A EPA
EPA/635/R-22/277Fa
www.epa.gov/iris
IRIS Toxicological Review of Perfluorobutanoic Acid (PFBA, CASRN 375-
22-4) and Related Salts
December 2022
Integrated Risk Information System
Center for Public Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
-------
Toxicological Review of PFBA and Related Salts
DISCLAIMER
This document has been reviewed by the U.S. Environmental Protection Agency, Office of
Research and Development and approved for publication. Any mention of trade names, products, or
services does not imply an endorsement by the U.S. government or the U.S. Environmental
Protection Agency. EPA does not endorse any commercial products, services, or enterprises.
ii
-------
Toxicological Review of PFBA and Related Salts
CONTENTS
AUTHORS| CONTRIBUTORS| REVIEWERS ix
EXECUTIVE SUMMARY xii
1. OVERVIEW OF BACKGROUND INFORMATION AND ASSESSMENT METHODS 1-1
1.1. BACKGROUND INFORMATION ON PERFLUOROBUTANOIC ACID (PFBA) 1-1
1.1.1. Physical and Chemical Properties 1-1
1.1.2. Sources, Production, and Use 1-3
1.1.3. Environmental Fate and Transport 1-3
1.1.4. Potential for Human Exposure and Populations with Potentially Greater
Exposure 1-5
1.2. SUMMARY OF ASSESSMENT METHODS 1-6
1.2.1. Literature Search and Screening 1-6
1.2.2. Evaluation of Individual Studies 1-9
1.2.3. Data Extraction 1-10
1.2.4. Evidence Synthesis and Integration 1-11
1.2.5. Dose-Response Analysis 1-13
2. LITERATURE SEARCH AND STUDY EVALUATION RESULTS 2-1
2.1. LITERATURE SEARCH AND SCREENING RESULTS 2-1
2.2. STUDY EVALUATION RESULTS 2-3
3. PHARMACOKINETICS, EVIDENCE SYNTHESIS, AND EVIDENCE INTEGRATION 3-1
3.1. PHARMACOKINETICS 3-1
3.1.1. Absorption 3-2
3.1.2. Distribution 3-2
3.1.3. Metabolism 3-8
3.1.4. Excretion 3-8
3.1.5. Summary 3-11
3.2. NONCANCER EVIDENCE SYNTHESIS AND INTEGRATION 3-14
3.2.1. Thyroid Effects 3-15
3.2.2. Hepatic Effects 3-28
3.2.3. Developmental Effects 3-50
3.2.4. Reproductive Effects 3-56
iii
-------
Toxicological Review of PFBA and Related Salts
3.2.5. Other Noncancer Health Effects 3-60
3.3. CARCINOGENICITY 3-63
4. SUMMARY OF HAZARD IDENTIFICATION CONCLUSIONS 4-1
4.1. SUMMARY OF CONCLUSIONS FOR NONCANCER HEALTH EFFECTS 4-1
4.2. SUMMARY OF CONCLUSIONS FOR CARCINOGENICITY 4-5
4.3. CONCLUSIONS REGARDING SUSCEPTIBLE POPULATIONS AND LIFESTAGES 4-5
5. DERIVATION OF TOXICITY VALUES 5-1
5.1. NONCANCER AND CANCER HEALTH EFFECT CATEGORIES CONSIDERED 5-1
5.2. NONCANCER TOXICITY VALUES 5-1
5.2.1. Oral Reference Dose (RfD) Derivation 5-1
5.2.2. Subchronic Toxicity Values for Oral Exposure (Subchronic Oral Reference Dose
[RfD]) Derivation 5-25
5.2.3. Inhalation Reference Concentration (RfC) 5-27
5.3. CANCER 5-27
5.3.1. Cancer Weight-of-Evidence Descriptor and Derivation of Cancer Risk Values 5-27
REFERENCES R-l
iv
-------
Toxicological Review of PFBA and Related Salts
TABLES
Table ES-1. Evidence integration judgements and derived toxicity values for PFBA xiii
Table 1-1. Predicted or experimental physicochemical properties of perfluorobutanoic acid
(PFBA; CASRN 375-22-4) and ammonium perfluorobutanoate (NH4+PFB;
CASRN 10495-86-0) 1-2
Table 1-2. Perfluorobutanoic acid (PFBA) levels in water, soil, and air at National Priority List
(NPL) sites 1-5
Table 1-3. Populations, Exposures, Comparators, and Outcomes (PECO) criteria 1-8
Table 3-1. Serum and liver concentrations of perfluorobutanoic acid (PFBA) following
subchronic or gestational exposure 3-4
Table 3-2. Summary of pharmacokinetics of serum perfluorobutanoic acid (PFBA)
(mean ± standard error) 3-12
Table 3-3. Percent change in thyroid hormones due to perfluorobutanoic acid (PFBA) exposure
in short-term and subchronic oral toxicity studies 3-17
Table 3-4. Incidence and severity of thyroid follicular hypertrophy/hyperplasia due to
perfluorobutanoic acid (PFBA) exposure in short-term and subchronic oral
toxicity studies 3-19
Table 3-5. Evidence profile table for thyroid effects 3-26
Table 3-6. Percent increase in relative liver weight due to perfluorobutanoic acid (PFBA)
exposure in short-term and subchronic oral toxicity studies 3-30
Table 3-7. Incidence and severity of liver histopathological lesions due to perfluorobutanoic acid
(PFBA) exposure in short-term and subchronic oral toxicity studies 3-35
Table 3-8. Evidence profile table for hepatic effects 3-46
Table 3-9. Developmental effects observed following perfluorobutanoic acid (PFBA) exposure in
a developmental toxicity study 3-52
Table 3-10. Evidence profile table for developmental effects 3-55
Table 3-11. Evidence profile table for reproductive effects 3-59
Table 4-1. Evidence integration summary for health effects for which evidence indicates a
hazard exists 4-3
Table 4-2. Hazard conclusions across published EPA PFAS human health assessments 4-4
Table 5-1. Endpoints considered for dose-response modeling and derivation of points of
departure 5-5
Table 5-2. Benchmark response levels selected for benchmark dose (BMD) modeling of
perfluorobutanoic acid (PFBA) health outcomes 5-6
Table 5-3. Rat, mouse, and human clearance values and data-informed dosimetric adjustment
factors 5-10
Table 5-4. Points of departure (PODs) considered for use in deriving candidate reference values
for perfluorobutanoic acid (PFBA) 5-13
Table 5-5. Uncertainty factors for the development of the candidate values for
perfluorobutanoic acid (PFBA) 5-14
Table 5-6. Comparison of liver-weight effects across species and durations of exposure 5-17
Table 5-7. Candidate values for perfluorobutanoic acid (PFBA) 5-20
Table 5-8. Confidence in the organ/system-specific oral reference doses (osRfDs) for
perfluorobutanoic acid (PFBA) 5-21
Table 5-9. Organ/system-specific oral reference dose (osRfD) values for perfluorobutanoic acid
(PFBA) 5-23
v
-------
Toxicological Review of PFBA and Related Salts
Table 5-10. Candidate subchronic oral reference dose (RfD) values for perfluorobutanoic acid
(PFBA) 5-25
FIGURES
Figure 1-1. Chemical structures of perfluorobutanoic acid (PFBA) and ammonium
perfluorobutanoate (NH4+PFB) 1-2
Figure 2-1. Literature search and screening flow diagram for perfluorobutanoic acid (PFBA) 2-2
Figure 2-2. Evaluation results for animal studies assessing effects of perfluorobutanoic acid
(PFBA; interactive data graphic for rating rationales) 2-4
Figure 2-3. Evaluation results for epidemiological studies assessing effects of perfluorobutanoic
acid (PFBA; interactive data graphic for rating rationales) 2-5
Figure 3-1. Evaluation results for animal studies assessing effects of perfluorobutanoic acid
(PFBA) exposure on the thyroid (see interactive data graphic for rating
rationales) 3-16
Figure 3-2. Thyroid hormone response to ammonium perfluorobutanoate (NH4+PFB) exposure
(see interactive data graphic and rationale for study evaluations for thyroid
hormone effects in Health Assessment Workspace Collaborative [HAWC]) 3-18
Figure 3-3. Thyroid histopathology and organ-weight responses to ammonium
perfluorobutanoate (NH4+PFB) exposure (see interactive data graphic and
rationale for study evaluations for other thyroid effects in Health Assessment
Workspace Collaborative [HAWC]) 3-19
Figure 3-4. Evaluation results for animal studies assessing effects of perfluorobutanoic acid
(PFBA) exposure on the liver (see interactive data graphic for rating rationales) 3-29
Figure 3-5. Liver-weight response to ammonium perfluorobutanoate (NH4+PFB) or
perfluorobutanoic acid (PFBA) exposure (see interactive data graphic and
rationale for study evaluations for liver-weight effects in Health Assessment
Workspace Collaborative [HAWC]) 3-33
Figure 3-6. Liver histopathology response to ammonium perfluorobutanoate (NH4+PFB) or
perfluorobutanoic acid (PFBA) exposure (see interactive data graphic and
rationale for study evaluation for liver histopathology effects in Health
Assessment Workspace Collaborative [HAWC]) 3-36
Figure 3-7. Evaluation results for animal studies assessing developmental effects of
perfluorobutanoic acid (PFBA) exposure (see interactive data graphic for rating
rationales) 3-51
Figure 3-8. Pre- and postnatal developmental responses to gestational ammonium
perfluorobutanoate (NH4+PFB) exposure (see interactive data graphic and
rationale for study evaluations for developmental effects in Health Assessment
Workspace Collaborative [HAWC]) 3-52
Figure 3-9. Reproductive responses to ammonium perfluorobutanoate (NH4+PFB) exposure (see
interactive data graphic and rationale for study evaluations for reproductive
effects in Health Assessment Workspace Collaborative [HAWC]) 3-57
vi
-------
Toxicological Review of PFBA and Related Salts
ABBREVIATIONS AND ACRONYMS
ACO
acyl-CoA oxidase
HAWC
Health Assessment Workspace
ADME
absorption, distribution, metabolism,
Collaborative
and excretion
HED
human equivalent dose
AFFF
aqueous film-forming foam
HERO
Health and Environmental Research
AIC
Akaike's information criterion
Online
ALP
alkaline phosphatase
HISA
highly influential scientific information
ALT
alanine aminotransferase
HPT
hypothalamic-pituitary-thyroid
AST
aspartate aminotransferase
IRIS
Integrated Risk Information System
atm
atmosphere
i.v.
intravenous
ATSDR
Agency for Toxic Substances and
IQ
intelligence quotient
Disease Registry
IQR
interquartile range
AUC
area-under-the-concentration curve
ISI
influential scientific information
BMD
benchmark dose
IUR
inhalation unit risk
BMDL
benchmark dose lower confidence limit
LLOQ
lower limit of quantitation
BMDS
Benchmark Dose Software
LN
log-normal
BMR
benchmark response
LOAEL
lowest-observed-adverse-effect level
BW
body weight
MBq
megabecquerel
Cavg
average concentration
MOA
mode of action
Cmax
maximum concentration
NCEA
National Center for Environmental
CA
Cochran-Armitage
Assessment
CAR
constitutive androstane receptor
NCV
nonconstant variance
CASRN
Chemical Abstracts Service registry
NIOSH
National Institute for Occupational
number
Safety and Health
CDR
Chemical Data Reporting
NIS
sodium-iodide symporter
CI
confidence interval
NOAEL
no-observed-adverse-effect level
CL
clearance
NPL
National Priority List
CLa
clearance in animals
NTP
National Toxicology Program
CLh
clearance in humans
OAT
organic anion transporter
CPAD
Chemical and Pollutant Assessment
OECD
Organisation for Economic Co-
Division
operation and Development
CPHEA
Center for Public Health and
OMB
Office of Management and Budget
Environmental Assessment
ORD
Office of Research and Development
CV
constant variance
OSF
oral slope factor
CYP
cytochrome P450 superfamily
PC
partition coefficient
DAF
dosimetric adjustment factor
PBPK
physiologically based pharmacokinetic
DNA
deoxyribonucleic acid
PBTK
physiologically based toxicokinetic
DNT
developmental neurotoxicity
PD
pharmacodynamic
DOD
Department of Defense
PECO
Populations, Exposures, Comparators,
EPA
Environmental Protection Agency
Outcomes
EOP
Executive Office of the President
PFAA
perfluoroalkyl acid
ER
extra risk
PFAS
per- and polyfluoroalkyl substances
FLR
full-litter resorption
PFBA
perfluorobutanoic acid
FTOH
fluorotelomer alcohol
PFBS
perfluorobutane sulfonate
GD
gestation day
PFCA
perfluoroalkyl carboxylic acid
GFR
glomerular filtration rate
PFDA
perfluorodecanoic acid
GGT
y-glutamyl transferase
PFHxA
perfluorohexanoic acid
GRADE
Grading of Recommendations
PFHxS
perfluorohexane sulfonate
Assessment, Development, and
PFNA
perfluorononanoic acid
Evaluation
PFOA
perfluorooctanoic acid
GSH
glutathione
PFOS
perfluorooctane sulfonate
PK
pharmacokinetic
vii
-------
Toxicological Review of PFBA and Related Salts
PND
postnatal day
TRI
Toxic Release Inventory
POD
point of departure
TSCA
Toxic Substances Control Act
PODhed
human equivalent dose POD
TSCATS
Toxic Substances Control Act Test
PPAR
peroxisome proliferator-activated
Submissions
receptor
TSH
thyroid-stimulating hormone
PQAPP
Programmatic Quality Assurance
TSHR
thyroid-stimulating hormone receptor
Project Plan
UCMR
Unregulated Contaminant Monitoring
PT
prothrombin time
Rule
PXR
pregnane X receptor
UDP-GT
uridine 5'-diphospho-
QA
quality assurance
glucuronosyltransferase
QAPP
Quality Assurance Project Plan
UF
uncertainty factor
QMP
Quality Management Plan
UFa
animal-to-human uncertainty factor
RBC
red blood cell
UFc
composite uncertainty factor
RD
relative deviation
UFd
database deficiencies uncertainty factor
RfC
inhalation reference concentration
UFh
human variation uncertainty factor
RfD
oral reference dose
UFl
LOAEL-to-NOAEL uncertainty factor
RS
Rao-Scott
UFs
subchronic-to-chronic uncertainty
SD
standard deviation
factor
S-D
Sprague-Dawley
Vd
volume of distribution
SE
standard error
VOC
volatile organic compound
TD
toxicodynamic
WOS
Web of Science
TH
thyroid hormone
Wy-
TK
toxicokinetic
14,643
pirinixic acid
TPO
thyroid peroxidase
viii
-------
Toxicological Review of PFBA and Related Salts
Assessment Managers (Lead Authors)
I. Allen Davis. M.S.P.H.
Michele M. Taylor. Ph.D.
U.S. EPA/Office of Research and Development/Center for
Public Health and Environmental Assessment
Authors
lason C. Lambert. Ph.D.
Elizabeth Radke. Ph.D.
Paul Schlosser. Ph.D.
U.S. EPA/Office of Research and Development/Center for
Public Health and Environmental Assessment
Contributors
Michelle Angrish. Ph.D.
Xabier Arzuaga. Ph.D.
Johanna Congleton, Ph.D.
Ingrid Druwe. Ph.D.
Mary Gilbert, Ph.D.
Christopher Lau, Ph.D.
Pamela Noves. Ph.D.
Katherine O'Shaughnessy, Ph.D.
Elizabeth Oesterling Owens, Ph.D.
Tammy Stoker, Ph.D.
Andre Weaver. Ph.D.
Amina Wilkins. M.P.H.
Michael Wright. Sc.D.
lav Zhao. Ph.D.
Chris Corton, Ph.D.
April Luke, M.S.
Kelly Garcia, B.S.
Andrew Greenhalgh, B.S.
Carolyn Gigot, B.A.
U.S. EPA/Office of Research and Development/Center for
Public Health and Environmental Assessment
U.S. EPA/Office of Research and Development/Center for
Computational Toxicology and Exposure
U.S. EPA/Office of Land and Emergency Management
Oak Ridge Associated Universities [ORAU] Contractor
Production Team
Maureen Johnson
Ryan Jones
Dahnish Shams
Jessica Soto-Hernandez
Vicki Soto
Samuel Thacker
Garland Waleko
U.S. EPA
Office of Research and Development
Center for Public Health and Environmental Assessment
Rebecca Schaefer
Oak Ridge Associated Universities (ORAU) Contractor
IX
-------
Toxicological Review of PFBA and Related Salts
Executive Direction
Wayne Cascio
V. Kay Holt
Samantha Jones
Kristina Thayer
Andrew Kraft
Janice Lee
Barbara Glenn
Viktor Morozov
CPHEA Center Director
CPHEA Deputy Center Director
CPHEA Associate Director
CPAD Division Director
CPAD Associate Division Director, IRIS PFAS Team Lead
CPHEA/CPAD/Toxic Effects Assessment (RTP) Branch Chief
CPHEA/CPAD (retired)
CPHEA/CPAD/Quantitative Assessment Branch Chief
Review
CPAD Executive Review Committee
Kristina Thayer
CPAD Division Director
Paul White
CPHEA/CPAD/Senior Science Advisor
Ravi Subramaniam
CPHEA/CPAD/Toxic Effects Assessment (DC) Branch Chief
Karen Hogan
CPH EA/ CPAD/Emeritus
Alan Stern
NJDEP (retired), Contractor
Agency Reviewers
This assessment was provided for review to scientists in EPA's program and regional offices.
Comments were submitted by:
Office of the Administrator/Office of Children's Health Protection
Office of Air and Radiation/Office of Air Quality Planning and Standards
Office of Chemical Safety and Pollution Prevention/Office of Pollution Prevention and Toxics
Office of Children's Health Protection
Office of Land and Emergency Management
Office of Water
Region 2, New York
Region 3, Philadelphia
Region 8, Denver
Interagency Reviewers
This assessment was provided for review to other federal agencies and the Executive Office of the
President (EOP). Comments were submitted by:
The White House
Office of Management and Budget
Department of Defense
Department of Health and Human Services
Agency for Toxic Substances and Disease Registry
National Institute for Occupational Safety and Health
National Institute of Environmental Health Sciences
This assessment was released for public comment on August 23, 2021, and comments were due on
November 22, 2021. The public comments are available on Regulations.gov. A summary and EPA's
disposition of the comments from the public is available in Appendix E. Comments were received
from the following entities:
3M Company
x
-------
Toxicological Review of PFBA and Related Salts
American Chemistry Council
Anonymous
Association of State Drinking Water Administrators
National Tribal Toxics Council
Natural Resources Defense Council
Neisa McMillin Graduate Student
State of New Jersey, Department of Environmental Protection
This assessment was peer reviewed by independent, expert scientists external to EPA convened by
ERG, a contractor to EPA. A peer-review meeting was held on February 22-23, 2022. The report of
the review of the EPA's Draft Toxicological Review of Perfluorobutanoic Acid and Related Salts
dated May 24, 2022, is available on the IRIS website. A summary and EPA's disposition of the
comments received is included in Appendix E.
Elaine M. Faustman, Ph.D., DABT (Chair)
Jeffrey W. Fisher, Ph.D.
Panagiotis G. Georgopoulos, Ph.D.
Joseph T. Haney, Jr., M.S.
Alan M. Hoberman, Ph.D., DABT
David A. Savitz, Ph.D.
R. Thomas Zoeller, Ph.D.
School of Public Health, University of Washington
ScitoVation
School of Public Health, Rutgers University
Texas Commission on Environmental Quality
Argus International, Inc.
School of Public Health, Brown University
Dep of Biology, University of Massachusetts, Amherst
xi
-------
Toxicological Review of PFBA and Related Salts
EXECUTIVE SUMMARY
Summary of Occurrence and Health Effects
Perfluorobutanoic acid (PFBA, CASRN 375-22-4)1 and its related salts are members of the
group of per- and polyfluoroalkyl substances (PFAS). This assessment applies to PFBA as well as
salts (including alkali metal salts) of PFBA that would be expected to fully dissociate in aqueous
solutions of pH ranging from 4-9 (e.g., in the human body). Thus, while this assessment would not
necessarily apply to non-alkali metal salts of PFBA (e.g., silver heptafluorobutyrate; CASRN 3794-
64-7) due to the possibility of PFBA-independent contributions of toxicity, it does apply to PFBA
salts including ammonium perfluorobutanoate (CASRN 10495-86-0), sodium perfluorobutanoate
(CASRN 2218-54-4), potassium heptafluorobutanoate [2966-54-3], and other non-metal or alkali
metal salts of PFBA. The synthesis of evidence and toxicity value derivation presented in this
assessment focuses on the free acid of PFBA and ammonium perfluorobutanoate given the
currently available toxicity data.
Concerns about PFBA and other PFAS stem from the resistance of these compounds to
hydrolysis, photolysis, and biodegradation, which leads to their persistence in the environment
PFAS are not naturally occurring in the environment; they are manmade compounds that have been
used widely over the past several decades in consumer products and industrial applications
because of their resistance to heat, oil, stains, grease, and water. PFBA is a breakdown product of
other PFAS that are used in stain-resistant fabrics, paper food packaging, and carpets; it was also
used for manufacturing photographic film, and it is used as a substitute for longer chain
perfluoroalkyl carboxylic acids (PFCAs) in consumer products. PFBA has been found to accumulate
in agricultural crops and has been detected in household dust, soils, food products, and surface,
ground, and drinking water. As such, exposure is possible via inhalation of indoor or outdoor air,
ingestion of drinking water and food, and dermal contact with PFBA-containing products.
Human epidemiological studies have examined possible associations between PFBA
exposure and health outcomes, such as thyroid hormones or disease, hepatic enzymes, birth
outcomes (e.g., birth weight, gestational duration), semen parameters, blood lipids, and blood
pressure. The ability to draw conclusions regarding these associations is limited due to the
methodological conduct of the studies (studies were generally considered low confidence for these
outcomes; two studies on congenital hypothyroidism and birth weight and gestational duration
1 The CASRN given is for linear PFBA; the source PFBA used in toxicity studies was reported to be 98% pure
fDas et al.. 20081and reagent grade fButenhoff et al.. 2012al. Neither study explicitly states that only the
linear form was used. Therefore, there is the possibility that a minor proportion of the PFBA used in the
studies were branched isomers and thus observed health effects may apply to the total linear and branched
isomers in a given exposure source.
xii
-------
Toxicological Review of PFBA and Related Salts
were considered uninformative); the small number of studies per health outcome; and the generally
null findings coincident with notable sources of study insensitivity due to lack of detecting
quantifiable levels of PFBA in blood samples or a narrow concentration range across exposure
groups. No studies were identified that evaluated the association between PFBA exposure and
carcinogenicity.
Animal studies of PFBA exposure in rats and mice have exclusively examined the oral route
(i.e., no inhalation or dermal studies were identified during the literature search) and have
examined noncancer endpoints only.
Altogether, the available evidence indicates that developmental, thyroid, and liver effects in
humans are likely caused by PFBA exposure in utero or during adulthood (see Sections 3.2.1, 3.2.2,
and 3.2.3). There was inadequate evidence to determine whether reproductive effects might
represent a potential human health hazard following PFBA exposure (see Section 3.2.4).
The few epidemiological studies did not inform the potential for effects in the thyroid, liver,
reproductive system, or developing offspring, and the evidence integration judgments are based on
PFBA studies in animals. Liver effects manifested as increased relative liver weight in adult animals
and increased incidence of hepatocellular hypertrophy (see Section 3.2.2 and Tables 3-6 and 3-7).
Thyroid effects in adult exposed rats were expressed through decreases in free and total thyroxine
(T4) and increased incidence of thyroid follicular hypertrophy and hyperplasia (see Section 3.2.1
and Tables 3-3 and 3-2). Developmental effects in exposed animals were expressed as the loss of
viable offspring (total litter resorption), and postnatal delays in postnatal developmental
milestones: eye opening, vaginal opening, and preputial separation (see Section 3.2.3 and Table 3-
9).
Table ES-1 summarizes the evidence integration judgments for health effects that had
enough evidence available to synthesize and draw hazard conclusions, and the toxicity values
derived for those health effects.
Table ES-1. Evidence integration judgements and derived toxicity values for
PFBA
Health system
Evidence
integration
judgment
Toxicity value
type
Value PFBA
(mg/kg-d)
Value IW
PFB (mg/kg-
d)a
Confidence in
Toxicity Valueb
UFCC
Basis
Hepatic
Evidence
indicates
(likely)
osRfD
1 X 10"3
1 X 10"3
Medium
l,000d
Increased
hepatocellular
hypertrophy in
adult rats
Subchronic
osRfD
1 x 10"2
1 x 10"2
Medium
100e
Increased
hepatocellular
hypertrophy in
adult rats
Thyroid
osRfD
1 x 10"3
1 x 10"3
Medium-low
l,000d
Decreased total
T4 in adult rats
xiii
-------
Toxicological Review of PFBA and Related Salts
Health system
Evidence
integration
judgment
Toxicity value
type
Value PFBA
(mg/kg-d)
Value IW
PFB (mg/kg-
d)a
Confidence in
Toxicity Valueb
UFCC
Basis
Evidence
indicates
(likely)
Subchronic
osRfD
1 X 10"2
1 X 10"2
Medium-low
100e
Decreased total
T4 in adult rats
Developmental
Evidence
indicates
(likely)
osRfD
6 x 10"3
7 x 10"3
Medium-low
100e
Developmental
delays in mice'
Subchronic
osRfD
6 x 10"3
7 x 10"3
Medium-low
100e
Developmental
delays in mice'
Reproductive
Evidence
inadequate
osRfD
Not derived
Not derived
NA
NA
NA
Subchronic
osRfD
Not derived
Not derived
NA
NA
NA
RfD
1 x 10"3
1 x 10"3
Medium
l,000d
Hepatic and
thyroid effects
Subchronic RfD
6 x 10"3
7 x 10"3
Medium-low
100e
Developmental
effects'
See Section 5.2.1 for full details on study and dataset selection, modeling approaches (including BMR selection),
uncertainty factor application, candidate value selection, and characterization of confidence in the osRfDs and
RfDs.
RfD = reference dose (in mg/kg-day) for lifetime exposure; subchronic RfD = reference dose (in mg/kg-d) for less-
than-lifetime exposure; osRfD = organ-specific oral reference dose (in mg/kg-d); UFc = composite uncertainty
factor; NA = not applicable.
aSee Tables 5-7 and 5-10 for details on how to calculate candidate values for salts of PFBA. The osRfDs presented
in this table have been rounded to 1 significant digit from the candidate values presented in Tables 5-7 and 5-10.
bThe overall confidence in the derived toxicity values is synthesized from confidence judgments regarding
confidence in the study used to derive the toxicity value, confidence in the evidence base supporting the hazard,
and confidence in the quantification of the point of departure; see Table 5-8 for full details for these confidence
judgments.
c See Table 5-5 for an explanation of the uncertainty factors applied to derive the osRfD and subchronic osRfD
values.
d UFc = 1000 comprised of UFa=3, UFh= 10, UFS= 10, UFl= 1, and UFD=3.
e U Fc = 100 com prised of U Fa = 3, U Fh = 10, U Fs = 1, U Fl = 1, and U Fd = 10.
fThe point of departure represents three types of developmental delays observed in the same study.
Chronic Oral Reference Dose (RfD) for Noncancer Effects
From the identified human health hazards of potential concern (liver, thyroid,
developmental toxicity), increased liver hypertrophy and decreased T4 in adult male rats after
subchronic exposure, as reported in Butenhoff et al. (2012a). were selected as the basis for the oral
reference dose (RfD) (see Section 5.2.1). A no-observed-adverse-effect level (NOAEL) of 6 mg/kg-
day NH4+PFB was identified for increased liver hypertrophy, and a NOAEL of 6 mg/kg-day NH4+PFB
was identified for decreased T4 (see Table 5-4). These values were used as the points of departure
(PODs). After converting the PODs from units of mg/kg-day NH4+PFB to units of mg/kg-day PFBA
(by multiplying by the ratio of the molecular weights of the free acid and the ammonium salt), the
ratio of serum clearance values between rats and humans was used to account for pharmacokinetic
differences between species (see Table 5-3), resulting in the human equivalent doses (PODhed) of
xiv
-------
Toxicological Review of PFBA and Related Salts
1.15 mg/kg-day and 1.27 mg/kg-day for increased liver hypertrophy and decreased T4,
respectively. The RfD for PFBA was calculated by dividing the PODhed values by a composite
uncertainty factor (UFc) of 1,000 to account for residual pharmacokinetic and pharmacodynamic
uncertainty in the extrapolation from rats to humans (UFa= 3), interindividual differences in
human susceptibility (UFh = 10), extrapolation from a subchronic-to-chronic exposure duration
(UFs = 10), and deficiencies in the toxicity database (UFd = 3) (see Table 5-5). The selected overall
RfD for PFBA derived based on liver and thyroid effects is 1 x 10~3 mg/kg-day.2,3
Confidence in the Oral Reference Dose (RfD)
The overall confidence in the RfD is medium, based on the confidence in the principal study,
confidence in the quantification of the PODs, and confidence in the evidence bases supporting the
thyroid and liver effects (see Table 5-8). The subchronic exposure toxicity study conducted by
Butenhoff et al. f2012al reported on administration of NH4+PFB by gavage to Sprague-Dawley (S-D)
rats for 90 days. This study is rated as high confidence with adequate reporting and appropriate
study design, methods, and conduct (see study evaluation analysis in Health Assessment
Workspace Collaborative [HAWC]).4 Confidence in the oral toxicity database for derivation of the
RfD is medium because consistent and coherent effects occurred within both individual organ
systems used to support the RfD, although important uncertainties remain. Confidence in the
quantification of the PODs supporting the RfD is medium, given (1) use of a NOAEL roughly
equivalent to BMDL (suggesting that this POD might not be more substantially more uncertain than
a BMD-based POD); (2) use of a NOAEL roughly equivalent with a decrease of one standard
deviation for thyroid effects (demonstrating that the NOAEL would be roughly equivalent to the
BMD, but higher than the BMDL, if BMD modeling had been conducted); and (3) dosimetric
adjustments using PFBA-specific pharmacokinetic information (see Table 5-8).
2 See Table 5-7 for details on how to calculate candidate values for salts of PFBA; briefly, the candidate values
for different salts of PFBA would be calculated by multiplying the candidate value for the free acid of PFBA by
the ratio of molecular weights. For example, for the ammonium salt the ratio would be: MW ammonium salt =
MW free acid,
231
— = 1.079. This same method of conversion can be applied to other salts of PFBA, such as the potassium or
sodium salts, using the corresponding molecular weights.
3 Note that the RfD for the free acid presented in this document and an RfD for the anion of PFBA
(perfluorobutanoate, C3F7CO2", CASRN 45048-62-2) would be practically identical given the molecular
weights between the two compounds differ by less than 0.5%, (i.e., by the weight of a single hydrogen atom).
4HAWC is a modular content management system designed to store, display, and synthesize multiple data
sources for the purpose of producing human health assessments of chemicals. This online application
documents the overall workflow of developing an assessment from literature search and systematic review to
data extraction (human epidemiology, animal bioassay, and in vitro assay), dose-response analysis, and
finally evidence synthesis and visualization. In order to view HAWC study evaluation results, visualizations,
etc., users must create first create a free account; see https: //hawcprd.epa.gov/about for more details.
xv
-------
Toxicological Review of PFBA and Related Salts
Noncancer Effects Observed Following Inhalation Exposure
No studies are available that examine toxicity in humans or experimental animals following
inhalation exposure, and no physiologically based pharmacokinetic (PBPK) models exist to allow a
route-to-route extrapolation; therefore, no inhalation reference concentration (RfC) was derived
(see Section 5.2.3).
Evidence for Carcinogenicity
Under EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA. 2005). EPA concluded
there is inadequate information to assess carcinogenic potential for PFBA by either oral or inhalation
routes of exposure (see Section 3.2.5). This conclusion precludes the derivation of quantitative
estimates for either oral (oral slope factor [OSF]) or inhalation (inhalation unit risk [IUR]) exposure
(see Section 5.3).
Subchronic Oral Reference Dose (RfD) for Noncancer Effects
In addition to providing organ/system-specific RfDs for lifetime exposures in multiple
systems, less-than-lifetime (subchronic) RfDs also were derived (see Section 5.2.2 and Tables 5-9
and 5-10). In the case of PFBA, all studies used to calculate the subchronic values were subchronic
or gestational in duration. Therefore, the method to calculate the organ/system-specific subchronic
RfDs is identical to that used for calculating the organ/system-specific RfDs, except in the
application of the UFs (e.g., the use of a UFs = 1 rather than 10 for subchronic studies given there is
no extrapolation to a chronic exposure duration). Thus, the individual organs and systems for
which specific subchronic RfD values were derived were the liver, thyroid, and developing fetus.
The value for the developing fetus was selected for the subchronic RfD. A BMDL of 3.8 mg/kg-day
NH4+PFB for increased time to vaginal opening in neonatal female mice was used as the basis for the
POD. After converting the PODs from units of mg/kg-day NH4+PFB to units of mg/kg-day PFBA (by
multiplying by the ratio of the molecular weights of the free acid and the ammonium salt), the HED
was calculated by multiplying the POD for the free acid by the ratio of serum clearance values
between mice and humans. The subchronic RfD for PFBA was calculated by dividing the PODhed of
0.62 mg/kg-day PFBA by a composite uncertainty factor of 100 to account for extrapolation from
rats to humans (UFa = 3), for interindividual differences in human susceptibility (UFh = 10), and
deficiencies in the toxicity database (UFd = 3). The subchronic RfD derived from the effects on
delayed time to vaginal opening, as representative of general developmental delays, was
6 x 10"3 mg/kg-day5.
5 See Table 5-10 for details on how to calculate subchronic candidate values for salts of PFBA; briefly, the
candidate values for different salts of PFBA would be calculated by multiplying the candidate value for the
free acid of PFBA by the ratio of molecular weights. For example, for the ammonium salt the ratio would be:
mw ammonium salt _ 231 _ ^ same method of conversion can be applied to other salts of PFBA, such as
MW free acid 214
the potassium or sodium salts, using the corresponding molecular weights.
xvi
-------
Toxicological Review of PFBA and Related Salts
1. OVERVIEW OF BACKGROUND INFORMATION
AND ASSESSMENT METHODS
A series of five PFAS assessments (PFBA, perfluorohexanoic acid [PFHxA], perfluorohexane
sulfonate [PFHxS], perfluorononanoic acid [PFNA], perfluorodecanoic acid [PFDA], and their
associated salts; (see December 2018 IRIS Outlook) is being developed by the Integrated Risk
Information System (IRIS) Program at the request of the U.S. Environmental Protection Agency
(EPA) national programs and regions. Appendix A is the systematic review protocol for these five
PFAS assessments. The protocol outlines the scoping and problem formulation efforts relating to
these assessments, including a summary of other federal and state reference values for PFBA. The
protocol also lays out the systematic review and dose-response methods used to conduct this
review (see also Section 1.2). This systematic review protocol was released for public comment in
November 2019 and was subsequently updated on the basis of those public comments. Appendix A
includes the updated version of the protocol, including a summary of the updates in the protocol
history section (see Appendix A, Section 12).
1.1. BACKGROUND INFORMATION ON PERFLUOROBUTANOIC ACID
(PFBA)
Section 1.1 provides a brief overview of aspects of the physicochemical properties, human
exposure, and environmental fate characteristics of perfluorobutanoic acid (PFBA,
CASRN 375-22-4) and its related salt ammonium perfluorobutanoate (NH4+ PFB, CASRN 10495-86-
0) that might provide useful context for this assessment This overview is not intended to provide a
comprehensive description of the available information on these topics. The reader is encouraged
to refer to source materials cited below, more recent publications on these topics, and the
assessment systematic review protocol (see Appendix A).
1.1.1. Physical and Chemical Properties
PFBA and its related salts are members of the group of per- and polyfluoroalkyl substances
(PFAS). Concerns about PFBA and other PFAS stem from the resistance of these compounds to
hydrolysis, photolysis, and biodegradation, which leads to their persistence in the environment
fSundstrom etal.. 20121. The specific chemical formula of PFBA is C4HF7O2 and the chemical
formula of NH4+PFB is C4H4F7NO2. More specifically, these PFAS are classified as perfluoroalkyl
carboxylic acids [PFCAs; OECD f20181], Because PFBA and NH4+PFB are PFCAs containing less than
seven perfluorinated carbon groups, they are considered short-chain PFAS. The specific chemical
formula of PFBA is C4HF7O2 and the chemical formula of NH4+PFB is C4H4F7NO2. More specifically,
1-1
-------
Toxicological Review of PFBA and Related Salts
these PFAS are classified as perfluoroalkyl carboxylic acids [PFCAs; OECD (2018)]. Because PFBA
and NH4+PFB are PFCAs containing less than seven perfluorinated carbon groups, they are
considered short-chain PFAS fATSDR. 2018al. The chemical structures of PFBA and NH4+PFB are
presented in Figure 1-1, and select physicochemical properties are provided in Table 1-1.
Figure 1-1. Chemical structures of perfluorobutanoic acid (PFBA) and
ammonium perfluorobutanoate (NH4+PFB).
Table 1-1. Predicted or experimental physicochemical properties of
perfluorobutanoic acid (PFBA; CASRN 375-22-4) and ammonium
perfluorobutanoate (NH4+PFB; CASRN 10495-86-0)
Property (unit)
Value
PFBA (free acid)
nh4+pfb
Molecular weight (g/mol)
214a
23 r
Melting point (°C)
-17.5a
ND
Boiling point (°C)
12 la
ND
Density (g/cm3)
1.65a
ND
Vapor pressure (mm Hg)
6.37a
ND
Henry's law constant (atm-m3/mole)
4.99 x 10"5a b
ND
Water solubility (mol/L)
2.09 x 10"3a
ND
PKa
0.08bc
ND
Octanol-water partition coefficient (Log Kow)
1.43a
ND
Soil adsorption coefficient (L/kg)
88.9ab
ND
Bioconcentration factor (BCF)
6.67a,b
ND
ND = no data.
aU.S. EPA (2018a) Chemicals Dashboard (PFBA DTXSID: 4059916): aU.S. EPA (2018a) Chemicals Dashboard (PFBA
DTXSID: 4059916): https://comptox.epa.gov/dashboard/dsstoxdb/results?utf8=%E2%9C%93&search=375-22-4.
Median or average experimental values used where available; otherwise, median, or average predicted values
used to depend on which was available.
Predicted.
c ATS PR (2018a).
1-2
-------
Toxicological Review of PFBA and Related Salts
1.1.2. Sources, Production, and Use
PFAS are not naturally occurring in the environment fATSDR. 2018al. They are synthetic
compounds that are or have been used widely over the past several decades in consumer products
and industrial applications because of their resistance to heat, oil, stains, grease, and water. PFBA is
a breakdown product of other PFAS used in stain-resistant fabrics, paper food packaging, and
carpets; it was also used for manufacturing photographic film fMDH. 20171. Shorter-chain PFAS like
PFBA are also being used as substitutes for longer chain PFAS in consumer products (LiuetaL
20141. Kotthoff et al. (20151 analyzed a variety of consumer products for PFAS. PFBA was detected
in nano- and impregnation-sprays, outdoor textiles, carpets, gloves, paper-based food contact
materials, ski wax, and leather.
The U.S. Environmental Protection Agency (EPA) has been working with companies in the
fluorochemical industry since the early 2000s to phase outthe production and use of long-chain
PFAS [ATSDR (2018a) https://www.epa.gov/assessing-and-managing-chemicals-under-tsca/risk-
management-and-polyfluoroalkyl-substances-PFAS], The production and use of these chemicals,
however, have resulted in their release to the environment through various waste streams (NLM.
2016. 20131. Also, because products containing PFAS are still in use, they could continue to be a
source of environmental contamination due to disposal or breakdown in the environment fKim and
Kannan. 20071.
No Chemical Data Reporting (CDR) on production volume for PFBA or its salts are available
in EPA's ChemView (U.S. EPA. 2019a). Also, because facilities manufacturing, processing, or
otherwise using PFAS are not required to report on releases to the environment, no quantitative
information on PFBA is available in EPA's Toxic Release Inventory [TRI (U.S. EPA. 2019a)].6
Wang etal. f20141 estimated global emission estimates of PFBA from direct and indirect
(i.e., degradation of precursors) sources between 1951 and 2030 to be between 15 and 915 metric
tons. The lower estimate assumes that producers cease production and use of long-chain PFCAs and
their precursors in line with global transition trends. The higher estimate assumes the emission
scenario in 2015 remains constant until 2030.
1.1.3. Environmental Fate and Transport
PFAS are stable and persistent in the environment fATSDR. 2018a: NLM. 2017. 2016. 20131
and many are found worldwide in the air, soil, groundwater, and surface water, and in the tissues of
plants and animals fhttps://www.epa.gov/assessing-and-managing-chemicals-under-tsca/risk-
management-and-polyfluoroalkyl-substances-PFASl.
PFAS released to air exist in the vapor phase in the atmosphere and resist photolysis, but
particle-bound concentrations also have been measured (NLM. 2017. 2016. 2013: Kim and Kannan.
6As part of the National Defense Authorization Act for Fiscal Year 2020 (Section 7321), 172 per- and
polyfluoroalkyl substances will be added to the TRI list; however, neither PFBA nor its ammonium salt is on
the list of PFAS subject to TRI reporting requirements for Reporting Year 2022.
1-3
-------
Toxicological Review of PFBA and Related Salts
20071. Wet and dry deposition are potential removal processes for particle-bound PFAS in air
(ATSDR. 2018b: Barton etal.. 2007: Prevedouros etal.. 2006: Hurley etal.. 20041. Vapor intrusion
may be a concern for PFBA given its vapor pressure and Henry's law constant, although no data
currently exist measuring inhalation exposures due to vapor intrusion of PFBA.
PFBA would not be expected to be mobile in soil based on its soil adsorption coefficient (see
Table 1-1). Zhao etal. f 20161 observed that shorter chain PFAS like PFBA were transported more
readily from the roots to the shoots of wheat plants than longer chain PFAS. Venkatesan and
Halden (20141 analyzed archived samples from outdoor mesocosms to investigate the fate over
3 years of PFAS in agricultural soil amended with biosolids. The mean half-life for loss of PFBA from
soils following biosolid application was estimated to be 385 days. PFBA would not be expected to
be mobile in soil based on its soil adsorption coefficient (see Table 1-1).
The potential for PFAS to bioconcentrate in aquatic organisms depends on their
bioconcentration factors (see Table 1-1), with longer chain PFAS accumulating to a greater degree.
Thus, the potential for PFBA to bioconcentrate is low compared with other PFAS (bioconcentration
factor of 7.61 vs. 789 and 752 for perfluorodecanoic acid [PFDA] and perfluorononanoic acid
[PFNA], respectively). PFBA has been found to bioaccumulate in foods grown on PFAS-containing
soil. Blaine etal. f20131 conducted a series of greenhouse and field experiments to investigate the
potential for PFAS to be taken up by lettuce, tomatoes, and corn when grown in industrially
impacted biosolids-amended soil and municipal biosolids-amended soil. PFBA was found to
bioaccumulate more readily than other PFAS (e.g., PFOA, PFOS, PFHxA, PFHxS, PFDA, and PFNA)
with bioaccumulation factors of 28.4-56.8 for lettuce and 68.4 for corn. PFBA had a
bioaccumulation factor of 12.2-18.2 for tomatoes, which was higher than all other PFAS studied
except perfluoropentanoic acid (bioaccumulation factor of 14.9-17.1).
PFBA has not been evaluated under the National Air Toxics Assessment program
fhttps: //www.epa.gov/national-air-toxics-assessment!. Likewise, although EPA conducted
monitoring for several PFAS in drinking water as part of the third Unregulated Contaminant
Monitoring Rule [UCMR; U.S. EPA (2019b)]. PFBA was not among the 30 contaminants monitored.
PFBA can be detected in most dust samples obtained from U.S. homes and vehicles,
however, and has been measured at higher levels in the soil and sediment surrounding
perfluorochemical industrial facilities, at U.S. military facilities, and at training grounds where
aqueous film-forming foam (AFFF) has been used for fire suppression (see Appendix A, Section 2.1).
PFBA also has been measured in the surface water and groundwater at military installations, AFFF
training grounds, and industrial sites, although data are sparse. PFBA levels in water at these sites
seem to exceed those identified in drinking water (see Appendix A, Section 2.1).
PFBA also can be detected in food. PFBA has been found in fish at 16% of sites sampled in
the U.S. Great Lakes (maximum concentration 1.3 ng/g) (Stahl etal.. 2014). Additionally, although
most of the available data are from samples from outside the U.S., PFBA has been detected in
1-4
-------
Toxicological Review of PFBA and Related Salts
grocery items including dairy products, meats and seafood, fruits and vegetables, food packaging,
and spices (see Appendix A, Section 2.1).
Specifically, regarding drinking water, PFBA concentrations ranged from 0.0855 to
2.04 |ig/L in seven municipal wells in Oakdale, Minnesota fU.S. EPA. 2019al. In a study of 23 public
water systems in New Jersey (out of over 1,000), only 3% of raw water samples contained PFBA,
and did so at concentrations much less than those reported in Minnesota [range from
nondetectable to 0.006 |ig/L; (Post etal.. 2013)]. Heo etal. (2014) detected PFBA in tap water and
bottled water in Korea at mean concentrations of 2.02 and 0.039 ng/L, respectively. The
concentrations of PFBA measured at National Priorities List (NPL) sites are provided in Table 1-2 ].
Heo etal. f 20141 detected PFBA in tap water and bottled water in Korea at mean concentrations of
2.02 and 0.039 ng/L, respectively. The concentrations of PFBA measured at National Priorities List
(NPL) sites are provided in Table 1-2 fATSDR. 20171.
Table 1-2. Perfluorobutanoic acid (PFBA) levels in water, soil, and air at
National Priority List (NPL) sites
Media
Value
Number of NPL sites with
detections
Water (ppb)
Median
Geometric mean
2.15
1.03
3
Soil (ppb)
Median
Geometric mean
1,600
1,600
2
Air (ppbv)
Median
Geometric mean
ND
ND
ND = No data.
Source: ATS PR (2017).
1.1.4. Potential for Human Exposure and Populations with Potentially Greater Exposure
The general population could be exposed to PFAS via inhalation of indoor or outdoor air
(with PFAS possibly being released to the atmosphere via manufacturing processes or via disposal,
i.e., incineration), ingestion of drinking water and food, and dermal contact with PFAS-containing
products fATSDR. 2018al. Exposure might also occur via hand-to-mouth transfer of materials
containing these compounds fATSDR. 2018al. The oral route of exposure has been considered the
most important one among the general population, however (Klaunig etal.. 2015). Contaminated
drinking water is likely to be a significant source of exposure. Due to the moderate water solubility
and mobility of PFAS in groundwater (and general lack of remediation technology at water
treatment facilities), populations consuming drinking water from any contaminated watershed
could be exposed to PFAS fSun etal.. 20161. Use of powdered granulated carbon is more efficient in
1-5
-------
Toxicological Review of PFBA and Related Salts
removing longer-chain PFAS (Sun etal.. 20161. Gebbink etal. (20151 modeled exposure to PFBA
among the adult general population using a number of exposure scenarios based on the 5th,
median, and 95th percentiles of all input exposure parameters. "Intermediate" exposure (i.e., based
on median inputs for all exposure parameters) from direct and indirect (i.e., precursor) sources was
estimated to be 19 pg/kg-day. Of the pathways evaluated (i.e., ingestion of dust, food, water;
inhalation of air), direct intake of PFBA in water accounted for the largest portion (approximately
90%-100%) of total exposure for all three exposure scenarios considered.
Several PFAS have been monitored in the human population as part of the National Health
and Nutrition Examination Survey [NHANES; CDC (2019)]. but PFBA was not among those
measured. PFBA has also been detected in breastmilk and baby food products, indicating a potential
additional route of exposure for infants. Antignac etal. T20131 reports that PFBA was detected in
17% (8 of 48) of breastmilk samples in a population of French mothers, with a mean concentration
of 0.081 ng/L. Lorenzo etal. (2016) further reported that PFBA was detected in breastmilk, infant
formulas, dry cereal baby food, and processed baby food in Valencia, Spain.
Although PFBA-specific exposure information is sparse, populations that might experience
exposures greater than those of the general population could include individuals in occupations
that require frequent contact with materials containing PFAS that break down into PFBA, such as
individuals working with stain-resistant fabrics, paper food packaging, ski wax, and carpets (see
Section 1.1.2). For example, Nilsson et al. f20101 observed a significant correlation between the
number of years individuals had worked as ski wax technicians and their blood levels of PFBA.
Populations living near fluorochemical facilities where environmental contamination to PFAS that
can break down into PFBA has occurred might also be more highly exposed.
1.2. SUMMARY OF ASSESSMENT METHODS
Section 1.2 summarizes the methods used for developing this assessment A more detailed
description of the methods for each step of the assessment development process is provided in the
systematic review protocol (see Appendix A). The protocol includes additional problem formulation
details, including the specific aims and key science issues identified for this assessment.
