093/toxsci/62.2.191.
12 Lyondell Chemical Co. (Lyondell Chemical Company). (2004). Reproductive and developmental
13 toxicity screening test in rats by oral gavage. (Document Control Number: 89-
14 040000106).
15 McComb, J; Goldstein, D. (1979a). Additive physical dependence: evidence for a common
16 mechanism in alcohol dependence. J Pharmacol Exp Ther 210: 87-90.
17 McComb, J; Goldstein, D. (1979b). Quantitative comparison of physical dependence on tertiary
18 butanol and ethanol in mice: Correlation with lipid solubility. J Pharmacol Exp Ther 208:
19 113-117.
20 Mcgregor, D; Cruzan, G; Callander, R; May, K; Banton, M. (2005). The mutagenicity testing of
21 tertiary-butyl alcohol, tertiary-butyl acetate and methyl tertiary-butyl ether in
22 Salmonella typhimurium. Mutat Res 565:181-189.
23 http://dx.doi.0rg/10.1016/i.mrgentox.2004.10.002.
24 McGregor, DB; Brown, A; Cattanach, P; Edwards, I; Mcbride, D; Caspary, WJ. (1988). Responses
25 of the L5178Y tk+/tk- mouse lymphoma cell forward mutation assay II: 18 coded
26 chemicals. Environ Mol Mutagen 11: 91-118.
27 Nihlen, A; Lof, A; Johanson, G. (1995). Liquid/air partition coefficients of methyl and ethyl t-
28 butyl ethers, t-amyl methyl ether, and t-butyl alcohol. J Expo Anal Environ Epidemiol 5:
29 573-582.
30 Nihlen. A: Lof. A: Johanson. G. (1998a). Controlled ethyl tert-butyl ether (ETBE) exposure of
31 male volunteers: I Toxicokinetics. Toxicol Sci 46: 1-10.
32 http://dx.doi.org/10.1006/toxs.1998.2516.
33 Nihlen, A: Lof, A: Johanson, G. (1998b). Experimental exposure to methyl tertiary-butyl ether: I
34 Toxicokinetics in humans. Toxicol Appl Pharmacol 148: 274-280.
35 Nihlen, A: Johanson, G. (1999). Physiologically based toxicokinetic modeling of inhaled ethyl
36 tertiary-butyl ether in humans. Toxicol Sci 51: 184-194.
37 http://dx.doi.0rg/10.1093/toxsci/51.2.184.
38 NIOSH (National Institute for Occupational Safety and Health). (2007). NIOSH pocket guide to
39 chemical hazards. (DHHS-2005-149. CBRNIAC-CB-112149). Cincinnati, OH.
40 http://www.cdc.gov/niosh/docs/2005-149/.
41 NSF International. (2003). t-Butanol: Oral Risk Assessment Document (CAS 75-65-0). Ann Arbor,
42 Ml.
This document is a draft for review purposes only and does not constitute Agency policy.
R-3 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 NTP (National Toxicology Program). (1995). Toxicology and carcinogenesis studies of t-butyl
2 alcohol (CAS no 75-65-0) in F344/N rats and B6C3F1 mice (Drinking water studies) (pp.
3 1-305). (NTPTR436). Research Triangle Park, NC.
4 NTP (National Toxicology Program). (1997). NTP technical report on toxicity studies of t-butyl
5 alcohol (CAS no 75-65-0) administered by inhalation to F344/N rats and B6C3F1 mice
6 (pp. 1-56, A51-D59). Research Triangle Park, NC.
7 http://ntp.niehs.nih.gov/ntp/htdocs/ST rpts7toxQ53.pdf.
8 OSHA (Occupational Safety & Health Administration). (2006). Table Z-l: Limits for air
9 contaminants. Occupational safety and health standards, subpart Z, toxic and hazardous
10 substances. (OSHA standard 1910.1000, 29 CFR). Washington, DC: U.S. Department of
11 Labor.
12 http://www.osha.gov/pls/oshaweb/owadisp.show document?p table=STANDARDS&p
13 id=9992.
14 Poet, TS; Valentine, JL; Borghoff, SJ. (1997). Pharmacokinetics of tertiary butyl alcohol in male
15 and female Fischer 344 rats. Toxicol Lett 92: 179-186.
16 Prah, J; Ashley, D; Blount, B; Case, M; Leavens, T; Pleil, J; Cardinal!, F. (2004). Dermal, oral, and
17 inhalation pharmacokinetics of methyl tertiary butyl ether (MTBE) in human volunteers.
18 Toxicol Sci 77: 195-205. http://dx.doi.org/10.1093/toxsci/kfh009.
19 Rao, HV; Ginsberg, GL. (1997). A physiologically-based pharmacokinetic model assessment of
20 methyl t-butyl ether in groundwater for a bathing and showering determination. Risk
21 Anal 17: 583-598.
22 Sgambato, A; lavicoli, I; De Paola, B; Bianchino, G; Boninsegna, A; Bergamaschi, A; Pietroiusti, A;
23 Cittadini, A. (2009). Differential toxic effects of methyl tertiary butyl ether and tert-
24 butanol on rat fibroblasts in vitro. Toxicol Ind Health 25: 141-151.