1.2.1. Literature Search and Screening
The detailed search approach, including the query strings and Populations, Exposures,
Comparators, and Outcomes (PECO) criteria (see Table 1-3), are provided in Appendix A, Section 4,
and Appendix B, respectively. The results of the current literature search and screening efforts are
documented below. Briefly, a literature search was first conducted in 2017 and regular updates are
performed (the literature searches will continue to be updated until shortly before release of the
document for public comment). The literature search queries the following databases (no date or
language restrictions were applied):
• PubMed (National Library of Medicine)
1-6
-------
Toxicological Review of PFBA and Related Salts
• Web of Science fThomson Reuters!
• Toxline (National Library of Medicine)7
• TSCATS (Toxic Substances Control Act Test Submissions)
In addition, relevant literature not found through database searching was identified by:
• Review of studies cited in any PFBA PECO-relevant studies and published journal reviews;
finalized or draft U.S. state, U.S. federal, and international assessments (e.g., the draft
Agency for Toxic Substances and Disease Registry [ATSDR] assessment released publicly in
2018). In addition, studies included in ongoing IRIS PFAS assessments (PFHxA, PFHxS,
PFNA, PFDA) were also scanned for any studies that met PFBA PECO criteria.
• Review of studies submitted to federal regulatory agencies and brought to the attention of
EPA. For example, studies submitted to EPA by the manufacturers in support of
requirements under the Toxic Substances Control Act (TSCA).
• Identification of studies during screening for other PFAS. For example, epidemiological
studies relevant to PFBA sometimes were identified by searches focused on one of the other
four PFAS currently being assessed by the Integrated Risk Information System (IRIS)
Program.
• Other gray literature (e.g., primary studies not indexed in typical databases, such as
technical reports from government agencies or scientific research groups; unpublished
laboratory studies conducted by industry; or working reports/white papers from research
groups or committees) brought to the attention of EPA.
All literature is tracked in the U.S. EPA Health and Environmental Research Online (HERO)
database fhttps://hero.epa.gov/hero/index.cfm/proiect/page/project id/26321. The PECO criteria
(see Table 1-3) identify the evidence that addresses the specific aims of the assessment and to focus
the literature screening, including study inclusion/exclusion.
7 Toxline has recently been moved into PubMed as part of a broad National Library of Medicine
reorganization. Toxline searches can now be conducted within PubMed.
1-7
-------
Toxicological Review of PFBA and Related Salts
Table 1-3. Populations, Exposures, Comparators, and Outcomes (PECO)
criteria
PECO
element
Evidence
Populations
Human: Any population and lifestage (occupational or general population, including children and
other sensitive populations). The following study designs will be included: controlled exposure,
cohort, case control, and cross-sectional. (Note: Case reports and case series will be tracked as
potential supplemental material.)
Animal: Nonhuman mammalian animal species (whole organism) of any lifestage (including
preconception, in utero, lactation, peripubertal, and adult stages).
Other: In vitro, in silico, or nonmammalian models of genotoxicity. (Note: Other in vitro, in silico,
or nonmammalian models will be tracked as potential supplemental material.)
Exposures
Human: Studies providing quantitative estimates of PFBA exposure based on administered dose
or concentration, biomonitoring data (e.g., urine, blood, or other specimens), environmental or
occupational-setting measures (e.g., water levels or air concentrations, residential location or
duration, job title, or work title). (Note: Studies that provide qualitative, but not quantitative,
estimates of exposure will be tracked as supplemental material.)
Animal: Oral or inhalation studies including quantified exposure to PFBA based on administered
dose, dietary level, or concentration. (Note: Nonoral and noninhalation studies will be tracked
as potential supplemental material.) PFBA mixture studies are included if they employ an
experimental arm that involves exposure to PFBA alone. (Note: Other PFBA mixture studies will
be tracked as potential supplemental material.)
Studies must address exposure to the following: PFBA (CASRN 375-22-4), or the ammonium salt
NH4+PFB (CASRN 10495-86-0). [Note: Although PFBAs are not metabolized or transformed in
the body, precursor compounds known to be bio-transformed to a PFAS are of interest,
e.g., 6:2 fluorotelomer alcohol is metabolized to multiple analytes including PFHxA and PFBA
(Russell et al., 2015b). Thus, studies of precursor PFAS that identify and auantifv PFBA will be
tracked as potential supplemental material (e.g., for ADME analyses or interpretations).]
Comparators
Human: A comparison or reference population exposed to lower levels (or no
exposure/exposure below detection levels) or for shorter periods of time.
Animal: Includes comparisons to historical controls or a concurrent control group that is
unexposed, exposed to vehicle-only or air-only exposures. (Note: Experiments including
exposure to PFBA across different durations or exposure levels without including one of these
control groups will be tracked as potential supplemental material [e.g., for evaluating key
science issues; Section 2.4 of the protocol].)
Outcomes
All cancer and noncancer health outcomes. (Note: Other than genotoxicity studies, studies
including only molecular endpoints [e.g., gene or protein changes; receptor binding or
activation] or other nonphenotypic endpoints addressing the potential biological or chemical
progression of events contributing toward toxic effects will be tracked as potential supplemental
material [e.g., for evaluating key science issues; Section 2.4 of the protocol].)
In addition to those studies meeting the PECO criteria and studies excluded as not relevant
to the assessment, studies containing supplemental material potentially relevant to the specific
1-8
-------
Toxicological Review of PFBA and Related Salts
aims of the assessment were inventoried during the literature screening process. Although these
studies did not meet PECO criteria, they were not excluded. Rather, they were considered for use in
addressing the identified key science issues (see Appendix A, Section 2.4) and other potential
scientific uncertainties identified during assessment development but unanticipated at the time of
protocol posting. Studies categorized as "potentially relevant supplemental material" included the
following:
• In vivo mechanistic or mode of action studies, including non-PECO routes of exposure
(e.g., intraperitoneal injection) and populations (e.g., nonmammalian models)
• In vitro and in silico models
• Absorption, distribution, metabolism, and excretion (ADME) and pharmacokinetic studies
(excluding models)8
• Exposure assessment or characterization (no health outcome) studies
• Human case reports or case series studies
The literature was screened by two independent reviewers with a process for conflict
resolution, first at the title and abstract level and subsequently the full-text level, using structured
forms in DistillerSR (Evidence Partners; https://distillercer.com/products/distillersr-systematic-
review-software/). Literature inventories for PECO-relevant studies and studies tagged as
"potentially relevant supplemental material" during screening were created to facilitate subsequent
review of individual studies or sets of studies by topic-specific experts.
1.2.2. Evaluation of Individual Studies
The detailed approaches used for the evaluation of epidemiological and animal toxicological
studies used in the PFBA assessment are provided in the systematic review protocol (see
Appendix A, Section 6). The general approach for evaluating PECO-relevant health effect studies is
the same for epidemiological and animal toxicological studies, although the specifics of applying the
approach differ; thus, they are described in detail in Appendices A, Sections 6.2 and 6.3,
respectively. Approaches for evaluating mechanistic evidence are described in detail in Appendix A,
Section 6.5.
The key concerns for the review of epidemiological and animal toxicological studies are
potential bias (systematic errors or deviations from the truth related to internal validity that affect
the magnitude or direction of an effect in either direction), and insensitivity (factors that limit the
ability of a study to detect a true effect; low sensitivity is a bias toward the null when an effect
8Given the known importance of ADME data, this supplemental tagging was used as the starting point for a
separate screening and review of toxicokinetics data (see Appendix A, Section 9.2 for details).
1-9
-------
Toxicological Review of PFBA and Related Salts
exists). In evaluating individual studies, two or more reviewers independently arrived at judgments
regarding the reliability of the study results (reflected as study confidence determinations; see
below) with regard to each outcome or outcome grouping of interest; thus, different judgments
were possible for different outcomes within the same study. The results of these reviews were
tracked within EPA's version of the Health Assessment Workplace Collaboration (HAWC). To
develop these judgments, each reviewer assigned a category of good, adequate, deficient (or not
reported, which generally carried the same functional interpretation as deficient), or critically
deficient (listed from best to worst methodological conduct; see Appendix A, Section 6 for
definitions) related to each evaluation domain representing the different characteristics of the
study methods that were evaluated on the basis of the criteria outlined in HAWC.
Once all evaluation domains were evaluated, the identified strengths and limitations were
collectively considered by the reviewers to reach a final study confidence classification:
• High confidence: No notable deficiencies or concerns were identified; the potential for bias
is unlikely or minimal, and the study used sensitive methodology.
• Medium confidence: Possible deficiencies or concerns were noted, but the limitations are
unlikely to be of a notable degree or to have a notable impact on the results.
• Low confidence: Deficiencies or concerns were noted, and the potential for bias or
inadequate sensitivity could have a significant impact on the study results or their
interpretation. Low confidence results were given less weight than high or medium
confidence results during evidence synthesis and integration (see Sections 1.2.4 and 1.2.5).
• Uninformative: Serious flaw(s) were identified that make the study results unusable.
Uninformative studies were not considered further, except to highlight possible research
gaps.
Using the HAWC platform (and conflict resolution by an additional reviewer, as needed), the
reviewers reached a consensus judgment regarding each evaluation domain and overall
(confidence) determination. The specific limitations identified during study evaluation were carried
forward to inform the synthesis (see Section 1.2.4) within each body of evidence for a given health
effect (i.e., study confidence determinations were not used to inform judgments in isolation).
1.2.3. Data Extraction
The detailed data extraction approach is provided in Appendix A, Section 8. Briefly, data
extraction and content management were carried out using HAWC. Data extraction elements that
were collected from epidemiological, controlled human exposure, animal toxicological, and in vitro
studies are described in HAWC f https://hawcprd.epa.gOv/about/I Not all studies that meet the
PECO criteria went through data extraction: studies evaluated as being uninformative were not
considered further and therefore did not undergo data extraction, and outcomes determined to be
1-10
-------
Toxicological Review of PFBA and Related Salts
less relevant during PECO refinement did not go through data extraction. The same was true for low
confidence studies when medium and high confidence studies (e.g., on an outcome) were available.
All findings are considered for extraction, regardless of the statistical significance of their findings.
The level of extraction for specific outcomes within a study could differ (i.e., ranging from a
narrative to full extraction of dose-response effect size information). For quality control, data
extraction was performed by one member of the evaluation team and independently verified by at
least one other member. Discrepancies in data extraction were resolved by discussion or
consultation within the evaluation team.
1.2.4. Evidence Synthesis and Integration
For the purposes of this assessment, evidence synthesis and integration are considered
distinct but related processes (see Appendix A, Sections 9 and 10 for full details). For each assessed
health effect, the evidence syntheses provide a summary discussion of each body of evidence
considered in the review that directly informs the integration across evidence to draw an overall
judgment for each health effect The available human and animal evidence pertaining to the
potential health effects are synthesized separately, with each synthesis providing a summary
discussion of the available evidence that addresses considerations regarding causation that are
adapted from fHill. 19651. Mechanistic evidence is also synthesized as necessary to help inform key
decisions regarding the human and animal evidence; processes for synthesizing mechanistic
information are covered in detail in Appendix A, Section 9.2.
The syntheses of the human and animal health effects evidence focus on describing aspects
of the evidence that best inform causal interpretations, including the exposure context examined in
the sets of studies. The evidence synthesis is based primarily on studies of high and medium
confidence. Low confidence studies could be used if few or no studies with higher confidence are
available to help evaluate consistency, or if the study designs of the low confidence studies address
notable uncertainties in the set of high or medium confidence studies on a given health effect. If low
confidence studies are used, a careful examination of the study evaluation and sensitivity with
potential effects on the evidence synthesis conclusions will be included in the narrative. When
possible, results across studies are compared using graphs and charts or other data visualization
strategies. The synthesis of mechanistic information informs the integration of health effects
evidence for both hazard identification (e.g., biological plausibility or coherence of the available
human or animal evidence; inferences regarding human relevance, or the identification of
susceptible populations and lifestages across the human and animal evidence) and dose-response
evaluation (e.g., selection of benchmark response levels, selection of uncertainty factors).
Evaluations of mechanistic information typically differ from evaluations of phenotypic evidence
(e.g., from routine toxicological studies). This is primarily because mechanistic data evaluations
consider the support for and involvement of specific events or sets of events within the context of a
broader research question (e.g., support for a hypothesized mode of action; consistency with
1-11
-------
Toxicological Review of PFBA and Related Salts
known biological processes), rather than evaluations of individual apical endpoints considered in
relative isolation.
Following the synthesis of human and animal health effects data, and mechanistic data,
integrated judgments are drawn across all lines of evidence for each assessed health effect. During
evidence integration, a structured and documented two-step process is used, as follows:
Building from the separate syntheses of the human and animal evidence, the strength of the
evidence from the available human and animal health effect studies are summarized in
parallel, but separately, using a structured evaluation of an adapted set of considerations
first introduced by Sir Bradford Hill fHill. 19651. This process is similar to that used by the
Grading of Recommendations Assessment, Development, and Evaluation (GRADE) (Morgan
etal.. 2016: Guvatt etal.. 2011: Schiinemann etal.. 2011). which arrives at an overall
integration conclusion based on consideration of the body of evidence. These summaries
incorporate the relevant mechanistic evidence (or mode-of-action [MOA] understanding)
that informs the biological plausibility and coherence within the available human or animal
health effect studies. The terms associated with the different strength of evidence
judgments within evidence streams are robust, moderate, slight, indeterminate, and
compelling evidence of no effect.
The animal, human, and mechanistic evidence judgments are then combined to draw an
overall judgment that incorporates inferences across evidence streams. Specifically, the inferences
considered during this integration include the human relevance of the animal and mechanistic
evidence, coherence across the separate bodies of evidence, and other important information
(e.g., judgments regarding susceptibility). Note that without evidence to the contrary, the human
relevance of animal findings is assumed. The final output is a summary judgment of the evidence
base for each potential human health effect across evidence streams. The terms associated with
these summary judgments are evidence demonstrates, evidence indicates (likely), evidence suggests,
evidence inadequate, and strong evidence of no effect. The decision points within the structured
evidence integration process are summarized in an evidence profile table for each considered
health effect.
As discussed in the protocol (see Appendix A), the methods for evaluating the potential
carcinogenicity of PFAS follow processes laid out in the EPA cancer guidelines U.S. EPA (2005) and
that the judgements described here for different cancer types are used to inform the evidence
integration narrative for carcinogenicity and selection of one of EPA's standardized cancer
descriptions. These are: (1) carcinogenic to humans, (2) likely to be carcinogenic to humans, (3)
suggestive evidence of carcinogenic potential, (4) inadequate information to assess carcinogenic
potential, or (5) not likely to be carcinogenic to humans. However, for PFBA, data relevant to cancer
were sparse and did not allow for such an evaluation (see Section 3.3).
1-12
-------
Toxicological Review of PFBA and Related Salts
1.2.5. Dose-Response Analysis
The details for the dose-response employed in this assessment can be found in Appendix A,
Section 11. Briefly, a dose-response assessment was performed for noncancer health hazards,
following exposure to PFBA via the oral route, as supported by existing data. For oral noncancer
hazards, oral reference doses (RfDs) are derived when possible. An RfD is an estimate, with
uncertainty spanning perhaps an order of magnitude, of an exposure to the human population
(including susceptible subgroups) that is likely to be without an appreciable risk of deleterious
health effects over a lifetime (U.S. EPA. 20021. The derivation of a reference value like the RfD
depends on the nature of the health hazard conclusions drawn during evidence integration. For
noncancer outcomes, a dose-response assessment was conducted for evidence integration
conclusions of evidence demonstrates or evidence indicates (likely). In general, toxicity values are not
developed for noncancer hazards with evidence suggests conclusions (see Appendix A, Section 10.2
for exceptions).
Consistent with EPA practice, the PFBA assessment applied a two-step approach for
dose-response assessment that distinguishes analysis of the dose-response data in the range of
observation from any inferences about responses at lower environmentally relevant exposure
levels flJ.S. EPA. 2012. 2005):
• Within the observed dose range, the preferred approach was to use dose-response
modeling to incorporate as much of the data set as possible into the analysis. This modeling
to derive a point of departure (POD) ideally includes an exposure level near the lower end
of the range of observation, without significant extrapolation to lower exposure levels.
• As derivation of cancer risk estimates and reference values nearly always involves
extrapolation to exposures lower than the POD; the approaches to be applied in these
assessments are described in more detail in Appendix A, Section 11.2.
When sufficient and appropriate human and laboratory animal data are available for the
same outcome, human data are generally preferred for the dose-response assessment because use
of human data eliminates the need to perform interspecies extrapolations. For reference values,
this assessment will derive a candidate value from each suitable data set Evaluation of these
candidate values will yield a single organ/system-specific value for each organ/system under
consideration from which a single overall reference value will be selected to cover all health
outcomes across all organs/systems. Although this overall reference value represents the focus of
these dose-response assessments, the organ/system-specific values can be useful for subsequent
cumulative risk assessments that consider the combined effect of multiple PFAS (or other agents)
acting at a common organ/system. For noncancer toxicity values, uncertainties in these estimates
are characterized and discussed.
For dose-response purposes, EPA has developed a standard set of models
(http://www.epa.gov/bmds) that can be applied to typical data sets, including those that are
1-13
-------
Toxicological Review of PFBA and Related Salts
nonlinear. In situations where alternative models with significant biological support are available
(e.g., pharmacodynamic models), those models are included as alternatives in the assessment(s)
along with a discussion of the models' strengths and uncertainties. EPA has developed guidance on
modeling dose-response data, assessing model fit, selecting suitable models, and reporting
modeling results [see the EPA Benchmark Dose Technical Guidance fU.S. EPA. 20121], Additional
judgment or alternative analyses are used if the procedure fails to yield reliable results; for
example, if the fit is poor, modeling might be restricted to the lower doses, especially if competing
toxicity at higher doses occurs. When alternative approaches fail or are not applicable, the
NOAEL/LOAEL approach is used for POD estimation. For each modeled response, a POD from the
observed data was estimated to mark the beginning of extrapolation to lower doses. The POD is an
estimated dose (expressed in human-equivalent terms) near the lower end of the observed range
without significant extrapolation to lower doses. The POD is used as the starting point for
subsequent extrapolations and analyses. For noncancer effects, the POD is used in calculating the
RfD.
1-14
-------
Toxicological Review of PFBA and Related Salts
2. LITERATURE SEARCH AND STUDY EVALUATION
RESULTS
2.1. LITERATURE SEARCH AND SCREENING RESULTS
As summarized in Section 1.2.1, the assessment used PECO criteria (see Table 8,
Appendix A) to identify the evidence that addresses the specific aims of the assessment and focuses
the literature screening, including study inclusion. In addition to those studies meeting the PECO
criteria, studies containing supplemental material potentially relevant to the specific aims of the
assessment were tagged during the literature screening process. Although these studies did not
meet PECO criteria, they were not excluded. Rather, they were considered for use in addressing the
identified key science issues and other major scientific uncertainties identified during assessment
development but unanticipated at the time of protocol posting. Studies categorized as "potentially
relevant supplemental material" included the following:
• In vivo mechanistic or mode-of-action studies, including non-PECO routes of exposure
(e.g., intraperitoneal injection) and non-PECO populations (e.g., nonmammalian models);
• In vitro and in silico models;
• Absorption, distribution, metabolism, and excretion (ADME) and pharmacokinetic (PK)
studies (excluding models);
• Exposure assessment or characterization (no health outcome) studies; and
• Human case reports or case-series studies.
The last literature search update prior to release of the draft Toxicological Review for public
comment was April 2020. As shown in Figure 2-1, the searches through 2020 yielded 610 unique
records, with 4 records identified from additional sources, such as Toxic Substances Control Act
(TSCA) submissions, posted National Toxicology Program (NTP) study tables, and review of
reference lists from other authoritative sources fATSDR. 20211. Of the 610 identified, 552 were
excluded during title and abstract screening, and 58 were reviewed at the full-text level. Of the 58
screened at the full-text level, 17 were considered to meet the PECO criteria. This included
eight epidemiological studies, nine animal studies (including one published study (Butenhoffetal..
2012a) that reported on two unpublished industry reports (van Otterdiik. 2007a) and (van
Otterdiik. 2007b). and one in vivo ge no toxicity study. No high-throughput screening data on
perfluorobutanoic acid (PFBA) were identified from ToxCastor Tox21. Additional literature
searches were conducted in April 2021 and 2022. Those studies were screened as described in the
protocol, resulting in the identification of two additional studies that met PECO criteria and are
2-1
DRAFT-DO NOT CITE OR QUOTE
-------
Toxicological Review ofPFBA and Related Salts
included in this revised assessment fGrandiean etal.. 2020: Zeng etal.. 20201. In addition, a table
compiling the published literature submitted in public comments received through the EPA docket
fhttps: //www.regulatiQns.gov/docket/EPA-HO-ORD-2020-06751 was provided to the external
peer review panel and posted to the docket. That table includes the full text screening decisions for
those studies (ultimately, none of the studies submitted to the docket that were not identified
through the literature search updates through 2022 were incorporated into this Toxicological
Review],
Literature Searches (through 2020)
1
TITLE AND ABSTRACT SCREENING
Title & Abstract Screening
(610 records after duplicate removal)
FULL TEXT SCREENING
Full-Text Screening
(n = 58)
1
*
Excluded (n = 552)
Not relevant to PECO (n = 552)
Excluded (n = 28)
not relevant to PECO (n = 14), review,
commentary, or letter (n = 8), abstract-only
(n = 1), other (n = 5)
Studies Meeting PECO {n = 17)
• Human health effects studies (n = 8)
• Animal health effect studies (n = 9)
• Genotoxicity studies (n = 1)
• PBPK models (n = 0)
• Accessory records, such as published
corrections for included studies (n = 0)
Tagged as Supplemental (n = 25)
• mechanistic or MOA (n = 8), ADME (n = 4),
exposure assessment or qualitative
exposure only (n = 4), mixture-only (n = 0),
non-PECO route of exposure (n = 2), ecotox
or zebrafish (n = 4), in silico or modeling (n
= 0), environmental occurrence (n = 0), case
report or case study (n = 0)
Figure 2-1. Literature search and screening flow diagram for
perfluorobutanoic acid (PFBA).
TheButenhoff et al. (2012a) study reported the findings of two unpublished industry reports: a 28-day and
90-day gavage study fully reported infvan Otterdiik. 2007a. b], All three of these references are listed here as
separate studies, but Figure 2-2 below only provides study quality determinations for fButenhoff et al.,
2012a],
2-2
DRAFT-DO NOT CITE OR QUOTE
-------
Toxicological Review of PFBA and Related Salts
2.2. STUDY EVALUATION RESULTS
Human and animal studies have evaluated potential effects to the thyroid, reproductive
systems, developing fetus, liver, urinary, and other organ systems (e.g., hematological) following
exposure to PFBA. The evidence base for these outcomes is presented in Sections 3.2.1-3.2.5.
The evidence base of all repeated-dose oral toxicity studies for PFBA and the related
compound ammonium perfluorobutanoate (NFU+PFB) that are potentially relevant for deriving oral
reference dose (RfD) values includes four short-term studies in rats and mice (Permadi et al.. 1993:
Permadi etal.. 1992: Tustetal.. 1989: Ikeda etal.. 19851. two 28-day studies in rats and mice
fButenhoff etal.. 2012a: Foreman etal.. 2009: van Otterdiik. 2007al. one subchronic-duration study
in rats fButenhoff etal.. 2012c: van Otterdiik. 2007bl. and one gestational exposure study in mice
fDas etal.. 20081. In addition, eight epidemiological studies were identified that report on the
association between PFBA and human health effects (Nian etal.. 2019a: Wang etal.. 2019: Song et
al.. 2018: Bao etal.. 2017a: Li etal.. 2017a: Li etal.. 2017b: Kim etal.. 2016: Fu etal.. 20141. The
available animal studies were generally well conducted (i.e., medium, or high confidence; see
Figure 2-2); thus, specific study limitations identified during evaluation are primarily discussed for
studies interpreted as low confidence, or when a limitation affects a specific inference for drawing
conclusions (e.g., in relation to a specific assessed endpoint within the health effects synthesis
sections below). No animal studies were considered uninformative. Thus, all animal studies meeting
PECO criteria during literature screening are included in the evidence synthesis and dose-response
analysis.
The study evaluations of the available epidemiological studies are summarized in
Figure 2-3, and rationales for each domain and overall confidence rating are available in HAWC (see
link in Figure 2-3). Based on the study evaluations, one human epidemiological study was
considered uninformative due to critical deficiencies in exposure measurement fKim etal.. 20161
this study is not discussed further in this assessment except to point out in more detail its critical
deficiencies in the relevant health effects section.
2-3
DRAFT-DO NOT CITE OR QUOTE
-------
Toxicological Review ofPFBA and Related Salts
Reporting quality
Allocation
Observational bias."blinding
Contoumhng.ivai-.aWe control
Selective reportag and attrition
Chemical administration and characterisation
Exposure liming frequency and duration
Endpoint sensitivity and specificity
Results presentation
Overall confidence
NR
NR
NR
NR
NR
D
D
1
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall
Deficient (metric) or Low confidence ioverall
Critically deficient (metnc) or Umnformative (overall
Not repotted
Figure 2-2. Evaluation results for animal studies assessing effects of
perfluorobutanoic acid (PFBA; interactive data graphic for rating rationales).
The following health outcome categories were investigated by the studies listed in Figure 2-2: thyroid effects
(Butenhoff et al., 2012a), liver effects (Butenhoff et al., 2012a; Foreman et si., 2009; Das et al., 2008; Permadi et
al„ 1993; Permadi et al.. 1992) developmental effects (Das et al., 2008), and reproductive effects (Butenhoff et al.,
2012a).
2-4
DRAFT-DO NOT CITE OR QUOTE
-------
Toxicological Review of PFBA and Related Salts
50^
Participant selection -
Exposure measurement -
Outcome ascertainment -
Confounding -
Analysis -
Sensitivity -
Selective Reporting -
Overall confidence -
•
•
-
-
-
-
+
~
+
B
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
++
++
-
-
+
++
++
-
+
-
+
++
-
+
-
++
+
-
B
+
-
+
¦
•
+
+
++
+
++
-
++
-
-
-
-
-
-
-
-
+
+
~
+
+
+
+
+
+
¦
-
-
+
¦
¦
Figure 2-3. Evaluation results for epidemiological studies assessing effects of
perfluorobutanoic acid (PFBA; interactive data graphic for rating rationales!.
The following health outcome categories were investigated by the studies listed in Figure 2-3: thyroid effects (Li et
al., 2017b; Kim et al., 2016), liver effects (Nian et al., 2019a), developmental effects (Li et al., 2017a) reproductive
effects (Song et al., 2018), blood lipids (Fu et al., 2014) hypertension/blood pressure (Bao et al., 2017b) and renal
function (Wang et al., 2019).
2-5
DRAFT-DO NOT CITE OR QUOTE
-------
Toxicological Review of PFBA and Related Salts
3. PHARMACOKINETICS, EVIDENCE SYNTHESIS,
AND EVIDENCE INTEGRATION
3.1. PHARMACOKINETICS
Pharmacokinetic studies have been conducted with dosing solutions prepared from PFBA
[e.g., fBurkemper et al.. 20171] and the ammonium and potassium salts [e.g., fChang et al.. 20081],
Some of the results evaluated below are semi-qualitative (e.g., that distribution is to all tissues of
the body), hence are described with reference to the acidic form, since given PFBA's pKa of 0.08 the
salts immediately dissociate after dissolution and analytic measurements are of the
perfluorobutanoate ion. These results are applicable independent of the form used to prepare
dosing solutions.
The one study for which quantitation of the pharmacokinetic parameters might depend on
the form is fChang etal.. 20081. Chang etal. f20081 was careful to identify the form used for each
part of their study so it is also clear that the chemical analysis used to measure concentrations in
serum used to determine pharmacokinetic parameters is that of the acid, PFBA. However,
calculation of the volume of distribution and clearance also involves the administered dose and
Chang etal. (20081 does not specify whether or not the doses were converted to dose of the acid
form. In a subsequent paper by the same research group evaluating the pharmacokinetics of PFHxS,
Sundstrom etal. (20121 explicitly state, "concentrations in serum, liver, urine and feces are
reported as PFHxS anion, and percent recoveries of administered dose in those matrices are
corrected for the potassium salt" Hence, we will presume that Chang etal. f20181 similarly
corrected either the applied dose or the serum concentrations to consistent units before reporting
their pharmacokinetic parameters. Since the key parameters, volume of distribution and clearance,
effectively involve the ratio of dose to serum concentration (or the area-under-the-concentration
curve), resulting in measures of volume per kg BW or volume per time that are independent of the
molecular weight, these results can be applied to analysis of PFBA per se, i.e., the acid or anion.
Conversion to corresponding doses of a given salt is applied before or after pharmacokinetic
analysis then provides the appropriate human equivalent doses for each form.
Animal evidence has shown that PFBA, like other perfluorinated chemicals, is well absorbed
following oral administration and distributes to all tissues examined (Burkemper etal.. 20171. A
study evaluating the volume of distribution concluded, however, that the empirical volume of
distribution is in the range typically associated with extracellular distribution (Chang etal.. 20081.
Because of its chemical resistance to metabolic degradation, PFBA appears to be primarily
eliminated unchanged in urine and feces.
3-1
-------
Toxicological Review of PFBA and Related Salts
Pharmacokinetic studies of PFBA in rats, mice, and monkeys have been performed,
providing information on the absorption, distribution, metabolism, and excretion (ADME) of PFBA
fBurkemper etal.. 2017: Chang etal.. 20081. Also, Russell etal. f2015al evaluated the metabolism of
6:2 fluorotelomer alcohol (6:2 FTOH) in mouse, rat, and human hepatocytes, showing that PFBA is a
metabolite of 6:2 FTOH, and evaluated PFBA pharmacokinetics (PK) after inhalation and oral
exposure of rats to 6:2 FTOH. The distribution of PFBA in human tissues also has been investigated
(Perez etal.. 20131. Information on the absorption and distribution of PFBA to the serum and liver
specifically has been investigated in several toxicological studies (Gomis etal.. 2018: Butenhoff et
al.. 2012a: Foreman etal.. 2009: Das etal.. 20081.
3.1.1. Absorption
Chang etal. f20081 conducted a set of pharmacokinetic experiments in which Sprague-
Dawley (S-D) rats three male and three female) were given either a single intravenous (i.v.) or oral
dose (30 mg/kg body weight via gavage) of ammonium perfluorobutanoate (NH4+PFB). The serum
area-under-the-concentration-curve (AUC) was 1,090 ± 78 and 239 ± 5 ([ig-h/mL) in male and
female rats, respectively, after i.v. dosing and 1,911 ±114 and 443 ± 42 in males and females,
respectively, after oral dosing. That the AUC after oral dosing was almost two times higher than
after i.v. dosing is theoretically impossible but might be a statistical result from the small sample
size (n = 3/group) or due to a problem in dosing. The result, however, indicates 100% oral
absorption.
In other experiments, Chang etal. (20081 orally administered 3-300 mg/kg to male and
female S-D rats via gavage. As expected, the concentration of PFBA in the serum increased with
dose in a fairly linear fashion up to 100 mg/kg PFBA; however, the serum concentration of PFBA in
rats dosed orally to 300 mg was approximately 60% the concentration at 100 mg/kg. Maximum
concentration (Cmax) values were similar in males and females following oral exposures to
30 mg/kg PFBA (131 ± 5 and 136 ± 12 [ig/mL, respectively), but the time to peak concentration
(Tmax) differed between sexes: 1.25 ± 0.12 hours for males and 0.63 ± 0.23 hours for females. Both
values, however, indicate that absorption to the serum was fairly rapid in rats.
Cmax values for male and female mice exposed to PFBA via oral gavage also were similar at
lower doses (10 mg/kg; 50.50 ± 5.81 and 52.86 ± 2.08 [ig/mL) but differed at 30 mg/kg
(119.46 ± 13.86 and 151.20 ± 6.92 ^g/mL) and 100 mg/kg (278.08 ± 20.38 and
187.97 ± 15.90 [ig/mL). Cmax and Tmax values for rats and mice at 30 mg/kg appear similar; however,
the Tmax was higher in female mice than in male mice (the opposite relationship compared to rats).
3.1.2. Distribution
Burkemper etal. (20171 investigated the distribution of PFBA in male CD-I mice (n = 4)
given a single i.v. dose of radiolabeled PSFJ-PFBA (~0.074 MBq/|iL). At 4 hours postinjection, the
PSFJ-PFBA was detected in every tissue investigated, with most of the dose found in the stomach
(~7.5% injected dose/g). All concentrations in the blood, lung, liver, kidney, intestines, and skin
3-2
-------
Toxicological Review of PFBA and Related Salts
were similar (~2%-3%). Compared with perfluorooctanoic acid (PFOA) and perfluorohexanoic
acid (PFHxA), the concentration of PFBA was much lower in the liver (~27% and ~20%,
respectively). Chang etal. f20081 estimated volumes of distribution (Vd, mL/kg) for NH4+PFB in
male and female rats (209 ± 10 and 173 ± 21 at 30 mg/kg orally), mice (152 and 107 at 10 mg/kg
orally; 296 and 134 at 30 mg/kg orally), and cynomolgus monkeys (526 ± 68 and 443 ± 59 at
10 mg/kg i.v.) (N = 3 animals/sex/dose group for all species); these values indicate thatNH4+PFB is
primarily distributed in the extracellular space.
Distribution in rats and mice was also examined in multiple toxicological studies of PFBA
(see Table 3-1). Although limited in scope (i.e., PFBA was measured only in the liver and blood
serum), these studies demonstrated consistently that PFBA does distribute to the liver
compartment in both species. Butenhoff et al. f2012al observed that liver concentrations of PFBA
([ig/g) were higher in male and female S-D rats exposed to PFBA for 28 days vs. rats exposed for
90 days. The ratio between liver concentrations (|J.g/g) and serum concentrations ([ig/mL) ranged
from 26% to 47% in the 28-day rats and 16% to 31% in the 90-day rats. In both exposure groups,
the concentration of PFBA in the serum or liver was drastically reduced following a 3-week
recovery period.
Das etal. f20081 investigated the distribution of PFBA to the liver in both pregnant and
nonpregnant mice and in postnatal day (PND) 1 and PND 10 pups. Serum levels and liver levels of
PFBA appeared to be lower in nonpregnant mice compared to pregnant mice in the lowest two dose
groups, with mean serum concentrations approximately twofold higher in pregnant mice compared
to nonpregnant mice in the 35 mg/kg-day and 175 mg/kg-day dose groups (see Table 3-1). This
pattern also was observed for liver concentrations where pregnant animals had approximately two
to three times the liver concentration of PFBA compared to nonpregnant animals in the 35 mg/kg-
day and 175 mg/kg-day dose groups. However, these differences were not statistically significant
and are based on only two or three nonpregnant mice at each dose level, and serum and liver levels
were essentially identical between pregnant and nonpregnant mice at the high dose (350 mg/kg)
level. Additionally, the serum and liver concentrations of PFBA were attenuated in high-dose
(350 mg/kg) animals. Possible explanations for this pattern (with both liver and serum levels being
lower in non-pregnant than pregnant animals) would be higher oral absorption, lower clearance, or
higher distribution to other tissues (including fetuses and placenta) at the intermediate doses but
not at the higher doses in the pregnant mice. However, serum concentrations in PND 1 pups were
about 7-fold lower than the pregnant dams (end of gestation) and liver concentrations were 6-7
fold lower, indicating that distribution to the fetuses was not higher than distribution to other
maternal tissues. Since PFBA absorption data (see Section 3.1.1) are consistent with close to 100%
absorption, similar to other PFAS, it is not possible for absorption to be increased during
pregnancy. It is possible that clearance is reduced during pregnancy due to hormonal changes
affecting renal transporters, increasing resorption and hence internal doses, with this effect being
neutralized by saturation of the transporters at the highest doses. Pharmacokinetic data from
3-3
-------
Toxicological Review of PFBA and Related Salts
Chang etal. (20081 (see Section 3.1.4) are consistent with saturation of renal resorption in the
range of 3-100 mg/kg doses in female mice, which supports this possible explanation. However,
given the small sample-size of the non-pregnant animals in Das etal. f20081 and the fact that the
animals were dosed for 17 days, compared to the single doses used in PK studies, additional
experiments would be needed to validate and more accurately quantify any pregnancy-related
differences.
As would be expected, both the serum and liver concentrations in PND 1 pups were much
greater than those in PND 10 pups, since dosing ceased on GD 18 (Das etal.. 20081. Das etal. (20081
corroborated the observations by Butenhoff et al. (2012al and Chang etal. (20081 that serum PFBA
concentrations are higher than liver concentrations. The ratios of liver to serum PFBA
concentration observed in Chang etal. f20081 were 22%-27% in male rats, 20%-23% in male
mice, and 15%-17% in female mice. Interestingly, minimal differences in liver/serum
concentrations also were observed in various genetic strains of mice exposed to 35-350 mg/kg
PFBA: 34%-47% in wild-type mice, 19%-37% in peroxisome proliferator-activated receptor alpha
(PPARa) null mice, and 22%-37% in humanized PPARa mice (Foreman etal.. 20091. These results
suggest that PPARa status has minimal effect on the distribution of PFBA between liver and serum.
Table 3-1. Serum and liver concentrations of perfluorobutanoic acid (PFBA)
following subchronic or gestational exposure
Dose group
(mg/kg-d)
Serum (pg/mL)
Liver (pg/g)
Serum (pg/mL)
Liver (pg/g)
Pregnant mice Das et al. (2008)
Nonpregnant female mice Das et al. (2008)
0
0.002 ± 0.001
0.003 ± 0.002
0.006 ± 0.003
0.038 ± 0.017
35
3.78 ± 1.01
1.41 ±0.42
1.96 ± 1.0
0.51 ±0.20
175
4.44 ±0.65
1.60 ±0.25
2.41 ± 1.65
0.86 ±0.55
350
2.49 ±0.60
0.96 ±0.18
2.67 ± 1.2
0.89 ±0.38
PD1 male and female mice Das et al. (2008)
PD10 male and female mice Das et al. (2008)
0
Not detected
0.004 ± 0.001
0.002 ± 0.002
0.003 ± 0.001
35
0.56 ±0.15
0.22 ±0.05
0.11 ±0.03
0.04 ± 0.01
175
0.61 ±0.39
0.29 ±0.14
0.14 ±0.07
0.04 ± 0.02
350
0.37 ±0.14
0.24 ± 0.08
0.12 ±0.05
0.04 ± 0.02
28-d male rats Butenhoff et al. (2012a)
90-d male rats Butenhoff et al. (2012a)
0
0.04 ± 0.05
<0.05
<0.01
<0.05
1.2
-
-
6.10 ±5.22
1.34 ± 1.24
6
24.65 ± 17.63
7.49 ± 4.46
13.63 ±9.12
3.07 ± 2.03
30
38.04 ±23.15
17.42 ±8.15
52.22 ± 24.89
16.09 ± 9.06
150
82.20 ±31.83
37.44 ± 18.12
-
-
3-4
-------
Toxicological Review of PFBA and Related Salts
Dose group
Serum (ng/mL)
Liver (ng/g)
Serum (|ig/mL)
Liver (ng/g)
(mg/kg-d)
Pregnant mice Das et al. (2008)
Nonpregnant female mice Das et al. (2008)
28-d female rats Butenhoff et al. (2012a)
90-d female rats Butenhoff et al. (2012a)
0
0.01 ±0.01
0.05 ± 0.03
0.07 ± 0.06
<0.05
1.2
-
-
0.23 ±0.14
0.05 ± 0.02
6
0.34 ±0.13
0.16 ±0.04
0.92 ±0.52
0.15 ±0.08
30
1.72 ±0.88
0.434 ±0.174
5.15 ±3.29
0.91 ±0.55
150
10.30 ±4.50
2.70 ± 1.47
-
-
28-d PPARa +/+ mice Foreman et al. (2009)*
28-d PPARa -/- mice Foreman et al. (2009)*
35
87 ±27
30 ± 1.6
67 ± 14
13 ± 1
175
108 ±7
51 ±5
99 ± 16
36 ±7
350
115 ± 11
46 ±4
81 ±20
28 ±5
28-d hPPARa mice Foreman et al. (2009)*
35
59 ± 12
21 ±4
175
146 ± 20
32 ±4
350
35 ±5
9 ± 2
* Foreman et al. (2009) analyzed tissue concentrations in male wild-type (PPARa +/+), PPARa -/- and humanized
PPARa (hPPARa) mice on an Sv/129 genetic background.* Foreman et al. (2009) analyzed tissue concentrations
in male wild-type (PPARa +/+), PPARa -/- and humanized PPARa (hPPARa) mice on an Sv/129 genetic background.
Perez etal. T20131 investigated the distribution of PFBA in multiple tissues in cadavers in
Tarragona County, Spain. PFBA was detected in liver, brain, lung, and kidney samples, but was
below the level of detection in bone. Lung and kidney samples by far had higher PFBA
concentrations (304 and 464 ng/g, respectively) than brain or liver samples (14 and 13 ng/g,
respectively). For both the lungs and kidneys, PFBA was detected in greater quantities than any of
the other 20 per- and polyfluoroalkyl substances (PFAS) analyzed. The observation that PFBA was
observed in the greatest quantities in kidney samples could be related to kidney reabsorption.
Chang etal. f20081 observed that rats given 300 mg/kg PFBA orally excreted substantially greater
amounts of PFBA in the urine than did rats given 100 mg/kg (90.16% ± 2.75% vs. 50.99% ± 4.35%),
and the authors suggested this as evidence of saturation of a renal tubular reabsorption process.
Abraham etal. (2021) analyzed PFBA levels in lung and tissue samples collected from
tumor patients in France and observed concentrations approximately three to four orders of
magnitude lower than Perez etal. (2013): 0.08-0.24 ng/g in lung (n = 7) and 0.04-0.19 ng/g in
kidney (n = 9). These were different individuals living in a different country, so some difference in
exposure levels is expected. Additionally, tissue samples were obtained from cancer patients versus
people that died from trauma or ischemic heart disease, further complicating the comparison. But
given the relatively rapid clearance of PFBA compared to other PFAS, one would expect its tissue
levels to be lower than other PFAS, not the highest, and one would have to assume that exposure to
3-5
-------
Toxicological Review of PFBA and Related Salts
the subjects of Abraham etal. (20211 to be thousands of times lower than the subjects of Perez etal.
(20131 order to otherwise explain the difference. Bangma etal. (20211 determined that a saturated
oxo-fatty acid as an analytic interferent with PFBA in the placenta, indicating that it could also have
given falsely high measurements in the Perez etal. f201311ung samples. It is noted that Perez et al.
f20131 describe careful and fairly comprehensive quality-assurance (QA) methods employed, such
as use of matrix-matched calibration, while Abraham etal. f20211 does not report what QA/quality-
control (QC) methods were used. Given the QA of (Perez etal.. 20131 an interferent would need to
be present in human lung but not pig lung (species source of tissues used for QA) to result in large
over-estimates. Resolution of this discrepancy is beyond the scope of this review and likely requires
additional tissue samples from a larger population, preferably one with known PFBA exposure
levels in the weeks immediately prior to sampling, given the short half-life.
Data are not available that can be used reliably to estimate the volume of distribution (Vd) in
humans, which effectively provides the total body burden based on observed blood or serum
concentrations. An estimation of human body distribution for other PFAS is provided by the PBPK
models for PFOA and PFOS of Loccisano etal. (20111 which assume identical tissue :blood partition
coefficients (PCs) in humans and monkeys, equal to the values measured using tissues from rats
(PFOA) and mice (PFOS). This assumption is common to many PBPK models, based on the
expectation that the biochemical properties of a given tissue, muscle for example, which determines
the relative affinity of a chemical for that tissue compared to blood, are similar across mammalian
species: mouse, rat, monkey, and human muscle are all similar in composition and the difference in
chemical distribution to muscle as a whole is determined by the difference in the volume of muscle
per kg BW between species.
PCs are the effective tissue specific Vd values because they determine the ratio of the
amount in a tissue vs. blood concentration at equilibrium. Based on this PBPK model Loccisano et
al. f20111. the Vd for PFOA predicted in monkeys and humans is 0.210 and 0.195 L/kg, respectively,
and for PFOS is 0.333 and 0.322 L/kg, respectively. These predictions are obtained by summing the
tissue fractions (ratios of tissue volumes/BW) multiplied by the corresponding PCs. In comparison,
based on the Loccisano etal. (20111 model for adult rats, the corresponding Vd values in that
species, for PFOA and PFOS, are 0.290 and 0.398, respectively. The difference between these rat
values and the human and monkey values is primarily due to the difference in physiology,
specifically the proportion of BW that is liver, kidney, and other tissues. Because of the
physiological similarities between humans and monkeys (more similar tissue fractions), the
predicted Vd values are within 7% of each other, although the difference between human and rat Vd
values is predicted to be 49% for PFOA and 24% for PFOS. They are much more similar between
humans and monkeys than between humans and rats, but the difference between humans and rats
is still less than a factor of 1.5.
Li etal. f20201 evaluated the transplacental transfer of multiple PFAS, including PFBA, in
human preterm vs. full-term births, and evaluated the data for correlation with the expression of
3-6
-------
Toxicological Review of PFBA and Related Salts
nine placental transporters. The transplacental transfer efficiency (TTE) was calculated as the ratio
of PFAS concentration in cord serum, collected at the time of birth, to the concentration in maternal
serum collected within 1 week of (prior to) birth. The median TTE for preterm births was 0.48, with
first and third quartiles of Q1 = 0.27 and Q3 =1.06 (n = 33) Li etal. f20201. hence the distribution in
the preterm fetus was predominantly less than one though it may not be significantly so. This result
is qualitatively consistent with the observations of Das etal. f20081 in mice, described above.
However, the human TTE was observed to increase to a median value of 1.06 in full-term deliveries,
with the difference between preterm and full-term indicated as significant (Li etal.. 2020). This
result is consistent with a possible loss of integrity of the placenta as a passive barrier to PFAS
transport occurring towards the end of pregnancy, as discussed by the authors, since the TTE did
not show a significant correlation with any of the transporters evaluated fLi etal.. 20201.
The extent to which the volume of distribution may change during pregnancy in humans
has not been evaluated. Based on data reported by Kuczmarski etal. (2000) an average woman
gains about 25% of her initial body-weight during pregnancy. If the data of Li etal. (2020) can be
interpreted as showing that distribution to this additional mass is about one half of distribution to
other maternal tissues, then the total volume of distribution in the pregnant mother (L) would
increase about 12.5% while her mass increases 25%, leading to Vd in late pregnancy of 112.5/125 =
90% of non-pregnant V& which is not a significant change. Even if distribution to the fetus was
much lower in the human fetus than the mother, her Vd would decrease by no more than 20%.
While such a change may be marginally significant, it is still well within the overall uncertainty for
estimates of Vd. Another factor during human pregnancy is the decrease in serum proteins, with the
decrease in albumin concentration being consistent with dilution of the protein into an increased
total plasma volume (Paabv. 1960). Such a decrease could lead to both an increased Vd of PFBA,
since a smaller fraction would be bound in blood, and an increase in clearance. Since the reduction
in protein concentration is on the order of 10%-20% fPaabv. 19601 like the potential change due to
the growth of the fetus discussed just above, the impact is not expected to be large, and it is in the
opposite direction of that effect Hence, while pregnancy-related factors specific to distribution may
cause some change in distribution, this change is not expected to be significant. On the other hand,
if hormonal changes increase renal resorption during pregnancy, as is suggested in mice
(discussion above), that could significantly increase maternal body burden during that time.
Measurements of clearance in pregnant vs. non-pregnant women (i.e., using matched blood and
urine samples) would be needed to determine if such a difference exists.
Based on this analysis for PFOA and PFOS, the most reasonable choice for estimation of Vd
for PFBA in humans is to assume that it is similar to the Vd estimated for PFBA in monkeys, rather
than values estimated for mice or rats.
It is recognized that the distribution of PFAS depends on the extent of binding to various
proteins and partitioning into phospholipid membranes. Chen and Guo (2009) measured the
binding of PFBA to human serum albumin and obtained a binding constant of (1.1 ± 0.1) x 106 M1
3-7
-------
Toxicological Review of PFBA and Related Salts
for albumin site I with no observed binding to the Trp site or site II. However, corresponding
measures of phospholipid partitioning and binding to cellular proteins are not available, so it is not
possible to estimate the extent to which these contribute to tissue partitioning.
3.1.3. Metabolism
PFBA has been shown to be a product of the metabolism of 6:2 FTOH in mice, rats, and
humans (Russell et al.. 2015b: Ruan etal.. 20141. No evidence of biotransformation for PFBA,
however, was found. PFBA, a short-chain (C4) of perfluoroalkyl acids (PFAAs), is expected to be
metabolically inert because its chemical stability is the same as longer chain PFAA chemicals,
including perfluorohexane sulfonate (PFHxS, C6), perfluorooctane sulfonate (PFOS, C8), and PFOA,
C8.
3.1.4. Excretion
In an overview of the toxicology of perfluorinated compounds, Lau (20151 briefly
summarized the excretion half-lives of seven compounds, including PFBA. All supporting data for
that review pertinent to PFBA are included in this analysis.
Chang etal. f20081 investigated the excretion of PFBA in S-D rats, CD-I mice, cynomolgus
monkeys, and workers occupationally exposed to PFBA, or compounds metabolized to PFBA. For
rats and monkeys, three animals per sex were used (rats: three animals each for i.v. and oral
dosing) at the single dose given to each. For mice, three animals per sex per time point were used at
each dose, or 15-18 animals/dose. OECD guidelines state that a minimum of four animals per sex
per dose should be used (OECD. 20101. Thus, the rat and monkey studies fall short of this standard.
For rats, however, the average clearance from the two routes of exposure is proposed to best
represent males and females of that species (details below), which is then based on data from six
animals per sex. For monkeys, the average volume of distribution for both males and females are
used as an estimate for that value in humans, again incorporating data from six animals. Therefore,
these data are presumed sufficient for the specific parameters being estimated. In S-D rats exposed
orally to 30 mg/kg PFBA, a marked difference was noted in the serum PFBA excretion constants (A)
between males and females, 0.075/hour and 0.393/hour, respectively, for oral exposure and
0.109/hour and 0.673/hour, respectively, for intravenous exposure (see Appendix C for a complete
discussion on whether the calculated elimination constants in various species are mono- or
biphasic). The difference in oral A resulted in half-lives (ti/2) of 9.22 and 1.76 hours, respectively,
for males and females. Chang etal. f20081 reported clearance (CL) values as mL/hour, not
normalized to BW, but the normalized average CL can be calculated as dose/AUC, using the AUC
values they reported. For oral doses in male and female rats, the result CLs are 0.38 and 1.6 L/kg-
day, respectively, while for i.v. doses they are 0.66 and 3.0 L/kg-day, respectively.
Russell etal. (2015b) attempted to evaluate the excretion of PFBA, formed as a metabolite
of 6:2 FTOH, after inhalation exposures in rats (strain not stated). In single-day studies, the animals
were exposed by inhalation for 6 hours and their blood levels monitored for 24 hours after start of
3-8
-------
Toxicological Review of PFBA and Related Salts
exposure. The decline in PFBA blood concentration was negligible, however, after 0.5 and 5 ppm
6:2 FTOH exposures in male rats and after 0.5 ppm exposure in female rats, precluding estimation
of half-life. An excretion half-life of 19 hours was estimated from the 5-ppm single-day data for
5 ppm in female rats. After a 2 3-day inhalation exposure to male rats, use of a PK model resulted in
estimation of a 27.7-hour half-life for that sex, which could explain the inability to estimate a half-
life from the single-day exposures. Both estimates depend on the estimated yield (percent of 6:2
FTOH metabolized to PFBA), however, which was 0.2% for male rats and 0.02% for female rats.
Given the low yields, small errors in the estimate of that parameter could result in significant errors
in the estimated half-life. Thus, the results of Chang etal. (20081 is used to represent excretion in
rats.
In male CD-I mice, the clearance was similar in mice exposed to 10 mg/kg
(0.35 ± 0.09 mL/hour) and 30 mg/kg PFBA (0.37 ± 0.80 mL/hour); however, clearance at
100 mg/kg was much higher (0.98 ± 0.14 mL/hour) (Chang etal.. 2008). Although the fit of the
simple one-compartment model used to describe the kinetic data appeared adequate for the two
lower doses, itunderpredicted the data at 24 and 48 hours for the 100 mg/kg dose, indicating it
was not sufficient for this highest exposure. In female mice clearance showed a similar, but less
strong pattern, with values of 0.76 ± 0.03, 0.87 ± 0.04, and 1.67 ± 0.08 mL/hour at 10, 30 and
100 mg/kg doses, respectively f Chang etal.. 20081. Unlike the data for male mice, the female mouse
data were fit well by the one-compartment PK model. For female data, the possible dose-
dependence can be resolved by using the average clearance for the lower two doses, which are
closer to the doses evaluated for point-of-departure (POD) determination. Because male mouse
endpoints are not considered for POD determination, an alternative PK analysis of these data is not
supported.
Using dose/AUC, the corresponding CL values are 0.23, 0.25, and 0.66 L/kg-day in male
mice at 10, 30, and 100 mg/kg, respectively, and 0.62, 0.72, and 1.36 L/kg-day in female mice,
respectively.
Cynomolgus monkeys (N = 3/sex) displayed a clear biphasic excretion pattern, with a rapid
decline in the initial (a) phase and a slower decline in the second ((3) phase (Chang etal.. 2008).
Notably, the (3 phase began at around 24 hours and was observed because samples also were taken
at 2, 4, 7, and 10 days, while in rodents, samples were reported only to 24 hours (rats and female
mice) or 48 hours (male mice). Whereas serum levels in female rats and mice dropped to less than
3% of peak concentration by 24 hours, indicating minimal longer-term elimination, the levels in
male mice and rats did not drop as quickly and are more suggestive of a (3 phase. Also noted is that
the mouse and rat PK plots in Chang etal. (2008) use a linear y-axis, while the monkey PK plots use
a log y-axis. That a (3 phase would have been clearly observed in male mice and rats is possible had
serum sampling been continued for a longer duration, and possibly in female mice and rats had the
data simply been plotted with a log y-axis. Serum excretion half-lives for the a and (3 phases in male
monkeys exposed to 10 mg/kg PFBA via i.v. injection were 1.61 ± 0.06 hours and
3-9
-------
Toxicological Review of PFBA and Related Salts
40.32 ± 2.36 hours, respectively; ti/2 values in female monkeys were 2.28 ± 0.14 hours and
41.04±4.71 hours, respectively.
Excretion of PFBA from the serum in humans also was investigated by Chang etal. f2008I
In the initial occupational study, baseline PFBA serum concentration was determined in male
workers [n = 3) exposed to either PFBA or related fluorinated compounds. Following voluntary
removal from the workplace, workers had blood samples taken over 8 days to estimate half-lives of
excretion. Given the small sample size of the initial occupational study, a second study was
conducted in which seven male and two female workers had blood samples taken immediately
before a vacation and upon returning to the production facility (minimum elapsed time was
7 days). For the male workers in the initial study, ti/2 of excretion from the serum ranged from 28.6
to 109.7 hours (1.2 to 4.6 days). For the nine workers in the second study, the ti/2 ranged from 44 to
152 hours (1.9 to 6.3 days), with an average value of 72 hours (95% confidence interval
[CI]: 1.8-4.2 days). Because these workers had been exposed previously for a significant duration
and the PK study was conducted over periods ranging from 7 to 11 days, the observed elimination
is reasonably presumed to represent (3-phase elimination, rather than the initial distribution phase.
Although only two female subjects were included in the second study (and their final PFBA serum
concentrations fell below the limit of detection), their estimated ti/2 values (118 hours and 56
hours) fell within the range of ti/2 values reported for males (44-152 hours). Therefore, although
sex differences in serum excretion in rodent species appear strong, the data in cynomolgus
monkeys and humans do not indicate such a difference.
Measurements for four of the subjects evaluated by Chang etal. (2008) fell below the lower
limit of quantification (LLOQ) when the second blood sample was taken, requiring the authors to
assume a value of LLOQ/V2 for those values. This approach introduces considerable uncertainty, so
the population half-life excluding those individuals was estimated as described in Appendix C.2 to
obtain a half-life of 67.9 hours (rather than the author-reported arithmetic mean value of 72 hours).
This revised estimate will be used for subsequent analysis.
Using an assumed BW0 75 scaling and standard species BWs of 0.25 kg in rats and 80 kg in
humans, the half-life in humans is predicted to be 4.2 times greater than in rats. Given half-lives of
9.22 and 1.76 hours, respectively, in male and female rats (oral dose values), one would then
predict half-lives of 37.8 hours in men and 7.2 hours in women. Although the value for men based
on the BW°75 scaling approach is within a factor of 2 of the value determined by Chang etal. f20081.
BW°75 scaling is not based on data for this class of chemicals (i.e., serum binding and clearance
mechanisms are known to occur for PFAS). For example, EPA's Recommended Use of Body Weight
3/4 as the Default Method in Derivation of the Oral Reference Dose (U.S. EPA. 2011) does not mention
serum binding; it does include references related to VOCs, drugs, and overall metabolism (with
metabolism a significant component in the clearance of many other toxic chemicals) but does it cite
papers evaluating the pharmacokinetics of PFAS. These results for PFBA indicate thatBW0 75 scaling
would lead to a lower prediction of human health risk at a given exposure than dosimetric scaling
3-10
-------
Toxicological Review of PFBA and Related Salts
based on the empirical data. Further, although only two women participated in the Chang et al.
(20081 study, that the observed elimination for them was 8 and 16 times slower than predicted by
BW0-75 is an unlikely occurrence—even given the small sample size—and using of BW0-75 scaling
(applied to the half-life in female rats) could underpredict the risk of exposure by an order of
magnitude. Therefore, use of BW0 75 as an alternative means of extrapolation is not considered
further here.
Excretion in the urine appears to be the major route by which PFBA is excreted from the
body. Female rodents (rats: 100.68%-112.37%; mice: 65.44%-67.98%) are observed to have
higher percentages of the dose excreted in urine at 24 hours compared to male rodents (rats:
50.99%-90.16%; mice: 34.58%-35.16%). This is consistent with evidence thatorganic anion
transporters (OAT) expressed in the kidneys of rodents reabsorb PFAS fWeaver etal.. 2010: Yang et
al.. 20091 and are more highly expressed in male rodents fLiuboievic etal.. 2007: Liuboievic etal..
2004: Buist etal.. 2002: Cerrutti etal.. 2002: Kato etal.. 20021. Both Yang etal. (20091 and Weaver
etal. (20101. however, observe that PFBA is not an active substrate of organic anion transporters
OAT1, OAT2, or OATPlal. Therefore, although the observed sex difference in urinary excretion of
PFBA is consistent with the literature for reabsorption of PFAS in general in the kidney in male
rodents, the mechanism for this reabsorption for PFBA specifically is not currently known. Sex
differences in urinary excretion rates are not observed in primates, with both female and male
cynomolgus monkeys having rates similar to those of male mice (36.2% and 41.69%, respectively)
Chang etal. (20081. The excretion of PFBA in feces in rats and mice was very low compared with
the excretion in urine, but higher in mice than in rats (4.10%-10.92% and 0.16%-2.99%,
respectively).
3.1.5. Summary
PFBA clearance (CL) data, which can be used to estimate the average blood concentration
for a given dose, are available for mice and rats. For mice, the average CL from PK experiments at
10 and 30 mg/kg is suggested for use in animal-human extrapolation. For rats, the average of
values estimated from i.v. and oral exposure to 30 mg/kg is suggested.
Direct comparison of animal and human data requires consideration of observed half-lives
because such data are available in humans, but CL cannot be directly estimated in humans.
Collectively, although the PFBA excretion half-lives for male and female rats appear shorter than for
male and female mice, respectively, data suggest a strong sex-specific pharmacokinetic difference
for both species (i.e., females appear to have a much faster excretion rate than males). Humans have
a longer serum excretion half-life (~day) than rodents (~hour). Although data in male mice and
rats might indicate a longer (3 phase elimination, the lower dose data in male mice are reasonably fit
using a single half-life (one-compartment model) as are the i.v. and oral data at the single dose
given to rats (30 mg/kg); the female mouse and rat data are likewise fit well by a one-compartment
model f Chang etal.. 20081. Therefore, although a longer elimination phase might be evident if
additional data were available, the estimated total clearance is unlikely to differ substantially from
3-11
-------
Toxicological Review of PFBA and Related Salts
the estimates provided here. The a-phase half-lives in monkeys (1.6-2.3 hours) are similar to the
half-life obtained for female mice (2.8-3.1 hours) and female rats (1-1.8 hours) but are
substantially shorter than the half-life observed in male mice (13-16 h at lower doses) and male
rats (6-9 hours). The (3-phase half-life in monkeys (1.7 days) is considerably longer than any of
these rodent values but is comparable to the lower end of the range for human subjects (1.8-2
days), although roughly one-half the average among humans (3 days). As noted above, these human
half-lives are expected to represent (3-phase, considering the period of observation vs. exposure.
Human CL can be estimated using the PK relationship, CL = Vd- ln(2)/to s. Because human
data do provide a value of ti/2, only a value of Vd is needed to determine CL. As discussed above,
however, one can reasonably anticipate that Vd in humans is similar to that in other primates based
on the similarity in physiology and assumptions common to PBPK modeling. This similarity is
illustrated on the basis of PBPK models for PFOA and PFOS Loccisano etal. f20111 from which Vd in
humans is predicted to be within 7% of the value for monkeys for those two PFAS. Thus, this choice
seems appropriate for estimating human clearance of PFBA. Using the average human half-life of
67.9 hours (2.8 days) from Appendix C.2 and average of male and female monkey Vd of 0.485 L/kg
from Chang etal. (2008) the resulting human clearance is 0.12 L/kg-day.
Table 3-2 provides a summary of PFBA pharmacokinetics.
Table 3-2. Summary of pharmacokinetics of serum perfluorobutanoic acid
(PFBA) (mean ± standard error)
Species/
sex
Study design
Excretion
half-life (h)
AUC
(pg-h/mL)
Clearance
(mL/h)
Clearance
(L/kg-d)a
Volume of
distribution
(mL/kg)
Rats
Male
30 mg/kg i.v. dose
6.38 ±0.53
1,090 ± 78
7.98 ±0.57
0.661
253 ±6
30 mg/kg oral dose
9.22 ±0.75
1,911 ± 114
4.63 ±0.28
0.377
209 ± 10
Female
30 mg/kg i.v. dose
1.03 ± 0.03
239 ±5
27.65 ±0.55
3.01
187 ±3
30 mg/kg oral dose
1.76 ±0.26
443 ± 42
14.32 ± 1.36
1.63
173 ± 21
Mice
Male
10 mg/kg oral dose
13.34 ±4.55
1,026 ± 248
0.35 ±0.09
0.234
152
30 mg/kg oral dose
16.25 ±7.19
2,869 ±6,116
0.37 ±0.80
0.251
296
100 mg/kg oral dose
5.22 ±2.27
3,630 ± 530
0.98 ±0.14
0.661
207
Female
10 mg/kg oral dose
2.87 ±0.30
387 ± 14
0.76 ±0.03
0.620
107
30 mg/kg oral dose
3.08 ±0.26
999 ± 42
0.87 ± 0.04
0.720
134
100 mg/kg oral dose
2.79 ±0.30
1,760 ± 88
1.67 ±0.08
1.36
207
Monkeys
Male
10 mg/kg i.v. dose
1.61 ±0.06 (a)
112 ±6
494 ± 61
2.14
526 ± 68
3-12
-------
Toxicological Review of PFBA and Related Salts
40.32 ±2.36 (P)
Female
10 mg/kg i.v. dose
2.28 ±0.14 (a)
41.04 ±4.71 (P)
159 ±8
224 ± 19
1.51
443 ± 59
Humans
Males and
females
NV
Study 1:
28.6-109.71
Study 2: 72
(mean)
NV
NV
NV
NV
AUC = area-under-the-concentration-curve, NV = not available.
All data from Chang et al. (2008).All data from Chang et al. (2008).
Calculated as dose (mg/kg) x (1,000 ng/mg) x (24 h/d) / ((AUC ng-h/mL) x (1,000 mL/L)).
The mouse PK data of Chang etal. f20081 clearly indicate nonlinear elimination, with more
rapid clearance at higher concentrations consistent with a mechanism of saturable renal resorption.
Since there is only a modest difference in clearance between the lowest two doses (10 and 30
mg/kg-day) it is reasonable to assume first-order elimination around and below these dose levels
in mice. However, use of the low-dose clearance for effects associated with higher doses is likely to
over-predict the corresponding HED, since mouse clearance is higher at higher exposures.
Unfortunately, the single dose used for PK in rats is not sufficient to demonstrate when
saturation might occur in that species. The data and model fits shown by Chang etal. f20081.
particular for the i.v. administration, appear quite consistent with first-order elimination assumed
in their analysis. Hence, for the purposes of the current analysis, it is assumed that the estimated CL
is applicable to 30 mg/kg-day doses or below and to avoid extrapolation above that dose.
Some mechanistic insight can be gained by comparing the clearance values described above
with species-specific glomerular filtration rate (GFR), with and without adjustment for serum
protein binding. Davies and Morris T19931 summarized GFR for multiple species. Considering the
time period when those data were collected, it seems appropriate to use the species average BW
values listed in Table III of Davies and Morris f 19931: 0.02 kg for the mouse, 0.25 kg for the rat, and
70 kg for the human. Using those, the GFR/BW for these species are 20.2 L/kg-day in mice, 7.55
L/kg-day in rats, and 2.57 L/kg-day in humans, which are, respectively, 83 and 32 times higher than
PFBA clearance in male and female mice (average of values at lowest two doses), 14.5 and 3.3 times
higher than the average for male and female rats from Chang etal. (20081. and 21 times higher than
the human PFBA clearance estimated above.
Binding to serum proteins plays a likely role in these very large differences. Chen and Guo
f20091 measured the binding of PFBA to human serum albumin and obtained a binding constant of
(1.1 ± 0.1) x 106 M1 for albumin site I with no observed binding to the Trp site or site II. Using a
representative serum albumin concentration of 40 mg/mL = 6 x 10 4 M, the predicted free fraction
of PFBA is/free = 0.0015. This binding may play a role in the limiting the rate of the renal excretion of
PFBA, in addition to the role played by renal transporters. Using this value, GFR x/free = 0.03 L/kg-
day in mice, 0.01 L/kg-day in rats, and 0.004 L/kg-day in humans. The measured CL for male mice
3-13
-------
Toxicological Review of PFBA and Related Salts
(0.23-0.66 L/kg-day) is 8- to 22-fold higher than GFRx/free and the CL for female mice (0.62-1.36
L/kg-day) is 21- to 45-fold above GFRx/free. Even more significantly, CL in rats is as much as 300-
fold higher than the corresponding GFRx/free, and the estimated CL in humans (0.12 L/kg-day) is
30-fold higher than the corresponding GFRx/free. The source of these apparent discrepancies is
unclear. It is reasonable to expect that plasma protein binding will limit the clearance of PFBA.
However, these results indicate that either/free is significantly under-estimated or that clearance is
not strictly limited to the free fraction (estimated from an in-vitro binding constant). Binding and
dissociation are dynamic processes, and it may be that as blood passes through the glomerulus and
filtration occurs, some portion of the albumin-bound PFBA is sufficiently labile to dissociate and
also be cleared. A mathematical model that incorporates the kinetics of plasma binding and release
to describe uptake of drugs by the brain has been previously described by Robinson and Rapoport
fl9861. but adaptation of this model to renal clearance of PFBA would require measurement of the
separate rates of association and dissociation, data which have not been reported.
Another possible explanation is from imperfect filtering of albumin by the glomerulus,
leading to some urinary excretion of albumin which may carry bound PFBA. Van Camp et al. (1990)
observed an albumin excretion rate in female rats on normal diets (i.e., control animals) of about 1
mg/day, which corresponds to a clearance of 0.025 mL/day given a serum albumin concentration of
40 mg/mL. The urine samples were collected at the mid-point of the experiment Based on the BW
reported on the first and final days of the experiment, the rats at this time were around 0.14 kg,
hence had an albumin CL of 1.8 x 10 4 L/kg-day; i.e., about four orders of magnitude lower than the
PFBA CL in female rats (see Table 3-2). While kidney damage is known to increase albumin
excretion (for example, a high phosphate diet increased albumin excretion in female rats about 50-
fold (Matsuzaki etal.. 2002)). an increase of 10,000-fold occurring within the 24-hour time-frame of
the PK experiments, when kidney toxicity has not been reported for PFBA exposure in rats, seems
rather unlikely. However, if only 5% of the bound PFBA is sufficiently labile to be available for
clearance, that would be consistent with the empirical data and estimated clearance rates.
3.2. NONCANCER EVIDENCE SYNTHESIS AND INTEGRATION
For each potential health effect discussed below, the synthesis describes the database of
available studies and the array of the experimental animal study results (the primary evidence
available for this PFAS) across studies. Effect levels presented in these arrays are based on
statistical significance9 or biological significance, or both. Examples relevant to interpretations of
biological significance include directionality of effect (e.g., statistically significantly decreased
cholesterol/triglycerides are of unclear toxicological relevance) and tissue-specific considerations
for magnitude of effect (e.g., statistically nonsignificant increase of >10% in liver weight might be
considered biologically significant). A significant finding at a single, lower dose level but not at
9In this review, "statistical significance" indicates a p-value < 0.05, unless otherwise noted.
3-14
-------
Toxicological Review of PFBA and Related Salts
multiple, higher dose levels might be interpreted as potentially spurious. For this section, evidence
to inform organ/system-specific effects of PFBA in animals following developmental exposure is
discussed in the individual organ/system-specific sections (e.g., liver effects after developmental
exposure are discussed in the liver effects sections). Evidence of other effects informing potential
developmental effects (e.g., vaginal opening, eyes opening) is discussed in the "Developmental
Effects" section.
3.2.1. Thyroid Effects
Human Studies
Two studies reported on the association between PFBA exposure and thyroid hormones or
disease. One study on congenital hypothyroidism was considered uninformative10 due to concerns
with participant selection, confounding, and exposure measurement fKim etal.. 20161. In one low
confidence study Li etal. (2017b) examining thyroid hormones among participants without thyroid
disease, inverse associations with thyroxine (T4), free triiodothyronine (T3), and
thyroid-stimulating hormone (TSH) were reported. Among the thyroid hormones measured, only
TSH demonstrated a statistically significant association (Pearson correlation coefficient = -0.348,
p<0.01).
Animal Studies
Two high confidence studies reported in two unpublished reports and one publication from
the same research group evaluated the effects of PFBA exposure on the thyroid, specifically
hormone levels, histopathology, and organ weight (Butenhoff etal.. 2012c: Butenhoff etal.. 2012a:
van Otterdiik. 2007a. b) following oral exposure (via gavage) of SD rats.11 Some outcome-specific
considerations for study evaluations were influential on the overall study rating for thyroid effects,
but none of these individual domain-specific limitations were judged likely to be severe or to have a
notable impact on the study results; all studies considered further in this section were rated as high
or medium confidence (see Figure 3-1). For more information on outcome-specific considerations
for study evaluations, please refer to the study evaluations in the HAWC PFBA project page.
10Clicking on the hyperlinked study evaluation determination will take users to the HAWC visualization for
that study evaluation review. From there, users can click on individual domains to see the basis for that
decision. In the subsequent hazard sections, hyperlinked endpoint names will take users to the HAWC
visualization for that endpoint, from which users can click on the endpoint or studies to see the response data
from which the visualization is derived.
1 'The Butenhoff et al. (2012a) study reported the findings of two unpublished industry reports: a 28-day and
90-day gavage study fully reported in (van Otterdiik. 2007a. b). These industry reports were conducted at the
same facility and largely by the same staff but independently of one another and at different times: July 2 6,
2006, through September 15, 2006, for the 28-day study and April 5,2007, through August 6,2007, for the
90-day study. Throughout the Toxicological Review, both fButenhoff et al.. 2012c: Butenhoff et al.. 2012a"! and
the relevant industry report are cited when discussing effects observed in these reports. Although only one
study evaluation was performed for this group of citations in HAWC, the overall confidence level of high
applies to both the 28-day and 90-day reports.
3-15
-------
Toxicological Review of PFBA and Related Salts
Reporting quality A
B Legend
Good (metric) or High confidence (overall)
+ Adequate (metric) or Medium confidence (overall)
Allocation A
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
Not reported
Confounding/variable control
Observational bias/blinding
Chemical administration and characterization A
Selective reporting and attrition A
Exposure timing, frequency and duration
Endpoint sensitivity and specificity
Results presentation -
Overall confidence A
Figure 3-1. Evaluation results for animal studies assessing effects of
perfluorobutanoic acid (PFBA) exposure on the thyroid (see interactive data
graphic for rating rationales! .
Organ weight
Absolute and relative thyroid weights were statistically significantly (p < 0.01) increased
(~2-fold) at the end of treatment in male rats exposed to 6 or 3 0 mg/kg-day via oral gavage for
28 days compared with controls. Organ weights, however, were increased only ~50% at
150 mg/kg-day, and this difference was not statistically significant. Thyroid weights were not
significantly increased in male rats following the recovery period or in female rats following the
treatment or recovery period. Thyroid weight was not measured in the rats exposed to NH4+PFB for
90 days fButenhoffetal.. 2012a: van Otterdiik. 2007b],
Thyroid hormones
Male rats exposed to NH4+PFB for 28 days via gavage exhibited significantly decreased total
thyroxine fT41 and free T4 ffT41 levels compared with controls (see Table 3-3 and Figure 3-2).
Total T4 was reduced 59%, 66%, and 79% and free T4 was reduced 46%, 50%, and 66% at 6, 30,
and 150 mg/kg-day, respectively fButenhoffetal.. 2012a: van Otterdiik. 2007a], Free T4
concentrations had returned to control levels at all doses 21 days after exposure ended, but total T4
levels remained decreased in the 150 mg/kg-day group (-23%). TSH levels were not affected by
NH4+PFB at any exposure level. No treatment-related effects on any of the thyroid hormone
measures were observed in female rats exposed for 28 days fButenhoffetal.. 2012a: van Otterdiik.
200Za)
3-16
-------
Toxicological Review of PFBA and Related Salts
Table 3-3. Percent change in thyroid hormones due to perfluorobutanoic acid
(PFBA) exposure in short-term and subchronic oral toxicity studies
Animal group
Dose (mg/kg-d)
1.2
6
30
150
Free T4
28 d; male S-D rats
Butenhoff et al. (2012a)
-46
-50
-66
28 d; female S-D rats
Butenhoff et al. (2012a)
-0.5
+18
-25
90 d; male S-D rats
Butenhoff et al. (2012a)
a
-9b
-30b
90 d; female S-D rats
Butenhoff et al. (2012a)
-6
+27
-15
Total T4
28 d; male S-D rats
Butenhoff et al. (2012a)
-59
-66
-79
28 d; female S-D rats
Butenhoff et al. (2012a)
-8
+27
-31
90 d; male S-D rats
Butenhoff et al. (2012a)
+13
-15
-39
90 d; female S-D rats
Butenhoff et al. (2012a)
+16
+14
-21
Bolded cells indicate statistically significant changes compared to controls (except for the 6 mg/kg-d and 30 mg/kg-
d dose groups for free T4 in male rats exposed for 90 d, tests for statistical significance in those cases were made
to the 1.2 mg/kg-d group [see footnote b]); shaded cells represent doses not investigated in the individual
studies.
aNo sample for the control group was available due to insufficient sample volume for assay.
bComparison is made to the 1.2 mg/kg-d dose group.
Decreased total T4 and free T4 levels also were observed in male rats exposed to NH4+PFB
via gavage for 90 days fButenhoffetal.. 2012a: van Otterdiik. 2007bl. Total T4 increased 13% and
decreased 15% following 1.2 and 6 mg/kg-day, respectively. In male rats exposed to the highest
dose tested (30 mg/kg-day NFU+PFB), total T4 was significantly reduced by 39%. Free T4 was also
reduced in the 30-mg/kg-day dose group, but comparison to a control group was not possible due
to insufficient sample volume in the control group. The decrease in free T4, however, appeared to
be monotonic with increasing dose, and the decrease in the 30-mg/kg-day group (30%) was
statistically significant compared with the free T4 concentration in the 1.2 mg/kg-day group. No
statistically significant treatment-related effects were observed in female rats exposed to NH4+PFB
for 90 days, although total T4 was nonsignificantly decreased at the highest dose [30 mg/kg-day;
fButenhoffetal.. 2012a: van Otterdiik. 2007b]].
3-17
-------
Toxicological Review ofPFBA and Related Salts
Study Name Endpoini N a me
Study Type
Animal Description
Observation Tune
BulenhofT. 2012,1289835 Total Thyroxine (TT4)
28 Day Oral
Rat, Sprague-Dawtey (5)
28.0 days
49.0 days
90 Day Oral
Rat, Sprague-Dawtey {£)
90.0 days
111.0 days
28 Day Oral
Rat, Sprague-Dawley (2)
28 0 days
49 0 days
90 Day Oral
Rat. Sprague-Dawley {£)
90.0 days
111.0 days
Free Thyroxine (T4)
28 Day Oral
Rat. Sprague-Dawley {£)
28.0 days
49 0 days
90 Day Oral
Rat, Sprague-Dawtey {£)
90.0 days
111.0 days
28 Day Oral
Rat, Sprague-Dawley {£)
28.0 days
49.0 days
90 Day Oral
Rat, Sprague-Dawtey {£)
90.0 days
111.0 days
Thyroid Stimulating Hormone (TSH)
28 Day Oral
Rat, Sprague-Dawtey {£)
28.0 days
49 0 days
90 Day Oral
Rat, Sprague-Dawtey {*)
90.0 days
111.0 days
28 Day Oral
Rat, Sprague-Dawley (2)
280 days
490 days
90 Day Oral
Rat, Sprague-Dawtey {£)
90 0 days
PFBA Thyroid Hormone Effects
•v
v
# Doses
A Treatment-Reieated Increase
V" Treatment-Related Decrease
hH Dose Range
••
••
-~
-•
-•
-•
-•
-•
60 SO 100
Axis label
Figure 3-2. Thyroid hormone response to ammonium perfluorobutanoate
(NH4+PFB) exposure (see interactive data graphic and rationale for study
evaluations for thvroid hormone effects in Health Assessment Workspace
Collaborative [HAWC]).
Histopathology
Butenhoff et al. (2012al: van Otterdiik (2007a. 2007b) also investigated thyroid
histopathological and histomorphological effects in male and female rats resulting from NH4+PFB
exposure (see Table 3-4 and Figure 3-3). Incidence of follicular hypertrophy /hyperplasia increased
in males exposed to 30 mg/kg-day (9/10) and 150 mg/kg-day (7/10) for 28 days compared with
control (3/10), with all observed lesions in the 30 mg/kg-day dose group graded by the study
authors as "minimal" severity (trend test p = 0.0498; Cochran-Armitage test, performed by EPA). In
the 150 mg/kg-day dose group, three of the seven affected animals were observed to have lesions
graded as "slight," a severity level greater than "minimal"; the remaining four affected animals were
graded as having "minimal" lesions. Female rats treated for 28 days with 150 mg/kg-day Nil rPFB
had 40% incidence (4/10) of minimal lesions compared with 3/10 minimal lesions observed in the
control group. Thyroid histopathology was not examined in the 30-mg/kg-day females and no
effects were noted in the 6-mg/kg-day group (although the thyroid of only one animal was available
for testing in this group). No treatment-related effects were observed in the recovery groups. In
contrast to the histopathological examination, the histomorphometric analysis reported no effects
on thyroid cell height or colloidal area in either the treatment or recovery groups. Follicular
hypertrophy/hyperplasia also was observed to increase in male rats exposed to 30 mg/kg-day
3-18
-------
Toxicological Review ofPFBA and Related Salts
(9/10) for 90 days compared to controls when considering all lesions (9/10 vs. 4/10; Cochran
Armitage trend p = 0.0108) and lesions were graded "slight" (5/10 vs. 0/10; Cochran Armitage
trendp < 0.0001).
Table 3-4. Incidence and severity of thyroid follicular
hypertrophy/hyperplasia due to perfluorobutanoic acid (PFBA) exposure in
short-term and subchronic oral toxicity studies
Animal group (n = 10 in
all groups)
Dose (mg/kg-d)
0
1.2
6
30
150
28 d; male S-D rats
Butenhoff et al. (2012a)
3 (min)
3 (min)
9 (min)
7
(4 min, 3 mild)
90 d; male S-D rats
Butenhoff et al. (2012a)
4 (min)
6 (min)
4 (min)
9
(4 min, 5 mild)
Bolded cells indicate statistically significant changes compared with controls; shaded cells represent doses not
investigated in the individual studies. Severity normalized to four points scaled as follows: min = minimal severity;
mild = mild/slight severity; mod = moderate severity; sev = marked severity.
Study Name
Endpoint Name
Butenhoff, 2012, 1289835 Follicular Hypertrophy/Hyperplasia 28 Day Ora
90 Day Ora
Thyroid Follicular Colloidal Area 28 Day Ora
90 Day Ora
Thyroid Follicular Epithelial Cell Height 28 Day Ora
90 Day Ora
Study Type Animal Description
Rat, Sprague-Dawley (•£
Rat, Sprague-Dawley (c
Rat, Sprague-Dawley (c
Rat, Sprague-Dawley
Rat. Sprague-Dawley (£
Rat. Sprague-Dawley (f
Observation Time
28.0 days
49.0 days
90.0 days
111.0 days
28.0 days
49.0 days
90.0 days
111.0 days
28.0 days
49.0 days
90.0 days
111.0 days
280 days
49.0 days
28.0 days
49.0 days
28.0 days
49.0 days
28.0 days
49 0 days
PFBA Other Thyroid Effects
Thyroid Weight, Absolute
Thyroid Weight, Relative
28 Day Oral Rat, Sprague-Dawley (;
Rat, Sprague-Dawley (f
28 Day Oral Rat, Sprague-Dawley (2
0 No significant change
Treatment-Related ii
^ Treatment-Related Decrease
Rat. Sprague-Dawley (;
I I II I III!
mg/kg-day
Figure 3-3. Thyroid histopathology and organ-weight responses to
ammonium perfluorobutanoate (NH4+PFB) exposure (see interactive data
graphic and rationale for study evaluations for other thvroid effects in Health
Assessment Workspace Collaborative [HAWC]).
Mechanistic Evidence and Supplemental Information
Thyroid effects observed in the PFBA database consist of increased thyroid weight,
increased incidence of follicular hypertrophy/hyperplasia, and decreased levels of thyroxine (total
and free T4). Overall, a pattern of decreased hormone levels with corresponding alterations in
tissue weight and histopathology in the absence of an increase in TSH was observed. However, the
3-19
-------
Toxicological Review of PFBA and Related Salts
coefficient of variation for TSH in controls in the 90-day study (Bute nhoff etal.. 2012a: van
Otterdiik. 2007a. b) ranged from 40%-55%, compared to 13%-25% for free T4. The lack of an
observation of increased TSH may be due to difficulties in detecting relatively small changes in TSH
given the assay used in the study. While there is uncertainty in the reliability of the TSH
measurements and patterns of TH changes in animals may not translate perfectly to human clinical
definitions, decreases in T4 alongside normal levels of TSH is consistent with the human clinical
condition referred to as hypothyroxinemia [see additional discussion in fU.S. EPA. 2018bl].
Although the PFBA database is limited to two adult exposure studies (28- and 90-d) (Bute nhoff et
al.. 2012a: van Otterdiik. 2007a. b) in rats the observed thyroid hormone effects are supported by
supplemental information from structurally related PFAS (PFBS and PFHxA).
Decreases in thyroid hormones (total T3, total T4, and free T4) were observed in
PFBS-exposed pregnant mice and gestationally exposed female mouse offspring at >200 mg/kg-day
fFeng etal.. 20171 and in adult female and male rats following short-term exposures of >62.6
mg/kg-day fNTP. 20191. Increased TSH was reported in mouse dams and in offspring during
development of the reproductive system (PND 30) following gestational exposure Feng et al.
(20171. but no changes were noted in rats exposed to PFBS as adults . Increased TSH was reported
in mouse dams and in offspring during development of the reproductive system (PND 30) following
gestational exposure Feng etal. (20171. but no changes were noted in rats exposed to PFBS as
adults fNTP. 20191. a pattern consistent with the observed changes following adult PFBA exposure.
Thyroid weight and histopathology were not changed after short-term exposure to PFBS in adult
male or female rats fNTP. 20191.
Although the available evidence for PFHxA appears to provide weaker support for
endocrine effects than studies on PFBA or PFBS (see public comment draft for PFHxA; (U.S. EPA.
2021b). the only study in the PFHxA database of animal toxicity studies to examine thyroid
hormone levels observed that short-term oral exposure to PFHxA altered thyroid hormone levels in
male but not female rats fNTP. 20181. Dose-dependent decreases in free and total T4 (25%-73%
and 20%-58%, respectively) and to a lesser degree T3 (18%-29%) were observed with no
concomitant increase in TSH fNTP. 20181.
Decreased serum T4 or T3 is a key event preceded by disrupted TH synthesis (via multiple
possible mechanisms, including thyroid stimulating hormone receptor [TSHR] binding and thyroid
peroxidase [TPO] or sodium-iodide symporter [NIS] inhibition) and results in a myriad of
downstream neurodevelopmental outcomes, including altered hippocampal anatomy/function and
hearing deficit Thyroid hormones are critically important for proper brain development fBernal.
2015: Miller etal.. 2009: Williams. 2008: Crofton. 2004: Morreale de Escobar etal.. 2004: Zoeller
and Rovet. 2004: Howdeshell. 20021 because they directly influence neurodevelopmental
processes, such as neurogenesis, synaptogenesis, and myelination (Puig-Domingo and Vila. 2013:
Stenzel and Huttner. 2013: Patel etal.. 2011). Early in gestation, TH is delivered to the developing
fetal brain via placental transfer from the mother to the fetus (Calvo etal.. 1990). The mother
3-20
-------
Toxicological Review of PFBA and Related Salts
imparts TH as its sole source until the fetal thyroid gland begins functioning. The fetal gland is
completely nonfunctional until late gestation (gestation day [GD] 17), having only minimal
functionality until near parturition (GD 22 fBernal. 2015: Obregon etal.. 2007: Morreale de Escobar
et al.. 200411. at this point, in rats, approximately 17% of fetal T4 is still derived from the maternal
source despite the presence of a newly functioning thyroid gland fG etal.. 19901. In humans, these
maternal-derived fetal T4 estimates range from 30% to 50% f Obregon etal.. 2007: Morreale de
Escobar et al.. 2004: Vulsma et al.. 1989).
Recent mechanistic data in human fetal tissue demonstrates the presence of thyroid
receptors and transporters in the brain which suggests the fetal brain has a direct sensitivity to
thyroid hormones and supports the decades of observational, genetic, and animal research fDiez et
al.. 2021: Lopez-Espindola etal.. 20191. In addition, given the importance of thyroid hormones in
neurodevelopment in humans and animals, low thyroid hormone status is associated with adverse
neurological effects (Stagnaro-Green and Rovet. 2016: Zoeller and Rovet. 2004). and is likely
associated with effects in numerous other organ systems, including the heart, bone, lung and
intestine (Mullur etal.. 2014: Bassettetal.. 2007: Mochizuki etal.. 2007: Wexler and Sharretts.
2007: Bizzarro and Gross. 2004). Butenhoff et al. (2012a) observed that PFBA not only reduced
thyroid function via decreased serum total and free T4 but also increased thyroid hormone action in
the liver. This pattern of the effects has been seen following exposure to polychlorinated biphenyls
(PCBs) and polybrominated diphenyl ethers (PBDEs). For instance, increased TH gene expression
in the liver has been shown with a corresponding, inverse reduction in serum total and free T4
(Giera etal.. 2011) and (Bansal etal.. 2014). Following PCB exposure, this complex pattern also
occurred with changes in thyroid hormone action in the brain (Bansal and Zoeller. 2008: Zoeller et
al.. 2000: Zoeller and Crofton. 2000) and (Mullur etal.. 2014). Increased thyroid hormone activation
in the liver is known to reduce serum cholesterol f Mullur etal.. 20141. Butenhoff et al. f2012al
reported decreased serum cholesterol following PFBA exposure; these effects are described in
section 3.2.2 "Hepatic Effects".
Cases of severe maternal and fetal hypothyroidism, which results from iodine deficiency,
Hashimoto's disease, or premature birth, further underscore the importance of maintaining thyroid
hormone homeostasis during pregnancy. Several human epidemiological studies have
demonstrated key relationships between decreased circulating levels of thyroid hormones, such as
T4 in pregnant women and in utero and early postnatal life neurodevelopmental status. For
example, neurodevelopmental and cognitive deficits have been observed in children who
experienced a 25% decrease in maternal T4 during the second trimester in utero fHaddowetal..
1999). Children born euthyroid but exposed to thyroid hormone insufficiency in utero (e.g., <10th
percentile free T4), present with cognitive impairments (e.g., decreased intelligence quotient [IQ],
increased risk of expressive language disorder) or concomitant abnormalities in brain imaging
fKorevaar etal.. 2016: Henrichs etal.. 2010: Lavado-Autric etal.. 2003: Mirabella et al.. 20001. This
level of T4 insufficiency (<10th percentile), defined as mild-to-moderate thyroid insufficiency, has
3-21
-------
Toxicological Review of PFBA and Related Salts
been shown to correspond to a 15%-30% decrease in T4 serum levels compared to median levels
(Finken etal.. 2013: Tulvez etal.. 2013: Roman etal.. 2013: Henrichs et al.. 20101. Animal toxicity
studies also have shown that decreases in mean maternal T4 levels of ~10%-17% during
pregnancy and lactation elicit neurodevelopmental toxicity in rat offspring f Gilbert etal.. 2016:
Gilbert. 20111. Human studies also observe that thyroid hormone insufficiency is associated with
cognitive deficits in children fCrofton and Zoeller. 2005: Crofton et al.. 20051.
There are very few human studies available to inform what percent decrease in T4 might
lead to other adverse outcomes. This is mainly due to the nature of epidemiological studies,
typically with representative samples analyzed post hoc; many also bin data by "hypothyroid,
euthyroid, hypothyroxinemic" based on reference ranges, and then correlate to adverse outcomes.
Specifically, three human studies Tansen etal. f20191: Levie etal. f20181: Korevaar etal. f20161
were identified that had sample sizes large enough to capture a wide range of TSH and/or T4
values, which were then correlated to various neurodevelopmental outcomes that could be
quantified. However, these studies still do not make direct comparisons from a percent decrease in
hormones that would lead to an adverse effect; rather, they stratify their hormone samples by
standard deviation to the mean/median, quartiles, etc. Therefore, it's difficult to make a conclusion
in humans regarding what percent of hormone dysfunction is adverse, as those kinds of data are
not generated. Additionally, in experimental animal models, there are no definitive values regarding
to what degree of T4 reduction is adverse. This is due to several factors, including the existence of
multiple thyroid-dependent processes in the brain, which likely have differing spatiotemporal
sensitivities. But there are studies that show how graded reductions in T4 can lead to neuronal
heterotopia (Gilbert etal.. 20141. synaptic transmission defects (Gilbert and Sui. 20081. now and
differential gene expression (O'Shaughnessv etal.. 20181 and (Sharlin etal.. 20101.
There are data gaps in the PFBA developmental toxicity database, including a lack of
information on the thyroid and nervous system following gestational exposure. Although short-
term PFBA exposure did not appear to alter thyroid hormone levels in nonpregnant adult female
rats, thyroid hormone levels fluctuate throughout normal gestation (O'Shaughnessv etal.. 2018:
Hassan etal.. 2017: Perez etal.. 2013: Calvo etal.. 1990: Fukuda etal.. 19801 as maternal demands
to provide the fetus with adequate thyroid hormones. Specifically, serum T4 and T3 normally
decline over the course of pregnancy and then rise during the postnatal period (O'Shaughnessv et
al.. 20181. Thus, although no changes in thyroid hormone levels occurred in nonpregnant rats, that
PFBA influences hormone homeostasis differently in pregnant rats during the perinatal period is
possible as maternal and fetal hormone demands fluctuate.
Overall, animal studies specific to PFBA and other potentially relevant PFAS provide
support for thyroid hormone disruptions which can potentially lead to other effects of concern (e.g.,
neurodevelopmental effects).
3-22
-------
Toxicological Review of PFBA and Related Salts
Evidence Integration Summary
Inverse associations between PFBA exposure and thyroid hormone levels were observed in
the one available informative human study fLi etal.. 2017bl Given the low confidence in the study
methods and the lack of biological coherence across the hormone changes, however, the available
human evidence did not notably contribute to the evidence integration judgment on PFBA-induced
thyroid effects (i.e., indeterminate evidence).
The animal evidence comes from two high confidence experiments conducted by the same
laboratory (Butenhoffetal.. 2012a: van Otterdiik. 2007a. b), which reported PFBA-induced
perturbation of the thyroid in one species and sex (male S-D rats) across two different exposure
durations. The reported PFBA exposure-induced effects across thyroid hormone measures
(i.e., adult males, reductions in total or free T4; T3 was not measured) were consistent, dose
dependent, and associated with increasing absolute and relative thyroid weights and
histopathology (follicular hypertrophy/hyperplasia). These decreases were large in magnitude
(>50% in some PFBA exposure groups), and perturbations in total T4 were shown to persist at
least 21 days after the termination of 90-day exposure to the highest dose (150 mg/kg-day) but not
lower doses (in fact, total T4 was increased at 30 mg/kg-day). No effects (e.g., increases) on TSH in
exposed rats were observed. The observed pattern of effects on the thyroid (i.e., decreased total and
free T4 without a compensatory increase in TSH) after PFBA exposure is consistent with thyroid
perturbations following exposure to other PFAS, including the structurally related compound
perfluorobutane sulfonate (U.S. EPA. 2021b). Taken together, the consistent changes in total and
free T4, thyroid weights, and histopathology across the two available oral PFBA exposure
experiments are biologically coherent and plausible.
Several aspects of the animal evidence base decrease the strength or certainty of the
evidence. Although there is coherence across different measures of thyroid toxicity in male rats,
some effects across durations of exposure are inconsistent: some effects occur in the 28-day study
but not in the 90-day study, and the magnitude of change of some effects is larger in the short-term
than in the subchronic study. Also, in male rats, for free T4 only, the lack of a control group in
animals exposed for 90 days complicates the interpretation of that endpoint
Although the organ-weight increases and histopathological effects (follicular hypertrophy)
observed in Butenhoff et al. f2012al are consistent with a scenario where serum T4 levels are low
but TSH levels are normal, the mechanism by which these changes occurred unclear. Rodents are
more sensitive to these histopathological changes (follicular cell hypertrophy), which then can
develop into follicular cell tumors (U.S. EPA. 1998a). Increased thyroid follicular cell hypertrophy
supports the finding that the thyroid hormone economy is perturbed. The observed changes are
likely due to increased metabolism or competitive displacement of T4. That no thyroid effects
(e.g., hormone or histopathological changes) were observed in adult nonpregnant females at any
dose or treatment duration might be related to PFBA pharmacokinetics because clearance rates in
3-23
-------
Toxicological Review of PFBA and Related Salts
rats are faster in females (compared to males, see Section 3.1.4). Taken together, the available
animal studies provided moderate evidence for thyroid effects.
Rodents and humans share many similarities in the production, regulation, and functioning
of thyroid hormones. Although differences exist, including the timing of in utero thyroid
development and hormone turnover rates, rodents are considered a good model for evaluating the
potential for thyroid effects in humans fZoeller et al.. 20071. More specifically, the observed
decreases in total or free T4 in the absence of increases in TSH are considered biologically relevant
to humans fCrofton. 2004: Lau etal.. 20031. TSH is an indicator that the thyroid system has been
perturbed, but it does not always change when serum T4 is decreased (Hood etal.. 19991. Adverse
neurological outcomes have been demonstrated following decreased T4 levels during the early
neonatal period with no changes in T3 or TSH fCrofton. 20041. The typical compensatory feedback
loop involves microsomal enzymes that induce uridine 5'-diphospho-glucuronosyltransferase
(UDP-GT), affecting the thyroid gland by increasing T4 glucuronidation, which in turn reduces
serum T4. In this case, the typical response to reduced serum free T4 is an increased production of
TSH (Hood and Klaassen. 20001. which can lead to thyroid hyperplasia or rat follicular tumors. In
that way, observation of thyroid histopathology can be an indication of perturbations in TSH levels
over time even in situations where increased TSH is not observed at the time histopathology is
measured fHood etal.. 19991. Rodents have been shown to have a unique sensitivity to thyroid
follicular hyperplasia (leading to development of follicular tumors), however, that is considered
less relevant to humans (U.S. EPA. 1998a). Nevertheless, the coherent and consistent perturbations
to thyroid hormone economy and the resultant increased thyroid histopathology indicates that
PFBA is exerting some effect on the thyroid of exposed male rats. Even considering the increased
sensitivity of rodents to thyroid follicular hyperplasia compared to humans, thyroid hormone
perturbations are considered relevant to humans and might be even more sensitive to change in
humans compared to rodents fU.S. EPA. 1998al.
A notable data gap exists for fuller interpretation of the reported thyroid effects. Studies
evaluating PFBA effects on neurodevelopment or thyroid measures after developmental exposure
(see Section 3.2.3 "Developmental Effects") were not identified, thus leaving uncertainty on the
potential for more sensitive developmental effects of PFBA exposure on the thyroid and nervous
systems. During developmental lifestages, such as gestational/fetal and postnatal/early newborn,
thyroid hormones are critical in a myriad of physiological processes associated with somatic
growth and maturation and survival mechanisms, such as thermogenesis, pulmonary gas exchange,
and cardiac development fSferruzzi-Perri etal.. 2013: Hillman etal.. 20121. That thyroid hormones
are at sufficient levels is essential during times critical to brain development and functioning and in
the growth, development, and functioning of numerous organ system processes, including basal
metabolism and reproductive, hepatic, sensory (auditory, visual) and immune systems (Forhead
and Fowden. 2014: Gilbert and Zoeller. 2010: Hulbert. 20001 (see Mechanistic Evidence and
Supplemental Information subsection above). Mammals are more susceptible during perinatal and
3-24
-------
Toxicological Review of PFBA and Related Salts
postnatal lifestages because their compensatory feedback responses are absent or not fully
developed and they have low thyroid hormone reserves (Morreale de Escobar etal.. 2004: Zoeller
and Rovet. 20041. Further, thyroid hormones are critically important in early neurodevelopment as
they directly influence neurogenesis, synaptogenesis, and myelination fPuig-Domingo and Vila.