25 http://dx.doi.org/10.1177/0748233709104867.
26 Snell, D. (1980). Impairment of avoidance behavior following short-term ingestion of ethanol,
27 tertiary-butanol, or pentobarbital in mice. Psychopharmacology 69: 53-57.
28 http://dx.doi.org/10.1007/BF00426521.
29 Spiteri, NJ. (1982). Circadian patterning of feeding, drinking and activity during diurnal food
30 access in rats. Physiol Behav 28: 139-147. http://dx.doi.org/10.1016/0031-
31 9384(82)90115-9.
32 Tang, G: Wang, J: Zhuang, Z. (1997). [Cytotoxicity and genotoxicity of methyl tert-butyl ether
33 and its metabolite to human leukemia cells]. Zhonghua Yufang Yixue Zazhi 31: 334-337.
34 Thurman. RG: Winn. K: Urquhart. B. (1980). Rat brain cyclic AMP levels and withdrawal
35 behavior following treatment with t-butanol. Adv Exp Med Biol 126: 271-281.
36 U.S. EPA (U.S. Environmental Protection Agency). (1991). Alpha-2u-globulin: Association with
37 chemically induced renal toxicity and neoplasia in the male rat. (EPA/625/3-91/019F).
38 Washington, DC: U.S. Environmental Protection Agency, National Center for
39 Environmental Assessment.
40 https://ntrl.ntis.gov/NTRL/dashboard/searchResults.xhtml?searchQuery=PB92143668.
41 U.S. EPA (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation
42 reference concentrations and application of inhalation dosimetry. (EPA/600/8-90/066F).
43 Research Triangle Park, NC: U.S. Environmental Protection Agency, Environmental
This document is a draft for review purposes only and does not constitute Agency policy.
R-4 DRAFT—DO NOT CITE OR QUOTE
-------
Supplemental Information—tert-Butyl Alcohol
1 Criteria and Assessment Office.
2 http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm7deich71993.
3 U.S. EPA (U.S. Environmental Protection Agency). (2000). Benchmark dose technical guidance
4 document [external review draft] [EPA Report]. (EPA/630/R-00/001). Washington, DC:
5 U.S. Environmental Protection Agency, Risk Assessment Forum.
6 http://www.epa.gov/raf/publications/benchmark-dose-doc-draft.htm.
7 U.S. EPA (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk
8 assessment. (EPA/630/P-03/001F). Washington, DC: U.S. Environmental Protection
9 Agency, Risk Assessment Forum, http://www.epa.gov/cancerguidelines/.
10 U.S. EPA (U.S. Environmental Protection Agency). (2011). Recommended use of body weight
11 3/4 as the default method in derivation of the oral reference dose. (EPA/100/R11/0001).
12 Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
13 http://www.epa.gov/raf/publications/interspecies-extrapolation.htm.
14 Videla, LA; Fernandez, V; de Marinis, A; Fernandez, N; Valenzuela, A. (1982). Liver
15 lipoperoxidative pressure and glutathione status following acetaldehyde and aliphatic
16 alcohols pretreatments in the rat. Biochem Biophys Res Commun 104: 965-970.
17 http://dx.doi.org/10.1016/0006-291X(82)91343-2.
18 Williams-Hill. D: Spears. CP: Prakash. S; Olah. GA: Shamma. T; Moin. T; Kim. LY: Hill. CK. (1999).
19 Mutagenicity studies of methyl-tert-butylether using the Ames tester strain TA102.
20 Mutat Res 446:15-21. http://dx.doi.org/10.1016/sl383-5718(99)00137-0.
21 Williams, TM; Borghoff, SJ. (2001). Characterization of tert-butyl alcohol binding to "alpha"2u-
22 globulin in F-344 rats. Toxicol Sci 62: 228-235.
23 http://dx.doi.Org/10.1093/toxsci/62.2.228.
24 Wood, J; Laverty, R. (1979). Physical dependence following prolonged ethanol or t-butanol
25 administration to rats. Pharmacol Biochem Behav 10: 113-119.
26 Yuan. Y; Wang. HF: Sun. HF: Du. HF: Xu. LH: Liu. YF: Ding. XF: Fu. DP: Liu. KX. (2007). Adduction
27 of DNA with MTBE and TBA in mice studied by accelerator mass Spectrometry. Environ
28 Toxicol 22: 630-635. http://dx.doi.org/10.1002/tox.20295.
29 Zeiger. E: Anderson. B: Haworth. S: Lawlor. T: Mortelmans. K: Speck. W. (1987). Salmonella
30 mutagenicity tests: III. Results from the testing of 255 chemicals. Environ Mutagen 9:1-
31 109. http://dx.doi.org/10.1002/em.2860090602.
32
This document is a draft for review purposes only and does not constitute Agency policy.
R-5 DRAFT—DO NOT CITE OR QUOTE
------- |