2013: Stenzel and Huttner. 20131. Although the PFBA database lacks information on thyroid
hormone levels in exposed pregnant animals or offspring exposed during gestation, these effects
have been observed following exposure of mice to the structurally related PFAS, PFBS fU.S. EPA.
2018b)- Decreases in total T4 andT3 were observed in dams at GD 20 and offspring atPND 1, 30,
and 60, clearly indicating that thyroid hormone levels were perturbed during periods of
neurological development Further, given the evidence is consistent with PFBA, the PFBS
assessment identifies developmental neurotoxicity as a database limitation due to the known
association between thyroid hormone insufficiency during gestation and developmental
neurotoxicity outcomes fU.S. EPA. 2018bl. Accordingly, given that developmental neurotoxicity
(due to thyroid hormone insufficiency) is a concern following exposure to PFBS, it follows that this
concern is relevant to exposure to PFBA during development because of the similarities in thyroid
effects across the two PFAS.
Taken together, the evidence indicates that PFBA exposure is likely to cause thyroid
toxicity in humans, given relevant exposure circumstances (see Table 3-5). This judgment is based
primarily on consistent and biologically coherent results from two high confidence studies (short-
term and subchronic study design) in male rats that indicate effects on thyroid hormone levels (T4
without compensatory effects on TSH). These effects on thyroid hormone levels generally occurred
at PFBA exposure levels >30 mg/kg-day, although some notable effects were observed after
exposure to 6 mg/kg-day.
3-25
-------
Toxicological Review of PFBA and Related Salts
Table 3-5. Evidence profile table for thyroid effects
Evidence Stream Summary and Interpretation
Inferences and Summary
Judgment
Evidence from studies of exposed humans (see Section 3.2.1: Human Studies)
Studies and
confidence
Summary of key
findings
Factors that increase
certainty
Factors that decrease
certainty
Judgments and
rationale
Evidence indicates (likely)
Thyroid Hormones
1 low confidence
study
• Single study
reporting inverse
associations with
free T4, free T3, and
TSH; only TSH was
statistically
significant
• No factors noted
• Lack of coherent
associations across
hormones
• Imprecision
ooo
Indeterminate
Primary basis:
Two high confidence studies in rats
ranging from short-term to
subchronic exposure; effects
observed at >6 mg/kg-d PFBA;
similar effects for related PFAS
Human relevance:
Effects in rats are considered
Evidence from in vivo animal studies (see Section 3.2.1: Animal Studies)
Studies and
confidence
Summary of key
findings
Factors that increase
certainty
Factors that decrease
certainty
Judgments and
rationale
potentially relevant to humans
based on conserved biological
processes, and the observed
Thyroid Hormones
2 hiqh confidence
studies in adult rats:
• 28-d
• 90-d
• Decrease in free
and total T4 in male
rats at >6 mg/kg-d
• Decrease in T4 with
no increase in TSH
• Consistent
increases in males
across all studies
• Dose-response
gradient
• Coherence of
decreased T4 with
histopathology
• Magnitude of
effect, up to 79%
• High confidence
studies
• Potential lack of
expected coherence
(no compensatory
TSH increase to T4
decrease)
®©o
Moderate
Findings considered
adverse based on
consistent and
biologically coherent
results for thyroid
hormone levels, organ
weights, and
pattern of changes is consistent
with potential neurological
outcomes following decreased T4
during development (see Section
3.2.1: Mechanistic Evidence and
Supplemental Information)
Cross-stream coherence:
N/A (human evidence
indeterminate)
3-26
-------
Toxicological Review of PFBA and Related Salts
Evidence Stream Summary and Interpretation
Inferences and Summary
Judgment
Histopathology
2 hiqh confidence
studies in adult rats:
• 28-d
• 90-d
• Follicular
hypertrophy/hyper
plasia observed in
male rates at
30 mg/kg-d
• No histopathological
effects at
150 mg/kg-d (after
short-term
exposure)
• Consistent follicular
hypertrophy/hyper-
plasia in male rats
across studies
• Coherence of
hypertrophy with
T4 decreases
• High confidence
studies
• Potential lack of
expected coherence
(no change in TSH
levels)
• Unexplained lack of
significant effects at
highest tested dose
histopathology. The
observation of effects
only in males might be
explained by
pharmacokinetics.
Uncertainties remain
as to how organ
weights and
histopathology are
affected in the
absence of TSH
increases.
Susceptible populations and
lifestages: The developing fetus
and children are susceptible to
altered thyroid hormone status;
the lack of data on thyroid or
nervous system effects following
gestational exposure is a data gap.
Organ Weight
1 hiqh confidence
study in adult rats:
• 28-d
• Increase in thyroid
weight (absolute
and relative) at 6
and 30 mg/kg-d
• No change in
thyroid weight at
150 mg/kg-d
• Magnitude of
effect, >2-fold
increases
• High confidence
study
• Potential lack of
expected coherence
(no change in TSH
levels)
• Unexplained lack of
significant effects at
highest tested dose
Mechanistic evidence and supplemental information (see subsection above)
Summary of key findings, interpretation, and limitations
Evidence stream judgment
Key findings and interpretation:
• PFBA-induced thyroid changes similar to those for related PFAS (i.e., PFBS and,
although the evidence is weaker, PFHxA)
• Findings for PFBS indicate the potential for effects of concern during development
Limitations: No PFBA-specific mechanistic evidence informing thyroid effects
Findings for related PFAS
support the plausibility of
findings for PFBA, and the
potential for effects of concern
with PFBA exposure during
development
3-27
-------
Toxicological Review of PFBA and Related Salts
3.2.2. Hepatic Effects
Human Studies
One epidemiological study reported on the relationship between PFBA exposure and serum
biomarkers of liver injury. This study Nian et al. f2019alThis study Nian etal. f2019al was cross-
sectional and was classified as medium confidence given minor concerns over participant selection,
outcome ascertainment, and confounding. Sensitivity was considered deficient due to limited
exposure contrast for PFBA (detected in 70%, median [interquartile range
(IQR)] = 0.15 ng/mL [0.01-0.51 ng/mL]), which likely reduced the study's ability to detect an effect.
The study found no association between serum levels of alanine aminotransferase (ALT), aspartate
aminotransferase (AST), total protein, alkaline phosphatase (ALP), y-glutamyl transferase (GGT),
total bilirubin, or cholinesterase with PFBA exposure, but given the sensitivity concerns, this is
difficult to interpret.
In addition, one low confidence cross-sectional study Fu etal. (2014) examined the
association between PFBA exposure and blood lipids. No association was reported; however, the
exposure levels in the study population were very low with narrow contrast (median [IQR] = 0.1
[0.03-0.2] ng/mL), so the study had poor sensitivity to detect an effect
Animal Studies
Hepatic effects were evaluated in multiple high and medium confidence, short-term and
subchronic studies in rats and mice (Bute nhoff etal.. 2012a: Foreman etal.. 2009: van Otterdiik.
2007a. b; Permadi etal.. 1993: Permadi etal.. 1992) and in one high confidence developmental
toxicity study in mice fDas etal.. 20081. Some outcome-specific considerations for study evaluations
were influential on the overall study rating for liver effects, but none of these individual domain-
specific limitations were judged as likely to be severe or have a notable impact on the study results,
and all studies considered further in this section were rated as high or medium confidence (see
Figure 3-4). For more information on outcome-specific considerations for study evaluations, please
refer to the study evaluations in the HAWC PFBA database.
3-28
-------
Toxicological Review of PFBA and Related Salts
rtX\0^' oOQ -0^- ^-A\.
Reporting quality
Allocation
Observational bias/blinding
Confounding/variable control
Selective reporting and attrition
Chemical administration and characterization
Exposure timing, frequency and duration
Endpoint sensitivity and specificity -
Results presentation
Overall confidence
Legend
Good (metric) or High confidence (overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
NRl Not reported
Figure 3-4. Evaluation results for animal studies assessing effects of
perfluorobutanoic acid (PFBA) exposure on the liver (see interactive data
graphic for rating rationales!.
One low confidence, short-term study also reported hepatic effects flkeda etal.. 19851. This
study was judged as low confidence given concerns over allocation of animals, reporting/attrition
concerns, characterization of the test compound, and endpoint sensitivity.
Endpoints evaluated in the studies reporting liver effects include liver weights,
histopathological changes, and serum biomarkers of effect.
Organ weight
Short-term and subchronic exposure studies consistently demonstrated increased liver
weight in rodents exposed to PFBA (see Table 3-6 and Figure 3-5). Liver weight is commonly
reported as either absolute weight or relative to body weight. In general, relative liver weight is the
preferred metric as it accounts for individual variations in body weight, either due to the exposure
being studied or to inter individual variability. Both absolute and relative liver weight are presented
in this section for the sake of completeness; results based on absolute liver weight closely track
those for relative liver weight
3-29
-------
Toxicological Review of PFBA and Related Salts
Table 3-6. Percent increase in relative liver weight due to perfluorobutanoic
acid (PFBA) exposure in short-term and subchronic oral toxicity studies
Animal group
Dose (mg/kg-d)
1.2
6
30
35
150
175
350
28 d; male S-D rats
Butenhoff et al. (2012a); van Otterdiik (2007a)
5
24
48
28 d; female S-D rats
Butenhoff et al. (2012a); van Otterdiik (2007a)
-1
0
-3
90 d; male S-D rats
Butenhoff et al. (2012a); van Otterdiik (2007a)
9
7
33
90 d; female S-D rats
Butenhoff et al. (2012a); van Otterdiik (2007a)
0
-3
3
28 d; PPARa wild-type male SV/129 mice
Foreman et al. (2009)
61
101
112
28 d; humanized PPARa male SV/129 mice
Foreman et al. (2009)
38
63
81
28 d; PPARa null male SV/129 mice
Foreman et al. (2009)
3
1
7
Pregnant Po female CD-I mice on GD 18
Das et al. (2008)
9
28
32
Nonpregnant Po female CD-I mice on GD 18
Das et al. (2008)
14
32
29
Fi male and female CD-I mice on PND 1
Das et al. (2008)
9
30
41
Bolded cells indicate statistically significant changes compared with controls; shaded cells represent doses not
investigated in the individual studies.
The only null study Ikeda etal. f!9851 reported that relative liver weight was not increased
over controls in male S-D rats exposed to 0.02% PFBA in the diet for 2 weeks (approximately
20 mg/kg-day). This study was judged low confidence, however, on the basis of concerns over
reporting, exposure characterization, and endpoint sensitivity/selectivity. Conversely, following
10 days of dietary exposure to 0.02% PFBA, relative liver weight was increased 38% in male
C57B1/6 mice in a medium confidence study fPermadi et al.. 19931. Twenty-eight days of daily
gavage exposure to >35 mg/kg-day PFBA significantly increased relative liver weights in adult male
wild-type (+/+) or humanized PPARa (hPPARa) Sv/129 male mice fForeman et al.. 20091. The
relative liver weight of wild-type male mice was increased by 61%, 101%, and 112% at 35,175, and
350 mg/kg-day, respectively. Increased relative liver weight was also observed in these same dose
groups in humanized PPARa (hPPARa) male mice, although they were somewhat less than those
observed in wild-type mice: 38%, 63%, and 81%. Relative liver weight was not changed in PPARa
null (-/-) mice fForeman et al.. 20091. A similar profile of increased relative liver weight also was
3-30
-------
Toxicological Review of PFBA and Related Salts
observed in male S-D rats exposed to >30 mg/kg-day NH4+PFB via oral gavage for 28 days
fButenhoff etal.. 2012a: van Otterdiik. 2007a). Relative liver weights were increased 24% and 48%
at 30 and 150 mg/kg-day. Relative liver weights in both dose groups were observed to return to
control levels following a 21-day recovery period. Female rats exposed at the same dose levels
experienced no increases in relative liver weights (l%-3% decrease).
Similar to increases following 28-day exposures, relative liver weights also were observed
to increase in male S-D rats exposed to NH4+PFB via oral gavage for 90 days fButenhoff etal.. 2012a:
van Otterdiik. 2007b). with relative liver weights increased 33% at 30 mg/kg-day. As with the
short-term exposure, relative liver weights returned to control values following a 21-day recovery
period after termination of subchronic exposure. As observed in the short-term study, exposure to
NH4+PFB for 90 days did not increase liver weights in female rats (3% decreases to 3% increases).
In a developmental toxicity study in CD-I mice, exposure to NH4+PFB via oral gavage increased
relative (to body weight) liver weights in pregnant (measured on GD 18) and nonpregnant Po
females at >175 mg/kg-day (Das etal.. 2008). Relative liver weights were increased by 28% and
32% at 175 and 350 mg/kg-day (respectively) in pregnant mice, whereas relative liver weights
were increased 32% and 29% in nonpregnant mice at the same dose levels. No effect on liver
weights was observed in the subset of dams followed until after weaning (PND 22). Similar
magnitudes of relative liver weight increase also were observed in Fi animals at PND 1: 30% and
41% at 175 and 350 mg/kg-day, respectively. In animals at PND 10, however, no change in relative
liver weights was observed. The lack of an effect on PND 10 in Fi or Po animals on PND 22 could be
because these animals were not exposed during lactation and therefore had a 10- or 22-day
recovery period compared with offspring or dams whose liver weights were measured on PND 1
and GD 17. This observation of no effect following a recovery period is consistent with the findings
of the subchronic and short-term exposures in adult animals fButenhoff etal.. 2012a: van Otterdiik.
2007a, b).
Although not an oral toxicity study, Weatherlv etal. f20211 also observed statistically
significant increases in relative liver weight (up to 60% increases) in mice dermally exposed to
PFBA.
In conclusion, effects on relative liver weights in adult male rats and mice were observed at
>30 or 35 mg/kg-day following subchronic or short-term exposures (respectively), whereas effects
in adult pregnant and nonpregnant female mice (exposed during pregnancy) and their offspring
were observed only at higher doses (>175 mg/kg-day). Adult female rats were only exposed up to
150 mg/kg-day in the subchronic study fButenhoff etal.. 2012a: van Otterdiik. 2007bl so whether
these animals would exhibit the same effects at the exposure levels used in the developmental
toxicity study Das etal. (2008) is unclear. Regardless, the data for relative liver weight seem to
indicate that male animals are more susceptible to this effect than female animals, possibly because
females have a much faster (5-6 times greater) excretion rate than males (see Section 3.1.4 for
details).
3-31
-------
Toxicological Review of PFBA and Related Salts
Changes in absolute liver weight across all studies were generally consistent with those
observed for relative liver weight. Following 10 days of dietary exposure to 0.02% (w/w) PFBA,
absolute liver weights were observed to be increased 64% in male C57B1/6 mice fPermadi etal..
1993: Permadi etal.. 19921. Absolute liver weights were also increased 27% and 45% following
28 days of exposure to 30 or 150 mg/kg-day NH4+PFB, respectively fButenhoff etal.. 2012a: van
Otterdiik. 2007al. No effects were observed in female rats following exposure or in male rats
following a 21-day recovery. Similar to increases following 28-day exposures, liver weights were
also observed to increase due to treatment in male S-D rats exposed to NH4+PFB for 90 days
fButenhoff etal.. 2012a: van Otterdiik. 2007b). with absolute liver weights increased by 23%. Liver
weights returned to control levels following a 21-day recovery period. As observed in the short-
term study, exposure to NH4+PFB for 90 days did not increase liver weights in female rats
(~3%-8% increases). In a developmental toxicity study in CD-I mice fDas etal.. 20081. exposure to
NH4+PFB increased absolute liver weights in pregnant and nonpregnant P0 females at
>175 mg/kg-day. Absolute liver weights were increased by 24% and 35% at 175 and
350 mg/kg-day, respectively, in pregnant mice, whereas absolute liver weights were increased 34%
and 21% at those same doses in nonpregnant P0 females. Similar magnitudes of absolute liver
weights increase (27% and 32%) also were observed in F1 animals at PND 1 at 175 and
350 mg/kg-day fDas etal.. 20081. As with relative liver weights, no effect was observed in offspring
at PND 10 or in pregnantPO animals atpostweaning (PND 22).
3-32
-------
Toxicological Review ofPFBA and Related Salts
Study Name
Endpoint Name
Study Type
Animal Description
Observation Time
PFBA Liver Weight Effects
Permadi 1993. 13324S2
Ltver Wetghl Relative
10 Day Oral
Mouse C57Bf6 (£)
100 days
•-A
Ikeda 1985 2325571
Liver Weight Relative
14 Day Oral
Rat. Sprague-Dawley (£)
14 0 days
• •
Foreman 2009, 2325387
Relative Liver Weight
28 Day Oral
Mouse 129/SV (£)
280 days
•-A—
A
—A
Mouse 129/SV PPARo nuH (£)
28 0 days
•—•—
•
—•
Mouse 129/Sv humanized PPARa (£]
28 0 days
• A
*
—A
Butenhoff. 2012, 1289835
Ltver Weight Relative
28 Day Oral
Rat. Sprague-Dawley (£)
280 days
»-A—
49 0 days
—
•
Rat, Sprague-Dawley <£)
28 0 days
(• •—
•
49 0 days
—
•
90 Day Oral
Rat. Sprague-Dawley (£)
90 0 days
(•A
111.0 days
Rat. Sprague-Dawley (5)
111.0 days
<••
90 0 days
<• •
Oas 2008 1290825
Relative Uver Weight (Pregnant)
17 Day Oral
P0 Mouse CD-I (=)
180 days
#—•—
A
—A
Relative Liver Weight (Non-pregnant)
17 Day Oral
P0 Mouse CD-I (£)
18 0 days
A
—A
Relative Liver Weighl (PND 1)
17 Day Oral
F1 Mouse CD-I (£2)
20 0 days
A
—A
Relative Liver Weighl (PND 10)
17 Day Oral
F1 Mouse CD-1 <£=)
29 0 days
#—•—
•
—•
Permadi 1993 1332452
Lrver Wetghl Absolute
10 Day Oral
Mouse C57BW6 (£)
10 0 days
• A
Bulenhoff 2012. 1289835
Liver Wetghl. Absolute
28 Day Oral
Rat. Sprague-Dawley (£)
28 0 days
<• A—
~
49 0 days
(••—
•
Rat, Sprague-Dawley (2)
28 0 days
—
•
49 0 days
(!-•
•
90 Day Oral
Rat. Sprague-Dawley (£)
90 0 days
(•^A
# Doses
111.0 days
A Treatment-Related Increase
Rat Sprague-Dawley (2)
90 0 days
^ Ireatment-Heiated Decrease
Ill.Odays
M Dose Range
Das 2008 1290825
Absolute Lrver Weighl (Pregnant)
17 Day Oral
P0 Mouse CD-I (=)
18 0 days
•—
~
—A
Absolute Liver Wetghl (Non-pregnant)
17 Day Oral
P0 Mouse CD-1 (5)
180 days
*
—A
Liver Weight Absolute (PND1)
17 Day Oral
F1 Mouse CD-I (£2)
20 0 days
A
—A
Lrver Weight Absolute (PND 10)
17 Day Oral
F1 Mouse CD-I (££)
29 0 days
—•
—•
-50 0 50 100 150 200 250 300 350 400
Axis label
Figure 3-5. Liver-weight response to ammonium perfluorobutanoate
(NH4+PFB) or perfluorobutanoic acid (PFBA) exposure (see interactive data
graphic and rationale for study evaluations for liver-weight effects in Health
Assessment Workspace Collaborative [HAWC]).
Histopathologv
Histopathological examination of the livers of mice and rats across three separate gavage
studies of 28-day fButenhoff etal.. 2012a: Foreman etal,, 2009: van Otterdiik. 2007a1 or 90-day
fButenhoff etal.. 2012a: van Otterdiik. 2007b] exposure duration revealed significant,
dose-dependent alterations and lesions (see Table 3-7 and Figure 3-6).
Both wild-type and hPPARa mice exposed to PFBA for 28 days developed hepatocellular
hypertrophy at doses >35 mg/kg-day (incidences of 100% in all doses), whereas PPARa null mice
did not develop hypertrophic lesions at any dose following 28-day exposures f Foreman etal..
2009). Although the incidence and severity of the hypertrophic lesions were similar between
wild-type and hPPARa mice at higher doses, hPPARa mice developed more severe lesions at
35 mg/kg-day than did the wild-type mice (5/10 severe lesions vs. 0/10, respectively).
Hepatocellular hypertrophy also was observed in 6/10 S-D rats exposed to 150 mg/kg-day PFBA
for 28 days fButenhoff etal.. 2012a: van Otterdiik. 2007al and 9/10 rats exposed to 30 mg/kg-day
PFBA for 90 days fButenhoff etal.. 2012a: van Otterdiik. 2007bl. In both cases, no lesions were
observed in animals following a 21-day recovery period.
3-33
-------
Toxicological Review of PFBA and Related Salts
hPPARa mice were much less susceptible to the development of hepatic focal necrosis
following a 28-day exposure to PFBA compared to wild-type mice. Wild-type mice developed
hepatic focal necrosis (with inflammatory cell infiltration) at 35 mg/kg-day (1/10), 175 mg/kg-day
(6/10) and 350 mg/kg-day (9/10), whereas focal necrosis was observed in only 1/10 (35 and 175
mg/kg-day) and 2/10 (350 mg/kg-day) hPPARa mice fForeman etal.. 20091. PPARa null mice only
developed focal necrosis in the 175 mg/kg-day (1/10) and 350 mg/kg-day (2/10) dose groups. For
all strains, most of the necrotic lesions were judged mild in severity. By comparison, in rats exposed
to PFBA for 28 days, no increase in hepatocellular coagulative necrosis (Bute nhoff etal.. 2012a: van
Otterdiik. 2007a) was observed. No effects on hepatocellular necrosis in rats were observed
following 90-day exposures to PFBA fBute nhoff etal.. 2012a: van Otterdiik. 2007bl.
Following exposure to 350 mg/kg-day for 28 days, centrilobular and periportal vacuolation
was observed in PPARa null and humanized mice, respectively, while no vacuolation was reported
for wild-type mice (Foreman et al.. 2009). Whether these effects occurred at lower doses was not
mentioned. Further, no quantitative data were reported for these effects, so examining the dose-
response or magnitude of effect across doses was not possible. The lack of vacuolation in wild-type
animals is consistent with the lack of vacuolation in rats exposed to PFBA for 90 days (Butenhoff et
al.. 2012b: van Otterdiik. 2007bl. where 4/10 control animals were reported to exhibit vacuolation,
but incidence dropped to 1/10 in the low-dose group and no vacuolation was observed at higher
doses.
All mice in all exposure groups were observed to develop hepatocellular hypertrophy
(characterized by increased cytoplasmic eosinophilia, decreased glycogen content, and increased
cellular volume) following dermal exposures of up to 15% v/v (Weatherlv etal.. 2021). Necrotic
lesions were not consistently observed following dermal exposure to PFBA, although genes
associated with necrosis were increased following exposure.
Although the number of studies was small, mice did seem more sensitive to development of
hepatocellular lesions compared to rats, possibly owing to the observed differences in
pharmacokinetics between the two species: Mice are observed to have serum excretion half-lives
approximately two times longer than rats at similar exposure levels (see Section 3.14 and Table 3-2
for details).
3-34
-------
Toxicological Review of PFBA and Related Salts
Table 3-7. Incidence and severity of liver histopathological lesions due to perfluorobutanoic acid (PFBA)
exposure in short-term and subchronic oral toxicity studies
Animal group (n = 10 in all groups)
Dose (mg/kg-d)
0
1.2
6
30
35
150
175
350
Hypertrophy
28 d; male rats
Butenhoff et al. (2012a); van Otterdiik (2007b)
0
0
0
6 (min)
90 d; male rats
Butenhoff et al. (2012a); van Otterdiik (2007b)
0
0
0
9 (5 min, 4
mild)
28 d; PPARa wild-type male mice
Foreman et al. (2009)
0
10 (4 mild, 6
mod)
10 (1 mild, 1
mod, 8 sev)
10 (sev)
28 d; hPPARa male mice
Foreman et al. (2009)
0
10 (1 mild, 4
mod, 5 sev)
10 (2 mod, 8
sev)
10 (sev)
28 d; PPARa null male mice
Foreman et al. (2009)
0
0
0
0
Coagulative necrosis
90 d; male rats
Butenhoff et al. (2012a); van Otterdiik (2007b)
0
0
0
0
Focal necrosis3
28 d; PPARa wild-type male mice
Foreman et al. (2009)
0
1 (mild)
6 (2 min, 4
mild)
9 (8 mild, 1
mod)
28 d; hPPARa male mice
Foreman et al. (2009)
0
1 (min)
1 (min)
2 (min)
28 d; PPARa null male mice
Foreman et al. (2009)
0
0
1 (min)
2 (min)
Vacuolation
None reported
28 d; hPPARa male mice
Foreman et al. (2009)
Periportal vacuolation reported to increase at 350 mg/kg-d, compared to controls (responses at 35 mg/kg-d
or 175 mg/kg-d were not reported by study authors)
28 d; PPARa null male mice
Foreman et al. (2009)
Centrilobular vacuolation reported to increase at 350 mg/kg-d, compared to controls (responses at 35
mg/kg-d or 175 mg/kg-d were not reported by study authors)
Bolded cells indicate statistically significant changes compared to controls; shaded cells represent doses not investigated in the individual studies. Severity
normalized to four points scaled as follows: min = minimal severity; mild = mild/slight severity; mod = moderate severity; sev = marked severity,
incidence of focal necrosis for the positive control of Wy-14,643 (a known PPARa/y activator) was 3 total (1 minimal, 2 mild) at 50 mg/kg-d exposure.
3-35
-------
Toxicological Review of PFBA and Related Salts
Study Name
Endpoint Name
Study Type
Animal Description
Observation Time
PFBA Liver Histopathology Effects
Foreman, 2009, 2325387
Hepatocellular Hypertrophy
28 Day Oral
Mouse. 129/SV (^)
28.0 days
•—A-
A
—A
Mouse. 129/SV PPARa null (^)
28.0 days
•—•—
4
—•
Mouse. 129/Sv humanized PPARa (£]
28.0 days
•—A-
A
—A
Butenhoff, 2012, 1289835
Hepatocellular Hypertrophy
28 Day Oral
Rat, Sprague-Dawley (£)
28.0 days
€#-•—
# Doses
49.0 days
•
A Treatment-Related Increase
90 Day Oral
Rat, Sprague-Dawley (^)
90.0 days
<• A
Y Treatment-Related Decrease
111.0 days
I—I Dose Range
Foreman, 2009, 2325387
Hepatic Focal Necrosis
28 Day Oral
Mouse. 129/SV (;?)
28.0 days
4
A
Mouse. 129/SV PPARa null (^)
28.0 days
•
—•
Mouse. 129/Sv humanized PPARa (£]
28.0 days
0
—•
Butenhoff, 2012, 1289835
Hepatocellular Coagulative Necrosis
28 Day Oral
Rat. Sprague-Dawley (J)
28.0 days
•
49.0 days
<#-#—
•
0 0 50
100 150
200 250 300
350
4
0
mgrtcg-day
Figure 3-6. Liver histopathology response to ammonium perfluorobutanoate
(NH4+PFB) or perfluorobutanoic acid (PFBA) exposure (see interactive data
graphic and rationale for study evaluation for liver histopathology effects in
Health Assessment Workspace Collaborative [HAWC]).
Serum biomarkers
Serum biomarkers associated with altered liver function or injury including ALT, AST, ALP,
total protein, albumin, and total bilirubin were not significantly changed in male or female S-D rats
exposed to up to 150 mg/kg-day PFBA for 28 days fButenhoff etal.. 2012a: van Otterdiik. 2007al.
However, prothrombin time (a measure of clotting time induced by the liver-produced
prothrombin protein) was decreased at 150 mg/kg-day in males and at 6 and 30 mg/kg-day in
females (but not at 150 mg/kg-day), although decreases were small (~5%-9% relative to control)
and were reported to be within the concurrent reference range for S-D rats. Prothrombin time,
however, was statistically significantly decreased (p < 0.01) in all dose groups in females after the
21-day recovery period. Some alterations in serum biomarkers were also observed in rats exposed
to PFBA for 90 days: ALP was increased 32% in male rats exposed to 30 mg/kg-day and bilirubin
was decreased 21% and 13% in male and female rats (respectively) exposed to 30 mg/kg-day
(Butenhoff etal.. 2012a: van Otterdiik. 2007a). ALT was not statistically significantly increased by
PFBA exposure in wild-type, PPARa null, or hPPARa mice (Foreman et al.. 2009). although it did
increase almost 3-fold at 350 mg/kg-day (20.28 U/I) compared to controls (7.39 U/I). Cholesterol
levels were significantly (p < 0.01) decreased 20% and 27% in male rats exposed to 30 and
150 mg/kg-day PFBA, respectively, for 28 days fButenhoff etal.. 2012a: van Otterdiik. 2007al.
Cholesterol levels returned to control values following recovery, and no effects on cholesterol were
observed in male rats exposed to PFBA for 90 days. No clear explanation exists to describe why
cholesterol levels might be changed after 28, but not 90, days of PFBA exposure.
In mice exposed to PFBA dermally (up to 15% v/v), several serum biomarkers including
serum cholesterol, glucose, and ALP were increased, and urea nitrogen was decreased, relative to
controls fWeatherlv etal.. 20211. Other serum biomarkers (ALT, total protein, albumin, or globulin)
were not increased due to exposure.
3-36
-------
Toxicological Review of PFBA and Related Salts
Mechanistic Evidence and Supplemental Information
The liver effects observed in the PFBA database consist of increased liver weight, increased
incidence of hepatocellular hypertrophy, and (to a lesser degree) hepatocellular necrosis. Increased
liver weight and hepatocellular hypertrophy can be associated with changes that are adaptive in
nature fHall etal.. 20121 and not necessarily indicative of adverse effects unless observed in
concordance with other clinical, pathological markers of overt liver toxicity (see PFBA Protocol;
Appendix A). The IRIS PFAS Assessment Protocol (which addresses PFBA) states the panel
recommendations from Hall etal. (2012) can be used to judge whether observed hepatic effects are
adverse or adaptive in nature. Given that Hall etal. f20121 was focused on framing noncancer liver
effects in the context of progression to liver tumors, however, the protocol further indicates that
"...consultation of additional relevant information will be considered to interpret the adversity of
noncancer liver effects over a lifetime exposure, taking into account that effects perceived as
adaptive can progress into more severe responses and lead to cell injury." For PFBA, the "additional
relevant information" consists of multiple in vitro mechanistic studies, an in vivo study
investigating PFBA-induced liver effects in wild-type humanized PPARa mice, and PPARa-null mice
(Foreman), as well as evidence from other PFAS that help elucidate possible MOAs of PFBA liver
toxicity.
Many of the hepatic effects caused by exposure to perfluoroalkyl acids such as PFBA have
been attributed to activation of the peroxisome proliferator-activated receptor alpha (PPARa12)
(Rosenmai etal.. 2018: Biork and Wallace. 2009: Foreman etal.. 2009: Wolf etal.. 2008). Due to
reported cross-species differences in PPAR signaling potency and dynamics, the potential human
relevance of some hepatic effects has been questioned, particularly as it relates to differences in
PPARa activation and activity across species. The goal of the qualitative analysis described in this
section is to evaluate the available mechanistic evidence for PFBA-induced liver effects and to
assess the biological relevance of effects observed in animal models to possible effects in humans.
Although the database is smaller for PFBA than for some other PFAS, in vitro studies
demonstrate that PFBA activates PPARa in both rodent and human cell lines. Studies using rodent
cell lines or COS-1 cells transfected to express rodent PPARa generally report that exposure to
PFBA consistently results in activation of PPARa and increased expression of PPARa-responsive
genes f Rosen etal.. 2013: Wolf etal.. 2012: Biork and Wallace. 2009: Wolf etal.. 20081. Although
PFAS generally have been shown to activate PPARa, however, shorter chain PFAS such as PFBA
appear to be weak activators. For example, Biork and Wallace f20091 showed PFBA is a weaker
activator of PPARa in primary rat and human hepatocytes than is either the six-carbon PFHxA or
the eight-carbon PFOA. PFBA is also one of the weakest mouse and human PPARa activators
compared with other longer chain PFAS [i.e., C5-C12; Rosen etal. (2013): Wolf etal. (2012): Wolfet
12PPARa is a member of the nuclear receptor superfamily that can be activated endogenously by free fatty
acid derivatives. PPARa plays a role in lipid homeostasis but is also associated with cell proliferation,
oxidative stress, and inflammation (NIDWOI. 2017: Angrish et al„ 2016: Mellor etal.. 2016: Hall et al.. 20121.
3-37
-------
Toxicological Review of PFBA and Related Salts
al. (2008)]. These studies also observed diminished effects and transcription levels in human cell
lines (primary hepatocytes) or COS-1 cells transfected with human PPARa compared to mice
(primary hepatocytes or transfected COS-1 cells). One study using the human hepatoma cell line
HepG2 also reported activation of PPARa after exposure to PFBA for 24 hours, further
demonstrating that the human PPARa can be activated by PFBA fRosenmai etal.. 20181.
Interestingly, when modeling the slope of PPARa activation in human hepatoma cells for various
PFAS, Rosenmai etal. (2018) observed PFBA (slope = 7.4 x 10~3) was a stronger activator than
PFOA (slope = 4.9 x 10"3). Foreman et al. (2009) investigated PPARa activation in the liver of mice
following in vivo exposure to PFBA. The PPARa-responsive gene CYP4A10 was activated to a
greater degree in wild-type mice than in humanized mice, but acyl-CoA oxidase (ACO, active in
(3-oxidation and lipid metabolism) appeared to be activated to a similar magnitude in both
wild-type and humanized mice. The known PPAR a/y activator Wy-14,643 activated CYP4A10 and
ACO to a similar magnitude in humanized PPARa mice compared to PFBA but to a lesser degree in
wild-type mice. Neither gene was activated following exposure to PFBA or Wy-14,643 in PPARa
null mice. Although the database is smaller for PFBA than for some other PFAS, in vitro studies
demonstrate that PFBA activates PPARa in both rodent and human cell lines. Studies using rodent
cell lines or COS-1 cells transfected to express rodent PPARa generally report that exposure to
PFBA consistently results in activation of PPARa and increased expression of PPARa-responsive
genes f Rosen etal.. 2013: Wolf etal.. 2012: Biork and Wallace. 2009: Wolf etal.. 20081. Although
PFAS generally have been shown to activate PPARa, however, shorter chain PFAS such as PFBA
appear to be weak activators. For example, Biork and Wallace (2009) showed PFBA is a weaker
activator of PPARa in primary rat and human hepatocytes than is either the six-carbon PFHxA or
the eight-carbon PFOA. PFBA is also one of the weakest mouse and human PPARa activators
compared with other longer chain PFAS [i.e., C5-C12; Rosen etal. f20131: Wolf etal. f20121: Wolfet
al. f20081]. These studies also observed diminished effects and transcription levels in human cell
lines (primary hepatocytes) or COS-1 cells transfected with human PPARa compared to mice
(primary hepatocytes or transfected COS-1 cells). One study using the human hepatoma cell line
HepG2 also reported activation of PPARa after exposure to PFBA for 24 hours, further
demonstrating that the human PPARa can be activated by PFBA (Rosenmai etal.. 2018).
Interestingly, when modeling the slope of PPARa activation in human hepatoma cells for various
PFAS, Rosenmai etal. f20181 observed PFBA (slope = 7.4 x 10-3) was a stronger activator than
PFOA (slope = 4.9 x 10-3). Foreman et al. f20091 investigated PPARa activation in the liver of mice
following in vivo exposure to PFBA. The PPARa-responsive gene CYP4A10 was activated to a
greater degree in wild-type mice than in humanized mice, but acyl-CoA oxidase [ACO, active in
(3-oxidation and lipid metabolism) appeared to be activated to a similar magnitude in both
wild-type and humanized mice. The known PPAR a/y activator Wy-14,643 activated CYP4A10 and
ACO to a similar magnitude in humanized PPARa mice compared to PFBA but to a lesser degree in
3-38
-------
Toxicological Review of PFBA and Related Salts
wild-type mice. Neither gene was activated following exposure to PFBA or Wy-14,643 in PPARa
null mice.
One in vivo study Foreman etal. f20091 provided evidence that oral PFBA exposure elicits
apical, toxicological effects in humanized PPARa mice. This study showed that increased liver
weight and hepatocellular hypertrophy were induced following exposure to >35 mg/kg-day PFBA
in wild-type and hPPARa mice. Although magnitude of liver-weight increases was larger for
wild-type mice, the effect on hypertrophy was the same for wild-type and hPPARa mice at higher
exposures. Conversely, hPPARa mice had more severe lesions at lower doses compared with
wild-type mice. Increased liver weight and hypertrophy also occurred in positive controls treated
with Wy-14,643.
Liver enlargement is one of the most common observations associated with chemical
exposures via the oral route in laboratory animals and humans. In addition to measured increases
in the mass of liver tissue, histological evaluation typically reveals isolated or multifocal areas of
hepatocellular hypertrophy. The swelling of hepatocytes could include accumulation of lipid
material (e.g., micro- or macrovesicular steatosis), organellar growth and proliferation
(e.g., peroxisomes, endoplasmic reticulum), increased intracellular protein levels (e.g., Phase I and
II enzymes), and altered regulation of gene expression (e.g., stress response, nuclear receptors) (for
review see, Batt and Ferrari f!99511. Importantly, hepatocellular hypertrophy alone is
morphologically indistinguishable between an adaptive or toxic response in the absence of
additional indicators of cell status Williams and Iatropoulos (2002). such as reduced glutathione
(GSH) levels, mitochondrial integrity, receptor-dependent or independent signal transduction
pathway activity (e.g., pro-survival vs. pro-cell death balance), or redox state, for example. Although
hepatocellular hypertrophy is commonly attributed to receptor-dependent organellar growth and
proliferation (e.g., PPAR mediated), the milieu of pathways involved in modulating hepatocyte
structural and functional response to chemicals are diverse fWilliams and Iatropoulos. 20021. For
example, hepatocyte swelling also has been associated with cell death processes, in particular
oncosis or oncotic necrosis (Kleiner et al.. 2012). Several liver diseases or conditions, such as
ischemia-reperfusion injury, drug-induced liver toxicity, and partial hepatectomy, have noted
oncosis (oncotic necrosis) upon cellular/tissue examination (for review see, Kass (2006): Taeschke
and Lemasters (2003)) and are not dependent on peroxisome proliferation or PPAR signaling.
Rather, cellular alterations such as a transition in mitochondrial membrane permeability and
caspase activation (especially Caspase-8) have been identified as key mediators or tipping points
for a shift from a hypertrophic (oncotic) hepatocellular phenotype to apoptotic or primary necrotic
cell death (Malhi etal.. 2006: Van Cruchten and Van Den Broeck. 2002). As such, an assumption that
chemical-induced hepatocellular hypertrophy is by default a distinctly proliferative/growth
response associated exclusively with PPAR signaling might not be accurate.
One study investigated the activation of PPARa and pregnane X receptor (PXR) in the livers
of exposed neonatal mice fDas etal.. 20081. This study showed the expression of genes associated
3-39
-------
Toxicological Review of PFBA and Related Salts
with either PPARa or PXR was not increased in the livers of neonatal male and female mice,
possibly indicating that the increased liver weights in these animals were associated with a non-
PPARa or PXR MOA. No other PFBA-specific studies investigated activation of other isoforms of
PPAR (e.g., PPARy) or additional pathways (e.g., constitutive androstane receptor [CAR] or
pregnane X receptor [PXR]); however, evidence from human cell culture experiments involving
PFOS and PFOA, two of the most heavily studied PFAS, suggest the possibility of other non-PPARa
MOAs operational in liver toxicity. For example, PFOA and PFOS exposure is associated with PPARy
activation (Beggs etal.. 2016: Buhrke etal.. 20151. and increased mRNA levels of CAR and PXR
responsive genes (Abe etal.. 2017: Zhang etal.. 20171. Activation of these hepatic nuclear receptors
plays an important role in regulating responses to xenobiotics and in energy and nutrient
homeostasis fdi Masi et al.. 20091. Animal studies of other PFAS also provide some evidence
suggesting that nuclear receptor pathways other than PPARa might be involved in PFAS-induced
liver effects. For example, two separate in vivo studies using PPARa null animal models report
increases in absolute and relative liver weight (Das etal.. 2017: Rosen etal.. 20171 and in
hepatocellular hypertrophy and lipid accumulation (Das etal.. 20171 following PFHxS or PFNA
exposure. Multiple in vivo studies have also evaluated activation of CAR and PXR in rodents
exposed to PFDA: PFDA exposure in wild-type C57BL6/6J mice led to increased nuclear
translocation of CAR and mRNA levels of CAR/PXR responsive genes [CYP2B10 and CYP3A11; Abe
etal. f20171]: these effects were not observed in CAR or PXR null mice. PFDA has also been
observed to activate PXR in human HepG2 cells (Zhang etal.. 20171 and increase mRNA levels of
CAR/PXR-regulated genes (CYP2B6 and CYP3A4) in primary human hepatocytes (Rosen etal..
20131.
In addition to hypertrophy, Foreman etal. (20091 also observed additional
histopathological effects. Hepatic focal necrosis was statistically significantly increased following
exposure of wild-type mice to >175 mg/kg-day PFBA. Although no statistically significant increases
in focal necrosis were observed at any dose in PPARa null or humanized mice, necrosis did increase
slightly in the highest dose compared to controls (2/10 vs. 0/10) in both strains; that exposure to
higher doses of PFBA would elicit increased necrotic effects in hPPARa or PPARa null mice is
possible. Foreman etal. (20091 suggest that, given the differences in pharmacokinetics between the
strains (see Section 3.1) and lower liver concentrations of PFBA in humanized and null mice, that
higher levels of exposure could possibly elicit a similar phenotype in these strains. Interestingly, no
statistically significant increase in focal necrosis was observed in any mouse strain treated with
Wy-14,643 in this study. That PFBA exposure resulted in statistically significant increases in liver
necrosis in wild-type mice, but not PPARa null mice, suggests that PPARa is required for the
development of this lesion. The observation that the positive control for PPARa activation, Wy-
14,643 also did not result in statistically significant increase in this lesion (in this study) further
supports that a PPARa-independent, complementary, or multifaceted MOA could be active in the
observed liver toxicity. Supporting this conclusion is the observation that centrilobular and
3-40
-------
Toxicological Review of PFBA and Related Salts
periportal vacuolation (i.e., lipid accumulation) was increased compared with controls in PPARa
null and humanized mice after exposure to 350 mg/kg-day PFBA, with greater vacuolation in
PPARa null mice than in humanized mice. Vacuolation was not reported in wild-type mice, and
results for the vacuolation endpoints were provided only for the control and high-dose groups for
the PPARa null and hPPARa mice. This result is consistent with Das etal. f20171 reported PFAS
increased accumulation and oxidation of lipids in the liver of exposed mice, with accumulation
occurring faster than oxidation. Thus, although vacuolation occurs in humanized PPARa mice,
oxidation is also induced (as evidenced by the upregulation of ACO), limiting lipid accumulation to a
degree. In PPARa null mice, however, accumulation of lipids in the liver of exposed animals must be
occurring through a PPARa-independent mechanism. Thus, PFBA appears to result in increased
lipid accumulation in the liver via a PPARa-independent mechanism, and although humanized mice
do exhibit an increase in (3-oxidation via ACO upregulation, this increase in lipid catabolism is not
sufficient to overcome the increased lipid deposition in the liver.
The observation of increased liver weight, increased incidence of hepatocellular
hypertrophy, vacuolation, and necrosis in wild-type and humanized PPARa mice is important when
considered in the context of the recommendations of the Hall etal. (20121 paper. In interpreting
"histological changes caused by an increase in liver weight"—exactly the situation observed in
PFBA-exposed hPPARa mice in Foreman etal. f20091—Hall etal. f20121 suggests that coincident
histological evidence of liver injury/damage can be used to support the conclusion that the liver
weight increases/histological changes (i.e., hypertrophy) are adverse. Among the histological
changes that Hall etal. (2012) identifies as sufficient supporting evidence is necrosis and steatotic
vacuolar degeneration, with the study authors further differentiating between macrovesicular
vacuolation (considered nonadverse) and microvesicular vacuolation. Microvesicular vacuolation is
described by the presence of hepatocytes partially or completely filled with multiple small vacuoles
without displacement of the nucleus fKleiner and Makhlouf. 20161. This pattern of vacuolation is
precisely what Foreman et al. f20091 observed in hPPARa mice exposed to PFBA. Additionally, focal
necrosis is observed in wild-type mice in Foreman etal. (2009). Thus, according to the Hall
recommendations, observation of liver weight increases, hypertrophy, microvesicular vacuolation,
and necrosis across wild-type and hPPARa mice is consistent with a determination that these
interconnected PFBA-induced liver effects meet the criteria for adversity.
Accumulation of lipids in the liver is an apical key event (decreased fatty acid efflux
resulting in lipid accumulation) leading to hepatic steatosis fAngrish etal.. 2016: Kaiser etal.. 20121
and has been observed in animal toxicological studies following exposure to numerous
environmental agents that ultimately cause steatosis (Toshi-Barve etal.. 2015: Wahlang etal..
2013). Sustained steatosis can progress to steatohepatitis and other adverse liver diseases such as
fibrosis and cirrhosis (Angrish et al.. 2016). Therefore, that vacuolation occurring in null PPARa
mice indicates a PPARa-independent mechanism for lipid accumulation in the liver, possibly as a
3-41
-------
Toxicological Review of PFBA and Related Salts
precursor to more severe forms of liver injury. The occurrence of vacuolation in humanized mice
further supports the human relevance of the observed hepatic toxicity.
Disrupted lipid metabolism due to PFBA exposure is supported by the findings of a dermal
toxicity study Weatherlv et al. f20211 that observed significant upregulation of genes associated
with steatosis (Cd36, Fasn, Lpl, Scdl), cholestasis [Abcd4, Abcc2, Abcc3), and phospholipidosis
[Fabpl, Hpn, Lss).
Overall, evidence specific to PFBA and from other potentially relevant PFAS provides
support for both PPARa dependent and independent pathway contributions to hepatic toxicity, and
further, that activation of humanized PPARa by PFBA can likewise result in hepatic effects of
concern. Additionally, application of the recommendations from Hall etal. f20121Additionallv.
application of the recommendations from Hall etal. f20121 clearly supports the conclusion that the
multiple and interconnected effects observed in the livers of exposed animals meet the criteria for
adversity.
Evidence Integration Summary
No association between PFBA and circulating levels of multiple serum biomarkers of
hepatic injury were observed in the only available, medium confidence epidemiological study with
reduced sensitivity fNian etal.. 2019bl. These null findings from a single study with low sensitivity
did not influence the evidence integration judgments, providing indeterminate evidence.
Hepatic effects associated with oral exposures to PFBA have been consistently observed in
high or medium confidence short-term and subchronic oral studies in adult mice or rats of both
sexes (Bute nhoff etal.. 2012a: Foreman et al.. 2009: van Otterdiik. 2007a. b; Permadi etal.. 1993:
Permadi etal.. 19921 and in an oral developmental toxicity study in mice (Das etal.. 20081.
Although there are hepatic effects observed in a single dermal toxicity study Weatherlv etal.
f20211. concerns over characterizing how dermal exposures relate to oral exposures preclude the
use of this study in evidence synthesis judgments. Overall, changes in liver weights and
histopathology (hepatocellular hypertrophy) were consistently observed across two species, with
effects occurring in male adult rats and mice, female pregnant or nonpregnant adult mice, and in
male and female neonatal mice. In particular, increases in liver weight and hepatocellular
hypertrophy incidence occurred at similar dose levels across species, occurred at multiple doses,
and appeared to be dose related (i.e., increasing magnitude of effect with increasing dose), as can be
seen in this interactive graphic on HAWC. Although uncertainties remain, given the consistency,
coherence, and inferred adversity (see below) of these findings, there is moderate animal evidence
for hepatic effects of PFBA exposure.
Increased liver weights were consistently observed in male, but not female, adult rats
following 28- or 90-day exposures (Butenhoff etal.. 2012a: van Otterdiik. 2007a. b) and in male
wild-type and hPPARa mice, pregnant and nonpregnant female mice, and neonatal male and female
mice on PND 1 fForeman et al.. 2009: Das etal.. 2008: Permadi etal.. 1993: Permadi et al.. 19921.
For male rodents, the doses at which effects occurred did not differ appreciably across species, but
3-42
-------
Toxicological Review of PFBA and Related Salts
wild-type PPARa mice seemed to exhibit greater magnitudes of effect vs. humanized PPARa mice or
rats. As noted above, female pregnant and nonpregnant mice, along with their offspring, exhibited
effects only at higher doses compared with adult male rats and mice, possibly relating to the
observation that female rodents eliminate PFBA much more rapidly than males (see Section 3.1.4).
Liver histopathology was also consistently observed across PFBA studies fButenhoffetal..
2012a: Foreman etal.. 2009: van Otterdiik. 2007a. b), although differences in the type or severity of
lesions differed somewhat across species and durations of exposure. Wild-type and hPPARa mice
were both observed to develop statistically significantly increased hepatocellular hypertrophy
following 28 days of oral exposure to PFBA, whereas only wild-type mice developed statistically
significantly increased hepatic focal necrosis fForeman et al.. 20091. PPARa null mice were not
observed to develop statistically significant increases in either of these lesions in response to
exposure. Adult male rats also were observed to develop hepatocellular hypertrophy, but not
coagulative necrosis, following 28 or 90 days of exposure fButenhoffetal.. 2012a: van Otterdiik.
2007a. b). Again, differences in pharmacokinetics might explain somewhat the differences in lesion
incidence across species, with rats eliminating PFBA much more rapidly than mice. Interestingly,
PPARa null and hPPARa mice were observed to develop centrilobular and periportal vacuolation,
whereas wild-type mice did not. This possibly indicates the accumulation of lipids within the liver.
Increased liver weights were concurrently observed at all doses with hepatocellular hypertrophy in
wild-type and hPPARa mice following short-term exposure fForeman et al.. 20091. In wild-type
mice, however, liver weight increases occurred at lower doses than did focal necrosis in the same
study Foreman etal. (20091 although focal necrosis was not observed in hPPARa mice in the
presence of liver weight changes at any dose. In male rats, changes in liver weight occurred at lower
doses than hepatocellular hypertrophy following 28-day exposures, whereas both effects were
observed at the same dose following 90-day exposures fButenhoffetal.. 2012a: van Otterdiik.
2007a. bl.
Changes in serum biomarkers of liver function or injury were not consistently observed
across exposure durations or concurrently with hepatocellular lesions. In the 28-day study in rats,
prothrombin time alterations were observed only at 150 mg/kg-day; no statistically significant
changes in ALT, AST, or ALP were observed. Although increased ALP and increased hepatocellular
hypertrophy were both observed in male rats exposed to 30 mg/kg-day for 90 days in the
subchronic study, no concurrent increase in ALT and AST was observed at this exposure level.
Further, the observed decreased bilirubin is inconsistent with what would be expected as a marker
of liver injury (i.e., an increase in bilirubin); therefore, this observation is of unclear toxicological
significance as a marker of liver injury. Lastly, cholesterol levels were decreased in a dose-
dependent manner following the 28-day, but not the 90-day, exposure. Although ALT was also not
statistically significantly increased in wild-type, hPPRAa, or PPARa mice following exposure to
PFBA, ALT was increased almost 3-fold in PPARa null mice in the high dose group (350 mg/kg-
day). As a whole, the various clinical chemistry endpoints, as measurements of liver toxicity, were
3-43
-------
Toxicological Review of PFBA and Related Salts
incoherent across some endpoints and durations of exposure. Thus, these data (i.e., on serum
biomarkers of liver function) were considered too uncertain and not influential to the overall
evidence integration judgments (i.e., the coherent, consistent, and biologically plausible liver weight
and histopathology endpoints were strong enough on their own to support the evidence integration
judgements).
One characteristic of the evidence base for PFBA is the sparsity of chemical-specific
mechanistic data to inform the human relevance of the observed increases in liver weight and
hypertrophic lesions in rats and mice. In the one study that does provide chemical-specific
information, Exposure of wild-type and hPPARa mice to PFBA increased both liver weights and
hepatocellular hypertrophy. Only wild-type mice were observed to have statistically significantly
increased focal necrosis following exposure, possibly indicating that activation of PPARa was a
necessary step in the MOA for developing this lesion. Hepatic focal necrosis, however, was not
statistically significantly increased in any group (wild-type, hPPARa, or PPARa null mice) exposed
to the positive control (the PPARa/y activator Wy-14,643). Further, increased vacuolation was
reported only in PPARa-null and hPPARa mice, an observation consistent with in vivo evidence for
longer chain PFAS (Das etal.. 2017). This observation (increased vacuolation) in PPARa-null and
humanized mice indicates that lipid accumulation in the liver occurs, at least in part, through a
PPARa-independent mechanism, and that either the lack, or attenuated activity, of PPARa-induced
lipid catabolism is not sufficient to overcome the increased accumulation. This strongly suggests a
complementary or multifaceted MOA for development of PFBA-induced hepatic effects. Indeed,
based on evidence from other PFAS chemicals, non-PPARa mechanisms relevant to hepatic effects
are apparent. In vivo and in vitro studies of PFOA, PFOS, PFDA, and PFNA demonstrate that PFAS
exposure can activate PPARy, CAR, and PXR (Abe etal.. 2017: Das etal.. 2017: Zhang etal.. 2017:
Beggs etal.. 2016: Buhrke etal.. 2015: Rosen etal.. 20131 and that activation of these receptors
results in the hepatic effects observed in PPARa null mice.
Thus, multiple lines of evidence, taken as a whole, indicate that the liver toxicity observed in
rodents due to PFBA exposure is likely adverse, relevant to humans, and dependent on multiple
biological pathways (i.e., both PPARa-dependent and independent pathways). Even considering a
PPARa-only MOA, human PPARa is observed to be activated by PFBA exposure in vitro, and
evidence in humanized PPARa mice (increased liver weight and increased hepatocellular
hypertrophy, which is observed to be more severe than that in wild-type mice) indicates the
PPARa-mediated components of the undefined MOA(s) appear relevant to human toxicity, given the
effects are observed in animals with human PPARa receptors. Further, the existing evidence base
also supports the operation of PPARa-independent pathways for other hepatotoxic effects, given
the direct observation of increased vacuolation in PPARa null mice in response to PFBA exposure,
an observation also occurring in humanized PPARa mice. Even in the absence of PPARa activity,
hepatic toxicity occurs that is possibly the precursor to more clearly adverse liver disease
(e.g., steatohepatitis, fibrosis, and cirrhosis). Thus, although there is uncertainty in relating the
3-44
-------
Toxicological Review of PFBA and Related Salts
sensitivity of hepatic changes observed in rodents to humans given the generally decreased
sensitivity of human responses to PPARa agonism, evidence from PFBA studies and studies on
other PFAS indicates that PPARa alone cannot be identified as the exclusive MOA for PFBA-induced
liver effects. Lastly, independent of conclusions regarding PPARa as the MOA, consideration of the
recommendations from Hall etal. T20121 also support a determination that the observed hepatic
effects in rodents are relevant to humans. Hall etal. f 20121 indicates coincident histological
evidence of liver injury/damage can be used to support the conclusion that liver
weight/hypertrophic effects are adverse. That PFBA induces a constellation of effects in the liver,
including increased liver weight, hypertrophy, vacuolation, and necrosis is clear from the in vivo
evidence in rodents. Therefore, according to Hall etal. f20121. these coincident effects are
consistent with the conclusion that PFBA-induced liver effects in rodents meet the criteria for
adversity.
The available animal evidence for effects on the liver includes multiple high and medium
confidence studies with consistent effects on liver weight and, separately, histopathology across
multiple species, sexes, exposure durations, and study designs (e.g., exposures during pregnancy); it
exhibits coherence between the effects on liver weights and histopathology and a clear biological
gradient (increasing effect with increasing dose); and the evidence is interpreted to be relevant to
humans. Taken together, the available evidence indicates that PFBA exposure is likely to cause
hepatic toxicity in humans (see Table 3-8), given relevant exposure circumstances. This judgment is
based primarily on a series of short-term, subchronic, and developmental studies in rats and mice,
generally exhibiting effects at PFBA exposure levels >30 mg/kg-day.
3-45
-------
Toxicological Review of PFBA and Related Salts
Table 3-8. Evidence profile table for hepatic effects
Evidence Stream Summary and Interpretation
Evidence Integration
Summary Judgment
Evidence from studies of exposed humans (see Section 3.2.2: Human Studies)
Studies, outcomes,
and confidence
Summary of key
findings
Factors that increase
certainty
Factors that decrease
certainty
Judgments and rationale
Evidence indicates (likely)
Serum Biomarkers
1 medium confidence
study;
1 low confidence study
• No association
between PFBA and
liver biomarkers or
blood lipids in studies
with poor sensitivity
• No factors noted
• No factors noted
ooo
Indeterminate
Primary basis:
Three high and one medium
confidence studies in male
adult rats and mice and
maternal and neonatal mice
(short-term, subchronic, and
gestational exposures) at
>30 mg/kg-d PFHxA
Evidence from in vivo animal studies (see Section 3.2.2: Animal Studies)
Studies, outcomes,
and confidence
Summary of key
findings
Factors that increase
certainty
Factors that decrease
certainty
Judgments and rationale
Human relevance:
Effects in rats are considered
relevant to humans (see
Section 3.2.2: Mechanistic
Evidence and Supplemental
Information)
Cross-stream coherence:
N/A (human evidence
indeterminate)
Susceptible populations and
lifestages:
None identified, although
those with preexisting liver
disease could be at greater
risk
Organ Weight
4 hiah, 2 medium, and
1 low confidence
studies in adult rats
and maternal and
neonatal mice:
• 14-d (x3)
• 28-d (x2)
• 90-d
• Gestational
• Increased liver
weight observed in:
o male adult rats at
>30 mg/kg-d
o female mice and
PND1 neonates at
>175 mg/kg-d
o male wild-type
PPARa and hPPARa
mice at >35 mg/kg-d
(no effects in PPARa
null mice)
• Reduced effects in
female rats could be
attributable to
pharmacokinetics
• Consistent
increases, across
most studies (one
null study)
• Dose-response in
most studies (one
null study)
• Coherence with
histopathology in
male rats and mice
(especially at high
dose)
• Magnitude of
effect, up to 112%
• High and medium
confidence studies
• No factors noted
®©o
Moderate
Findings were
considered adverse (as
determined using Hall
et al. (2012) criteria (see
Section 3.2.2:
Mechanistic Evidence
and Supplemental
Information),
consistent, dose
dependent, and
biologically coherent
3-46
-------
Toxicological Review of PFBA and Related Salts
Evidence Stream Summary and Interpretation
Evidence Integration
Summary Judgment
Nictrtnat hn Inou
• Hepatocellular
• Consistent cellular
• No factors noted
across multiple
Other inferences: the MOA
nibiupdinuiugy
hypertrophy
hypertrophy or
measures of hepatic
for liver effects is not fully
2 hiah and 1 medium
observed in:
focal necrosis
toxicity (i.e., liver weight
established, although
confidence studies in
o male adult rats at
across studies and
and histopathological
available evidence indicates
adult rats and mice:
30 mg/kg-d
species
changes). PPARa-
that multiple pathways are
• 28-d (x2)
(subchronic)
• Coherence with
dependence appears
likely involved
• 90-d
o male wild-type
liver weight effects
likely for some effects
PPARa and hPPARa
(especially at high
(focal necrosis) but not
mice at >35 mg/kg-d
doses)
others (vacuolation)
(short-term)
• Dose-response
Findings were
• Focal necrosis
• High and medium
considered adverse (as
observed in male
confidence studies
determined using Hall
wild-type PPARa
et al. (2012) criteria (see
mice exposed to
Section 3.2.2:
>175 mg/kg-d (short-
Mechanistic Evidence
term)
and Supplemental
• Vacuolation
Information),
observed in male
consistent, dose
PPARa-null and
dependent, and
hPPARa mice at
biologically coherent
350 mg/kg-day
across multiple
(short-term)
measures of hepatic
• Reduced effects in
toxicity (i.e., liver weight
female rats could be
and histopathological
attributable to
changes). PPARa-
pharmacokinetics
dependence appears
3-47
-------
Toxicological Review of PFBA and Related Salts
Evidence Stream Summary and Interpretation
Evidence Integration
Summary Judgment
Serum Biomarkers
2 high confidence
studies in adult rats:
• 28-d
• 90-d
• Increased ALP and
decreased bilirubin in
male or male and
female rats,
respectively, at
30 mg/kg-d
• High confidence
studies
• Incoherent
observations (e.g.,
increased ALP but with
no clear increases in
ALT or AST, and
bilirubin decreased not
increased as expected)
likely for some effects
(focal necrosis) but not
others (vacuolation)
Mechanistic evidence and supplemental information (see subsection above)
Biological events or
pathways
Summary of key findings, interpretation, and limitations
Evidence stream judgment
Molecular Initiating
Events—PPARa
Key findings and interpretation:
• In vitro increased expression of PPARa-responsive genes in primary
rata and human hepatocytes and cells transfected with rat or human
PPARa.
• In vivo increased expression of PPARa-responsive genes in wild-type
and hPPARa mice.
Limitations: small database investigating PPARa activation, some
inconsistencies regarding the strength of activation or interspecies
differences.
Overall, studies in rodent
and human in vitro and in
vivo models suggest that
PFBA induces hepatic
effects, at least in part,
through PPARa. The
evidence also suggests a
role for PPARa-
independent pathways in
Molecular Initiating
Events—Other
Pathways
Key findings and interpretation:
• Indirect evidence of alternative pathways following observation of
effects in humanized PPARa and PPARa null mice exposed to PFBA.
• Direct evidence from other PFAS (PFOA, PFOS, PFDA, PFHxA, PFHxS)
that multiple non-PPARa pathways (PPARy, CAR, PXR) activated
following exposure.
Limitations: No PFBA-specific in vitro data; only one in vivo study providing
indirect evidence.
liver effects of PFBA.
Organ Level Effects
Key findings and interpretation:
• Observation of increased liver weight and increased hepatocellular
hypertrophy/vacuolation in humanized PPARa mice.
3-48
-------
Toxicological Review of PFBA and Related Salts
Evidence Stream Summary and Interpretation
Evidence Integration
Summary Judgment
• Concurrent observation that a known PPARa/y activator (Wy-14,643)
did not elicit the same statistically significant increased effects (focal
necrosis) as PFBA exposure in wild-type mice.
Limitations: Only one in vivo study.
3-49
-------
Toxicological Review of PFBA and Related Salts
3.2.3. Developmental Effects
This section describes studies of PFBA exposure and potential early life effects or
developmental delays and effects attributable to developmental exposure. The latter includes all
studies where exposure is limited to gestation or early life. As such, this section has some overlap
with evidence synthesis and integration summaries for other health systems where studies
evaluated the effects of developmental exposure (see Sections 3.2.2 and 3.2.4 on potential "Hepatic
Effects" and "Reproductive Effects," respectively). Synthesis descriptions of studies across sections
can vary in detail, depending on the impact the data have on summarizing the evidence relevant to
that hazard; typically, earlier hazard sections will include a more detailed discussion that is then
cited in later sections.
Human Studies
The one epidemiological study that investigated developmental effects (birth weight,
gestational age) Li etal. (2017a) was cross-sectional study based on umbilical cord PFBA exposure
deemed low confidence primarily due to concerns over participant selection and exposure
measurement Li etal. f2017al reported a mean birth weight deficit of-46 grams (95%CI: -111,
19) in the overall population per each unit (ng/mL) PFBA increase; this was driven by the
association in boys (-86 grams; 95%CI: -180, 9) as the results were null in girls. The exposure
range in this study, however, is quite small and a one-unit increase is beyond the bounds of the
exposure range in this population. Thus, when expressed on an IQR unit change, they reported
small birth weight deficits (-4 grams (95%CI: -10, 2) per each PFBA IQR unit change (0.09 ng/mL)
and in boys (-8 grams; 95%CI: -16,1). No association was observed with gestational age in weeks.
Animal Studies
A standardized suite of potential developmental effects was evaluated in one high
confidence developmental toxicity study in mice fDas etal.. 20081. Some outcome-specific
considerations for study evaluations were influential on the overall study rating for developmental
effects, but none of these individual domain-specific considerations were judged deficient, and the
Das etal. (2008) study considered further in this section was rated as high confidence (see
Figure 3-7). Endpoints evaluated in the study included time to eye opening, full litter resorption,
postnatal survival, vaginal opening, preputial separation, body weights, and morphological
evaluations (see Table 3-9 and Figure 3-8).
3-50
-------
Toxicological Review of PFBA and Related Salts
Reporting quality -
Allocation -
Observational bias/blinding -
Confounding/variable control -
Selective reporting and attrition -
Chemical administration and characterization
Exposure timing, frequency and duration
Endpoint sensitivity and specificity
Results presentation
Overall confidence
Legend
| Good (metric) or High confidence {overall)
Adequate (metric) or Medium confidence (overall)
Deficient (metric) or Low confidence (overall)
Critically deficient (metric) or Uninformative (overall)
I NR| Not reported
Figure 3-7. Evaluation results for animal studies assessing developmental
effects of perfluorobutanoic acid (PFBA) exposure (see interactive data
graphic for rating rationales).
Oral exposure via gavage from GD 1 to 17 of CD-I mice (male and female offspring were
evaluated) to NH4+PFB resulted in delayed eye opening by 1.1,1.4, and 1.5 days compared to
controls at 30,175, and 350 mg/kg-day, respectively fDas etal.. 20081. Significantly increased full
litter resorptions also occurred at 350 mg/kg-day (28% vs. 7% in controls), although no effects
were observed on the number of implants or live fetuses. Additionally, although not statistically
significant, postnatal survival was consistently reduced atPNDs 7,14, and 21 by approximately 5%.
The male and female reproductive developmental landmarks (preputial separation and vaginal
opening, respectively) were delayed. Preputial separation was delayed by 2.3 days at
350 mg/kg-day although vaginal opening was delayed 3.3 and 3.6 days (175 and 350 mg/kg-day,
respectively). No changes were observed in neonatal or postweaning body weight Anatomical
changes were observed (renal dilation, fetal hydronephrosis, and absent testis) but were randomly
distributed among the treatment groups, including controls, and thus were not attributable to PFBA
exposure.
3-51
-------
Toxicological Review ofPFBA and Related Salts
Table 3-9. Developmental effects observed following perfluorobutanoic acid
(PFBA) exposure in a developmental toxicity study
Animal group
Dose (mg/kg-d)
0
35
175
350
Full-litter resorptions; pregnant Po female CD-I mice on GD 18
Das et al. (2008)
2/29
1/29
4/28
8/29
Survival to PND 1 (%); ft male and female CD-I mice on PND 1
Das et al. (2008)
91.7 ±2.1
90.2 ±2.4
92.9 ± 1.6
87.9 ±2.6
Survival to PND 7 (%); Fi male and female CD-I mice on PND 7
Das et al. (2008)
90.9 ±2.3
90.0 ±2.3
90.0 ±3.1
86.4 ±2.7
Survival to PND 14 (%); Fi male and female CD-I mice on PND 14
Das et al. (2008)
90.9 ±2.3
89.7 ±2.4
89.6 ±3.2
85.7 ±3.0
Survival to PND 21 (%); Fi male and female CD-I mice on PND 21
Das et al. (2008)
90.9 ±2.3
88.7 ±2.4
89.6 ±3.2
85.7 ±3.0
Delayed eye opening (d); Ft male and female CD-I mice
Das et al. (2008)
16.28 ± 1.19
17.38 ± 0.79
17.69 ± 0.68
17.8 ± 0.83
Delayed vaginal opening (d); Fi female CD-I mice
Das et al. (2008)
31.25 ±2.62
33.71 ±2.59
34.57 ± 2.59
34.92 ± 2.23
Delayed preputial separation (d); Fi male CD-I mice
Das et al. (2008)
29.55 ± 1.14
30.21 ± 1.99
30.56 ± 1.84
31.88 ± 1.72
Bolded cells indicate statistically significant changes compared to controls.
Study Name
Endpotnt Name
Study Type
Animal Description
Observation Time
Das 2008 1290825
Fun Utter Resorption (FIR) Number
17 Day Oral
P0 Mouse CD-I
z)
18 0 days
Full Letter Resorption iFLR) Litter Implants
17 Da/ Oral
P0 Mouse CD-1
2)
18 0 days
Live Fetuses
17 Day Oral
FI Mouse CD-I
£ =)
18 0 days
Fetal Weight
17 Day Oral
FI Mouse CD-I
£1)
18 0 days
Fetal Renal Dilation
17 Day Oral
FI Mouse CD-I
= 2)
18 0 days
Fetal Hydronephrosis
17 Day Oral
FI Mouse CO-1
= S>
18 0 days
Fetal Absent Testis
17 Day Oral
FI Mouse CD-I
= )
18 0 days
Live Litter implants
17 Day Oral
FI Mouse CD-1
= = )
19 0 days
Live Births
17 Day Oral
FI Mouse CD-I
='=)
19.0 days
Live Births/implants
17 Day Oral
FI Mouse CD-I
£1)
190 days
Survival to PND 1
17 Day Oral
FI Mouse CD-I
£1)
20 0 days
Survival to PND 7
17 Day Oral
FI Mouse CD-I
£1)
26 0 days
Survival to PND 14
17 Day Oral
FI Mouse. CD-1
£1)
33 0 days
Survival to PN0 21
17 Day Oral
F1 Mouse CD-I
£1)
40 0 days
Eye Opening
17 Day Oral
FI Mouse CD-I
£1)
29 0 days
Vaginal Opening
17 Day Oral
FI Mouse CD-I
S)
450 days
Preputial Separation
17 Day Oral
FI Mouse CD-I
(£)
450 days
PFBA Developmental Effects
150 200
Axis label
0 Doses
A Treatmenl-Related Increase
^ Treatment-Related Decrease
H Dose Range
250 300 350 400
Figure 3-8. Pre- and postnatal developmental responses to gestational
ammonium perfluorobutanoate (NH4+PFB) exposure (see interactive data
graphic and rationale for study evaluations for developmental effects in
Health Assessment Workspace Collaborative [HAWC]).
3-52
-------
Toxicological Review of PFBA and Related Salts
Evidence Integration Summary
One low confidence human study reported lower birth weight in boys with higher PFBA
exposure. No association was observed with gestational age. The lack of additional studies with
lower risk of bias reduces the interpretability of these findings. Overall, the evidence on potential
developmental effects from studies of humans exposed to PFBA was indeterminate.
Coherent effects on developmental maturation were observed in one high confidence study
in mice Das etal. (20081 following in utero exposure to PFBA. The developmental effects of PFBA
exposure in this study included delayed eye opening, full-litter resorption, decreased survival, fetal
absent testis, and delays in vaginal opening and preputial separation, although pup growth and
body weight were unaffected. These effects indicate that PFBA appears to disrupt the normal
gestational and postnatal development of exposed fetuses. One factor increasing the strength of
evidence is that effects on the developing offspring (e.g., delayed eye opening, delays in the
development of the male and female reproductive systems) are seen following exposure to other
PFAS, most notably the structurally related compound perfluorobutane sulfonate (U.S. EPA. 2018b)
but other, longer chain PFAS as well. Following exposure to >200 mg/kg-day PFBS (U.S. EPA.
2018b) or 5 mg/kg-day perfluorooctanoic acid [PFOA; Lau etal. f20061] or perfluorooctane
sulfonate [PFOS; Lau etal. f2004-1]. similar delays in eye opening (~1.5 days) were observed in
mice. Similarly, following exposure to >200 mg/kg-day PFBS, time to vaginal opening was increased
by >3 days (Feng etal.. 2017) and time to vaginal patency was increased ~3 days in mice exposed
to 20 mg/kg-day PFOA (Lau etal.. 2006) and ~2 days in rats exposed to 30 mg/kg-day PFOA
(Butenhoff etal.. 2004). Time to reach reproductive milestones was also altered in male rodents
exposed to PFOA: preputial separation accelerated 2-4 days in mice exposed to doses up to 10
mg/kg-day and delayed ~1.5 days in mice exposed to 20 mg/kg-day fLau etal.. 20061. In rats
exposed to 30 mg/kg-day PFOA, preputial separation was delayed ~3.5 days fButenhoffetal..
20041. Thus, qualitatively, a consistent pattern of altered reproductive milestones is observed
following exposure to other PFAS, including the structurally related PFBS, increasing certainty in
the similar findings available for PFBA.
The onset of puberty in humans is driven by surges in the levels of estrogen in females and
testosterone in males, so the timing of puberty can be altered by exposure to endocrine disrupting
chemicals that mimic or antagonize these hormones. In female rodents, pubertal markers include
vaginal opening (indicative of the first ovulation in rats, but not mice) and the subsequent first
estrus and onset of regular estrous cyclicity (rats and mice) fPrevot. 20141. Since vaginal opening
isn't indicative of first ovulation in mice, the delayed vaginal opening in mice reported by Das et al.
(2008), not a direct correlate to puberty in humans. However, it is assuredly a marker of sexual
and/or reproductive development As the EPA's Reproductive Guidelines (U.S. EPA. 1996) state that
both accelerations and delays in the timing of puberty should be considered adverse, it is
reasonable to extend this conclusion to developmental milestones that are more broadly indicative
of sexual and/or reproductive developmental, Further, the absence of effects on body weight in
3-53
-------
Toxicological Review of PFBA and Related Salts
PFBA-exposed offspring or maternal toxicity strengthens the confidence that the observed
developmental delays are biologically significant, adverse effects. Taken together, the available
animal studies provided moderate evidence of potential developmental effects.
Data gaps in the developmental toxicity database include a lack of information on the
thyroid and nervous system following gestational exposure. Given that PFBS alters thyroid
hormone levels following gestational exposure and that PFBA induces changes in thyroid hormone
levels in exposed adult animals, PFBA also might alter normal thyroid hormone action in the
developing fetus. As both PFBA and PFBS evidence bases lack studies on developmental
neurotoxicity, a potential consequence of altered thyroid hormone action during development, this
represents an important unknown.
Thus, considering the coherent suite of developmental effects, primarily developmental
delays, observed following PFBA exposure in one high confidence study, and similar effects
observed following exposure to multiple other PFAS (including the structurally similar PFBS), the
evidence indicates PFBA exposure is likely to cause adverse developmental effects in humans (see
Table 3-10), given relevant exposure circumstances. The basis for this judgmentis a single high
confidence gestational exposure study in mice, with multiple adverse effects occurring at PFBA
exposure levels >175 mg/kg-day (with delays in eye opening occurring at >35 mg/kg-day). Notably,
even in the absence of evidence informing potential similarities of effects between PFBA and other
PFAS regarding gestational thyroid hormone action, the available PFBA-specific developmental
effects alone support this judgment
3-54
-------
Toxicological Review of PFBA and Related Salts
Table 3-10. Evidence profile table for developmental effects
Evidence Stream Summary and Interpretation
Evidence Integration
Summary Judgment
Evidence from studies of exposed humans (see Section 3.2.3: Human Studies)
0®Q
Evidence indicates (likely)
Primary basis:
One high confidence gestational
study in mice, with effects
observed at >35 mg/kg-d PFBA
Human relevance:
In the absence of evidence to
the contrary, the
developmental effects observed
in mice are considered relevant
to humans based on conserved
biological processes
Cross-stream coherence:
N/A (human evidence
indeterminate)
Susceptible populations and
lifestages:
Pregnancy and early life
Other inferences:
PFBA-induced developmental
effects are consistent with
effects seen for other PFAS (see
Section 3.2.3: Evidence
Integration Summary
Studies, outcomes, and
confidence
Summary of key findings
Factors that increase
certainty
Factors that decrease
certainty
Judgments and
rationale
Birth Weight
1 low confidence study
• Birth weight deficit with
higher PFBA exposure in
boys (nonstatistically
significant)
• No factors noted
• Low confidence study
• Imprecision
ooo
Indeterminate
Evidence from in vivo animal studies (see Section 3.2.3: Animal Studies)
Studies, outcomes, and
confidence
Summary of key findings
Factors that increase
certainty
Factors that decrease
certainty
Judgments and
rationale
Developmental
Milestones
1 hiqh confidence
gestational study in
mice
• Dose-dependent delays
in developmental
milestones in:
o Eye opening in males
and females at >
35 mg/kg-d
o Preputial separation
in males at 350
mg/kg-d
o Vaginal opening in
females at 175 and
350 mg/kg-d
• Increased full litter
resorption at 350 mg/kg-
d
• No effects on pup weight
• Dose-response
gradient
• Coherence across
developmental
milestones
• Magnitude of
effect, up to 12%
increase in time to
milestone and
4-fold increase in
full litter
resorptions
• High confidence
study
• No factors noted
®©o
Moderate
Coherent delays in
developmental
milestones, with
multiple alterations
observed at
>35 mg/kg-d
3-55
-------
Toxicological Review of PFBA and Related Salts
3.2.4. Reproductive Effects
Human Studies
One low confidence cross-sectional study Song et al. f20181 examined the association
between PFBA exposure and semen parameters. No evidence of an association between PFBA
exposure and decreased semen quality was found (correlation coefficients were -0.03 for semen
concentration and 0.2 for progressive motility), although issues were noted during study evaluation
regarding the ability of this study to detect an effect due to the small sample size [n = 58) and risk of
outcome misclassification, which makes the null finding difficult to interpret. Other study
deficiencies including the potential for selection bias and confounding were noted in the study
evaluation, but the direction of these biases is unknown.
Animal Studies
Two high confidence studies reported in three publications from the same research group
Butenhoff et al. (2012a): van Otterdiik (2007a. 2007b) evaluated the effects of PFBA exposure on
reproductive organ weights in rats (see Figure 3-9). In addition, one high confidence developmental
toxicity study Das etal. f20081 reported several delays in reproductive system development
(e.g., vaginal opening, preputial separation) after gestational exposure in mice. These latter results
are synthesized and integrated with other studies examining developmental outcomes (see Section
3.2.3) given the apparent coherence of findings of developmental delays after PFBA exposure and
the general lack of other studies or effects on reproduction, including an absence of studies on
functional measures (see discussion below).
Organ weight
Short-term exposure (28 d) to PFBA in male S-D rats increased absolute epididymis weight
(note: absolute organ weights are typically preferred for these reproductive organ measures) 10%
compared to controls, but only at the lowest dose [6 mg/kg-day; Butenhoff et al. (2012a): van
Otterdiik (2007a)]. In a separate cohort, this effect was not observed following a 3-week recovery
period (at 49 days) from exposure at any dose (6, 30, or 150 mg/kg-day). Changes in absolute or
relative testis weight were not observed in rats following either 28 days of exposure or during the
recovery period. Similarly, no changes in absolute or relative ovary weight were observed in rats
following short-term (28 days) PFBA exposure and none arose during the recovery period
fButenhoff etal.. 2012a: van Otterdiik. 2007al.
3-56
-------
Toxicological Review ofPFBA and Related Salts
Study Name Endpoint Name
Study Type
Animal Description
Observation Time
Butenhoff, 2012, 1289535 Testes Weight, Absolute
28 Day Oral
Rat Sprague-Davriey (£)
28 0 days
49 0 days
Epididymis Weight. Absolute
28 Day Oral
Rat Sprague-Davriey {£)
28 0 days
49.0 days
Testes Weight. Relative
28 Day Oral
Rat Sprague-Davriey {£)
280 days
49 0 days
Epididyrnidts Weight. Relative
28 Day Oral
Rat Sprague-Davriey (£)
28Qdays
49 0 days
Ovary Weight Absolute
28 Day Oral
Rat. Sprague-Davriey (3)
28.0 days
Ovary Weight Relative
28 Day Oral
Rat Sprague-Davriey (2)
28 0 days
490 days
Ovary Weight Absolute
28 Day Oral
Rat Sprague-Davriey (5)
280 days
PFBA Reproductive Effects
0 Doses
A Treatment-Reiated Increase
^Treatment-Related Decrease
H Dose Range
40 60 SO
Axis label
Figure 3-9. Reproductive responses to ammonium perfluorobutanoate
(NH4+PFB) exposure (see interactive data graphic and rationale for study
evaluations for reproductive effects in Health Assessment Workspace
Collaborative |HAWC|).
Evidence Integration Summary
The database of studies examining the potential for PFBA exposure to elicit effects on
reproductive parameters is limited to one human and one animal study. There is evidence for
delayed development of the reproductive system (i.e., delayed vaginal opening and preputial
separation) following gestational PFBA exposure fDas etal.. 20081. These latter results are
synthesized and integrated in the developmental effects section (see Section 3.2.3] where the
human relevance of delayed development of the reproductive system observed in mice are outlined
and not discussed further in this section.
In the only available human study (a low confidence study), no association was observed
between semen quality and PFBA exposure. Null findings in a single study with low sensitivity
(biased toward the null] are not interpreted to influence the evidence integration judgments, and
thus the human evidence was indeterminate.
The available animal evidence is sparse, limited to evaluations of reproductive
organ-weight measurements in a high confidence short-term experiment reported in three
publications from the same research group (Butenhoff etal.. 2012a: van Otterdiik. 2007a. b).
Specifically, the authors evaluated reproductive organ weights in a cohort of rats immediately after
exposures ended and another cohort 21 days postexposure, both of which were largely null. Given
the limited interpretability of these data, the animal evidence was indeterminate.
Given the sparsity of evidence on potential reproductive effects, the relative insensitivity of
the outcome measures (organ weights] in animals, and the largely null findings, there is
inadequate evidence to determine whether PFBA exposure has the potential to cause reproductive
effects in humans (other than the developmental delays discussed in Section 3.2.3; see Table 311].
This determination is consistent with the lack of convincing evidence for reproductive effects for
3-57
-------
Toxicological Review of PFBA and Related Salts
several other PFAS, including Gen X, PFOA, PFOS, and PFBS fU.S. EPA. 2021a. 2018b. 2016a. b) and
CU.S. EPA. 2021bl.
3-58
-------
Toxicological Review of PFBA and Related Salts
Table 3-11. Evidence profile table for reproductive effects
Evidence Stream Summary and Interpretation
Evidence Integration Summary
Judgment
Evidence from studies of exposed humans (see Section 3.2.4: Human Studies)
OOO
Inadequate Evidence
Primary basis:
One high confidence study in rats
Human relevance:
Organ weight changes in rats are
considered relevant to humans in
the absence of evidence to the
contrary
Cross-stream coherence:
N/A (human evidence
indeterminate)
Susceptible populations and
lifestages:
None identified
Studies, outcomes,
and confidence
Summary of key
findings
Factors that increase
certainty
Factors that decrease
certainty
Judgments and
rationale
Birth Weight
1 low confidence
study
• No association
between PFBA
exposure and
semen quality
• No factors noted
• Low confidence
study
ooo
Indeterminate
Evidence from in vivo animal studies (see Section 3.2.4: Animal Studies)
Studies, outcomes,
and confidence
Summary of key
findings
Factors that increase
certainty
Factors that decrease
certainty
Judgments and
rationale
Organ weights
1 hiqh confidence
28-d study in rats
• Increased
epidydimal weight
in rats at 6 mg/kg-d
but not higher
doses
• No changes in
testis or ovary
weights
• No factors noted
• Lack of dose-
response
• Lack of coherence
across
reproductive organ
weights
OOO
Indeterminate
Largely null findings in
in the only available
study that examined
reproductive organ
weights
3-59
-------
Toxicological Review of PFBA and Related Salts
3.2.5. Other Noncancer Health Effects
In addition to the potential health effects outlined above, some epidemiological studies have
examined the potential for associations between PFBA exposure and immunosuppression, blood
pressure, and renal function, while several experiments in rats and mice have examined potential
effects of PFBA exposure on body weight (note: these data were used to inform interpretation of
the health effects discussed in prior sections), hematological effects, and ocular effects. Given the
paucity of studies available and the lack of consistent or coherent effects of PFBA exposure, there is
inadequate evidence to determine whether any of these evaluated outcomes might represent
potential human health hazards of PFBA exposure. Additional studies on these health effects could
modify these interpretations.
Human Studies
Two studies examined the association between PFBA exposure and immunosuppression.
One medium confidence study examined severity of COVID-19 illness in Denmark using biobank
samples and national registry data (Grandjean etal.. 20201. There was some concern for selection
bias in this study due to the expectation that biobank samples were more likely to be available for
individuals with chronic health concerns. In addition, severity of COVID-19 is not a direct measure
of immune suppression as other factors may contribute to illness severity. This study reported
higher odds of severe disease (based on hospitalization, admission to intensive care and/or death)
with higher exposure to PFBA (OR = 1.57, 95% CI = 0.96, 2.58 for >LOD vs. LOD in all participants;
OR = 2.10, 95% CI = 1.02, 4.33 in only participants with exposure measured at the time of diagnosis,
which reduces concern for selection bias). In addition, one low confidence cross-sectional study in
China analyzed hepatitis B surface antibody fZeng etal.. 20201. This study was downgraded due to
concerns for exposure misclassification resulting from lack of temporality between the exposure
and outcome. There is no way to determine when participants were exposed to hepatitis B;
additionally, vaccination status was not considered. This study reported an inverse association
between PFBA exposure and hepatitis B surface antibody (mean difference = -0.18 log mlU/mL,
95% CI = -0.28, -0.08). Overall, both available studies report findings consistent with immune
suppression with greater PFBA exposure. However, there is residual uncertainty in both studies,
and, in the absence of additional corroboration (see animal evidence discussion below), they do not
support a stronger judgment than concluding that there is inadequate evidence of
immunosuppression. In addition, neither study is suitable for dose-response modeling due to
dichotomous exposure modeling Grandjean etal. (2020) and study limitations Zeng etal. (20201.In
addition, neither study is suitable for dose-response modeling due to dichotomous exposure
modeling Grandjean etal. (20201 and study limitations Zeng etal. (20201.
One medium confidence cross-sectional study examined the association between PFBA
exposure and blood pressure and reported statistically significant increased odds of hypertension
(OR = 1.10 [95%CI: 1.04-1.17 per ln-PFBA, ng/mL]) and increased systolic blood pressure
3-60
-------
Toxicological Review of PFBA and Related Salts
(P = 0.80 mm HG [95%CI: 0.25-1.34 per ln-PFBA, ng/mL]). This is despite narrow exposure
contrast (median 0.16 ng/mL, IQR 0.01-0.54). Although this was a medium confidence study,
potential for bias remains; this includes outcome misclassification resulting from the volatility of
blood pressure and its measurement at a single time point and the cross-sectional design. In the
absence of additional confirmatory epidemiological studies, or other supportive findings (e.g., from
animal studies), the results of this observational study alone are interpreted as "inadequate
evidence."
One low confidence cross-sectional study Wang etal. (2019) examined the association
between PFBA exposure and renal function. They reported statistically significant lower estimated
glomerular filtration rate ((3: -0.5, 95%CI: -0.8, -0.1 [change in GFR (mL/min/1.73 m2) per
1 ln-serum PFAS (ng/mL)]) and higher, though not significant, odds of chronic kidney disease
(OR: 1.1, 95%CI: 1.0,1.2). There is potential for reverse causation in this association, however. In
essence, as described in Watkins etal. (2013), decreased renal function (as measured by decreased
GFR or other measures) could plausibly lead to higher levels of PFAS, including PFBA, in the blood.
This hypothesis is supported by data presented by Watkins etal. (2013), although the conclusions
are somewhat uncertain because of the use of modeled exposure data as a negative control and the
potential for the causal effect to occur in both directions. Consequently, there is considerable
uncertainty in interpreting the results of studies of this outcome.
Animal Studies
Body-weight changes were evaluated in multiple high and medium confidence short-term
and subchronic-duration studies in rats and mice (Butenhoff et al., 2012a; Foreman et al., 2009;
Das etal.. 2008: van Otterdiik. 2007a. b). In general, no PFBA-related effects on body weight were
observed in any study. Foreman etal. (2009) reported that body weighs were not affected in any
exposure group of Sv/129 mice. Initial and final body weights were statistically significantly lower
in humanized PPARa (hPPARa) Sv/129 mice exposed to 350 mg/kg-day PFBA compared to all
other groups, but this was explained by random assignment of animals; body weights in this group
actually increased slightly during the study, indicating the lower measured body weights were not
treatment related. The change in body weight across the duration of the study was not changed at
any dose in any group of animals, however, indicating PFBA exposure had no deleterious effect on
adult body weight in mice. Maternal, preweaning, and postweaning body weights were not altered
by PFBA exposure in CD-I mice fDas etal.. 20081. Adult body weights were not altered in S-D rats
exposed to PFBA for either 28 or 90 days fButenhoffet al.. 2012a: van Otterdiik. 2007a. bl. PFBA
appears not to affect body weight across multiple species, exposure durations, or lifestages.
Some evidence of effects on the hematological system was observed in S-D rats exposed to
PFBA. Following 28 days of exposure, no effects other than on prothrombin time (PT; a measure of
clotting potential) were observed (van Otterdiik. 2007a. b). In males, PT was statistically
significantly decreased 6% following exposure to 150 mg/kg-day PFBA, whereas in females,
statistically significant decreases of 4% and 5% were observed in the 6- and 30-mg/kg-day dose
3-61
-------
Toxicological Review of PFBA and Related Salts
groups, respectively. PT was decreased 4% in the 150-mg/kg-day dose group in females, but the
decrease was not statistically significant. Following the recovery period, no statistically significant
decreases in PT were found in male rats, but consistent statistically significant 7%-8% decreases in
PT were observed in all exposed female dose groups (p < 0.01). Hematological effects were more
pronounced following 90-day exposures. In males, red blood cell counts, hemoglobin, and
hematocrit were decreased 4%, 6%, and 5%, respectively, and red blood cell distribution width was
increased 5% following exposure to 30 mg/kg-day PFBA. Although the number of RBCs and the
RBC distribution width were observed to return to control values following recovery, hemoglobin
and hematocrit remained decreased 5% relative to control. Mean corpuscular hemoglobin and
mean corpuscular hemoglobin concentration were decreased 2%-3% in female rats exposed to
30 mg/kg-day PFBA. These effects returned to control levels following recovery. Taken as a whole,
although some hematological effects were observed in exposed rats, the effect sizes were quite
small, they generally returned to control levels following a recovery period, and no consistency of
effects across exposure durations or sexes were found.
Immunotoxicity were observed in mice dermally exposed to up to 15% v/v PFBA
(Weatherlv etal.. 20211. Following 28 days of dermal exposure, the frequency of CD4+ and CD8+ T-
cells were decreased in the draining lymph nodes, whereas B-cells, dendritic cells, and CDllb+ cell
numbers were increased. Cell frequencies were also changed in the ear pinna following dermal
exposure: CD45+, CD4 T-cell, CD8 T-cell, NK cell, eosinophils, neutrophils, and dendritic cell
numbers were all increased. Relatively fewer changes were noted in the spleen, however, where
total cells, B-cells, CD4 T-cells, neutrophils, and CDllb dendritic cells were all decreased. It is not
clear that these data relate to the uncertain evidence of potential immunosuppressive effects
observed in the two human studies, and thus they do not strengthen that evidence. Further, no oral
PFBA toxicity studies that investigated immunotoxicity were available. Overall, this single animal
study does not provide evidence supporting a stronger evidence integration judgment and overall,
there is inadequate evidence to determine whether PFBA exposure can cause immunological
effects in humans.
Ocular effects also were observed in rats exposed to PFBA for 28 or 90 days (van Otterdiik.
2007a. b). In male rats exposed for 28 days, a delayed bilateral pupillary reflex was observed at
150 mg/kg-day. Although examination of neuronal tissue (including the optic nerve) revealed no
histopathological effects, ocular histological effects were observed. Outer retinal degeneration,
characterized as a loss of 25%-30% of photoreceptors, was observed along with a decrease
(20%-35%) in retinal thickness. Ocular effects also were also observed in the 90-day subchronic
study: Delays in pupillary dilation were observed at weeks 8 and 12 in rats exposed to
30 mg/kg-day. These delays were reported to be unilateral, not consistent across the treatment
period, and low incidence. No ocular histopathological results were observed in the 90-day
subchronic study. Thus, although some ocular effects were observed following PFBA exposure,
3-62
-------
Toxicological Review of PFBA and Related Salts
effects across durations were somewhat inconsistent, with greater effects following short-term
exposures than in subchronic exposures. This limited the interpretability of the observed effects.
3.3. CARCINOGENICITY
No human or animal studies were available to inform the potential for PFBA exposure to
cause cancer. Only one study Crebelli et al. (20191 investigated PFBA-induced genotoxicity: No
evidence of DNA damage or micro nucleus formation was observed in male mice exposed to PFBA
via drinking water for 5 weeks. As shown in Table 4-2, EPA's carcinogenicity conclusion for the
closely related PFAS, PFBS, is inadequate information to assess carcinogenic potential. While
there is some evidence of carcinogenicity for PFOA and PFOS, the ability to relate the findings for
these longer chain PFAS to PFBA is currently too uncertain to influence the carcinogenicity
judgment for PFBA.
Overall, there is inadequate information to assess the carcinogenic potential of PFBA exposure.
3-63
-------
Toxicological Review of PFBA and Related Salts
4. SUMMARY OF HAZARD IDENTIFICATION
CONCLUSIONS
4.1. SUMMARY OF CONCLUSIONS FOR NONCANCER HEALTH EFFECTS
The currently available evidence indicates hazards likely exist with respect to the potential
for thyroid, liver, and developmental effects in humans, given relevant PFBA exposure conditions.
These judgments are based on data from short-term (28-day exposure), subchronic (90-day
exposure), and developmental (17-day gestational exposure) oral-exposure studies in rodents.
Further characterizations of the exposure conditions relevant to the identified hazards are
provided in Section 5. A summary of the justifications for the evidence integration judgments for
each of the main hazard sections is provided below, organized by health effect, and further
summarized in Table 4-1.
The hazard identification judgment that the evidence indicates PFBA exposure is likely to
cause thyroid toxicity in humans (given relevant circumstances) is based primarily on a short-term
and subchronic study in male rats reporting a consistent and coherent pattern of hormonal, organ
weight, and histopathological changes, generally at PFBA exposure levels >30 mg/kg-day, although
some notable effects were observed at 6 mg/kg-day. For effects on the thyroid in exposed animals,
PFBA-induced perturbations were observed in one species and sex (male rats) across two different
exposure durations (short-term and subchronic). Consistent, dose-dependent decreases in total and
free T4 were observed independent of any effect on TSH. Additionally, increased thyroid weights
and increases in thyroid follicular hypertrophy were observed. Although the observed thyroid
histopathological changes support the potential for PFBA to disrupt the thyroid hormone
homeostasis, however, rodents are uniquely sensitive to the development of thyroid follicular
hypertrophy and tumor development U.S. EPA (1998b) compared with humans. Because of the
similarities in the production and regulation of thyroid hormone homeostasis between rodents and
humans and the consistency of the observed pattern of effects with changes observed in humans,
the effects in rodents were considered relevant to humans. A detailed discussion of thyroid effects
is included in Section 3.2.1.
The hazard identification judgment that the evidence indicates PFBA exposure is likely to
cause hepatic toxicity in humans, given relevant exposure circumstances, is based primarily on a
series of short-term, subchronic, and developmental studies in rats and mice, generally exhibiting
effects at PFBA exposure levels >30 mg/kg-day. The PFBA-induced effects were observed in two
species (rats and mice), in males and females, and across multiple exposure durations (short-term,
subchronic, and gestational). Consistent, coherent, dose-dependent, and biologically plausible
effects were observed for increased liver weights and increased incidences of hepatic
4-1
-------
Toxicological Review of PFBA and Related Salts
histopathological lesions. Supporting the biological plausibility and human relevance of these
effects is mechanistic information that suggests non-PPARa MOAs could explain some of the
observed effects in exposed rodents and that observed effects might be precursors to clearly
adverse health outcomes such as steatosis. Supporting this conclusion is evidence from other PFAS
that have consistently shown that longer chain PFAS can activate non-PPARa nuclear receptors,
including PPARy, CAR, and PXR, although there is uncertainty in inferring a similar relationship for
the short-chain PFB A.
The hazard identification judgment that the evidence indicates PFBA exposure is likely to
cause developmental effects in humans (given relevant exposure circumstances), including
increased prenatal effects (full-litter resorptions) and delays in developmental milestones (days to
eye opening, vaginal opening, and preputial separation) without effects on fetal (pup) growth is
based on a single study in mice exposed gestationally to PFBA. Although the observed
developmental effects due to PFBA exposure were investigated in only one high confidence study,
they demonstrate a constellation of effects affecting the developing organism that is internally
coherent (within-study) and consistent across related PFAS compounds, including PFBS, PFOA, and
PFOS.
There was inadequate evidence to determine whether PFBA exposure has the potential to
cause reproductive toxicity (in adults), effects on hematological or clinical chemistry markers,
ocular effects, changes in blood pressure, or effects on renal function in humans. Other potential
health outcomes have not been evaluated in the context of PFBA exposure. Most notably, potential
for PFBA exposure to affect the immune system, thyroid or nervous system in developing
organisms, or mammary glands represent important data gaps given the associations observed for
other PFAS, such as PFBS, PFOA, PFOS, and GenX (U.S. EPA, 2021a; MDH, 2020, 2019, 2018; U.S.
EPA. 2018b. 2016a. b). See Table 4-2 for a comparison of the noncancer hazard judgments drawn
for PFBA with the judgments in the final EPA assessments for PFBS, PFOA, PFOS, and GenX.
4-2
-------
Toxicological Review of PFBA and Related Salts
Table 4-1. Evidence integration summary for health effects for which evidence
indicates a hazard exists
Evidence stream scenarios
Evidence in
studies of
humansa
Evidence in
animal studies3
Evidence basis
O
'i_
ro
c
a»
(j
uo
ro
cu
s_
+-»
uo
Q)
U
c
Q)
~a
Q)
CU)
c
o
+-»
UO
V
Wo Studies, or Low
Confidence or
Conflicting
Evidence
Strong
Mechanistic
Evidence Alone
One High or
Medium
Confidence Apical
Study without
Supporting or
Conflicting
Evidence
Multiple High or
Medium
Confidence Apical
Studies with Some
Inconsistency or
Important
Uncertainties
Multiple High or
Medium
Confidence Apical
Studies with
Strong Support
(e.g., MOA
understanding
supporting
biological
plausibility)
Developmental
Hepatic
Thyroid
Developmental
Thyroid
Hepatic
Developmental
• No human studies
• Coherent observations of delays in developmental
milestones (eye opening, vaginal opening,
preputial separation) and fetal mortality in one
high confidence study of mice exposed
gestationally
• Consistent with findings for related PFAS
• No MOA information
• Human relevance presumed
Thyroid
• Single low confidence study in humans
• Consistent and biologically coherent results for
thyroid hormone levels (T4 without compensatory
changes in TSH), organ weights, and
histopathology from two high confidence studies
(short-term, subchronic) in male rats
• Consistent with findings for related PFAS
• No MOA information
• Human relevance presumed
Hepatic
• Two null studies (one medium and one low
confidence) in humans with poor sensitivity
• Consistent, dose-dependent, and biologically
coherent effects on liver weights and
histopathology from seven high or medium
confidence studies in adult male rats and mice
(short-term and subchronic) and adult and female
mice exposed as adults or gestationally
• PPARa-dependence observed for some effects
(focal necrosis) but other effects (vacuolation)
occur in animals lacking PPARa activity (null mice)
or in animals with human PPARa (humanized mice)
• Involvement of both PPARa-dependent and
independent mechanisms, including hypertrophic
responses in humanized PPARa mice
• MOA information supports human relevance
aCan include consideration of studies informing biological plausibility: For studies in humans, this includes studies
of human tissues or cells, and other relevant simulations; for animal studies, this includes ex vivo and in vivo
experiments and other relevant simulations.
4-3
-------
Toxicological Review of PFBA and Related Salts
Table 4-2. Hazard conclusions across published EPA PFAS human health
assessments
Health Outcome
EPA PFAS Assessmentsa b c
PFBA
PFBS
GenX Chemicals
PFOAd
PFOSd
Thyroid
+
+
_e
Human: +
Animal: +/-
Human: +/-
Animal: +/-
Liver
+
-
+
Human: +
Animal: +
Human: -
Animal: +
Developmental
+
+
+/-
Human: +
Animal: +
Human: +
Animal: +
Reproductive
-
-
+/-
Human: -
Animal: +/-
_e
Immunotoxicity
-
-
+/-
Human: +
Animal: +
Human: +/-
Animal: +
Renal
-
+
+/-
Human: +/-
Animal: +/-
_e
Hematological
-
_e
+/-
_e
_e
Ocular
-
_e
_e
_e
_e
Serum Lipids
_e
-
_e
Human: +
Animal: +
Human: +
Hyperglycemia
_e
_e
_e
Human: -
Animal: -
Animal: +/-
Nervous System
_e
_e
_e
Human: -
Animal: -
Animal: +/-
Cardiovascular
_e
-
_e
_e
_e
Cancer
-
-
+/-
+/-
+/-
a Assessments used multiple approaches to summarizing their non-cancer hazard conclusions; for comparison
purposes, the conclusions are presented as follows:'+' =evidence demonstrates or evidence indicates (e.g., PFBA), or
evidence supports (e.g., PFBS);=suggestive evidence;= inadequate evidence (e.g., PFBA) or equivocal evidence
(e.g., PFBS); and = sufficient evidence to conclude no hazard (no assessment drew this conclusion).
b The assessments all followed the EPA carcinogenicity guidelines (2005) U.S. EPA (2005) a similar presentation to that
used to summarize the noncancer judgments is applied for the cancer hazard conclusions, as follows:'+' =
carcinogenic to humans or likely to be carcinogenic to humans;= suggestive evidence of carcinogenic potential;
' = inadequate information to assess carcinogenic potential; and = not likely to be carcinogenic to humans (no
assessment drew this conclusion).
c The hazard conclusions for the various EPA PFAS assessments presented in this table were not considered during
evidence integration and thus did not inform the evidence integration conclusions presented in the PFBA assessment.
dThe U.S. EPA PFOA U.S. EPA (2016b) and PFOS U.S. EPA (2016a) assessments did not use structured language to
summarize the noncancer hazard conclusions. The presentation in in this table was inferred from the hazard
summaries found in the respective assessments; however, this is for comparison purposes only and should not be
taken as representative of the conclusions from these assessments. Those interested in the specific noncancer hazard
conclusions for PFOA and PFOS must consult the source assessments. Note that new assessments for PFOA and PFOS
are currently being finalized to support a National Primary Drinking Water Regulation; hazard conclusions in these
updated assessments may differ from those presented in this table.
e No data available for this outcome for this PFAS, so 'entered by default.
4-4
-------
Toxicological Review of PFBA and Related Salts
4.2. SUMMARY OF CONCLUSIONS FOR CARCINOGENICITY
No human or animal studies were available to inform the potential for PFBA exposure to
cause cancer and the single study of genotoxicity did not observe effects. Overall, there is
inadequate information to assess the carcinogenic potential of PFBA exposure. See Table 4-2 for
a comparison of the carcinogenicity conclusion drawn for PFBA with the carcinogenicity
conclusions in the final EPA assessments for PFBS, PFOA, PFOS, and GenX.
4.3. CONCLUSIONS REGARDING SUSCEPTIBLE POPULATIONS AND
LIFESTAGES
No human studies were available to inform the potential for PFBA exposure to affect
sensitive subpopulations or lifestages.
In adult animals exposed subchronically, PFBA exposure was consistently observed to elicit
stronger responses in male rats compared with female rats. The reason for this sex dependence is
most likely due to differences in pharmacokinetics between males and females. The serum half-life
of PFBA following a single oral dose of 30 mg/kg-day is approximately 9 hours in male rats,
compared to 2 hours in female rats (see Table 3-2). Differences in half-life values is similar in mice,
with male mice having much longer half-lives than females at 30 mg/kg-day (16 hours vs. 3 hours).
Urinary excretion rates are much faster in female rodents compared to male rodents
(approximately 50%-90% faster), possibly due to renal reabsorption of PFBA in male rats by
organic anion transporters (OATs). However, as noted in Section 3.1.4, PFBA is not an active
substrate of OAT1, OAT2, or OATPlal which are expressed in the kidney and active towards other
PFAS, and as described atthe end of Section 3.1.5, it seems unlikely thaturinary excretion of
albumin (which is not sex-dependent in control rats (Matsuzaki etal.. 2002)) could explain the
observation.
Further, and specifically relevant to hepatic effects, the liver concentrations of PFBA
following subchronic exposure to 30 mg/kg-day is approximately 16-fold higher in male rats than
in female rats [16.09 vs. 0.91 mg/kg-day; Butenhoff et al. f2012al: van Otterdiik f2007a. 2007b}],
and responses for liver weight are 11-fold higher (33% vs 3% increase, see Table 3-6). 0.91 mg/kg-
day; Butenhoff et al. (2012a): van Otterdiik (2007a. 2007b)]. and responses for liver weight are 11-
fold higher (33% vs 3% increase, see Table 3-6). On the other hand, the estimated clearance in male
rats is only 4.5-fold lower than in female rats (see Table 3-2). Thus, PFBA clearance (primarily in
the urine) partly explains the sex difference in internal dose and liver weight effect, suggesting that
additional sex-related differences that impact the distribution to the liver contribute to the overall
difference in response for this endpoint. However, reductions in free and total T4 at 150 mg/kg-day
after 28 days and at 30 mg/kg-day after 90 days were 2- to 3-fold less severe in female rats
compared to male rats (see Table 3-3), which is somewhat less than the relative clearance (4.5-
fold), but sufficiently similar to suggest that the difference in clearance is a primary factor in the
difference in response.
4-5
-------
Toxicological Review of PFBA and Related Salts
No difference in serum half-lives was observed in monkeys exposed to a single i.v. dose of
10 mg/kg: 1.61 hours for males vs. 2.28 hours in females (Chang etal.. 20081. Also, although
quantitative data were not provided, serum excretion half-lives were reported not to differ between
males and females in the one occupational study available fChang etal.. 20081. Additionally, effects
on liver weight were observed in pregnant and nonpregnant mice fDas etal.. 20081. Developmental
effects also were observed in female fetuses/neonates (full litter resorption, delayed eye opening,
delayed vaginal opening) and male fetuses/neonates [full litter resorption, delayed eye opening,
delayed preputial separation; Das etal. (20081]. with no clear difference in sensitivity. Therefore,
although there does appear to be a clear sex dependence for some PFBA-induced health effects in
adult rodents, the observed lack of sex-specific sensitivity for other effects in adult and immature
rodents and the apparent lack of pharmacokinetic differences between sexes in primates (and a
single human occupational study) preclude the identification of males as a broadly sensitive
subpopulation for PFBA-induced health effects in humans.
Lastly, given the effects observed in pregnant mice (increased liver weights, full-litter
resorptions) and the developing organism (fetal/postnatal death and delays in time to eye opening,
vaginal opening, and preputial separation), that pregnancy and early life represent two sensitive
lifestages to PFBA exposure is possible.
4-6
-------
Toxicological Review of PFBA and Related Salts
5. DERIVATION OF TOXICITY VALUES
5.1. NONCANCER AND CANCER HEALTH EFFECT CATEGORIES
CONSIDERED
The available evidence indicates that oral exposure to PFBA is likely to cause adverse
thyroid, hepatic, and developmental effects in humans based on multiple high and medium
confidence animal toxicity studies (Bute nhoff etal.. 2012a: Foreman etal.. 2009: Das etal.. 2008:
van Otterdiik. 2007a. b; Permadi etal.. 1993: Permadi etal.. 19921.
No human or animal toxicity studies are available to inform the potential for PFBA to cause
adverse effects via inhalation. Likewise, no human or animal studies are available to inform the
potential for oral or inhalation exposure to cause genotoxicity or cancer.
5.2. NONCANCER TOXICITY VALUES
The noncancer oral toxicity values (i.e., reference doses) derived in this section are
estimates of an exposure for a given duration to the human population (including susceptible
subgroups and lifestages) that is likely to be without an appreciable risk of adverse health effects
over a lifetime. The RfD derived in Section 5.2.1 corresponds to chronic, lifetime exposure and is the
primary focus of this document In addition, RfDs specific to each organ or system are provided
(organ/system-specific RfDs), as these toxicity values might be useful in some contexts (e.g., when
assessing the potential cumulative effects of multiple chemical exposures occurring
simultaneously). Less-than-lifetime, subchronic toxicity values (including the subchronic RfD and
organ/system-specific subchronic RfDs), which are derived in Section 5.2.2, correspond to exposure
durations between 30 days and 10% of the life span in humans. These subchronic toxicity values
are presented because they might be useful for certain decision purposes (e.g., site-specific risk
assessments with less-than-lifetime exposures). Section 5.2.3 discusses that no information exists
to inform the potential toxicity of inhaled PFBA.
5.2.1. Oral Reference Dose (RfD) Derivation
Study Selection
Given the identified hazards relating to thyroid, liver, and developmental effects, two high
confidence studies reporting these effects were selected for the purpose of deriving an oral
reference dose (RfD). The subchronic Butenhoff et al. (2012a) and developmental Das etal. (2008)
studies were selected to support RfD derivation given the ability of these study designs to estimate
potential effects of lifetime exposure, as compared to short-term or acute studies. Both studies used
rats or mice as the laboratory animal species and used vehicle-exposed controls. Animals were
5-1
-------
Toxicological Review of PFBA and Related Salts
exposed to reagent-grade NH4+PFB (reported as >98% pure or as a 28.9% solution in distilled
water; impurities not reported) via a relevant route (oral administration via gavage) and for a
relevant duration (90 days or GD 1-17) of exposure.
Also available in the PFBA database are two short-term (i.e., 28-day) studies that provide
information on the hepatic and thyroid effects of PFBA fButenhoffetal.. 2012a: Foreman etal..
2009: van Otterdiik. 2007al. Although these studies were used for qualitative hazard identification
purposes (they supported the final evidence integration judgments for these endpoints and thus
were critical for identifying these endpoints for dose-response analysis), they ultimately were not
considered for use as the basis for the quantitative dose-response analyses. When developing a
lifetime reference value, chronic or subchronic studies (and studies of developmental exposure) are
generally preferred over short-term or acute studies. Likewise, subchronic and developmental
studies are preferred when developing a subchronic RfD. Although short-term studies were not
used for the identification of points of departure (PODs), however, they were deemed relevant to
decisions regarding the application of uncertainty factors for deriving toxicity values (see
"Derivation of Candidate Toxicity Values" below).
In the liver, a pattern of adverse effects has been observed in mice and rats, with PFBA
exposure resulting in increased liver weights (absolute and relative) in adult exposed animals
Butenhoff et al. f2012al: Das etal. f20081: van Otterdiik f2007bl in conjunction with
histopathological lesions [i.e., hepatocellular hypertrophy; Butenhoff et al. f2012al: van Otterdiik
(2007b)]. As discussed in Section 3.2.2, the observed effects in the livers of exposed experimental
animals are judged relevant to human health as evidenced by the observation of increased liver
weights and increased hepatocellular hypertrophy in mice expressing human PPARa and increased
vacuolation in humanized-PPARa and PPARa null mice. This strongly suggests a multifaceted mode
of action (MOA) for liver effects consisting, in part, of non-PPARa mechanisms operant in humans
(noting that activation of human PPARa by PFBA also results in hepatic changes). Further, the
observation of vacuolation specifically indicates the observed effects are possible precursors to
clearly adverse downstream effects such as steatohepatitis, fibrosis, and cirrhosis. Thus, the
observed pattern of liver effects in PFBA-exposed animals are judged to be adverse, relevant to
human health, and appropriate to consider for reference value derivation. For the purposes of dose-
response modeling, relative liver weights were chosen over absolute liver weights. Although body
weights were not affected on average in any PFBA study, relative liver weights are still preferred
because this measure of effect accounts for any changes in body weights that occur in individual
animals (changes in body and liver weights are associated). For liver hypertrophy, severity
information in addition to raw incidence was available. Therefore, both total incidence of lesions
and incidence of "slight" severity lesions were considered for dose-response analysis.
A pattern of adverse effects in the thyroid also is observed in exposed rats that consists of
decreased free and total T4 levels and increased incidence of thyroid follicular hypertrophy and
hyperplasia fButenhoffetal.. 2012a: van Otterdiik. 2007bl. Decreased thyroid hormone levels are
5-2
-------
Toxicological Review of PFBA and Related Salts
judged relevant to human health, given the many similarities in the production, regulation, and
functioning of thyroid hormones between rodents and humans. For effects on T4, total T4 was
chosen for dose-response modeling over free T4, on the basis of lack of data in the control group for
free T4 (given insufficient volume for the assay). In addition, rodents are more sensitive to
increases in thyroid follicular hypertrophy and hyperplasia, and thus changes in thyroid hormone
levels are considered more relevant for deriving human health toxicity values. For this reason, the
increases in thyroid hypertrophy/hyperplasia were not considered further for RfD derivation. Note,
however, that decreased total T4 was observed at 6 mg/kg-day in rats exposed to PFBA for 28 days,
but not in rats exposed for 90 days (where it was observed only at 30 mg/kg-day). This discrepancy
can be explained, however, by the difference in serum concentrations following 28- and 90-day
exposures. Serum free T4 concentrations were higher in the 6 mg/kg-day dose group following
28-day exposures (24.7 ng/mL) vs. 90-day exposures (6.1 ng/mL). This difference was reversed in
the 30 mg/kg-day dose group for the 28-day and 90-day animals, being 38.0 ng/mL vs. 52.2 |ig/mL,
respectively. Because serum concentrations following chronic exposures likely will resemble those
following subchronic exposures (more so than serum concentrations following short-term
exposures), the effects on total T4 following subchronic exposure are deemed most appropriate for
deriving lifetime and subchronic toxicity values.
Effects on the developing reproductive system included delays in vaginal opening and
preputial separation fDas etal.. 20081. EPA's Reproductive Toxicity Guidelines U.S. EPA f!9961
states that significant effects in the development of the male and female reproductive systems
"either early or delayed, should be considered adverse..." and thus supports considering these
endpoints for reference value derivation. Delayed eye opening, also found following PFOA
exposure, is identified as a "simple, but reliable" indicator of impaired postnatal development by
Das etal. f20081. Further, a delay of eye opening is a form of visual deprivation that prevents ocular
visual signals from reaching the brain during a critical period of development fWiesel. 19821. A
time-sensitive critical period in the development of the visual system is when the architecture of
the visual cortex is established Espinosa and Strvker (20121. and accordingly, any alterations of the
visual system during that time is considered adverse. Evidence in humans further supports the
adversity of this endpoint, given that infants born with congenital cataracts that interfere with the
processing of visual signals have permanent visual defects if the cataracts are removed after the
critical window for visual development fWiesel. 19821. Therefore, any delay in the development of
sight or development of the visual neurological system results in permanent functional decrements
and is relevant to human health.
Full litter resorption (FLR), a clear indicator of postimplantation embryo/fetal mortality,
was increased twofold and fourfold in pregnant mice exposed to 175 mg/kg-day or 350 mg/kg-day
(respectively) during pregnancy. In the uteri of dams without full resorptions, there was additional
evidence of fetal resorptions. In addition, in a separate cohort of gestationally exposed dams that
were allowed to deliver litters and were killed after their pups were weaned on lactation day 22,
5-3
-------
Toxicological Review of PFBA and Related Salts
there was an indication of decreased pre- and postnatal survival of the offspring (as determined by
a comparison of the number of maternal implantation sites to the number of pups delivered), the
magnitude of which is considered biologically significant (discussed below). Taken together, the
potential coherence of decreased pre- or postnatal survival with other effects on early fetal
mortality and developmental maturation (i.e., delays in eye opening and reproductive milestones)
supports consideration of all these developmental endpoints for deriving PODs.
Individual animal data were obtained from the study authors, which allowed for a thorough
consideration of pre- and postnatal mortality data. When the FLR data were combined with data for
prenatal mortality from litters without FLR to provide a more complete assessment of embryo/fetal
mortality, the response was statistically significant (p = 0.012) using the Cochran-Armitage trend
test with a Rao-Scott adjustment (CA/RS) method fRao and Scott. 19921. Although the embryo/fetal
mortality observed as FLR is presumed to have occurred much earlier in pregnancy than fetal
mortality in non-FLR litters and could involve different or overlapping contributing mechanisms,
combining these endpoints provides information on pregnancy loss and fetal mortality over the
entire gestational period, corresponding to the period of PFBA exposure. This was deemed more
appropriate than modeling FLR and non-FLR fetal mortality separately. Combining the data in this
way has the added benefit of allowing the data to be modeled with the nested dichotomous models
and avoids the lower resolution of modeling the FLR data as dam incidence per dose group.
The individual litter data obtained from the study authors also allowed for consideration of
modeling postnatal mortality (i.e., number of neonatal deaths compared to the number of
implantation sites). Analysis of the individual litter data revealed a nonmonotonic dose-response
for postnatal mortality, with response rates of 0.38%, 1.04%, 2.93%, and 1.2% at 0, 35,175, and
350 mg/kg-day, respectively, and the CA/RS trend test for the datasetwas not statistically
significant (p = 0.09). Further, the data for postnatal mortality clearly indicates it is a weaker
response compared to prenatal mortality. Given that postnatal mortality was a weaker response
than prenatal morality, it failed to achieve statistical significance, and prenatal mortality is more
closely aligned with the period of exposure, postnatal mortality was not considered further for POD
derivation.
The studies (excluding the short-term studies) and outcomes relevant to the identified
hazards were selected and advanced for POD derivation as presented in Table 5-1. These selected
datasets were evaluated for toxicity value derivation as described below and in Appendix D.
5-4
-------
Toxicological Review of PFBA and Related Salts
Table 5-1. Endpoints considered for dose-response modeling and derivation
of points of departure
Endpoint
Exposure duration
Species, sex
POD derivation3
Reference15
Liver
Increased relative liver weight
Subchronic
S-D rat, male
Yes
Butenhoff et al.
(2012a)
Gestational
CD-I mouse, female
Yes
Increased absolute liver weight
Subchronic
S-D rat, male
No
Gestational
CD-I mouse, female
No
Increased liver hypertrophy
Subchronic
S-D rat, male
Yes
Thyroid
Decreased total T4
Subchronic
S-D rat, male
Yes
Butenhoff et al.
(2012a)
Decreased free T4
Subchronic
S-D rat, male
No
Increased thyroid follicular
hypertrophy
Subchronic
S-D rat, male
No
Developmental
Embryo/fetal mortality
Gestational
CD-I mouse, male
and female
Yes
Das et al. (2008)
Postnatal mortality
Gestational
CD-I mouse, male
and female
No
Delayed eye opening
Gestational
CD-I mouse, male
and female
Yes
Delayed vaginal opening
Gestational
CD-I mouse, female
Yes
Delayed preputial separation
Gestational
CD-I mouse, male
Yes
a See text for rationale for inclusion/exclusion from POD derivation.
b Both the Butenhoff et al. (2012a) and Das et al. (2008) studies were rated as high confidence.
Estimation or Selection of Points of Departure (PODs)
Consistent with EPA's Benchmark Dose Technical Guidance U.S. EPA f20121. the BMD and
95% lower confidence limit on the BMD (BMDL) were estimated using a BMR to represent a
minimal, biologically significant level of change. The BMD technical guidance U.S. EPA (20121 sets
up a hierarchy by which BMRs are selected, with the first and preferred approach using a biological
or toxicological basis to define what minimal level of response or change is biologically significant.
If that biological or toxicological information is lacking, the BMD technical guidance recommends
BMRs that can be used instead, specifically a BMR of 1 standard deviation (SD) from the control
mean for continuous data or a BMR of 10% extra risk (ER) for dichotomous data. The BMRs
selected for dose-response modeling of PFBA-induced health effects are listed in Table 5-2 along
with the rationale for their selection.
5-5
-------
Toxicological Review of PFBA and Related Salts
Table 5-2. Benchmark response levels selected for benchmark dose (BMD)
modeling of perfluorobutanoic acid (PFBA) health outcomes
Endpoint
BMR
Rationale
Liver
Increased relative
liver weight
10% relative
deviation
A 10% increase in liver weight has generally been considered a minimally biologically
significant response.
Increased liver
hypertrophy
10% extra
risk
A 10% extra risk is a commonlv used BMR for dichotomous endooints U.S. EPA (2012) in the
absence of information for a biologically based BMR; the endpoint is not considered a frank
effect and does not support using a lower BMR.A 10% extra risk is a commonly used BMR for
dichotomous endooints U.S. EPA (2012) in the absence of information for a biologically based
BMR; the endpoint is not considered a frank effect and does not support using a lower BMR.
Thyroid
Decreased total T4
1 standard
deviation
Toxicological evidence that would support identification of a minimally biologically significant
response is lacking in adult animals. Further, evidence for the level of response in thyroid
hormones associated with neurodevelopmental effects is inconsistent, with decreases of
10%-25% identified in human and rodent studies (Gilbert et al.. 2016: Gilbert. 2011: Haddow
et al.. 1999).The BMD technical guidance (U.S. EPA. 2012) recommends a BMR eaual to 1
standard deviation for continuous endpoints when biological information is not sufficient to
identify the BMR. In this case, the BMR based on 1 SD from the Butenhoff et al. (2012a) study
corresponds to a ~13% decrease, consistent with the levels of decreased T4 associated with
neurodevelopmental decrements, thus strengthening the rationale for using a BMR = 1 SD for
this endpoint.
Developmental
Embryo/fetal
morality
1% extra risk
For quantal endpoints, the BMG Technical Guidance states "[f]from a statistical standpoint,
most reproductive and developmental studies with nested study designs support a BMR of
5%" and "[b]iological considerations may warrant the use of a BMR of 5% or lower for some
types of effects (e.g., frank effects)...". As increased treatment-related embryo/fetal mortality
is clearly a frank effect, BMRs of 5% and 1% were considered. Given that the study employed
a nested design with individual animal data available that allow the use of the nested
dichotomous models (to account for intra-litter similarity), and the effect of interest was a
frank effect (supporting a BMR 5% or lower), a BMR of 1% extra risk was ultimately selected
for derivation of the POD to account for the biological severity of these endpoints (i.e.,
mortality) and the robust statistical power of the study.
Delayed eye opening
5% relative
deviations
Biological evidence supports identification of a minimally significant decrease of visual input
(1-d delayed eve opening) during a critical period of retinal development (Espinosa and
Strvker, 2012). Delays of 1 d in eve opening reduces the time available for visual cortex
development related to orientation selectivity by approximately 20% Espinosa and Strvker
Delayed vaginal
opening
Delayed preputial
separation
(2012) and corresoonds to ~6% change in Das et al. (2008). Further, delavs in vaginal ooening
greater than or equal to 2 d have been used previously to define biologically relevant
responses U.S. EPA (2013), and this magnitude in delay in Das et al. (2008) is also ~6%. Both
levels of response are consistent with a 5% relative deviation. Lastly, a 5% change in other
markers of growth/development in gestational studies (e.g., fetal weight) has generally been
considered a minimally biologically significant response level.
When modeling was feasible, the estimated BMDLs were used as points of departure (PODs,
see Table 5-4). Further details, including the modeling output and graphical results for the model
selected for each endpoint, can be found in Appendix D. When dose-response modeling was not
feasible, or adequate modeling results were not obtained, NOAEL or LOAEL values were identified
based on biological rationales when possible and used as the POD. For example, for liver weight, a
5-6
-------
Toxicological Review of PFBA and Related Salts
NOAEL would be chosen as the dose below which causes at least a 10% change, consistent with the
rationale for the selecting the BMR for that endpoint. If no biological rationale for selecting the
NOAEL/LOAEL is available, statistical significance was used as the basis for selection. The PODs
(based on BMD modeling or NOAEL/LOAEL selection) for the endpoints advanced for dose-
response analysis are presented in Table 5-4.
Approach for Animal-Human Extrapolation of Perfluorobutanoic Acid (PFBA) Dosimetry
The PFAS protocol (Appendix A) recommends the use of physiologically based
pharmacokinetic (PBPK) models as the preferred approach for dosimetry extrapolation from
animals to humans, while allowing for the consideration of data-informed extrapolations (such as
the ratio of serum clearance values) for PFAS that lack a scientifically sound and sufficiently
validated PBPK model. If chemical-specific information is not available, the protocol then
recommends that doses be scaled allometrically using body weight (BW)3/4 methods. This
hierarchy of recommended approaches for cross-species dosimetry extrapolation is consistent with
EPA's guidance on using allometric scaling for deriving oral reference doses (U.S. EPA. 2011). This
hierarchy preferentially prioritizes adjustments that result in reduced uncertainty in the dosimetric
adjustments (i.e., preferring chemical-specific values to underpin adjustments vs. use of default
approaches).
No PBPK model is available for PFBA. But as pharmacokinetic data for PFBA exist in
relevant animals (rats, mice, and monkeys) and humans, a data-informed extrapolation approach
for estimating the dosimetric adjustment factor (DAF) can be used. Briefly, the ratio of the clearance
(CL) in humans to animals, CLh:CLa, can be used to convert an oral dose rate in animals
(mg/kg-day) to a human equivalent dose rate. Assuming the exposure being evaluated is low
enough to be in the linear (or first-order) range of clearance, the average blood concentration (Cavg)
that results from a given dose is calculated as:
Cavg (mg/mL) = [5-1)
avo v &/ j CL (mL/kg/h) 1 J
where/abs is the fraction absorbed and dose is average dose rate expressed at an hourly rate.
If humans are exposed to a regular (daily) dose, D, then use of an estimated human
clearance (CLh) leads to a prediction of an ongoing blood concentration equal to D/CLh; i.e., that is
the steady-state or average blood concentration, Cavg, given the daily dose, D. Hence, this evaluation
assumes that the steady-state level increases or decreases in direct proportion to D, with 1 /CLh
being the proportionality constant
Assuming equal toxicity given equal Cavg in humans as mice or rats, and that/abS is the same
in humans as animals, the equitoxic dose (i.e., the human dose that should yield the same blood
concentration [Cavg] as the animal dose from which it is being extrapolated) is then calculated as
follows:
5-7
-------
Toxicological Review of PFBA and Related Salts
(5-2)
Thus, the DAF is simply CLh:CLa, the ratio of clearance in humans to clearance in the animal
from which the POD is obtained. Note that although this evaluation of relative internal dose (Cavg)
assumes that internal dose increases linearly with exposure (as does default allometric scaling),
nonlinearity is usually observed only at relative high exposure levels. Further, although clearance of
PFBA could be biphasic, it is still linear: A two-compartment classical PK model still uses all linear
rate equations, and the predicted Cavg from a two-compartment model still increases linearly with
exposure or applied dose.
Clearance values, however, are not reported for humans in the one pharmacokinetic study
available for PFBA f Chang etal.. 20081. As clearance is a measure of average excretion, to calculate
it, one also needs to evaluate a companion variable, the volume of distribution [Vd], which in turn
requires a measure of total exposure or dose. Chang etal. f20081 did not report the Vd for humans.
Chang etal. (2008) did report Vd for cynomolgus monkeys, however, and as summarized above in
Section 3.1.5, the data suggest a difference in Vd between rodents and monkeys. For comparison, the
Vd values for PFOA and PFOS estimated from the PBPK parameters of Loccisano etal. (2011) are
approximately 0.2 and 0.3 L/kg, respectively, although that obtained from monkeys for PFBA is
approximately 0.5 L/kg. This value of Vd for PFBA was obtained from standard analysis of the
empirical PK data, which is not influenced by any preliminary chemical-specific assumptions, but as
stated by the authors, "Volume of distribution estimates indicated primarily extracellular
distribution" (Chang etal.. 2008). The difference between Vd for PFBA and those for PFOA and PFOS
indicates slightly more intracellular distribution by PFBA. As described in Section 3.1.2
Distribution, Vd for humans is expected to be similar to the value for monkeys, thus the average
value for male and female monkeys from Chang etal. (2008) will be used. Human clearance,
normalized to body weight, can be calculated as follows:
Note that in equation (5-3), BW normalization is embedded in the fact that Vd is a volume per kg
BW. For example, the average blood concentration, Cavg (mg/mL), can then be estimated using
equation (5-1) for any given dose (mg/kg/h = (mg/kg/d)/(24 h/d)), independent of specific BW.
As ti/2 is required in the calculation of CL, these values must be determined from the data
presented for humans in (Chang etal.. 2008). Chang etal. (2008) reported values for human
subjects from two 3M facilities: Cottage Grove, Minnesota and Cordova, Illinois. Cottage Grove had
three subjects, which were not identified by gender. Cordova had nine subjects, two of which were
identified as female. The half-lives for those two women fell among the values of the other subjects
(Cottage Grove and men from Cordova). Considering the minimal difference in ti/2 observed
CLfiuman (mL/kg-h) = ln(2) x
l
^ ^d,monkey
(mL/kg) (5-3)
ti/2,human(h)
5-8
-------
Toxicological Review of PFBA and Related Salts
between male and female monkeys, the available data were assumed insufficient to distinguish
male and female humans. The analytic method used replaced concentration measurements below
the lower limit of quantitation (LLOQ) with LL0Q/V2. For individuals where only two
measurements were made, the resulting half-life estimate was then highly sensitive to this
assumption. The two known female subjects (Cordova), one male subject from Cordova, and one
subject from Cottage Grove fell into this category; half-lives for these four subjects were not used.
Additionally, the last time point for Subject 2 from Cottage Grove was below the LLOQ and was also
excluded from ti/2 estimation. The mean and median ti/2 values estimated from these data (8 total
subjects, 20 observations) were 81.8 and 67.5 hours, respectively. Mixed-effects modeling
confirmed this half-life, estimating an approximate half-life of 67.9 hours when accounting for
clustering (see Appendix C). Other details of the human half-life data are described in Section 3.1.4,
Excretion.
As discussed in Section 3.1.4, using the common assumption of BW0-75 scaling of clearance
and standard species BWs of 0.25 kg in rats and 80 kg in humans, the half-life in humans would be
predicted to be 4.2 times greater than rats. Given half-lives of 9.22 and 1.76 hours in male and
female rats, one would then predict half-lives of 38.7 hours in men and 7.4 hours in women.
Although the value for men is in the range of results for humans, the value for women is much less
than that estimated using the human data available from Chang etal. f2008I DAFs based on BWO-75
scaling for rats and a standard BW of 0.03 kg for mice are presented in Table 5-3. EPA's guidance on
use of BWO75 as the default method for derivation of an oral reference dose states, however, "EPA
endorses a hierarchy of approaches to derive human equivalent oral exposures from data from
laboratory animal species." It goes on to state that, although use of PBPK models is preferred,
"Other approaches may include using chemical-specific information, without a complete
physiologically-based pharmacokinetic model" (i.e., the approach described here, using relative
clearance) and that use of BWO75 Ss endorsed, "In lieu of data to support either of these types of
approaches" (U.S. EPA. 2011). Thus, because data are available to support a chemical-specific
approach, it is clearly preferred.
Using a value of 484.5 mL/kg for Vd for humans [average of male and female Vd values in
monkeys, 526 and 443 mL/kg, respectively, Table 4, Chang etal. f20081] and 67.9 hours for ti/2 in
male humans, CL in humans is estimated to be 4.95 mL/kg-hour = 0.12 L/kg-day. See Table 5-3 for
the DAFs for converting rat and mice PODs to human equivalent doses (HEDs).
5-9
-------
Toxicological Review of PFBA and Related Salts
Table 5-3. Rat, mouse, and human clearance values and data-informed
dosimetric adjustment factors
Sex
Species
Animal CL (mL/kg-h)
Human CL (mL/kg-h)
DAF (CIhiCIa)
DAF (BW075)d
Male
Rat
21.61a
4.95°
0.229
0.236
Mouse
10.10b
0.490
0.139
Female
Rat
96.62a
0.051
0.236
Mouse
27.93b
0.177
0.139
Data from Tables 2, 3, 5, and 6 of Chang et al. (2008).Data from Tables 2, 3, 5, and 6 of Chang et al. (2008).
aAverage of CL = dose/AUC (area-under-the-concentration-curve) was calculated using values reported for oral and
i.v. exposures reported in Table 2 of Chang et al. (2008) see Table 3-2.
bAverage of CL = dose/AUC was calculated using values reported for the 10- and 30-mg/kg dose groups reported in
Table 3 of Chang et al. (2008) see Table 3-2. CL for the 100-mg/kg dose group was excluded, as it was "threefold
and ~twofold higher for males and females, respectively, than the values reported at 10 or 30 mg/kg. This could
be due to saturation of renal absorption or serum binding.
CCL value for humans (male and female) as described above.
dDAFs based on assumption that elimination scales as BW075, hence clearance (elimination/BW) scales as BW"°25,
using standard BWs of 0.03,0.25, and 80 kg for mice, rats, and humans, respectively.
Therefore, human equivalent dose (HED) for considered health effects was calculated as
follows, using relative liver weight observed in male rats in the subchronic Butenhoff et al. f2012al
study as an example. Note that the concentration of the ammonium salt first needs to be converted
to the concentration of the free acid before HED calculation:
HED PFBA = POD N/tfPFB (mg/kg-d) "" PF"A X " (""¦/W0 [5_4)
MW NH% PFB CL animal (mL/kg-h.)
. 214 g/mol 4.95 (mL/kg-h)
HED = 9.6 (mg/kg-d) x ^ x n 61 (mL/ke_h) = 2.04 (mg/kg-d)
As discussed in Section 3.1.5 (Summary of Pharmacokinetics), the assumed linearity in PK
(constant clearance) is likely to be valid for animal POD values of 30 mg/kg-day and below, but
these DAFs should not be applied to higher PODs.
Uncertainty of Animal-to-Human Extrapolation of PFBA Dosimetry
There is uncertainty in applying this dosimetric approach given the volume of distribution
(Vd) was not measured in humans and the human Vd was assumed equal to that in monkeys to
estimate clearance in humans. An alternative approach to using the ratio of clearance values for
animal:human dosimetric adjustments is to use the measured serum concentrations from
toxicological studies as BMD modeling inputs and then use the estimated human clearance values to
calculate the HED. This approach, compared to the ratio of the clearance values approach, however,
is interpreted to have even greater uncertainty. First, the measured serum concentrations were
reported to have been taken 24 hours after the last exposure in the developmental toxicity study
5-10
-------
Toxicological Review of PFBA and Related Salts
Das etal. (20081 and likely were similarly taken in the subchronic toxicity study (Bute nhoff etal..
2012a: van Otterdiik. 2007a). Given the relatively short half-life of PFBA measured in mice and rats,
this end-of-exposure measurement of serum concentrations likely did not reflect the average serum
concentrations exposed animals experienced. For example, the reported serum levels (see
Section 2.1.1) in female mice in the Das etal. f2008I Also, to estimate the HED without a validated
PBPK model, the resulting POD (in units of serum concentrations) would need to be multiplied by
the estimated human clearance value. Thus, in addition to the uncertainty in using end-of-exposure
serum concentrations not reflective of average exposures, this approach would be characterized by
the same uncertainty as the assumption that human and monkey volumes of distribution are equal
and the uncertainty in the human half-life. Therefore, the ratio of clearance values is considered to
have less uncertainty than either serum concentration-based BMD modeling or use of default
allometric dosimetric adjustments. Thus, the approach based on clearance values is the one used
here.
That only a single study reported PFBA PK data in rats or mice (or monkeys) introduces
qualitative uncertainty, because these results were not validated in independent experiments.
Results from different studies cannot be compared quantitatively. In the Chang etal. (2008) study,
some results have relatively tight standard errors (SEs), indicating high confidence, but others
(especially for mice), indicate high variability/uncertainty. Although the results for AUC in rats have
relatively small SEs, they surprisingly show higher AUC (hence lower clearance) following oral
doses than following i.v. doses (30 mg/kg). Oral absorption or bioavailability can range between
near zero and 100%, but why the blood concentrations after an oral dose are higher than when the
same dose is injected directly into the blood is puzzling. The data and plot of the PK model shown in
Figures 1 and 2 of Chang et al. (2008) indicate the absorption and clearance phases are well
characterized and described by the model, so the uncertainty does not appear to be due to the study
design or analysis method. The almost twofold difference in clearance rates estimated from the oral
vs. i.v. rat data thus indicate a comparable degree of uncertainty.
Compared to the results for rats, the Chang etal. (2008) clearance estimates at the two
lower oral doses in male and female mice are much closer, with only an 8% difference between the
two doses for males and a 16% difference for females. The results for both male and female mice
show a dose-dependent increase in clearance across all dose levels, consistent with the hypothesis
of saturable renal resorption. Although the increase only seems significant with the increase from
30 to 100 mg/kg, the differences between 10 and 30 mg/kg could result from the same mechanism.
Thus, those differences might reflect a biological mechanism as much as experimental or analytic
variability. The lack of i.v. data in mice at the same dose as any of the oral doses, however, means
that one cannot fully compare the apparent self-consistency of the mouse data to the inconsistency
noted above for rats.
If the oral vs. i.v. discrepancy in rats is interpreted as indicating an overall factor of 2
uncertainty in the animal clearance values, that can be considered a moderate degree of
5-11
-------
Toxicological Review of PFBA and Related Salts
uncertainty. ORD's Umbrella quality assurance project plan (QAPP) for dosimetry and mechanism-
based models fU.S. EPA. 20201 states that PBPK models are expected to match the corresponding
data within a factor of 2 to be considered sufficient for use in risk assessment and similarly IPCS
f20101 states that PBPK models can be considered adequate when predictions that are, on average,
within a factor of two of experimental data. Hence, this level of uncertainty is considered
acceptable in PK analyses. Although the human half-life estimates vary just over fivefold from
highest to lowest, this much variability in a human population is not surprising, and with results
from just 12 subjects to characterize the mean, uncertainty in that mean can, again, be considered
moderate. Given that the physiological fractions of different tissue types is similar in humans and
primates and that the blood serum:tissue portioning is reasonably expected to be similar across
mammals, the assumption that the volume of distribution in humans is similar to monkeys is
considered to have low uncertainty. Considering all these factors, the overall uncertainty in HED
calculations using equation (5-4) with the parameters estimated here is considered moderate, that
is, within a factor of 3.
Application of Animal-Human Extrapolation of PFB A Dosimetry
Table 5-4 presents the PODs and estimated PODhed values for the thyroid, liver, and
developmental toxicity endpoints.
5-12
-------
Toxicological Review of PFBA and Related Salts
Table 5-4. Points of departure (PODs) considered for use in deriving
candidate reference values for perfluorobutanoic acid (PFBA)
Endpoint/reference
Species/strain
/sex
POD
type/model
POD NH4+ PFB
(mg/kg-d)
POD PFBA
(mg/kg-
d)a
PODhed
PFBAb
(mg/kg-d)
Increased relative liver weight
Butenhoff et al. (2012a)
S-D rat, male
BMDLiord
Exp3 (LN-CV)
9.6
8.89
2.04
Increased relative liver weight
Das et al. (2008)
CD-I mouse, Po
female
BMDLiord
Exp4 (CV)
15
13.9
2.46
Increased liver hypertrophy
Butenhoff et al. (2012a)
S-D rat, male
NOAELb
(0% response)
6
5.56
1.27
Decreased total T4
Butenhoff et al. (2012a)
S-D rat, male
NOAEL0
(15% decrease)
6
5.56
1.27
Embryo/fetal mortalityd
Das et al. (2008)
CD-I mouse, Fi
male/female
BMDLier
Nested-Logistic
5.7
5.28
0.93
Delayed eyes openingd
Das et al. (2008)
CD-I mouse, Fi
male/female
BMDL5RD
Hill (CV)
4.9
4.54
0.80
Delayed vaginal openingd
Das et al. (2008)
CD-I mouse, Fi
female
BMDL5RD
Hill (CV)
3.8
3.52
0.62
Delayed preputial separationd
Das et al. (2008)
CD-I mouse, Fi
male
BMDL5RD
Exp3 (CV)
179.1
165.92
n/ae
BMDL = 95% lower limit on benchmark dose, RD = relative deviation, LN = log-normal, CV = constant variance,
ER = extra risk, NOAEL = no-observed-adverse-effect level.
a Both of these studies used the ammonium salt of PFBA as the test article. To calculate a POD for the free acid of
PFBA from any PFBA salt, multiply the POD of interest by the ratio of molecular weights of the salt and the free
acid. For example, to convert from the ammonium salt of PFBA to the free acid, multiply the ammonium salt POD
by 0.926: MW free acld— = 111= 0.926.
MW ammonium salt 231
See discussion in Section 5.2.1, Approach for Animal-Human Extrapolation of PFBA Dosimetry, for details on HED.
b NOAEL approach used as responses are only seen in the high dose group at levels much higher (90%) than the
BMR.
c No models provided adequate fit to the mean when using constant or nonconstant variance with the normal
distribution or constant variance with the log-normal distribution.
d All HED calculations used DAF for female mice, given exposures were to pregnant animals.
e As noted previously, linearity in clearance values is only valid up to approximately 30 mg/kg-d and the DAF based
on a ratio of clearance values should not be applied to PODs greater than 30 mg/kg-d. Therefore, given that the
POD for preputial separation is above this limit, and other PODs are below that limit, preputial separation is not
considered further for use in estimating a candidate toxicity value for developmental delays. Instead, PODs for
delays in vaginal opening and eye opening are advanced for this purpose.
Derivation of Candidate Toxicity Values for the Oral Reference Dose (RfDJ
Under EPA's A Review of the Reference Dose and Reference Concentration Processes U.S. EPA
f20021 and Methods for Derivation of Inhalation Reference Concentrations and Application of
Inhalation Dosimetry U.S. EPA fl9941 five possible areas of uncertainty and variability were
considered in deriving the candidate values for PFBA. An explanation of these five possible areas of
5-13
-------
Toxicological Review of PFBA and Related Salts
uncertainty and variability and the values assigned to each as designated UFs to be applied to the
candidate PODhed values are listed in Table 5-5. As discussed below, the short-term studies of
thyroid and hepatic effects after PFBA exposure were considered for use in UF selection.
Table 5-5. Uncertainty factors for the development of the candidate values for
perfluorobutanoic acid (PFBA)
UF
Value
Justification
UFa
3
A UFa of 3 (10°5 = 3.16 ~3) is applied to account for uncertainty in characterizing the
pharmacokinetic and pharmacodynamic differences between mice or rats and humans following
oral NH4+PFB/PFBA exposure. Some aspects of the cross-species extrapolation of pharmacokinetic
processes have been accounted for by calculating an HED through application of a DAF based on
animal and human half-lives; however, some residual pharmacokinetic uncertainty and
uncertainty regarding pharmacodynamics remains. Available chemical-specific data further
support the selection of a UF of 3 for PFBA; see text below for further discussion.
UFh
10
A UFh of 10 is applied for interindividual variability in the absence of quantitative information on
the pharmacokinetics and pharmacodynamics of NhVPFB/PFBA in humans.
UFs
10
1
A UFs of 10 is applied to endpoints observed in the subchronic studv Butenhoff et al. (2012a); van
Otterdiik (2007a) for the purposes of deriving chronic toxicity values. See additional discussion on
this decision below.
A UFs of 1 is applied to endpoints observed in the developmental toxicity studv Das et al. (2008)
the developmental period is recognized as a susceptible lifestage where exposure during certain
time windows (e.g., pregnancy and gestation) is more relevant to the induction of developmental
effects than lifetime exposure (U.S. EPA, 1991).
UFl
1
A UFl of 1 is applied for LOAEL-to-NOAEL extrapolation when the POD is a BMDL or NOAEL
UFd
3
A UFd of 3 is applied because, although the PFBA database is relatively small, high confidence
subchronic and developmental toxicity studies are available in mice and rats. Although these high
confidence studies are available for PFBA, the database has some deficiencies, including the lack
of information on developmental neurotoxicity and other endpoints; see the text below for
further discussion.
UFc
Table
5-7
Composite uncertainty factor = UFa x UFh x UFs x UFl x UFd.
As described in EPA's A Review of the Reference Dose and Reference Concentration Processes
U.S. EPA f20021. the interspecies uncertainty factor (UFa) is applied to account for extrapolation of
animal data to humans; it accounts for uncertainty regarding the pharmacokinetic and
pharmacodynamic differences across species. As is usual in the application of this uncertainty
factor, the pharmacokinetic uncertainty is mostly addressed through the application of dosimetric
approaches for estimating human equivalent doses (see Section 4.2.2). This leaves some residual
uncertainty around the pharmacokinetics and the uncertainty surrounding pharmacodynamics.
Typically, a threefold UF is applied for this uncertainty in the absence of chemical-specific
information. This is the case for the thyroid and developmental endpoints. For the liver endpoints,
chemical-specific information should be considered further in determining the most appropriate
value for the UFa to account for the uncertainty.
5-14
-------
Toxicological Review of PFBA and Related Salts
Foreman etal. (20091 investigated the response to PFBA exposure in PPARa wild-type,
PPARa null, and hPPARa mice for hepatic effects and observed either that effects were generally
equivalent in wild-type vs. humanized mice (liver weight, liver hypertrophy, see Table 3-6 and
Table 3-7), that wild-type mice exhibited effects that humanized mice did not (focal hepatic
necrosis, based on statistical significance), and that PPARa null mice generally did not exhibit
hepatic effects except for vacuolation. Additionally, in vitro studies suggest that human cells or cells
transfected with human PPARa were less sensitive to PPAR activation than rodent cells or rodent
PPARa ("Rosen et al.. 2013: Wolf etal.. 2012: Biork and Wallace. 2009: Wolf etal.. 20081. If PPARa
were the only operant MOA for noncancer effects in the liver, this observation might support
reducing the remaining portion of the UFa to 1, as it could be argued that humans are not as
sensitive as wild-type rats to the hepatic effects of PFBA exposure (note: without evidence to the
contrary, as mentioned in the previous paragraph, the pharmacodynamic portion of this UF is
typically assigned a value of 3 assuming responses manifest in humans could be more sensitive
than those observed in animals). Additional evidence presented in Foreman etal. (20091 and other
studies (see Section 2.2.5), however, indicates that non-PPARa MOAs appear to be active in the
livers of exposed rats. Specifically, from Foreman etal. (20091 vacuolation is reported in the livers
of PPARa null and humanized mice, but not in wild-type mice, although the degree to which null or
humanized mice are more susceptible to this effect is difficult to characterize given the results are
presented qualitatively. Vacuolation (i.e., the accumulation of lipids) is an important precursor
event in the development of steatosis, which itself is a precursor to other adverse conditions such
as steatohepatitis, fibrosis, and cirrhosis. As discussed in Section 2.2.5, this observation of
PFBA-induced effects independent of PPARa activation is supported by in vitro and in vivo data
that show other PFAS can activate other forms of PPAR (i.e., PPARy) and additional pathways
(i.e., constitutive androstane receptor [CAR] or pregnane X receptor [PXR]). Given the observation
of apical liver effects in humanized PPARa mice and the observation that other MOAs appear to
contribute to potential liver toxicity, the observation that humanized PPARa mice exhibit
diminished responses for some hepatic effects attributable to PPARa activation cannot alone
determine the appropriate value of the pharmacodynamic portion of the UFa. Therefore, given the
remaining uncertainty in additional MOAs that appear active in PFBA-induced liver effects, and the
relative contribution of these MOAs to toxicity in humans as compared with rodents, the value of
UFa was set to 3 for the purposes of deriving toxicity values for hepatic effects. No MOA information
is available for thyroid or developmental effects; in the absence of information suggesting
otherwise, as noted above, a UFa (3) is also applied to these endpoints to account for any residual
pharmacokinetic and pharmacodynamic uncertainty.
The short-term studies of Butenhoff etal. (2012a). van Otterdiik (2007a). and Foreman et
al. (2009) were considered for potential use in informing the selection of the UFS. More specifically,
for several outcomes from which PODs were derived, comparisons between short-term exposure
and subchronic exposure appeared possible (i.e., because of the inherent similarities in study
5-15
-------
Toxicological Review of PFBA and Related Salts
design and experimental conduct). When comparing short-term to subchronic PFBA exposure for
liver weight and thyroid hormone measures, there was no apparent increased sensitivity with
longer exposure duration in terms of the magnitude of the observed effects at the same tested doses
or the lowest doses at which effects were observed. In addition, given the pharmacokinetics of
PFBA, steady-state levels in potential target tissues might not substantially increase with increasing
exposure duration fButenhoff etal.. 2012a: van Otterdiik. 2007a. b). In these studies, the latter
conclusion seemed dose dependent, as PFBA levels actually decreased with longer exposures when
comparisons are made at 6 mg/kg-day (~25 to 14 |J.g/mL in serum and ~7.5 to 3.1 |ig/g in liver
comparing 28 to 90 days of exposure), whereas levels were either increased slightly or were similar
when comparisons are made at 30 mg/kg-day (~38 to 52 |J.g/mL in serum and ~17.4 to 16.1 |ig/mL
in liver comparing 28 to 90 days of exposure). This indicates perhaps that steady-state conditions
have been reached in the livers of exposed rats after only 28 days of exposure. Preliminarily, this
indicates that increased durations of exposure might not elicit increased effects in the target tissue,
as the LOAEL for liver weights is 30 mg/kg-day for male rats exposed to either 28 or 90 days. When
also considering results from the 28-day exposure study by Foreman et al. (2009) and the
gestational exposure study by Das etal. (2008) basing comparisons on human equivalent external
concentrations (see Table 5-6 below for modeling results and application of dosimetric
adjustments), liver weight appears affected at similar doses across mice and rats across these three
different exposure durations (i.e., gestational, short-term, and subchronic). However, it should be
noted that these data indicating no increase of effect when comparing subchronic exposures to
short-term exposures (an increase in duration of approximately 3-fold) is not considered sufficient
evidence to convincingly argue that effects would not worsen following chronic exposures (an
increase of approximately 8-fold increase in duration, compared to subchronic). While it is true that
pharmacokinetic data suggests liver concentrations may reach steady-state conditions rapidly (i.e.,
following 28 days of exposure), it is reasonable to assume that prolonged exposure to those levels
of tissue-specific concentrations over the course of multiple years could result in an increased
magnitude of effect or effects evident at lower doses. Contributing to this assumption is the
consideration of the impact of PFBA exposure duration on related liver effects.
In fact, the lack of increasing effect with increasing duration is not the case for all liver
effects. Histopathological evaluations of the liver in male rats exposed to PFBA for 90 days show
that hepatocellular hypertrophy occurs at 30 mg/kg-day, whereas hypertrophy occurs only at
150 mg/kg-day in male rats exposed for 28 days fButenhoff etal.. 2012a: van Otterdiik. 2007a. b).
Thus, although liver concentrations are equivalent following 28- or 90-day exposures, that
prolonged exposure (i.e., 90 days vs. 28 days) elicits adverse effects in the liver is readily apparent.
Taking into account the increased potential for some effects in the liver with increasing durations of
exposure, and the large uncertainty associated with the lack of data on whether the effects
observed in the subchronic study worsen after chronic exposure, the UFs were therefore set to 10
for the purposes of the liver endpoints.
5-16
-------
Toxicological Review of PFBA and Related Salts
Table 5-6. Comparison of liver-weight effects across species and durations of
exposure
Reference
Species/strain/
sex
Duration
POD
type/model
POD NH4+ PFB
(mg/kg-d)
POD PFBA
(mg/kg-d)
PODhed PFBA
(mg/kg-d)
Relative liver weight
Butenhoff et al. (2012a)
S-D rat, male
90 d
BMDLiord
Exp3 (LN-CV)
9.6
8.89
2.04
Relative liver weight
Butenhoff et al. (2012a)
S-D rat, male
28 d
BMDLio, Exp4
(NCV)
6.34
5.87
1.3
Relative liver weight
Foreman et al. (2009)
Sv/129 WT
mouse, male
28 d
LOAEL3
35
32.42
1.59b
Relative liver weight
(Foreman et al., 2009)
Sv/129 hPPARa
mouse, male
28 d
BMDLio, Hill
(NCV)
4.41
4.09
2.00
Relative liver weight
Das et al. (2008)
CD-I mouse, Po
female
Gestational
BMDLiord
Exp4 (CV)
15
13.9
2.46
a Data is highly supralinear and BMD modeling guidance recommends against modeling this type of dose-response pattern.
bAs this data set only supported identification of a LOAEL, the LOAEL-to-NOAEL uncertainty factor was applied to facilitate
comparison to the other HEDs for liver-weight effects.
Regarding thyroid endpoints, effects on total T4 following subchronic exposures were not
worse compared to effects following 28-day exposures in the Butenhoff et al. (2012a) study,
owever, for thyroid hypertrophy/hyperplasia, although total incidence of hypertrophy/hyperplasia
was not observed to worsen with increasing duration (the LOAEL was 30 mg/kg-day for both
exposure durations), there is evidence that the severity of the observed lesions worsened after 90-
day exposures. As Table 3-4 shows, nine out of ten animals developed thyroid
hypertrophy/hyperplasia following 28-day exposures, with all animals displaying minimally severe
lesions. However, following 90-day exposures, while the total incidence was the same (9/10), four
animals had minimally severe lesions while 5 animals developed mild lesions. While not conclusive,
this evidence suggests that damage to the thyroid organ specifically might worsen with increasing
duration. Given this potential concern for more severe effects on the thyroid with longer exposures
and the small number of studies (e.g., one) available to inform this interpretation for either thyroid
histopathology or levels of circulating THs, the default UFs of 10 was also retained for thyroid
endpoints.
As described in EPA's A Review of the Reference Dose and Reference Concentration Processes
U.S. EPA (2002). the database uncertainty factor is applied to account for the potential of deriving
an underprotective reference value as a result of incomplete characterization of a chemical's
toxicity. The PFBA database is relatively small but contains high confidence subchronic and
developmental toxicity studies investigating effects in multiple organ systems in male and female
rats and mice.
For PFBA, given the small number of available studies, both a UFd = 10 or a UFd = 3 were
considered due to the limited database (most specifically the lack of a two-generation
5-17
-------
Toxicological Review of PFBA and Related Salts
developmental/reproductive toxicity study, but also a lack of studies on potential
neurodevelopmental or developmental immune effects), and a UFd = 3 ultimately was applied.
Typically, the specific study types lacking in a chemical's database that influence the value of the
UFd to the greatest degree are developmental toxicity and multigenerational reproductive toxicity
studies. The PFBA database does include a high confidence Das etal. f20081 developmental toxicity
study in mice. Despite its quality, however, that study fails to cover endpoints related to potential
transgenerational impacts of longer-term exposures evaluated in a two-generation study. The 1994
Reference Concentration Guidance U.S. EPA (19941 and 2002 Reference Dose Report U.S. EPA
(20021 support applying a UFd in situations when such a study is missing. The 2002 Reference Dose
Report U.S. EPA f20021 states that "[i]f the RfD/RfC is based on animal data, a factor of 3 is often
applied if either a prenatal toxicity study or a two-generation reproductive study is missing."
Consideration of the PFBA, PFBS (a short-chain perfluoroalkane sulfonic acid with a 4-carbon
backbone like PFBA), PFHxA (a short-chain perfluoroalkyl carboxylic acid; see public comment
draft for PFHxA; ),13 and PFHxS (a long-chain perfluoroalkane sulfonic acid) databases together,
however, diminish the concern that the availability of a multigenerational reproductive study
would result in reference values lower than those currently derived for PFBA. Although limited in
their ability to assess reproductive health or function, measures of possible reproductive toxicity,
including reproductive organ weights (i.e., epididymis, testis, and ovary weights) were unaffected
when measured after exposure to PFBA for 28 days fButenhoff etal.. 2012a: van Otterdiik. 2007al.
Likewise, the available data on reproductive toxicity in the PFBS database is consistent with this
general lack of sensitive reproductive effects: No biologically significant changes were observed in
male mating and fertility parameters, reproductive organ weights, reproductive hormone levels, or
altered sperm parameters (U.S. EPA. 2018b). The female reproductive effects that were observed
(e.g., altered estrous cyclicity) occurred at doses equal to or higher than those that resulted in
effects in other organ systems (e.g., thyroid, liver), thus indicating they were not more sensitive
markers of toxicity. Further, no notable male or female reproductive effects were observed in
epidemiological or toxicological studies investigating exposure to PFHxA (see public comment draft
for PFHxA, (U.S. EPA. 2021b) and (Luz etal.. 2019: NTP. 2019: Klaunigetal.. 2015: Chengelis etal..
2009) or PFHxS (MDH. 2019). Therefore, when considering the limited chemical-specific
information alongside information gleaned from structurally related compounds, the lack of a
multigenerational reproductive study is not considered a major concern relative to UFd selection.
Another gap in the PFBA database is the lack of measures of thyroid toxicity in gestationally
exposed offspring and the lack of a developmental neurotoxicity study. The potential for
neurodevelopmental effects, whether thyroid hormone-dependent or independent, remains an
13The systematic review protocol for PFBA (see Appendix A) defines perfluoroalkyl carboxylic acids with
seven or more perfluorinated carbon groups and perfluoralkane sulfonic acids with six or more
perfluorinated carbon groups as 'long-chain" PFAS. Thus, PFHxA is considered a short-chain PFAS, whereas
PFHxS is considered a long-chain PFAS.
5-18
-------
Toxicological Review of PFBA and Related Salts
uncertainty for PFBA. Thyroid hormones are critical in myriad physiological processes and must be
maintained at sufficient levels during times of brain development in utero and after birth. Although
no PFBA-specific data on thyroid hormone levels following gestational exposure are available, total
T4 is reduced in both pregnant mice and their offspring following whole-gestation oral exposure to
PFBS, with effects evident in offspring at PNDs 1, 30, and 60. Therefore, anticipating that effects due
to PFBA exposure also could have been observed had thyroid hormone levels been measured in the
Das etal. (20081 developmental study is reasonable. For PFBS, the PODs for effects in dams and
offspring on PND 1 were almost identical, indicating that thyroid hormone homeostasis is
perturbed at equivalent exposure levels in both pregnant animals and developing offspring. Thus,
although some concern remains that thyroid insufficiency during in utero and perinatal
development could be a more sensitive effect of PFBA exposure than insufficiency in adults, this
concern is mitigated on the basis of data from other PFAS. Likewise, given that neurodevelopmental
effects due to thyroid hormone insufficiency would be downstream effects, application of a UFd
(and derivation of reference values) addressing the potential for developmental thyroid
insufficiency would presumably be protective of any potential neurodevelopmental endpoints
related to that mechanism. The potential for neurodevelopmental effects independent of a thyroid
hormone-related mechanism remains an uncertainty for PFBA.
Lastly, the potential for immunotoxicity (including developmental immunotoxicity, in
particular) and mammary gland effects represents an area of concern across several constituents of
the larger PFAS family (primarily long-chain PFAS). No studies have evaluated these outcomes
following PFBA oral exposure or following oral exposure to the structurally related PFBS described
above. However, one dermal toxicity study Weatherlv etal. (2021) did observe altered cellularity
for multiple immune cell types in the draining lymph nodes and ear pinna of exposed animals,
raising the concern for immunotoxicity following oral exposures. However, without reported
internal serum levels, it is difficult to ascertain whether the exposure levels in the dermal study are
equipotent to the oral exposures used in the subchronic and developmental toxicity studies and
whether the hepatic, thyroid, or developmental endpoints observed in those studies would be
protective of immunotoxicity endpoints. Overall, no chemical-specific information is available to
judge the degree to which the existing endpoints in the PFBA Toxicological Review would be
protective of mammary gland or immune (including developmental immune) effects after oral
exposure.
Given the residual concerns for potentially more sensitive effects outlined above, a database
uncertainty factor is considered necessary. Specifically, a value of 3 was selected for the UFd to
account for the uncertainty surrounding the lack of a multigenerational reproductive study,
developmental neurotoxicity study (or information on thyroid hormone perturbation in utero and
postnatally), immunotoxicity (and developmental immunotoxicity, in particular), or mammary
gland effects. A UFd of 10 was not applied, given that multiple lines of chemical-specific information
or data from structural analogs are available to partially mitigate the concern that additional study
5-19
-------
Toxicological Review of PFBA and Related Salts
would possibly result in reference values one order of magnitude lower than the one currently
derived. Thus, a UFd value of 3 was applied because currently available lines of evidence do not fully
eliminate this concern.
The candidate values (see Table 5-7) are derived by dividing the PODhed by the composite
uncertainty factor. For example, for relative liver weight in adult rats from Butenhoff et al. f2012al.
the candidate value is calculated as:
Candidate value for PFBA = BMDL10 h- UFc (5-5)
Candidate value = 2.04 d) ~=~ 1,000
Candidate value = 0.002 (^/kg-d)
Candidate value = 2.0 x 10-3 (m^/kg-d)
Table 5-7. Candidate values for perfluorobutanoic acid (PFBA)
Endpoint
PODhed
PFBA
(mg/kg-d)
UFa
UFh
UFs
UFl
UFd
UFC
Candidate
value PFBA
(mg/kg-d)
Candidate
value NH4+
PFB
(mg/kg-d)a
Increased relative liver weight
Butenhoff et al. (2012a)
2.04
3
10
10
1
3
1,000
2.0 x 10"3
2.2 x 10"3
Increased relative liver weight
Das et al. (2008)
2.46
3
10
10
1
3
1,000
2.5 x 10"3
2.7 x 10"3
Increased liver hypertrophy
Butenhoff et al. (2012a)
1.27
3
10
10
1
3
1,000
1.3 x 10"3
1.4 x 10"3
Decreased total T4
Butenhoff et al. (2012a)
1.27
3
10
10
1
3
1,000
1.3 x 10"3
1.4 x 10"3
Embryo/fetal mortality
Das et al. (2008)
0.93
3
10
1
1
3
100
9.5 x 10"3
1.0 x 10"2
Delayed eyes opening
Das et al. (2008)
0.80
3
10
1
1
3
100
8.0 x 10"3
8.6 x 10"3
Delayed vaginal opening
Das et al. (2008)
0.62
3
10
1
1
3
100
6.2 x 10"3
6.7 x 10"3
a To calculate candidate values for salts of PFBA, multiply the candidate value of interest by the ratio of molecular
weights of the free acid and the salt. For example, for the ammonium salt of PFBA, the RfD would be calculated by
.... . . . . MW ammonium salt 231 . „ „ „ . . ....
multiplying the free acid RfD by 1.079: = — = 1.079. This same conversion can be applied to
MW free acid 214
other salts of PFBA, such as the potassium or sodium salts.
5-20
-------
Toxicological Review of PFBA and Related Salts
Selection of Lifetime Toxicity Value(s)
Selection of organ/system-specific oral reference doses fosRfDsl
From among the candidate values presented in Table 5-7, organ/system-specific RfDs
(osRfDs) are selected for the individual organ systems identified as hazards in Section 3. The osRfD
values selected were associated with increased liver hypertrophy for liver effects, decreased total
T4 for thyroid effects, and developmental delays (based on the candidate value for delayed time to
vaginal opening) for developmental effects. The confidence decisions about the study, evidence
base, quantification of the POD, and overall RfD for these organ/system-specific values are fully
described in Table 5-8, along with the rationales for selecting those confidence levels. In deciding
overall confidence, confidence in the evidence base is prioritized over the other confidence
decisions. The overall confidence in the osRfD for liver effects is medium, whereas the confidence in
the osRfDs for thyroid effects and developmental effects is medium-low. Selection of the overall RfD
is described in the following section.
Table 5-8. Confidence in the organ/system-specific oral reference doses
(osRfDs) for perfluorobutanoic acid (PFBA)
Confidence
categories
Designation
Discussion
Liver RfD = 1 x 10"3 mg/kg-d PFBA; 1 x 10"3 mg/kg-d NH4+ PFB
Confidence in
study3 used to
derive osRfD
High
Confidence in the studv Butenhoff et al. (2012a); van Otterdiik (2007b) is hiah given
the study evaluation results (i.e., rating of good or adequate in all evaluation
categories) and characteristics that make it suitable for deriving toxicity values,
including the relevance of the exposure paradigm (route, duration, and exposure
levels), use of a relevant species, and the study size and design.
Confidence in
evidence base
supporting this
hazard
Medium
Confidence in the evidence base for liver effects is medium because there are
consistent, dose-dependent, and biologically coherent effects on organ weight and
histopathology observed in multiple high and medium confidence studies. Although
the available mechanistic evidence also supports the human relevance of observed
effects, there is a sparsity of chemical-specific information. One in vivo PFBA study
Foreman et al. (2009) is available that indicates non-PPARa modes-of-action are
active in the development of liver effects, but no PFBA-specific studies investigated
activation of other PPAR isoforms or additional pathways. Another limitation of the
database for PFBA-induced liver effects is the lack of a chronic duration study.
Confidence in
quantification
of the PODhed
Medium
Confidence in the quantification of the POD and osRfD is medium given the POD was
based on a NOAEL (BMD modeling not supported given that responses are only
observed in the high dose group at levels (90%) much greater than the BMR) and
dosimetric adjustment was based on PFBA-specific pharmacokinetic information, the
latter of which introduces some uncertainty. Generally, the use of a NOAEL for the
POD would result in a reduced confidence rating. However, in this case, the NOAEL of
6 mg/kg-d is very close to the BMDL (5.4 mg/kg-d) that would be selected had BMD
modeling been supported. Therefore, this NOAEL is not interpreted as likely to be
substantially more uncertain than a BMD-based POD. This supports a determination
that the confidence in the quantification of the POD is medium.
5-21
-------
Toxicological Review of PFBA and Related Salts
Confidence
categories
Designation
Discussion
Overall
confidence in
osRfD
Medium
The overall confidence in the osRfD is medium and is primarily driven by medium
confidence in both the evidence base supporting this hazard and the quantification of
the POD from a high confidence study.
Thyroid RfD = 1 x 10"3 mg/kg-d PFBA; 1 x 10"3 mg/kg-d NH4+ PFB
Confidence in
study3 used to
derive osRfD
High
Confidence in the studv Butenhoff et al. (2012a); van Otterdiik (2007b) is hiah given
the study evaluation results (i.e., rating of good or adequate in all evaluation
categories) and characteristics that make it suitable for deriving toxicity values,
including the relevance of the exposure paradigm (route, duration, and exposure
levels), use of a relevant species, and the study size and design.
Confidence in
evidence base
supporting this
hazard
Medium
Confidence in the evidence base for thyroid effects is medium because there were
consistent and coherent effects on hormone levels, organ weights, and
histopathology in a single high confidence study. Confidence is decreased by the lack
of coherence between histopathology and TSH, as well as the increased sensitivity of
rodents for developing thyroid hypertrophy compared to humans. Another limitation
of evidence base for thyroid effects is the lack of a chronic-duration or developmental
study.
Confidence in
quantification
of the PODhed
Medium-low
Confidence in the quantification of the POD and osRfD is medium-low given the POD
was based on a NOAEL (BMD modeling did not provide an adequate fit to the data)
and dosimetric adjustment was based on PFBA-specific pharmacokinetic information,
the latter of which introduces some uncertainty. Although a 15% decrease in total T4
levels, upon which the NOAEL was based, is consistent with a 13% decrease in total
T4 that would correspond to a response level at a BMR of 1SD (i.e., the BMD), there
is uncertainty regarding how much lower a BMDL would be as compared to the
NOAELb. Therefore, while this NOAEL is not likely to be substantially more uncertain
than a BMD, it is higher than a BMDL-based POD would be. This introduces some
additional uncertainty and supports a determination that the confidence in the
quantification of the POD is medium-low.
Overall
confidence in
osRfD
Medium-low
The overall confidence in the osRfD is medium-low and is primarily driven by medium
confidence in the evidence base; however, the medium-to-low confidence in the
quantification of the POD does warrant decreasing the overall confidence in the
osRfD.
Developmental RfD = 6 x 10 3 mg/kg-d PFBA; 7 x 10 3 mg/kg-d NhV PFB
Confidence in
study3 used to
derive osRfD
High
Confidence in the studv Das et al. (2008) is hiah given the studv evaluation results
(i.e., rating of good or adequate in all evaluation categories) and characteristics that
make it suitable for deriving toxicity values, including the relevance of the exposure
paradigm (route, duration, and exposure levels), use of a relevant species, and the
study size and design.
Confidence in
evidence base
supporting this
hazard
Medium
Confidence in the evidence base for developmental effects is medium. Although data
are only available in gestationally exposed animals in a single high confidence
developmental toxicity study, there were coherent delays in multiple developmental
milestones (general and reproductive development).
5-22
-------
Toxicological Review of PFBA and Related Salts
Confidence
categories
Designation
Discussion
Confidence in
quantification
of the PODhed
Medium-low
Confidence in the quantification of the POD and osRfD is medium-to-low given the
POD was based on BMD modeling and dosimetric adjustment was based on PFBA-
specific pharmacokinetic information, the latter of which introduces some
uncertainty. Other sources of uncertainty are the use of dosimetric adjustments
based on the ratio of adult pharmacokinetic parameters, and that the derived BMDL
is approximately ninefold below the observed range of the data.
Overall
confidence in
osRfD
Medium-low
The overall confidence in the osRfD is medium-low and is primarily driven by the
medium-to-low confidence in the quantification of the POD given the extrapolation
below the range of the observed data. Modeling data from a high confidence study in
a medium-confidence evidence base does not fully mitigate the medium-to-low
confidence in the actual modeling results in this case.
a All study evaluation details can be found on HAWC.
b Note that the BMDL would be considerably less than an order of magnitude lower given that the next lower dose
tested was only 5-fold lower than the NOAEL and a non-significant increase in T4 was observed at that dose.
Selection of overall oral reference dose fRfDl and confidence statement
Organ/system-specific RfD values for PFBA selected in the previous section are summarized
in Table 5-9.
Table 5-9. Organ/system-specific oral reference dose (osRfD) values for
perfluorobutanoic acid (PFBA)
System
Basis
POD
UFC
OSRfD
PFBA
(mg/kg-d)
OSRfD
NH4+ PFB
(mg/kg-d)b
Confidence
Hepatic
Increased
hepatocellular
hypertrophy in
adult male S-D
rats
BMDLhed from
Butenhoff et al.
(2012a)
1,000
1 x 10"3
1 X 10"3
Medium
Thyroid
Decreased total
T4 in adult male
S-D rats
NOAELhed from
Butenhoff et al.
(2012a)
1,000
1 x 10"3
1 x 10"3
Medium-low
Developmental
Developmental
delays after
gestational
exposure in CD1
mice3
BMDLhed from
Das et al. (2008)
100
6 x 10"3
7 x 10"3
Medium-low
a POD based on delayed vaginal opening used to represent three developmental delays observed in the study.
b See Table 5-7 for details on how to calculate candidate values for salts of PFBA; the osRfDs presented in this table
have been rounded to 1 significant digit from the candidate values presented in Table 5-7.
From the identified human health hazards of PFBA exposure and the derived osRfDs for
effects in the liver, thyroid, and developing organism, an overall RfD ofl x 10~3 mg/kg-day PFBA
5-23
-------
Toxicological Review of PFBA and Related Salts
based on increased liver hypertrophy and decreased total T4 is selected. The selected RfD for the
ammonium salt of PFBA is also 1 x 10-3 mg/kg-day. These osRfDs are selected as the overall RfD as
they represent effects in two different organ systems with the same osRfD value., including the
osRfD with the highest confidence of all osRfDs derived (i.e., the hepatic osRfD, with medium
confidence). The other available osRfD (for developmental effects) was interpreted with medium-
low confidence and had a higher osRfD value; thus, it was not selected. Although the overall
confidence in the individual liver and thyroid osRfDs do differ slightly (medium for increased liver
hypertrophy and medium-low for decreased total T4), an overall confidence of medium is selected
for the final RfD. This confidence level of medium is supported given the two osRfDs come from the
same high confidence study and that the evidence bases for both organ systems were rated as
medium. The difference in the overall confidence for the two osRfDs was driven primarily by the
confidence in the quantification of the PODheds: medium for increased liver hypertrophy and
medium-low for decreased total T4. As noted in Table 5-8, the medium-low confidence in the thyroid
PODhed reflects that the selected NOAEL would be greater than the BMDL that would be derived if
BMD modeling were possible, although this difference would be considerably less than an order of
magnitude (see Table 5-8) which reduces the level of concern for this uncertainty. This uncertainty
is further mitigated when taken together with the medium confidence in the PODhed for the co-
critical effect on the liver. Altogether, this supports the determination of medium confidence in the
overall RfD based on liver and thyroid effects.
Another consideration in selecting the overall RfD is the difference in composite uncertainty
factors across the three candidate osRfDs. The composite UF for the liver and thyroid osRfDs was
greater than that for developmental effects (1,000 vs. 100), stemming from not applying a UFs for
the developmental effects. Application of the larger composite UF for liver and thyroid effects
results in osRfDs that are fivefold lower than the developmental osRfD and thus protective of PFBA-
induced effects on the developing organism. If the osRfD for developmental effects were chosen as
the overall RfD on the basis of the application of a smaller composite UF, this would raise concerns
that it would not be protective against potential liver and thyroid effects. Lastly, the selection of the
overall RfD based on liver and thyroid effects is further supported by the fact that the confidence in
that RfD is medium, compared with medium-low for developmental effects. Selection of the RfD
based on liver and thyroid effects is presumed to be protective of possible developmental effects in
humans, although uncertainty in the database currently available for PFBA remains including a lack
of information on the potential for sensitive transgenerational, neurodevelopmental, or
developmental immune effects of PFBA exposure (see discussion on UFd selection above).
Increased liver hypertrophy and decreased total T4 was observed only in male rats exposed
to PFBA, thus possibly identifying males as a susceptible population. As discussed in Section 3.3,
however, this observation in rats could be driven primarily by the observed sex-dependent
differences in pharmacokinetics in rats. No compelling information is available that supports a
5-24
-------
Toxicological Review of PFBA and Related Salts
similarly strong sex dependence in pharmacokinetics in humans. Therefore, this RfD is presumed
equally applicable to both male and female humans.
5.2.2. Subchronic Toxicity Values for Oral Exposure (Subchronic Oral Reference Dose [RfD])
Derivation
In addition to providing RfDs for lifetime exposures in multiple systems, this document also
provides an RfD for less-than-lifetime, subchronic-duration exposures. In the case of PFBA, all
studies used to calculate the RfDs were subchronic or gestational in duration. Therefore, the
method to calculate the subchronic RfDs is identical to that used for calculating the RfDs, minus the
application of a 10-fold UFs for the subchronic studies (see Table 5-6). The individual organs and
systems for which specific candidate subchronic RfD values were derived were the liver, thyroid,
and the developing organism (see Table 5-10).
Table 5-10. Candidate subchronic oral reference dose (RfD) values for
perfluorobutanoic acid (PFBA)
Endpoint
PODhed
PFBA
(mg/kg-d)
UFa
UFh
UFs
UFl
UFd
UFC
Candidate
value PFBA
(mg/kg-d)
Candidate
value NH4+
PFB
(mg/kg-d)a
Increased relative liver weight
Butenhoff et al. (2012a)
2.04
3
10
1
1
3
100
2.0 x 10"2
2.2 x 10"2
Increased relative liver weight
Das et al. (2008)
2.46
3
10
1
1
3
100
2.5 x 10"2
2.7 x 10"2
Increased liver hypertrophy
Butenhoff et al. (2012a)
1.15
3
10
1
1
3
100
1.1 x 10"2
1.2 x 10"2
Decreased total T4
Butenhoff et al. (2012a)
1.27
3
10
1
1
3
100
1.3 x 10"2
1.4 x 10"2
Embryo/fetal mortality
Das et al. (2008)
0.93
3
10
1
1
3
100
9.3 x 10"3
1.0 x 10"2
Delayed eyes opening
Das et al. (2008)
0.80
3
10
1
1
3
100
8.0 x 10"3
8.6 x 10"3
Delayed vaginal opening
Das et al. (2008)
0.62
3
10
1
1
3
100
6.2 x 10"3
6.7 x 10"3
a To calculate subchronic candidate values, osRfDs, or the subchronic RfD for salts of PFBA, multiply
the value of interest by the ratio of molecular weights of the free acid and the salt. For example, for
the ammonium salt of PFBA, the RfD would be calculated by multiplying the free acid RfD by 1.079:
MW ammonium salt 231 . „„„ . . . . .
= — = 1.079. This same method of conversion can be applied to other salts of
MW free acid. 214
PFBA, such as the potassium or sodium salts, using the corresponding molecular weights.
5-25
-------
Toxicological Review of PFBA and Related Salts
From the identified human health hazards of PFBA exposure and the derived candidate
RfDs, osRfDs of 1 x 10"2 mg/kg-day are selected for liver effects (increased liver hypertrophy) and
thyroid effects (decreased total T4) (corresponding osRfD of 1 x 10~2 mg/kg-day for the ammonium
salt), and an osRfD of 6 x 10~3 mg/kg-day PFBA is selected for developmental effects
(developmental delays based on the candidate value for delayed vaginal opening) (corresponding
osRfD of 7 x 10"3 mg/kg-day for the ammonium salt). The selection of these candidate values over
other candidates and the confidence in these subchronic osRfDs are identical to the confidence in
the osRfDs discussed in the previous section and presented in Table 5-8. Note, specifically for
developmental delays, the candidate value for delayed eye opening was not selected as the osRfD as
it was 33% larger than the candidate value for vaginal opening and thus inadequately protective of
human health.
From these subchronic osRfDs, an overall subchronic RfD of 6 x 10~3 mg/kg-day PFBA
based on developmental delays is selected (the corresponding overall subchronic RfD is
7 x 10"3 mg/kg-day for the ammonium salt). This osRfD is selected as the overall subchronic RfD, as
it is the lowest osRfD among the derived subchronic osRfDs, even though it is not the osRfD
interpreted with the highest confidence. In the case of the subchronic RfD, selection need not
consider differences in the composite UF, as a value of 100 is applied to all PODs. This is because all
the studies considered for the subchronic RfD are subchronic or gestational duration studies. This
results in the osRfD for developmental delays being approximately 50% lower than the osRfD for
liver or thyroid effects. Although the overall confidence in the osRfD for developmental delays
[medium-low) is lower than for liver effects (medium confidence, see derivation of RfD section),
selection of the developmental osRfD as the overall subchronic RfD is presumed protective of
possible effects in other organ systems. Selection of the liver osRfD, although having a stronger
overall confidence determination, as the overall subchronic RfD would be considered inadequate to
protect against potential developmental effects. Also, although the subchronic RfD is intended to
protect health during a less-than-lifetime exposure to PFBA, developmental delays are appropriate
endpoints on which to base a subchronic RfD. First, as discussed above (Study Selection
subsection), given delayed reproductive milestones occuring during critical periods of
development, EPA's Reproductive Toxicity Guidelines U.S. EPA (1996) state that significant effects
on puberty (and thus by inference, the development of the male and female reproductive systems
more broadly) "either early or delayed, should be considered adverse...". Further, delays in
reaching developmental milestones are not phenomena that can be resolved (e.g., after PFBA
exposure is removed), and they can result from short (less-than-lifetime) exposures during discrete
windows of development. More importantly, the consequences of these delays can have permanent
impacts on health (e.g., delays in eye opening leading to permanent decrements in visual acuity). So,
although the delay itself might occur only over a short portion of lifetime, the functional
consequences are permanent
5-26
-------
Toxicological Review of PFBA and Related Salts
5.2.3. Inhalation Reference Concentration (RfC)
No published studies investigating the effects of subchronic, chronic, or gestational
exposure to PFBA in humans or animals have been identified. Therefore, an RfC is not derived.
5.3. CANCER
5.3.1. Cancer Weight-of-Evidence Descriptor and Derivation of Cancer Risk Values
No studies were identified that evaluated the carcinogenicity of PFBA in humans or animals.
In accordance with the Guidelines for Carcinogen Risk Assessment M.S. EPA f20051. EPA concluded
that there is inadequate information to assess carcinogenic potential for PFBA (or salts of PFBA) for
any route of exposure. This conclusion precludes the derivation of quantitative estimates for either
oral (oral slope factor [OSF]) or inhalation (inhalation unit risk [IUR]) exposure.
5-27
-------
Toxicological Review of PFBA and Related Salts
REFERENCES
Abe. T: Takahashi. M: Kano. M: Amaike. Y: Ishii. C: Maeda. K: Kudoh. Y: Morishita. T: Hosaka. T:
Sasaki. T: Kodama. S: Matsuzawa. A: Koiima. H: Yoshinari. K. (2017). Activation of nuclear
receptor CAR by an environmental pollutant perfluorooctanoic acid. Arch Toxicol 91: 2365-
2374. http://dx.doi.org/10.1007/s00204-016-1888-3.
Abraham. K: El-Khatib. AH: Schwerdtle. T: Monien. BH. (2021). Perfluorobutanoic acid (PFBA): No
high-level accumulation in human lung and kidney tissue. Int J Hyg Environ Health 237:
113830. http://dx.doi.Org/10.1016/i.iiheh.2021.113830.
Angrish. MM: Kaiser. TP: Mcqueen. CA: Chorlev. BN. (2016). Tipping the balance: Hepatotoxicity and
the 4 apical key events of hepatic steatosis [Review], Toxicol Sci 150: 261-268.
http: / /dx. doi. o r g /10.10 9 3 /toxsci /kfwO 18.
Antignac. TP: Vevrand. B: Kadar. H: Marchand. P: Oleko. A: Le Bizec. B: Vandentorren. S. (2013).
Occurrence of perfluorinated alkylated substances in breast milk of French women and
relation with socio-demographical and clinical parameters: Results of the ELFE pilot study.
Chemosphere 91: 802-808. http://dx.doi.Org/10.1016/i.chemosphere.2013.01.088.
ATSDR (Agency for Toxic Substances and Disease Registry). (2017). Substance priority list (SPL)
resource page. Retrieved from
https://www.atsdr.cdc.gov/spl/resources/ATSDR 2017 Full SPL Spreadsheetxlsx
ATSDR (Agency for Toxic Substances and Disease Registry). (2018a). An overview of perfluoroalkyl
and polyfluoroalkyl substances and interim guidance for clinicians responding to patient
exposure concerns. Interim guidance revised 5/7/2018.
https://stacks.cdc.gov/view/cdc/77114.
ATSDR (Agency for Toxic Substances and Disease Registry). (2018b). Toxicological profile for
perfluoroalkyls. Draft for public comment [ATSDR Tox Profile], Atlanta, GA: U.S. Department
of Health and Human Services, Centers for Disease Control and Prevention.
https://www.atsdr.cdc.gov/toxprofiles/tp200.pdf.
ATSDR (Agency for Toxic Substances and Disease Registry). (2021). Toxicological profile for
perfluoroalkyls [ATSDR Tox Profile], Atlanta, GA: U.S. Department of Health and Human
Services, Public Health Service. http://dx.doi.Org/10.15620/cdc:59198.
Bangma. IT: Reiner. 1: Fry. RC: Manuck. T: Mccord. 1: Strvnar. Ml. (2021). Identification of an
Analytical Method Interference for Perfluorobutanoic Acid in Biological Samples. Environ
SciTechnol Lett 8: 1085-1090. http://dx.doi.org/10.1021/acs.estlett.lcQ0901.
Bansal. R: Tighe. D: Danai. A: Rawn. DFK: Gaertner. DW: Arnold. PL: Gilbert. ME: Zoeller. RT. (2014).
Polybrominated Diphenyl Ether (DE-71) Interferes With Thyroid Hormone Action
Independent of Effects on Circulating Levels of Thyroid Hormone in Male Rats.
Endocrinology 155: 4104-4112. http://dx.doi.org/10.1210/en.2Q14-1154.
Bansal. R: Zoeller. RT. (2008). Polychlorinated biphenyls (Aroclor 1254) do not uniformly produce
agonist actions on thyroid hormone responses in the developing rat brain. Endocrinology
149: 4001-4008. http://dx.doi.org/l 0.1210/en.2007-l 774.
Bao. S: Zhang. 1: Lin. Z: Su. K: Mo. I: Hong. L: Oian. S: Chen. L: Sun. F: Wen. C: Wu. 0: Hu. L: Lin. G:
Wang. X. (2017a). Serum metabolic changes in rats after intragastric administration of
dextromethorphan. Biomed Chromatogr 31: e3814. http://dx.doi.org/10.1002 /bmc.3814.
R-l
-------
Toxicological Review of PFBA and Related Salts
Bao. WW: Oian. ZM: Geiger. SD: Liu. E: Liu. Y: Wang. SO: Lawrence. WR: Yang. BY: Hu. LW: Zeng. XW:
Dong. GH. (2017b). Gender-specific associations between serum isomers of perfluoroalkyl
substances and blood pressure among Chinese: Isomers of C8 Health Project in China. Sci
Total Environ 607-608: 1304-1312. http://dx.doi.Org/10.1016/i.scitotenv.2017.07.124.
Barton. CA: Kaiser. MA: Russell. MH. (2007). Partitioning and removal of perfluorooctanoate during
rain events: the importance of physical-chemical properties. J Environ Monit 9: 839-846.
http://dx.doi.org/10.1039/b7Q3510a.
Bassett. THD: Clement-Lacroix. P: Samarut. 1: Chassande. 0: Williams. GR: O'Shea. PI:
Sriskantharaiah. S: Rabier. B: Bovde. A: Howell. PGT: Weiss. R. ovE: Roux. TP: Malaval. L. uc.
(2007). Thyroid hormone excess rather than thyrotropin deficiency induces osteoporosis in
hyperthyroidism. Mol Endocrinoll095-1107. http://dx.doi.org/10.1210/me.2007-0033.
Batt. AM: Ferrari. L. (1995). Manifestations of chemically-induced liver-damage. Clin Chem 41:
1882-1887.
Beggs. KM: Mcgreal. S. R: Mccarthv. A: Gunewardena. S: Lampe. IN: Lau. C: Apte. U. (2016). The role
of hepatocyte nuclear factor 4-alpha in perfluorooctanoic acid- and perfluorooctanesulfonic
acid-induced hepatocellular dysfunction. Toxicol Appl Pharmacol 304: 18-29.
http://dx.doi.Org/10.1016/i.taap.2016.05.001.
Bernal. 1. (2015). Thyroid hormones in brain development and function. In LJ De Groot; G Chrousos;
K Dungan; KR Feingold; A Grossman; JM Hershman; C Koch; M Korbonits; R McLachlan; M
New; J Purnell; R Rebar; F Singer; A Vinik (Eds.), NCBI Bookshelf. South Dartmouth, MA:
MDText.com, Inc.
Bizzarro. MI: Gross. I. (2004). Effects of hormones on fetal lung development [Review], Obstet
Gynecol Clin North Am 31: 949-961, xii. http://dx.doi.Org/10.1016/i.ogc.2004.08.001.
Biork. TA: Wallace. KB. (2009). Structure-Activity Relationships and Human Relevance for
Perfluoroalkyl Acid-Induced Transcriptional Activation of Peroxisome Proliferation in Liver
Cell Cultures. Toxicol Sci 111: 89-99. http://dx.doi.org/10.1093/toxsci/kfp093.
Blaine. AC: Rich. CD: Hundal. LS: Lau. C: Mills. MA: Harris. KM: Higgins. CP. (2013). Uptake of
perfluoroalkyl acids into edible crops via land applied biosolids: field and greenhouse
studies. Environ Sci Technol 47: 14062-14069. http://dx.doi.org/10.1021/es403Q94q.
Buhrke. T: Kriiger. E: Pevnv. S: RolSler. M: Bitter. K: Lampen. A. (2015). Perfluorooctanoic acid
(PFOA) affects distinct molecular signalling pathways in human primary hepatocytes.
Toxicology 333: 53-62. http://dx.doi.Org/10.1016/i.tox.2015.04.004.
Buist. SC: Cherrington. NT: Choudhuri. S: Hartley. DP: Klaassen. CD. (2002). Gender-specific and
developmental influences on the expression of rat organic anion transporters. J Pharmacol
Exp Ther 301: 145-151. http://dx.doi.Org/10.1124/ipet.301.l.145.
Burkemper. TL: Aweda. TA: Rosenberg. AT: Lunderberg. DM: Peaslee. GF: Lapi. SE. (2017).
Radiosynthesis and biological distribution of F-18-labeled perfluorinated alkyl substances.
Environ Sci Technol Lett 4: 211-215. http://dx.doi.org/10.1021/acs.estlett.7b00042.
Butenhoff. TL: Biork. TA: Chang. SC: Ehresman. DT: Parker. GA: Das. K: Lau. C: Lieder. PH: van
Otterdiik. FM: Wallace. KB. (2012a). Toxicological evaluation of ammonium
perfluorobutyrate in rats: twenty-eight-day and ninety-day oral gavage studies. Reprod
Toxicol 33: 513-530. http://dx.doi.Org/10.1016/i.reprotox.2011.08.004.
Butenhoff. TL: Chang. SC: Olsen. GW: Thomford. PI. (2012b). Chronic dietary toxicity and
carcinogenicity study with potassium perfluorooctanesulfonate in Sprague Dawley rats.
Toxicology 293: 1-15. http://dx.doi.Org/10.1016/i.tox.2012.01.003.
Butenhoff. TL: Chang. SC: Olsen. GW: Thomford. PI. (2012c). Chronic dietary toxicity and
carcinogenicity study with potassium perfluorooctanesulfonate in Sprague Dawley rats:
Supplemental data. Toxicology 293.
R-2
-------
Toxicological Review of PFBA and Related Salts
Butenhoff. TL: Kennedy. GL: Frame. S. R.: O'Connor. TC: York. RG. (2004). The reproductive
toxicology of ammonium perfluorooctanoate (APFO) in the rat. Toxicology 196: 95-116.
http://dx.doi.Org/10.1016/i.tox.2003.ll.005.
Calvo. R: Obregon. Ml: Ruiz de Ona. C: Ferreiro. B: del Rev. FE: Morreale de escobar. G. (1990).
Thyroid hormone economy in pregnant rats near term: a "physiological" animal model of
non thyroidal illness. Endocrinology 127: 10-16. http://dx.doi.org/0013-7227 i90/1271-
0010$02.00/0.
CDC (Centers for Disease Control and Prevention). (2019). Fourth National Report on Human
Exposure to Environmental Chemicals, Updated Tables, January 2019, Volume 1. Centers for
Disease Control and Prevention, National Health and Nutrition Examination Survey.
https://www.cdc.gov/exposurereport/pdf/FourthReport UpdatedTables Volumel Tan201
9-508.pdf.
Cerrutti. TA: Brandoni. A: Ouaglia. NB: Torres. AM. (2002). Sex differences in p-aminohippuric acid
transport in rat kidney: Role of membrane fluidity and expression of 0AT1. Mol Cell
Biochem 233: 175-179. http://dx.doi.Org/10.1023/A:1015563021602.
Chang. S: Butenhoff. TL: Parker. GA: Coder. PS: Zitzow. ID: Krisko. RM: Biork. TA: Wallace. KB: Seed.
TG. (2018). Reproductive and developmental toxicity of potassium
perfluorohexanesulfonate in CD-I mice. Reprod Toxicol 78: 150-168.
http://dx.doi.Org/10.1016/i.reprotox.2018.04.007.
Chang. S: Das. K: Ehresman. DT: Ellefson. ME: Gorman. GS: Hart. TA: Noker. PE: Tan. Y: Lieder. PH:
Lau. C: Olsen. GW: Butenhoff. TL. (2008). Comparative pharmacokinetics of
perfluorobutyrate in rats, mice, monkeys, and humans and relevance to human exposure via
drinking water. Toxicol Sci 104: 40-53. http: //dx.doi.org/10.lQ93 /toxsci/kfn057.
Chen. YM: Guo. LH. (2009). Fluorescence study on site-specific binding of perfluoroalkyl acids to
human serum albumin. Arch Toxicol 83: 255-261. http://dx.doi.org/10.1007/s00204-008-
0359-x.
Chengelis. CP: Kirkpatrick. IB: Radovskv. A: Shinohara. M. (2009). A 90-day repeated dose oral
(gavage) toxicity study of perfluorohexanoic acid (PFHxA) in rats (with functional
observational battery and motor activity determinations). Reprod Toxicol 27: 342-351.
http://dx.doi.Org/10.1016/i.reprotox.2009.01.006.
Crebelli. R: Caiola. S: Conti. L: Cordelli. E: De Luca. G: Dellatte. E: Eleuteri. P: Iacovella. N: Leopardi.
P: Marcon. F: Sanchez. M: Sestili. P: Siniscalchi. E: Villani. P. (2019). Can sustained exposure
to PFAS trigger a genotoxic response? A comprehensive genotoxicity assessment in mice
after subacute oral administration of PF0A and PFBA. Regul Toxicol Pharmacol 106: 169-
177. http://dx.doi.Org/10.1016/i.yrtph.2019.05.005.
Crofton. KM. (2004). Developmental disruption of thyroid hormone: Correlations with hearing
dysfunction in rats. Risk Anal 24: 1665-1671. http: //dx.doi.0rg/lO.llll /i.0272-
4332.2004.00557.x.
Crofton. KM: Craft. ES: Hedge. TM: Gennings. C: Simmons. IE: Carchman. RA: Carter. WH. Tr: Devito.
Ml. (2005). Thyroid-hormone-disrupting chemicals: Evidence for dose-dependent additivity
or synergism. Environ Health Perspect 113: 1549-1554.
http: / /dx. doi. o r g /10.12 89 /ehp. 819 5.
Crofton. KM: Zoeller. RT. (2005). Mode of action: neurotoxicity induced by thyroid hormone
disruption during development-hearing loss resulting from exposure to PHAHs [Review],
Crit Rev Toxicol 35: 757-769. http://dx.doi.Org/10.1080/10408440591007304.
Das. KP: Grey. BE: Zehr. RD: Wood. CR: Butenhoff. TL: Chang. SC: Ehresman. DT: Tan. YM: Lau. C.
(2008). Effects of perfluorobutyrate exposure during pregnancy in the mouse. Toxicol Sci
105: 173-181. http://dx.doi.org/10.1093/toxsci/kfn099.
R-3
-------
Toxicological Review of PFBA and Related Salts
Das. KP: Wood. CR: Lin. MT: Starkov. AA: Lau. C: Wallace. KB: Corton. TC: Abbott. BP. (2017).
Perfluoroalkyl acids-induced liver steatosis: Effects on genes controlling lipid homeostasis.
Toxicology 378: 37-52. http://dx.doi.Org/10.1016/i.tox.2016.12.007.
Davies. B: Morris. T. (1993). Physiological parameters in laboratory animals and humans [Review],
Pharm Res 10: 1093-1095. http://dx.doi.org/10.1023 /A:l 018943613122.
di Masi. A: De Marinis. E: Ascenzi. P: Marino. M. (2009). Nuclear receptors CAR and PXR: Molecular,
functional, and biomedical aspects [Review], Mol Aspects Med 30: 297-343.
http://dx.doi.Org/10.1016/i.mam.2009.04.002.
Diez. D: Morte. B: Bernal. 1. (2021). Single-cell transcriptome profiling of thyroid hormone effectors
in the human fetal neocortex: expression of SLC01C1, DI02, and THRB in specific cell types.
Thyroid 31: 1577-1588. http://dx.doi.org/10.1089/thv.2021.0057.
Espinosa. IS: Strvker. MP. (2012). Development and plasticity of the primary visual cortex [Review],
Neuron 75: 230-249. http://dx.doi.Org/10.1016/i.neuron.2012.06.009.
Feng. X: Cao. X: Zhao. S: Wang. X: Hua. X: Chen. L: Chen. L. (2017). Exposure of pregnant mice to
perfluorobutanesulfonate causes hypothyroxinemia and developmental abnormalities in
female offspring. Toxicol Sci 155: 409-419. http://dx.doi.org/10.1093/toxsci/kfw219.
Finken. MT: van Eiisden. M: Loomans. EM: Vriikotte. TG: Rotteveel. 1. (2013). Maternal
hypothyroxinemia in early pregnancy predicts reduced performance in reaction time tests
in 5- to 6-year-old offspring. J Clin Endocrinol Metab 98: 1417-1426.
http://dx.doi.org/10.1210/ic.2012-3389.
Foreman. IE: Chang. SC: Ehresman. DT: Butenhoff. TL: Anderson. CR: Palkar. PS: Kang. BH: Gonzalez.
FT: Peters. TM. (2009). Differential Hepatic Effects of Perfluorobutyrate Mediated by Mouse
and Human PPAR-alpha. Toxicol Sci 110: 204-211.
http://dx.doi.org/10.1093/toxsci/kfp077.
Forhead. AT: Fowden. AL. (2014). Thyroid hormones in fetal growth and prepartum maturation
[Review], J Endocrinol 221: R87-R103. http://dx.doi.org/10.1530/TQE-14-0025.
Fu. Y: Wang. T: Fu. 0: Wang. P: Lu. Y. (2014). Associations between serum concentrations of
perfluoroalkyl acids and serum lipid levels in a Chinese population. Ecotoxicol Environ Saf
106: 246-252. http://dx.doi.Org/10.1016/i.ecoenv.2014.04.039.
Fukuda. H: Ohshima. K: Mori. M: Kobavashi. I: Greer. MA. (1980). Sequential changes in the
pituitary-thyroid axis during pregnancy and lactation in the rat. Endocrinology 107: 1711-
1716. http://dx.doi.org/0013-7227/80/1076-1711$02.00/0.
G. MDE: R. C: 1. OM: F. EDR (1990). CONTRIBUTION OF MATERNAL THYROXINE TO FETAL
THYROXINE POOLS IN NORMAL RATS NEAR TERM. Endocrinology 126: 2765-2767.
Gebbink. WA: Berger. U: Cousins. IT. (2015). Estimating human exposure to PFOS isomers and PFCA
homologues: the relative importance of direct and indirect (precursor) exposure. Environ
Int74: 160-169. http://dx.doi.Org/10.1016/i.envint2014.10.013.
Giera. S: Bansal. R: Ortiz-Toro. TM: Taub. DG: Zoeller. RT. (2011). Individual polychlorinated
biphenyl (PCB) congeners produce tissue- and gene-specific effects on thyroid hormone
signaling during development. Endocrinology 152: 2909-2919.
http://dx.doi.org/10.1210/en.2010-1490.
Gilbert. ME. (2011). Impact of low-level thyroid hormone disruption induced by propylthiouracil on
brain development and function. Toxicol Sci 124: 432-445.
http://dx.doi.org/10.1093/toxsci/kfr244.
Gilbert. ME: Ramos. RL: Mccloskev. DP: Goodman. TH. (2014). Subcortical band heterotopia in rat
offspring following maternal hypothyroxinaemia: Structural and functional characteristics. J
Neuroendocrinol 26: 528-541. http: //dx.doi.org/10.Ill 1/ine. 12169.
Gilbert. ME: Sanchez-Huerta. K: Wood. C. (2016). Mild thyroid hormone insufficiency during
development compromises activity-dependent neuroplasticity in the hippocampus of adult
male rats : Supplemental data [Supplemental Data], Endocrinology 157: 774-787.
R-4
-------
Toxicological Review of PFBA and Related Salts
Gilbert. ME: Sui. L. (2008). Developmental exposure to perchlorate alters synaptic transmission in
hippocampus of the adult rat. Environ Health Perspect 116: 752-760.
http://dx.d0i.0rg/l 0.12 89 /ehp. 11089.
Gilbert. ME: Zoeller. RT. (2010). Thyroid hormones-impact on the developing brain: Possible
mechanisms of neurotoxicity. In GJ Harry; HA Tilson (Eds.), Neurotoxicology (3rd ed., pp.
79-111). New York, NY: Informa Healthcare.
Gomis. MI: Vestergren. R: Borg. D: Cousins. IT. (2018). Comparing the toxic potency in vivo of long-
chain perfluoroalkyl acids and fluorinated alternatives. Environ Int 113: 1-9.
http://dx.doi.0rg/lO.lOl6/i.envint.2Ol8.Ol.Oll.
Grandiean. P: Timmermann. CAG: Kruse. M: Nielsen. F: Vinholt. PI: Boding. L: Heilmann. C: M0lbak.
K. (2020). Severity of COVID-19 at elevated exposure to perfluorinated alkylates. PLoS ONE
15: e0244815. http://dx.doi.org/10.1371/iournal.pone.0244815.
Guvatt. G: Oxman. AD: Akl. EA: Kunz. R: Vist. G: Brozek. 1: Norris. S: Falck-Ytter. Y: Glasziou. P:
DeBeer. H: Taeschke. R: Rind. D: Meerpohl. I: Dahm. P: Schiinemann. HI. (2011). GRADE
guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin
Epidemiol 64: 383-394. http://dx.doi.Org/10.1016/i.iclinepi.2010.04.026.
Haddow. IE: Palomaki. GE: Allan. WC: Williams. TR: Knight. GT: Gagnon. 1: O'Heir. CE: Mitchell. ML:
Hermos. RT: Waisbren. SE: Faix. ID: Klein. RZ. (1999). Maternal thyroid deficiency during
pregnancy and subsequent neuropsychological development of the child. N Engl J Med 341:
549-555. http://dx.d0i.0rg/l 0.1056/NETM199908193410801.
Hall. AP: Elcombe. CR: Foster. TR: Harada. T: Kaufmann. W: Knippel. A: Kiittler. K: Malarkev. DE:
Maronpot. RR: Nishikawa. A: Nolte. T: Schulte. A: Strauss. V: York. Ml. (2012). Liver
hypertrophy: a review of adaptive (adverse and non-adverse) changes-conclusions from
the 3rd International ESTP Expert Workshop [Review], Toxicol Pathol 40: 971-994.
http://dx.doi.org/10.1177/0192623312448935.
Hassan. I: El-Masri. H: Kosian. PA: Ford. 1: Degitz. ST: Gilbert. ME. (2017). Neurodevelopment and
thyroid hormone synthesis inhibition in the rat: quantitative understanding within the
adverse outcome pathway framework. Toxicol Sci 160: 57-73.
http://dx.doi.org/10.lQ93/toxsci/kfxl63.
Henrichs. I: Bongers-Schokking. II: Schenk. II: Ghassabian. A: Schmidt. HG: Visser. TT: Hooiikaas. H:
de Muinck Keizer-Schrama. SM: Hofman. A: Taddoe. W: Visser. W: Steegers. EA: Verhulst.
FC: de Riike. YB: Tiemeier. H. (2010). Maternal thyroid function during early pregnancy and
cognitive functioning in early childhood: The generation R study. J Clin Endocrinol Metab
95: 4227-4234. http://dx.d0i.0rg/l 0.1210/ic.2010-0415.
Heo. II: Lee. TW: Kim. SK: Oh. IE. (2014). Foodstuff analyses show that seafood and water are major
perfluoroalkyl acids (PFAAs) sources to humans in Korea. J Hazard Mater 279: 402-409.
http://dx.doi.Org/10.1016/i.ihazmat.2014.07.004.
Hill. AB. (1965). The environment and disease: Association or causation? Proc R Soc Med 58: 295-
300.
Hillman. NH: Kallapur. SG: lobe. AH. (2012). Physiology of transition from intrauterine to
extrauterine life. Clin Perinatol 39: 769-783.
Hood. A: Hashmi. R: Klaassen. CD. (1999). Effects of microsomal enzyme inducers on thyroid-
follicular cell proliferation, hyperplasia, and hypertrophy. Toxicol Appl Pharmacol 160: 163-
170. http://dx.doi.org/10.1006/taap.1999.8752.
Hood. A: Klaassen. CD. (2000). Differential effects of microsomal enzyme inducers on in vitro
thyroxine (T4) and triiodothyronine (T3) glucuronidation. Toxicol Sci 55: 78-84.
http: / /dx. doi. o r g /10.109 3 /toxsci /55.1.78.
Howdeshell. KL. (2002). A model of the development of the brain as a construct of the thyroid
system [Review], Environ Health Perspect 110 Suppl 3: 337-348.
http://dx.d0i.0rg/l 0.12 89 /ehp. 02110s3 3 3 7.
R-5
-------
Toxicological Review of PFBA and Related Salts
Hulbert. AT. (2000). Thyroid hormones and their effects: A new perspective [Review], Biol Rev
Camb Philos Soc 75: 519-631.
Hurley. MP: Andersen. MPS: Wallington. TT: Ellis. DA: Martin. TW: Maburv. SA. (2004). Atmospheric
chemistry of perfluorinated carboxylic acids: Reaction with OH radicals and atmospheric
lifetimes. J Phys Chem A 108: 615-620. http://dx.doi.org/10.1021/ip036343b.
Ikeda. T: Aiba. K: Fukuda. K: Tanaka. M. (1985). The induction of peroxisome proliferation in rat
liver by perfluorinated fatty acids, metabolically inert derivatives of fatty acids. J Biochem
98: 475-482. http://dx.doi.org/10.1093/oxfordiournals.ibchem.al35302.
IPCS (International Programme on Chemical Safety). (2010). Characterization and application of
physiologically based pharmacokinetic models in risk assessment (Harmonization Project
Document No 9). Geneva, Switzerland: World Health Organization.
http://www.inchem.org/documents/harmproi/harmproi/harmproi9.pdf.
Taeschke. H: Lemasters. II. (2003). Apoptosis versus oncotic necrosis in hepatic
ischemia/reperfusion injury. Gastroenterology 125:1246-1257.
http://dx.doi.org/10.1016/S0016-5085r03101209-5.
Tansen. TA: Korevaar. TIM: Mulder. TA: White. T: Muetzel. RL: Peeters. RP: Tiemeier. H. (2019).
Maternal thyroid function during pregnancy and child brain morphology: a time window-
specific analysis of a prospective cohort. Lancet Diabetes Endocrinol 7: 629-637.
http://dx.doi.org/10.1016/S2213-8587C19130153-6.
Toshi-Barve. S: Kirpich. I: Cave. MC: Marsano. LS: McClain. CI. (2015). Alcoholic, nonalcoholic, and
toxicant-associated steatohepatitis: Mechanistic similarities and differences [Review],
CMGH 1: 356-367. http://dx.doi.org/10.1016/i.icmgh.2015.05.006.
Tulvez. 1: Alvarez-Pedrerol. M: Rebagliato. M: Murcia. M: Forns. 1: Garcia-Esteban. R: Lertxundi. N:
Espada. M: Tardon. A: Riano Galan. I: Sunver. I. (2013). Thyroxine levels during pregnancy
in healthy women and early child neurodevelopment Epidemiology 24: 150-157.
http://dx.doi.org/10.1097/EDE.0b013e318276ccd3.
lust. WW: Gorgas. K: Hartl. FU: Heinemann. P: Salzer. M: Schimassek. H. (1989). Biochemical effects
and zonal heterogeneity of peroxisome proliferation induced by perfluorocarboxylic acids
in rat liver. Hepatology 9: 570-581. http: //dx.doi.org/10.1002 /hep.1840090411.
Kaiser. TP: Lipscomb. TC: Wesselkamper. SC. (2012). Putative mechanisms of environmental
chemical-induced steatosis. Int J Toxicol 31: 551-563.
http://dx.doi.org/10.1177/1091581812466418.
Kass. GEN. (2006). Mitochondrial involvement in drug-induced hepatic injury. Chem Biol Interact
163: 145-159. http://dx.doi.org/10.1016/i.cbi.2006.06.007.
Kato. Y: Kuge. K: Kusuhara. H: Meier. PI: Sugivama. Y. (2002). Gender difference in the urinary
excretion of organic anions in rats. J Pharmacol Exp Ther 302: 483-489.
http://dx.doi.org/10.1124/ipetl02.033878.
Kim. DH: Kim. UT: Kim. HY: Choi. SD: Oh. IE. (2016). Perfluoroalkyl substances in serum from South
Korean infants with congenital hypothyroidism and healthy infants - Its relationship with
thyroid hormones. Environ Res 147: 399-404.
http://dx.doi.Org/10.1016/i.envres.2016.02.037.
Kim. SK: Kannan. K. (2007). Perfluorinated acids in air, rain, snow, surface runoff, and lakes:
relative importance of pathways to contamination of urban lakes. Environ Sci Technol 41:
8328-8334. http://dx.doi.org/10.1021/es072107t.
Klaunig. IE: Shinohara. M: Iwai. H: Chengelis. CP: Kirkpatrick. IB: Wang. Z: Bruner. RH. (2015).
Evaluation of the chronic toxicity and carcinogenicity of perfluorohexanoic acid (PFHxA) in
Sprague-Dawley rats. Toxicol Pathol 43: 209-220.
http://dx.doi.org/10.1177/0192623314530532.
R-6
-------
Toxicological Review of PFBA and Related Salts
Kleiner. DE: Makhlouf. HR. (2016). Histology of Nonalcoholic Fatty Liver Disease and Nonalcoholic
Steatohepatitis in Adults and Children. Clin Liver Dis 20: 293-+.
http://dx.doi.Org/10.1016/i.cld.2015.10.011.
Kleiner. DE: Weerasinghe. P: Buia. LM. (2012). Oncosis: An important non-apoptotic mode of cell
death. Exp Mol Pathol 93: 302-308. http://dx.doi.Org/10.1016/i.yexmp.2012.09.018.
Korevaar. TIM: Muetzel. R: Medici. M: Chaker. L: Taddoe. VWV: de Riike. YB: Steegers. EAP: Visser.
TT: White. T: Tiemeier. H: Peeters. RP. (2016). Association of maternal thyroid function
during early pregnancy with offspring IQ and brain morphology in childhood: A population-
based prospective cohort study. Lancet Diabetes Endocrinol 4: 35-43.
http://dx.doi.org/10.1016/S2213-8587ri5100327-7.
Kotthoff. M: Miiller. 1: liirling. H: Schlummer. M: Fiedler. D. (2015). Perfluoroalkyl and
polyfluoroalkyl substances in consumer products. Environ Sci Pollut Res Int 22: 14546-
14559. http://dx.doi.org/10.1007/sl 1356-015-4202-7.
Kuczmarski. RT: Ogden. CL: Grummer-Strawn. LM: Flegal. KM: Guo. SS: Wei. R: Mei. Z: Curtin. LR:
Roche. AF: Tohnson. CL. (2000). CDC growth charts: United States. Hyattsville, MD: U.S.
Department of Health and Human Services, Centers for Disease Control and Prevention,
National Center for Health Statistics, https: //www.cdc.gov/nchs/data/ad/ad314.pdf.
Lau. C. (2015). Perfluorinated compounds: An overview. In JC DeWitt (Ed.), Toxicological effects of
perfluoroalkyl and polyfluoroalkyl substances (pp. 1-21). New York: Springer.
http://dx.doi.org/10.1007/978-3-319-15518-0 1.
Lau. C: Butenhoff. TL: Rogers. TM. (2004). The developmental toxicity of perfluoroalkyl acids and
their derivatives [Review], Toxicol Appl Pharmacol 198: 231-241.
http://dx.doi.Org/10.1016/i.taap.2003.ll.031.
Lau. C: Thibodeaux. TR: Hanson. RG: Narotskv. MG: Rogers. TM: Lindstrom. AB: Strvnar. Ml. (2006).
Effects of perfluorooctanoic acid exposure during pregnancy in the mouse. Toxicol Sci 90:
510-518. http: / /dx. doi. o r g /10.109 3 /toxsci /kfi 10 5.
Lau. C: Thibodeaux. TR: Hanson. RG: Rogers. TM: Grey. BE: Stanton. ME: Butenhoff. TL: Stevenson. LA.
(2003). Exposure to perfluorooctane sulfonate during pregnancy in rat and mouse. II:
postnatal evaluation. Toxicol Sci 74: 382-392. http: / /dx.doi.org/10.1093/toxsci /kfgl 22.
Lavado-Autric. R: Auso. E: Garcia-Velasco. IV: del Carmen Arufe. M: Escobar del Rev. F: Berbel. P:
Morreale de Escobar. G. (2003). Early maternal hypothyroxinemia alters histogenesis and
cerebral cortex cytoarchitecture of the progeny. J Clin Invest 111: 1073-1082.
http://dx.doi.org/10.1172/TCI16262.
Levie. D: Korevaar. TIM: Bath. SC: Dalmau-Bueno. A: Murcia. M: Espada. M: Dineva. M: Ibarluzea. TM:
Sunver. I: Tiemeier. H: Rebagliato. M: Ravman. MP: Peeters. RP: Guxens. M. (2018). Thyroid
Function in Early Pregnancy, Child IQ, and Autistic Traits: A Meta-Analysis of Individual
Participant Data. J Clin Endocrinol Metab 103: 2967-2979.
http://dx.doi.org/10.1210/ic.2018-00224.
Li. T: Cai. D: Chu. C: Li. 00: Zhou. Y: Hu. LW: Yang. BY: Dong. GH: Zeng. XW: Chen. D. (2020).
Transplacental Transfer of Per- and Polyfluoroalkyl Substances (PFASs): Differences
between Preterm and Full-Term Deliveries and Associations with Placental Transporter
mRNA Expression. Environ SciTechnol 54: 5062-5070.
http://dx.doi.org/10.1021/acs.est.0c00829.
Li. M: Zeng. XW: Oian. ZM: Vaughn. MG: Sauve. S: Paul. G: Lin. S: Lu. L: Hu. LW: Yang. BY: Zhou. Y:
Oin. XD: Xu. SL: Bao. WW: Zhang. YZ: Yuan. P: Wang. I: Zhang. C: Tian. YP: Nian. M: Xiao. X:
Fu. C: Dong. GH. (2017a). Isomers of perfluorooctanesulfonate (PFOS) in cord serum and
birth outcomes in China: Guangzhou Birth Cohort Study. Environ Int 102: 1-8.
http://dx.doi.Org/10.1016/i.envint.2017.03.006.
R-7
-------
Toxicological Review of PFBA and Related Salts
Li. Y: Cheng. Y: Xie. Z: Zeng. F. (2017b). Perfluorinated alkyl substances in serum of the southern
Chinese general population and potential impact on thyroid hormones. Sci Rep 7: 43380.
http://dx.doi.org/10.1038/srep4338Q.
Liu. X: Guo. Z: Krebs. KA: Pope. RH: Roache. NF. (2014). Concentrations and trends of perfluorinated
chemicals in potential indoor sources from 2007 through 2011 in the US. Chemosphere 98:
51-57. http://dx.doi.Org/10.1016/i.chemosphere.2013.10.001.
Liuboievic. M: Balen. D: Breliak. D: Kusan. M: Anzai. N: Bahn. A: Burckhardt. G: Sabolic. I. (2007).
Renal expression of organic anion transporter 0AT2 in rats and mice is regulated by sex
hormones. Am J Physiol Renal Physiol 292: F361-F372.
http://dx.doi.Org/10.1152/aiprenal.00207.2006.
Liuboievic. M: Herak-Kramberger. CM: Hagos. Y: Bahn. A: Endou. H: Burckhardt. G: Sabolic. I.
(2004). Rat renal cortical 0AT1 and 0AT3 exhibit gender differences determined by both
androgen stimulation and estrogen inhibition. Am J Physiol Renal Physiol 287: F124-F138.
http://dx.doi.Org/10.1152/aiprenal.00029.2004.
Loccisano. AE: Campbell. TL. Tr: Andersen. ME: Clewell. HI. III. (2011). Evaluation and prediction of
pharmacokinetics of PFOA and PFOS in the monkey and human using a PBPK model. Regul
Toxicol Pharmacol 59: 157-175. http://dx.doi.Org/10.1016/i.yrtph.2010.12.004.
Lopez-Espindola. D: Garcia-Aldea. A: Gomez de la Riva. I: Rodriguez-Garcia. AM: Salvatore. D: Visser.
TT: Bernal. 1: Guadano-Ferraz. A. (2019). Thyroid hormone availability in the human fetal
brain: novel entry pathways and role of radial glia. Brain Struct Funct 224: 2103-2119.
http://dx.doi.Org/10.1007/s00429-019-01896-8.
Lorenzo. M: Farre. M: Blasco. C: Onghena. M: Pico. Y: Barcelo. D. (2016). Perfluoroalkyl substances
in Breast milk, infant formula and baby food from Valencian Community (Spain). Environ
Nanotechnol Monit Manag 6: 108-115. http://dx.doi.Org/10.1016/i.enmm.2016.09.001.
Luz. AL: Anderson. IK: Goodrum. P: Durda. 1. (2019). Perfluorohexanoic acid toxicity, part I:
Development of a chronic human health toxicity value for use in risk assessment Regul
Toxicol Pharmacol 103: 41-55. http://dx.doi.Org/10.1016/i.yrtph.2019.01.019.
Malhi. H: Gores. GT: Lemasters. 11. (2006). Apoptosis and necrosis in the liver: A tale of two deaths?
[Review], Hepatology 43: S31-S44. http://dx.doi.org/10.1002/hep.21062.
Matsuzaki. H: Katsumata. S: Masuvama. R: Uehara. M: Nakamura. K: Suzuki. K. (2002). Sex
differences in kidney mineral concentrations and urinary albumin excretion in rats given
high-phosphorus feed. Biosci Biotechnol Biochem 66: 1737-1739.
http://dx.doi.org/10.1271/bbb.66.1737.
MDH (Minnesota Department of Health). (2017). PFBA and Drinking Water.
https://www.health.state.mn.us/communities/environment/risk/docs/guidance/gw/pfbai
nfo.pdf.
MDH (Minnesota Department of Health). (2018). Health based guidance for water:
Perfluorooctanoic acid toxicological summary. Health Risk Assessment Unit, Environmental
Health Division, Minnesota Department of Health.
https://www.health.state.mn.us/communities/environment/risk/docs/guidance/gw/pfoa.
pdf.
MDH (Minnesota Department of Health). (2019). Health Based Guidance for Water: Toxicological
Summary for: Perfluorohexane sulfonate (PFHxS).
https://www.health.state.mn.us/communities/environment/risk/docs/guidance/gw/pfhx
s.pdf.
MDH (Minnesota Department of Health). (2020). Health Based Guidance for Water: Toxicological
Summary for: Perfluorooctane Sulfonate (PFOS).
https://www.health.state.mn.us/communities/environment/risk/docs/guidance/gw/pfos.
pdf.
R-8
-------
Toxicological Review of PFBA and Related Salts
Mellor. CL: Steinmetz. FP: Cronin. MT. (2016). The identification of nuclear receptors associated
with hepatic steatosis to develop and extend adverse outcome pathways [Review], Crit Rev
Toxicol 46: 138-152. http://dx.doi.org/10.3109/10408444.2015.1089471.
Miller. MP: Crofton. KM: Rice. DC: Zoeller. RT. (2009). Thyroid-disrupting chemicals: interpreting
upstream biomarkers of adverse outcomes [Review], Environ Health Perspect 117: 1033-
1041. http://dx.doi.org/10.1289/ehp.0800247.
Mirabella. G: Feig. D: Astzalos. E: Perlman. K: Rovet. IF. (2000). The effect of abnormal intrauterine
thyroid hormone economies on infant cognitive abilities. J Pediatr Endocrinol Metab 13:
191-194.
Mochizuki. K: Yagi. E: Sakaguchi. N: Mochizuki. H: Takabe. S: Kuranuki. S: Suzuki. T: Shimada. M:
Goda. T. (2007). The critical period for thyroid hormone responsiveness through thyroid
hormone receptor isoform alpha in the postnatal small intestine. Biochim Biophys Acta
1770: 609-616. http: / /dx.doi.org/10.1016/i.bbagen.2006.12.011.
Morgan. RL: Thayer. KA: Bero. L: Bruce. N: Falck-Ytter. Y: Ghersi. D: Guvatt. G: Hooiimans. C:
Langendam. M: Mandrioli. D: Mustafa. RA: Rehfuess. EA: Roonev. AA: Shea. B: Silbergeld. EK:
Sutton. P: Wolfe. MS: Woodruff. TT: Verbeek. TH: Hollowav. AC: Santesso. N: Schiinemann. HI.
(2016). GRADE: Assessing the quality of evidence in environmental and occupational health.
Environ Int92-93: 611-616. http://dx.doi.Org/10.1016/i.envint.2016.01.004.
Morreale de Escobar. G: Obregon. MT: Escobar del Rev. F. (2004). Role of thyroid hormone during
early brain development [Review], Eur J Endocrinol 151 Suppl 3: U25-U37.
http://dx.doi.Org/10.1530/eie.0.151u025.
Mullur. R: Liu. YY: Brent. GA. (2014). Thyroid hormone regulation of metabolism [Review], Physiol
Rev 94: 355-382. http://dx.doi.org/10.1152/physrev.00030.2013.
Nian. M: Li. 00: Bloom. M: Oian. ZM: Svberg. KM: Vaughn. MG: Wang. SO: Wei. 0: Zeeshan. M:
Gurram. N: Chu. C: Wang. 1: Tian. YP: Hu. LW: Liu. KK: Yang. BY: Liu. RO: Feng. D: Zeng. XW:
Dong. GH. (2019a). Liver function biomarkers disorder is associated with exposure to
perfluoroalkyl acids in adults: Isomers of C8 Health Project in China. Environ Res 172: 81-
88. http://dx.doi.Org/10.1016/i.envres.2019.02.013.
Nian. M: Li. 00: Bloom. M: Oian. ZM: Svberg. KM: Vaughn. MG: Wang. SO: Wei. 0: Zeeshan. M:
Gurram. N: Chu. C: Wang. 1: Tian. YP: Hu. LW: Liu. KK: Yang. BY: Liu. RO: Feng. D: Zeng. XW:
Dong. GH. (2019b). Liver function biomarkers disorder is associated with exposure to
perfluoroalkyl acids in adults: Isomers of C8 Health Project in China: Supplemental material
[Supplemental Data], Environ Res 172.
Nilsson. H: Karrman. A: Westberg. H: Rotander. A: van Bavel. B: Lindstrom. G. (2010). A time trend
study of significantly elevated perfluorocarboxylate levels in humans after using fluorinated
ski wax. Environ Sci Technol 44: 2150-2155. http://dx.doi.org/10.1021/es9034733.
NTDWOI (New Jersey Drinking Water Quality Institute). (2017). Public review draft: Health-based
maximum contaminant level support document: Perfluorooctane sulfonate (PFOS) (CAS #:
1763-23-1; Chemical Formula: C8HF1703S). NJDWQI Health Effects Subcommittee.
https://cswab.org/wp-content/uploads/2018/09/New-lersey-Scientific-Documentation-
Supporting-PF0S-13-ppt.pdf.
NLM (National Library of Medicine). (2013). HSDB: Perfluoro-n-nonanoic acid. Available online at
https://pubchem.ncbi.nlm.nih.gOv/source/hsdb/8040 (accessed
NLM (National Library of Medicine). (2016). TOXNET Toxicology Data Network, Hazardous
Substances Database (HSDB): Perfluorohexanoic acid. Available online at
https://toxnetnlm.nih.gov/cgi-bin/sis/search/a?dbs+hsdb:@term+@DOCNO+8299
(accessed
NLM (National Library of Medicine). (2017). HSDB: Perfluorohexanesulfonic acid. Available online
at https://pubchem.ncbi.nlm.nih.gov/source/hsdb/8274 (accessed
R-9
-------
Toxicological Review of PFBA and Related Salts
NTP (National Toxicology Program). (2018). 28-day evaluation of the toxicity (C20613) of
perfluorohexanoic acid (PFHxA) (307-24-4) on Harlan Sprague-Dawley rats exposed via
gavage. Available online at
https://tools.niehs.nih.gov/cebs3/views/?action=main.dataReview&.bin id=3879 (accessed
NTP (National Toxicology Program). (2019). NTP technical report on the toxicity studies of
perfluoroalkyl carboxylates (perfluorohexanoic acid, perfluorooctanoic acid,
perfluorononanoic acid, and perfluorodecanoic acid) administered by gavage to Sprague
Dawley (Hsd:Sprague Dawley SD) rats [NTP], (Toxicity Report97). Research Triangle Park,
NC. https://ntp.niehs.nih.gov/ntp/htdocs/st rpts/tox097 508.pdf.
O'Shaughnessv. KL: Kosian. PA: Ford. TL: Oshiro. WM: Degitz. ST: Gilbert. ME. (2018). Developmental
Thyroid Hormone Insufficiency Induces a Cortical Brain Malformation and Learning
Impairments: A Cross-Fostering Study. Toxicol Sci 163: 101-115.
http://dx.doi.org/10.1093/toxsci/kfy016.
Obregon. Ml: Calvo. RM: Del Rev. FE: de Escobar. GM. (2007). Ontogenesis of thyroid function and
interactions with maternal function [Review], Endocr Dev 10: 86-98.
http://dx.doi.org/10.1159/0000106821.
OECD (Organisation for Economic Co-operation and Development). (2010). TestNo. 417:
Toxicokinetics. In OECD guidelines for the testing of chemicals, section 4. Paris, France:
OECD Publishing, http://dx.doi.org/10.1787/9789264070882-en.
OECD (Organisation for Economic Co-operation and Development). (2018). Toward a new
comprehensive global database of per- and polyfluoroalkyl substances (PFASs): Summary
report on updating the OECD 2007 list of per- and polyfluoroalkyl substances (PFASs).
(ENV/JM/MONO(2018)7). Paris, France, http://www.oecd.org/chemicalsafety/portal-
perfluorinated-chemicals/.
Paabv. P. (1960). Changes in serum proteins during pregnancy. J Obstet Gynaecol Br Emp 67: 43-55.
http://dx.doi.Org/10.llll/i.1471-0528.1960.tb06949.x.
Patel. I: Landers. K: Li. H: Mortimer. RH: Richard. K. (2011). Thyroid hormones and fetal
neurological development [Review], J Endocrinol 209: 1-8. http://dx.doi.org/10.1530/TQE-
10-0444.
Perez. F: Nadal. M: Navarro-Ortega. A: Fabrega. F: Domingo. TL: Barcelo. D: Farre. M. (2013).
Accumulation of perfluoroalkyl substances in human tissues. Environ Int 59: 354-362.
http://dx.doi.Org/10.1016/i.envint.2013.06.004.
Permadi. H: Lundgren. B: Andersson. K: Depierre. TW. (1992). Effects of perfluoro fatty acids on
xenobiotic-metabolizing enzymes, enzymes which detoxify reactive forms of oxygen and
lipid peroxidation in mouse liver. Biochem Pharmacol 44: 1183-1191.
http://dx.doi.Org/10.1016/0006-2952f92190383-T.
Permadi. H: Lundgren. B: Andersson. K: Sundberg. C: Depierre. TW. (1993). Effects of perfluoro fatty
acids on peroxisome proliferation and mitochondrial size in mouse liver: dose and time
factors and effect of chain length. Xenobiotica 23: 761-770.
http://dx.doi.Org/10.3109/00498259309166782.
Post. GB: Louis. IB: Lippincott. RL: Procopio. NA. (2013). Occurrence of perfluorinated compounds
in raw water from New Jersey public drinking water systems. Environ Sci Technol 47:
13266-13275. http://dx.doi.org/10.1021 /es402884x.
Prevedouros. K: Cousins. IT: Buck. RC: Korzeniowski. SH. (2006). Sources, fate and transport of
perfluorocarboxylates [Review], Environ Sci Technol 40: 32-44.
http://dx.doi.org/10.1021/es0512475.
Prevot. V. (2014). Chapter 30 - puberty in mice and rats. In Knobil and Neill's Physiology of
Reproduction (Fourth Edition) (4 ed.). Lille, France: Laboratory of Development and
Plasticity of the Neuroendocrine Brain, Jean-Pierre Aubert Research Center, School of
Medicine, University of Lille. http: //dx.doi.org/10.1016/B978-0-12-397175-3.00030-2.
R-10
-------
Toxicological Review of PFBA and Related Salts
Puig-Pomingo. M: Vila. L. (2013). The implications of iodine and its supplementation during
pregnancy in fetal brain development [Review], Curr Clin Pharmacol 8: 97-109.
Rao. INK: Scott. AT. (1992). A simple method for the analysis of clustered binary data. Biometrics 48:
577-585. http://dx.doi.org/10.2307/2532311.
Robinson. PI: Rapoport. SI. (1986). Kinetics of protein binding determine rates of uptake of drugs
by brain. Am J Physiol Regul Integr Comp Physiol 251: R1212-R1220.
http://dx.doi.Org/10.1152/aipregu.1986.251.6.R1212.
Roman. GC: Ghassabian. A: Bongers-Schokking. 11: Taddoe. VW: Hofman. A: de Riike. YB: Verhulst.
FC: Tiemeier. H. (2013). Association of gestational maternal hypothyroxinemia and
increased autism risk : Supplementary materials. Ann Neurol 74: 733-742.
Rosen. MB: Das. KP: Roonev. 1: Abbott. B: Lau. C: Corton. TC. (2017). PPARa-independent
transcriptional targets of perfluoroalkyl acids revealed by transcript profiling. Toxicology
387: 95-107. http: //dx.doi.org/10.1016/i.tox.2017.05.013.
Rosen. MB: Das. KP: Wood. CR: Wolf. CI: Abbott. BP: Lau. C. (2013). Evaluation of perfluoroalkyl acid
activity using primary mouse and human hepatocytes. Toxicology 308: 129-137.
http://dx.doi.Org/10.1016/i.tox.2013.03.011.
Rosenmai. AK: Ahrens. L: le Godec. T: Lundqvist. 1: Oskarsson. A. (2018). Relationship between
peroxisome proliferator-activated receptor alpha activity and cellular concentration of 14
perfluoroalkyl substances in HepG2 cells. J Appl Toxicol 38: 219-226.
http: / /dx. doi. o r g /10.10 0 2 /i at. 3 515.
Ruan. T: Sulecki. LM: Wolstenholme. BW: Tiang. G: Wang. N: Buck. RC. (2014). 6:2 Fluorotelomer
iodide in vitro metabolism by rat liver microsomes: comparison with [1,2-(14)C] 6:2
fluorotelomer alcohol. Chemosphere 112: 34-41.
http://dx.doi.Org/10.1016/i.chemosphere.2014.02.068.
Russell. 11: Heaton. TC: Underwood. T: Boughtflower. R: Mccallev. DV. (2015a). Performance of
charged aerosol detection with hydrophilic interaction chromatography. J Chromatogr A
1405: 72-84. http://dx.doi.Org/10.1016/i.chroma.2015.05.050.
Russell. MH: Himmelstein. MW: Buck. RC. (2015b). Inhalation and oral toxicokinetics of 6:2 FTOH
and its metabolites in mammals. Chemosphere 120:328-335.
http://dx.doi.Org/10.1016/i.chemosphere.2014.07.092.
Schiinemann. H: Hill. S: Guvatt. G: Akl. EA: Ahmed. F. (2011). The GRADE approach and Bradford
Hill's criteria for causation. J Epidemiol Community Health 65: 392-395.
http://dx.doi.org/10.1136/iech.2010.119933.
Sferruzzi-Perri. AN: Vaughan. OR: Forhead. AT: Fowden. AL. (2013). Hormonal and nutritional
drivers of intrauterine growth [Review], Curr Opin Clin Nutr Metab Care 16: 298-309.
http://dx.doi.Org/10.1097/MCO.0b013e32835e3643.
Sharlin. PS: Gilbert. ME: Taylor. MA: Ferguson. DC: Zoeller. RT. (2010). The nature of the
compensatory response to low thyroid hormone in the developing brain. J Neuroendocrinol
22: 153-165. http://dx.doi.org/10.1111 /i.l 365-2826.2009.01947.x.
Song. X: Tang. S: Zhu. H: Chen. Z: Zang. Z: Zhang. Y: Niu. X: Wang. X: Yin. H: Zeng. F: He. C. (2018).
Biomonitoring PFAAs in blood and semen samples: Investigation of a potential link between
PFAAs exposure and semen mobility in China. Environ Int 113: 50-54.
http://dx.doi.0rg/lO.lOl6/i.envint.2Ol8.Ol.OlO.
Stagnaro-Green. A: Rovet. 1. (2016). Pregnancy: Maternal thyroid function in pregnancy - A tale of
two tails [Comment], Nat Rev Endocrinol 12: 10-11.
http://dx.doi.org/10.1038/nrendo.2015.212.
Stahl. LL: Snyder. BP: Olsen. AR: Kincaid. TM: Wathen. IB: Mccarty. HB. (2014). Perfluorinated
compounds in fish from U.S. urban rivers and the Great Lakes. Sci Total Environ 499: 185-
195. http://dx.doi.Org/10.1016/i.scitotenv.2014.07.126.
R-ll
-------
Toxicological Review of PFBA and Related Salts
Stenzel. D: Huttner. WB. (2013). Role of maternal thyroid hormones in the developing neocortex
and during human evolution. Frontiers in Neuroanatomy 7: 9.
http://dx.doi.org/10.3389/fnana.2013.00Q19.
Sun. M: Arevalo. E: Strvnar. M: Lindstrom. A: Richardson. M: Kearns. B: Pickett. A: Smith. C: Knappe.
DRU. (2016). Legacy and emerging perfluoroalkyl substances are important drinking water
contaminants in the Cape Fear River Watershed of North Carolina. Environ Sci Technol Lett
3: 415-419. http://dx.doi.org/10.1021/acs.estlett6b00398.
Sundstrom. M: Bogdanska. 1: Pham. HV: Athanasios. V: Nobel. S: Mcalees. A: Eriksson. 1: Depierre.
TW: Bergman. A. (2012). Radiosynthesis of perfluorooctanesulfonate (PFOS) and
perfluorobutanesulfonate (PFBS), including solubility, partition and adhesion studies.
Chemosphere 87: 865-871. http://dx.doi.Org/10.1016/i.chemosphere.2012.01.027.
U.S. EPA (U.S. Environmental Protection Agency). (1991). Guidelines for developmental toxicity risk
assessment Fed Reg 56: 63798-63826.
U.S. EPA (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation
reference concentrations and application of inhalation dosimetry [EPA Report],
(EPA600890066F). Research Triangle Park, NC.
https://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=71993&CFID=51174829&CFTOKE
N=25006317.
U.S. EPA (U.S. Environmental Protection Agency). (1996). Guidelines for reproductive toxicity risk
assessment Fed Reg 61: 56274-56322.
U.S. EPA (U.S. Environmental Protection Agency). (1998a). Assessment of thyroid follicular cell
tumors [EPA Report] (pp. 1-51). (EPA/630/R-97/002). Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment Forum.
https://www.epa.gov/sites/production/files/2014-ll/documents/thyroid.pdf.
U.S. EPA (U.S. Environmental Protection Agency). (1998b). Guidelines for neurotoxicity risk
assessment [EPA Report] (pp. 1-89). (ISSN 0097-6326
EISSN 2167-2520
EPA/630/R-95/001F). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment
Forum, http://www.epa.gov/risk/guidelines-neurotoxicity-risk-assessment.
U.S. EPA (U.S. Environmental Protection Agency). (2002). A review of the reference dose and
reference concentration processes. (EPA630P02002F). Washington, DC.
https://www.epa.gov/sites/production/files/2014-12/documents/rfd-final.pdf.
U.S. EPA (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk assessment
[EPA Report], (EPA630P03001F). Washington, DC.
https://www.epa.gov/sites/production/files/2Q13-
09/documents/cancer guidelines final 3-25-05.pdf.
U.S. EPA (U.S. Environmental Protection Agency). (2011). Recommended use of body weight 3/4 as
the default method in derivation of the oral reference dose. (EPA100R110001). Washington,
DC. https://www.epa.gov/sites/production/files/2013-09/documents/recommended-use-
of-bw34.pdf.
U.S. EPA (U.S. Environmental Protection Agency). (2012). Benchmark dose technical guidance [EPA
Report], (EPA100R12001). Washington, DC: U.S. Environmental Protection Agency, Risk
Assessment Forum, https: //www.epa.gov/risk/benchmark-dose-technical-guidance.
U.S. EPA (U.S. Environmental Protection Agency). (2013). U.S. Environmental Protection Agency
endocrine disruptor screening program SAP review of EDSP tier 1 screening: Assay and
battery performance.
U.S. EPA (U.S. Environmental Protection Agency). (2016a). Health effects support document for
perfluorooctane sulfonate (PFOS) [EPA Report], (EPA 822-R-16-002). Washington, DC: U.S.
Environmental Protection Agency, Office of Water, Health and Ecological Criteria Division.
R-12
-------
Toxicological Review of PFBA and Related Salts
https://www.epa.gov/sites/production/files/2016-
05/documents/pfos hesd final 508.pdf.
U.S. EPA (U.S. Environmental Protection Agency). (2016b). Health effects support document for
perfluorooctanoic acid (PFOA) [EPA Report], (EPA 822-R-16-003). Washington, DC: U.S.
Environmental Protection Agency, Office of Water, Health and Ecological Criteria Division.
https://www.epa.gOv/sites/production/files/2016-05/documents/pfoa hesd final-
plain.pdf.
U.S. EPA (U.S. Environmental Protection Agency). (2018a). Chemistry Dashboard. Washington, DC.
Retrieved from https: //comptox.epa.gov/dashboard
U.S. EPA (U.S. Environmental Protection Agency). (2018b). Human health toxicity values for
perfluorobutane sulfonic acid (CASRN 375-73-5) and related compound potassium
perfluorobutane sulfonate (CASRN 29420-49-3) (public comment draft). (EPA-823-R-18-
307). Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). (2019a). ChemView [Database], Retrieved from
https://chemview.epa.gov/chemview
U.S. EPA (U.S. Environmental Protection Agency). (2019b). EPA Actions to Address PFAS. Available
online at https://www.epa.gov/pfas/epa-actions-address-pfas (accessed
U.S. EPA (U.S. Environmental Protection Agency). (2020). Umbrella quality assurance project plan
(QAPP) for dosimetry and mechanism-based models. (EPA QAPP ID Number: L-CPAD-
0032188-QP-1-2). Research Triangle Park, NC.
U.S. EPA (U.S. Environmental Protection Agency). (2021a). Human health toxicity values for
hexafluoropropylene oxide (HFPO) dimer acid and its ammonium salt (CASRN 13252-13-6
and CASRN 62037-80-3). Also known as "GenX chemicals." Final report [EPA Report], (EPA-
822R-21-010). Washington, DC: U.S. Environmental Protection Agency, Office of Water.
U.S. EPA (U.S. Environmental Protection Agency). (2021b). Human health toxicity values for
perfluorobutane sulfonic acid (CASRN 375-73-5) and related compound potassium
perfluorobutane sulfonate (CASRN 29420-49-3) [EPA Report], (EPA/600/R-20/345F).
Washington, DC: U.S. Environmental Protection Agency, Office of Research and
Development. https://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=350888.
Van Camp. I: Ritskes-Hoitinga. 1: Lemmens. AG: Bevnen. AC. (1990). Diet-induced nephrocalcinosis
and urinary excretion of albumin in female rats. Lab Anim 24: 137-141.
http://dx.doi.Org/10.1258/002367790780890086.
Van Cruchten. S: Van Den Broeck. W. (2002). Morphological and biochemical aspects of apoptosis,
oncosis and necrosis [Review], AnatHistol Embryol 31: 214-223.
http://dx.doi.Org/10.1046/i.1439-0264.2002.00398.x.
van Otterdiik. FM. (2007a). Repeated dose 28-day oral toxicity study with MTDID-8391 by daily
gavage in the rat, followed by a 21-day recovery period. (Study Number 06-226).
Maple wood, MN: 3M.
van Otterdiik. FM. (2007b). Repeated dose 90-day oral toxicity study with MTDID 8391 by daily
gavage in the rat followed by a 3-week recovery period. (Study Number 06-398).
Maple wood, MN: 3M.
Venkatesan. AK: Halden. RU. (2014). Loss and in situ production of perfluoroalkyl chemicals in
outdoor biosolids-soil mesocosms. Environ Res 132: 321-327.
http://dx.doi.Org/10.1016/i.envres.2014.04.024.
Vulsma. T: Gons. MH: de Viilder. 11. (1989). Maternal-fetal transfer of thyroxine in congenital
hypothyroidism due to a total organification defect or thyroid agenesis. N Engl J Med 321:
13-16. http://dx.doi.Org/10.1056/NETM198907063210103.
Wahlang. B: Beier. II: Clair. HB: Bellis-Tones. HI: Falkner. K: Mcclain. CI: Cave. MC. (2013). Toxicant-
associated steatohepatitis [Review], Toxicol Pathol 41: 343-360.
http://dx.doi.org/10.1177/0192623312468517.
R-13
-------
Toxicological Review of PFBA and Related Salts
Wang. T: Zeng. XW: Bloom. MS: Oian. Z: Hinvard. LT: Belue. R: Lin. S: Wang. SO: Tian. YP: Yang. M:
Chu. C: Gurram. N: Hu. LW: Liu. KK: Yang. BY: Feng. D: Liu. RO: Dong. GH. (2019). Renal
function and isomers of perfluorooctanoate (PFOA) and perfluorooctanesulfonate (PFOS):
Isomers of C8 Health Project in China. Chemosphere 218: 1042-1049.
http://dx.doi.Org/10.1016/i.chemosphere.2018.ll.191.
Wang. Z: Cousins. IT: Scheringer. M: Buck. RC: Hungerbiihler. K. (2014). Global emission inventories
for C4-C14 perfluoroalkyl carboxylic acid (PFCA) homologues from 1951 to 2030, Parti:
production and emissions from quantifiable sources [Review], Environ Int 70: 62-75.
http://dx.doi.Org/10.1016/i.envint.2014.04.013.
Watkins. DT: Tosson. 1: Elston. B: Bartell. SM: Shin. HM: Vieira. VM: Savitz. DA: Fletcher. T: Wellenius.
GA. (2013). Exposure to perfluoroalkyl acids and markers of kidney function among
children and adolescents living near a chemical plant. Environ Health Perspect 121: 625-
630. http://dx.doi.org/10.1289/ehp.1205838.
Weatherlv. LM: Shane. HL: Lukomska. E: Baur. R: Anderson. SE. (2021). Systemic toxicity induced
by topical application of heptafluorobutyric acid (PFBA) in a murine model. Food Chem
Toxicol 156: 112528. http: //dx.doi.org/10.1016/i.fct.2021.112528.
Weaver. YM: Ehresman. DT: Butenhoff. TL: Hagenbuch. B. (2010). Roles of rat renal organic anion
transporters in transporting perfluorinated carboxylates with different chain lengths.
Toxicol Sci 113: 305-314. http: / / dx. do i. o r g /10.10 9 3/toxsci/kfp275.
Wexler. TA: Sharretts. 1. (2007). Thyroid and bone [Review], Endocrinol Metab Clin North Am 36:
673-705, vi. http://dx.doi.Org/10.1016/i.ecl.2007.04.005.
Wiesel. TN. (1982). The postnatal development of the visual cortex and the influence of
environment [Review], Biosci Rep 2: 351-377. http://dx.doi.org/10.1007/bf01119299.
Williams. GM: Iatropoulos. Ml. (2002). Alteration of liver cell function and proliferation:
Differentiation between adaptation and toxicity [Review], Toxicol Pathol 30: 41-53.
http://dx.doi.org/10.1080/01926230252824699.
Williams. GR. (2008). Neurodevelopmental and neurophysiological actions of thyroid hormone
[Review], J Neuroendocrinol 20: 784-794. http: / /dx.doi.org/10.1111 /i. 1365-
2826.2008.01733.x.
Wolf. CI: Schmid. IE: Lau. C: Abbott. BP. (2012). Activation of mouse and human peroxisome
proliferator-activated receptor-alpha (PPARa) by perfluoroalkyl acids (PFAAs): further
investigation of C4-C12 compounds. Reprod Toxicol 33: 546-551.
http://dx.doi.Org/10.1016/i.reprotox.2011.09.009.
Wolf. CI: Takacs. ML: Schmid. IE: Lau. C: Abbott. BP. (2008). Activation of mouse and human
peroxisome proliferator-activated receptor alpha by perfluoroalkyl acids of different
functional groups and chain lengths. Toxicol Sci 106: 162-171.
http://dx.doi.org/10.1093/toxsci/kfnl66.
Yang. CH: Glover. KP: Han. X. (2009). Organic anion transporting polypeptide (Oatp) lal-mediated
perfluorooctanoate transport and evidence for a renal reabsorption mechanism of Oatplal
in renal elimination of perfluorocarboxylates in rats. Toxicol Lett 190: 163-171.
http://dx.doi.Org/10.1016/i.toxlet2009.07.011.
Zeng. XW: Li. 00: Chu. C: Ye. WL: Yu. S: Ma. H: Zeng. XY: Zhou. Y: Yu. HY: Hu. LW: Yang. BY: Pong.
GH. (2020). Alternatives of perfluoroalkyl acids and hepatitis B virus surface antibody in
adults: Isomers of C8 Health Project in China. Environ Pollut 259: 113857.
http://dx.doi.Org/10.1016/i.envpol.2019.113857.
Zhang. YM: Pong. XY: Fan. LT: Zhang. ZL: Wang. 0: Tiang. N: Yang. XS. (2017). Poly- and
perfluorinated compounds activate human pregnane X receptor. Toxicology 380: 23-29.
http://dx.doi.Org/l 0.1016/i .tox.2 017.01.012.
R-14
-------
Toxicological Review of PFBA and Related Salts
Zhao. H: Ou. B: Guan. Y: liang. T: Chen. X. (2016). Influence of salinity and temperature on uptake of
perfluorinated carboxylic acids (PFCAs) by hydroponically grown wheat (Triticum aestivum
L.). SpringerPlus 5: 541. http://dx.doi.org/10.1186/s40064-016-2Q16-9.
Zoeller. RT: Crofton. KM. (2000). Thyroid hormone action in fetal brain development and potential
for disruption by environmental chemicals [Review], Neurotoxicology 21: 935-945.
Zoeller. RT: Dowling. ALS: Vas. AA. (2000). Developmental exposure to polychlorinated biphenyls
exerts thyroid hormone-like effects on the expression of RC3/neurogranin and myelin basic
protein messenger ribonucleic acids in the developing rat brain. Endocrinology 141: 181-
189. http://dx.doi.Org/10.1210/endo.141.l.7273.
Zoeller. RT: Rovet. 1. (2004). Timing of thyroid hormone action in the developing brain: Clinical
observations and experimental findings [Review], J Neuroendocrinol 16: 809-818.
http://dx.doi.org/10.1111 /i.l 365-2826.2004.01243.x.
Zoeller. RT: Tan. SW: Tvl. RW. (2007). General background on the hypothalamic-pituitary-thyroid
(HPT) axis [Review], Crit Rev Toxicol 37: 11-53.
http://dx.doi.Org/10.1080/10408440601123446.
R-15
------- |